Methamphetamine-Induced Dopamine-Independent Alterations in Striatal Gene Expression in the 6-Hydroxydopamine Hemiparkinsonian Rats

Unilateral injections of 6-hydroxydopamine into the medial forebrain bundle are used extensively as a model of Parkinson's disease. The present experiments sought to identify genes that were affected in the dopamine (DA)–denervated striatum after 6-hydroxydopamine-induced destruction of the nigrostriatal dopaminergic pathway in the rat. We also examined whether a single injection of methamphetamine (METH) (2.5 mg/kg) known to cause changes in gene expression in the normally DA-innervated striatum could still influence striatal gene expression in the absence of DA. Unilateral injections of 6-hydroxydopamine into the medial forebrain bundle resulted in METH-induced rotational behaviors ipsilateral to the lesioned side and total striatal DA depletion on the lesioned side. This injection also caused decrease in striatal serotonin (5-HT) and 5-hydroxyindoleacetic acid (5-HIAA) levels. DA depletion was associated with increases in 5-HIAA/5-HT ratios that were potentiated by the METH injection. Microarray analyses revealed changes (± 1.7-fold, p<0.025) in the expression of 67 genes on the lesioned side in comparison to the intact side of the saline-treated hemiparkinsonian animals. These include follistatin, neuromedin U, and tachykinin 2 which were up-regulated. METH administration caused increases in the expression of c-fos, Egr1, and Nor-1 on the intact side. On the DA-depleted side, METH administration also increased the expression of 61 genes including Pdgf-d and Cox-2. There were METH-induced changes in 16 genes that were common in the DA-innervated and DA-depleted sides. These include c-fos and Nor-1 which show greater changes on the normal DA side. Thus, the present study documents, for the first time, that METH mediated DA-independent changes in the levels of transcripts of several genes in the DA-denervated striatum. Our results also implicate 5-HT as a potential player in these METH-induced alterations in gene expression because the METH injection also caused significant increases in 5-HIAA/5-HT ratios on the DA-depleted side.

Methamphetamine (METH) is an indirect agonist that induces release of DA and serotonin (5-HT) in the brain [21,22,[24][25][26].Repeated injections of large METH doses also cause delayed increases in glutamate release in the striatum [23,27,28].In addition, METH administration influences striatal gene expression in animals with normal dopaminergic innervation [29][30][31][32][33].The METH-induced transcriptional changes depend on stimulation of DA and glutamate receptors [29,31,33].However, the extent to which METH might also induce changes in gene expression in the absence of DA innervation has not been clarified.Studies that have examined the effects of indirect agonists on striatal gene expression in hemiparkinsonian rodents have measured the expression of only a few genes or proteins.For example, repeated injections of amphetamine (5 mg/kg) to rats with unilateral nigral 6-OHDA lesions caused increased striatal dynorphin-like immunoreactivity on the intact but not on the lesioned side [34].Neither did amphetamine administration induce any changes in metenkephalin expression on the lesioned side [34].In contrast, Chritin et al. [35] reported that amphetamine caused up-regulated Fos protein expression on both sides of hemiparkinsonian rats, with the increases being of smaller magnitude on the DAdenervated side.
Given the importance of DA in the mediation of striatal synaptic plasticity and striatum-dependent behaviors, we thought it is likely that DA depletion might be associated with changes in the expression of a larger number of genes than those described so far in intrinsic striatal cells [13][14][15].The possibility also existed that indirect DA agonists including METH, which causes release of other neurotransmitters such as 5-HT [22], might affect the expression of some genes in a DA-independent fashion.Therefore, the present study was undertaken to analyze global gene expression in the DA-denervated striatum and to quantify striatal METH-induced transcriptional responses after 6-OHDA-induced lesions of the nigrostriatal dopaminergic pathway.Our results show that there are other genes that are affected in the DAdepleted striatum in addition to those previously reported [13][14][15].We also found that METH administration does indeed cause changes in the expression of several genes in the DA-depleted striatum.Our observations further suggest that METH-induced increased 5-HT turnover might, in part, be responsible for the later changes.

Animals
Male Sprague-Dawley rats (Charles Rivers Laboratories, Raleigh, NC), weighing 270-300 g at the beginning of the experiments were used in the present study.Animals were housed in a humidity-and temperature-controlled room and were given free access to food and water.All animal procedures were performed according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Animal Care and Use Committee of the National Institute on Drug Abuse, Intramural Research Program.The research was conducted under Animal Study Protocol #09-CNRB-25.

6-OHDA lesioning
Unilateral lesions were performed under anesthesia with chloral hydrate (400 mg/kg, i.p.).After immobilization on a stereotaxic frame (model 940; David Kopf Instruments), a hole was drilled in the skull for injections of 6-OHDA in the medial forebrain bundle (MFB).6-OHDA (2 mg/ml 65 ml in 0.9% NaCl containing 0.2 mg/ml ascorbic acid) was unilaterally injected into the left MFB (24.4 mm AP, 1.2 mm ML relative to bregma and 8.4 mm below skull) over 4 minutes.At the end of each injection, the micropipette was left in place for an additional 5 min and then withdrawn slowly to prevent reflux of the solution.

Rotation
Rotational behavior was evaluated using a multichannel rotometer system (RotoMax, AccuScan Instruments, Inc).Contraversive or ipsiversive rotational behaviors were induced by subcutaneous injection of apomorphine (APO) (0.05 mg/kg) on day 24 or (6) METH HCl (2.5 mg/kg, s.c) on day 31, respectively, after the 6-OHDA injection.Each animal was placed in a cylindrical test chamber for 90 min.The highest consecutive clockwise and counter-clockwise rotations over 60 min were used for analysis.

METH treatment and tissue collection
One week after measuring METH-induced rotation, the animals were divided into two groups based on rotational behavior.The two matched groups of animals were injected with either saline or METH (2.5 mg/kg, i.p.) and then euthanized 2 hours after the injection.Additional control animals that did not get 6-OHDA injections were also used.Their brains were quickly removed, tissues were dissected on ice, snap frozen on dry ice, and stored at 280uC until used in HPLC, microarray, and quantitative PCR (qPCR) experiments.The experimental groups were: salinetreated controls (SC), METH-treated controls (MC), non-lesioned side of saline-treated 6-OHDA-injected rats (SNL), lesioned side of saline-treated 6-OHDA-injected rats (SL), non-lesioned side of METH-treated 6-OHDA-injected animals (MNL), and lesioned side of METH-treated 6-OHDA-injected animals (ML).

RNA extraction, microarray hybridization, and data analysis
Total RNA was isolated using RNeasy Midi kit (Qiagen, Valencia, CA).RNA integrity was assessed using an Agilent 2100 Bioanalyzer (Agilent, Palo Alto, CA) and showed no degradation.Microarray hybridization was carried out using RatRef-12 Expression BeadChips arrays (22, 523 probes) (Illumina Inc., San Diego, CA).In brief, a 600 ng aliquot of total RNA from each striatal sample was amplified using Illumina RNA Amplification kit (Ambion, Austin, TX).Single-stranded RNA (cRNA) was generated and labeled by incorporating biotin-16-UTP (Roche Diagnostics, Indianapolis, IN).750 ng of each cRNA sample were hybridized to Illumina arrays at 55uC overnight according to the Whole-Genome Gene Expression Protocol for BeadStation (Illumina Inc.).Hybridized biotinylated cRNA was detected with Cyanine3-streptavidin (GE Healthcare, Piscataway, NJ) and quantified using Illumina's BeadStation 500GX Genetic Analysis Systems scanner.
The microarray data reported in the manuscript are in accordance with MIAME guidelines.The raw data for the analyses have been deposited in the NCBI GEO database, series record GSE24233 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc = GSE24233).The Illumina BeadStudio software was used to measure fluorescent hybridization signals.Data were extracted by BeadStudio (Illumina) and analyzed using GeneSpring software v. 7.3.1 (Silicon Genetics, Redwood City, CA).Raw data were imported into GeneSpring and normalized using global normalization.The normalized data were used to identify changes in gene expression in the 4 group comparisons: SL vs SNL, ML vs SNL, MNL vs SNL, and ML vs SL.A gene was identified as significantly changed if it showed increased or decreased expression according to an arbitrary cut-off of 1.7-fold changes at p,0.025, according to GeneSpring statistical package (unpaired t-test).In previous studies, genes identified by similar criteria were consistently validated by qPCR [31,38,39].qPCR Total RNA extracted from the striatum was used to confirm the expression of genes of interest by qRT-PCR.In brief, unpooled total RNA obtained from 8-12 striata per group was reversetranscribed with oligo dT primers and RT for PCR kit (Clontech, Palo Alto, CA).PCR experiments were done using the Chroma4 RT-PCR Detection System and iQ SYBR Green Supermix (BioRad, Hercules, CA).Sequences for gene-specific primers corresponding to PCR targets were obtained using LightCycler Probe Design software (Roche).The primers were synthesized and HPLC-purified at the Synthesis and Sequencing Facility of Johns Hopkins University (Baltimore, MD).The primers are listed in Table S1.qPCR values were normalized using light chain of clathrin and quantified.The results are reported as relative changes calculated as the ratios of normalized gene expression data of each group compared to the control group injected with saline (SC).

Statistical Analysis
Statistical analysis for the HPLC and PCR data was performed using analysis of variance (ANOVA) followed by Fisher's protected least significant difference post-hoc comparison (StatView 4.02, SAS Institute, Cary, NC).Values are shown as means 6 SEM.The null hypothesis was rejected at p,0.05.

Functional characterization of the unilateral 6-OHDA nigrostriatal lesion
Rotation.We assessed the effectiveness of the 6-OHDA lesions by both behavioral and biochemical means.Figure 1 shows the effects of APO and METH on contraversive and ipsiversive rotational behaviors, respectively.We found time-dependent changes in APO-induced turns, which peaked within the first 6 min after injection of the drug (Figure 1A).APO-induced contralateral turns were stable for the first 30 min and then rapidly tapered to being nonexistent by 54 min after drug.METHinduced ipsiversive turns became apparent within the first 12 min after the drug injection and peaked at around 30 min.METHinduced rotations remained prominent for about 1 hour and were still measurable at 90 min after METH injection (Figure 1B).
Biochemistry.We found no differences in the levels of either DA or its metabolites, DOPAC and HVA on the intact side of hemiparkinsonian rats in comparison to control animals (Table 1).Unilateral injections of 6-OHDA in the MFB caused complete loss of DA, DOPAC, and HVA (Table 1).Injection of METH prior to euthanizing the animals did not affect DA, DOPAC, and HVA levels.METH did not cause any significant changes in DOPAC/ DA or HVA/DA ratios in any of the groups (data not shown).Table 1 also shows the effects of unilateral 6-OHDA lesions on 5-HT and 5-HIAA levels as well as 5-HIAA/5-HT ratios in the striatum.We found significant decreases in 5-HT concentrations on the side of the 6-OHDA injections in the saline-(237%) and in METH-treated (251%) rats.5-HIAA levels were also affected on the lesioned side of the saline-(226.5%)and METH-treated (226.4%)animals.In addition, injections of METH caused significant increases in 5-HIAA levels in the control group (+33%) and on the intact side of the hemiparkinsonian rats (+24%).We also found increases in 5-HIAA/5-HT ratios on the 6-OHDAlesioned side of the saline-(+24.4%)and METH-treated animals (+54.7%).The increases in 5-HIAA/5-HT ratios on the DAdepleted side of animals treated with METH were significantly higher than the ratios in all other groups.These findings are consistent with those of other investigators who have reported increases in 5-HIAA/5-HT ratios in animals that had suffered greater than 90% loss of striatal DA [40].However, the potentiating effects of METH on 5-HIAA/5-HT ratios in the lesioned striatum have not been reported before.

Dopamine depletion and METH-induced changes in mRNA levels in the striatum
Microarray analysis.In order to provide a panoramic view of METH effects on gene expression in the DA-depleted striatum, we used large scale microarray analysis with rat Illumina arrays that contain about 23, 000 genes. Figure 2 shows a Venn diagram of the comparisons between 4 groups of interest: SL vs SNL, ML vs SNL, MNL vs SNL, and ML vs SL.The identity of the affected genes is given in Tables 2, 3, 4, and 5 which list the full names of the genes.The Venn diagram shows 67 genes that were affected in the DA-denervated side in comparison to the intact side (SL vs SNL).Of these, 45 were up-regulated whereas 22 were downregulated (Table 2).Up-regulated genes include Ldhc, Stab2, Nmu, Fst, Nts, and Tac2 whereas down-regulated transcripts included Tac1 (Table 2).METH injection caused changes in the expression of 86 genes in the intact side (MNL) in comparison to the SNL side (Table 3).Genes of interest that were up-regulated include transcription factors c-fos, Junb, Egr2, and Nr4a3 (Nor-1).METH administration caused alterations in mRNA levels for 98 genes on the lesioned side in comparison to the SNL side (ML vs SNL) (Table 4).These include 61 up-regulated and 37 down-regulated genes, with only 16 of these genes being also affected in the MNL vs SNL comparison (Tables 3 and 4).Of the 98 genes affected in the ML vs SNL group, 28 were also found in the SL vs SNL comparison, with their expression, for the most part, not being further affected by the METH administration (Tables 2 and 4).Genes of interest found only in the ML vs SNL comparison were Syt10 and Cox-2, which were up-regulated by the METH injection.One of the common genes between the two sets of comparisons (MNL vs SNL and ML vs SNL) is Nr4a3 (Nor-1) which shows higher expression in the MNL vs SNL group.When we evaluated gene expression in the ML versus SL group, another set of 55 genes was identified, with 30 being up-regulated and 25 being down-regulated (Table 5).As shown in the Venn diagram, only 5 of these transcripts were also found in the ML vs SNL group.These include c-fos and Hcst (Tables 4 and 5).We also compared gene expression between the MNL and ML groups and found that several transcription factors including c-fos, Egr1, Egr2, and Junb were significantly up-regulated in the MNL group (data not shown, see Table 6 for qPCR validation).
qPCR. Figure 3 shows the results of the qPCR validation for some of the genes affected on the lesioned side of the hemiparkinsonian rats treated with saline (SL vs SNL comparison).Consistent with the array data, DA depletion was associated with significant decreases (268% of SC group) in Tac1 (also called tachykinin 1, preprotachykinin A, substance P) expression on the DA depleted side (SL).The injection of METH did not significantly influence Tac1 expression on the lesioned side (250% of SC group).In contrast, the single METH injection caused small increases in normal animals (1.4-fold) and on the intact side (1.3-fold) of the 6-OHDA-treated rats (Figure 3A).In contrast to the effects on Tac1, DA depletion resulted in significant increases (2.2-fold) in the expression of Tac2 (neurokinin B) (Figure 3B).METH administration did not induce any changes in Tac2 expression in the control animals nor in the intact side of 6-OHDA lesioned animals.METH did not potentiate the effects of 6-OHDA on Tac2 expression on the lesioned side (compare SL to ML groups in Figure 3B).There were also significant 6-OHDA-induced decreases in Pdyn expression in the SL group (Figure 3C).These 6-OHDA-induced changes were not affected by METH.In contrast, METH injection caused increases in Pdyn mRNA levels in the control animals and in the intact side of 6-OHDA-injected rats.Striatal DA depletion also induced significant increases (2.86-fold) in Nts mRNA levels (Figure 3D).The METH injection caused increases (2.21-fold) in Nts mRNA in the control animals (MC group), confirming previous results, which showed that METH induced a dramatic increase in striatal Nts mRNA levels [41].METH injection did not cause further increases in Nts expression on the side of the lesion when compared to the saline-treated 6-OHDA-injected animals (compare SL to ML group).Nevertheless, the data for the ML group were significantly higher than those measured for the MC group (Figure 3D).There were also significant DA depletioninduced increases (4.36-fold) in Nmu mRNA levels in the 6-OHDA-treated animals (Figure 3E).METH injection did not change Nmu expression in the control animals (MC group) nor on the intact side of the lesioned animals.However, METH treatment enhanced the increases (6.04-fold) in Nmu expression on the DA-depleted side (compare SL to ML group).The qPCR experiment also replicated the array data for Fst expression, which showed significant increases (3.39-fold) on the DA-depleted side of hemiparkinsonian rats (Figure 3F).The METH-treated animals also showed increases (4.78-fold) in the lesioned side, with these changes being significantly higher than those observed in the saline-treated 6-OHDA-lesioned animals.The observations that the small METH dose caused further increases in Fst mRNA levels on the lesioned side of 6-OHDA-treated rats suggest that these METH-induced alterations in Fst expression might be independent of METH-induced DA release.
Table 6 shows the qPCR validation of the microarray data which identified multiple transcription factors affected on the intact side of the 6-OHDA-treated rats after METH injection (see Table 3).The effects of METH on the expression of immediateearly genes in the normally DA innervated striatum have been the subjects of several reports ( [31] and references therein).However, because these studies were conducted in animals with normal neurotransmitter contents, conclusions drawn from these investigations might not help to fully characterize the role of DA and other neurotransmitters in these transcriptional responses.Therefore, for the purpose of thoroughness, we performed qPCR on other genes that fall within similar classes of transcription factors and that had been identified as affected by injections of higher METH doses [30,31].Except for Fra1, c-jun, and Nr4a2 (Nurr1), METH injection caused significant increases in the expression of the other members of these classes of transcription factors.We found that METH caused increases in c-fos mRNA levels on both the intact (2.7-fold) and on the DA-denervated (1.7-fold) side (Table 6).Another gene of interest is Jund.METH injection caused similar increases in Jund expression on both the intact (2.2-fold) and lesioned (1.9-fold) striata of the hemiparkinsonian rats (Table 6).We found significant increases in the expression of Nr4a1 (Nur77) in the denervated striatum in comparison to the intact side (SL vs SNL) of hemiparkinsonian rats.These data are different from the findings of microarray analysis that did not identify Nr4a1 as being changed in the SL group.These might be related to the somewhat stringent criteria used in the array analysis and/or to the greater sensitivity of qPCR.The values in the SL group were also comparable to those measured on the lesioned side after METH (ML group) administration, implying that METH did not influence Nr4a1 expression in the absence of DA.Other transcription factors of interest are the EGRs [42].Egr2 expression was substantially more induced by METH injection in the intact side (6.1-fold) of the hemiparkinsonian rats in comparison to control animals (3.8-fold) (Table 6).Similar differential responses were observed for Egr3 expression after METH administration to the control and 6-OHDA-injected animals.This is in contrast to the effects of METH on Egr1 which were almost identical in the control and lesioned rats.
Figure 4 illustrates the expression of some genes that showed METH-induced changes in the DA-depleted striatum.METH injection caused increases in Inhba in the intact (2.35-fold) (MNL) and denervated (2.21-fold) (ML) sides of the striatum in hemiparkinsonian rats (Figure 4A).METH injection also induced increases in Inhba in the control animals but these changes were of smaller magnitude than the increases in the lesioned animals (compare the MC to the MNL and ML groups).Figure 4B shows the effects of METH on the activin receptor, Acvr1 [43].Although not found on the array, we wanted to know if there were changes in this receptor in view of METH-induced changes in Inhba.We found that METH caused significant increases in Acvr1 mRNA This partial list of genes was generated from the SL vs SNL comparison.To be included the genes had to meet the inclusion criteria: 61.7-fold at p,0.025.The genes are listed in descending order according to fold changes.doi:10.1371/journal.pone.0015643.t002levels in the controls (MC, 1.79-fold), as well as on the intact (MNL, 1.93-fold) and lesioned (ML, 1.79-fold) sides of hemiparkinsonian rats.There were also significant increases (1.57-fold) in Acvr1 expression in the intact side of saline-injected 6-OHDAtreated rats.The changes in the lesioned side of these animals did not reach statistical significance (1.49-fold, p = 0.06).We also found significant METH-induced increases in Pdgf-d mRNA levels on both the intact (1.39-fold) and denervated (1.39-fold) striatum in comparison to the SC group and the intact (SNL) side of 6-OHDA-treated rats (Figure 4C).There were no significant changes in METH-treated control group (MC) and in 6-OHDA-treated animals injected with saline (SL).There were no significant effects of the 6-OHDA lesion on striatal Cox-2 expression (Figure 4D).However, METH injection caused increases (MC, 1.83-fold) in the control group, as well as in the intact (MNL, 1.66-fold) and lesioned (ML, 3.19-fold) sides of the The data were generated from the MNL vs SNL comparison.The genes are listed in descending order according to fold changes in comparison to the SNL group.doi:10.1371/journal.pone.0015643.t003 hemiparkinsonian rats.The METH-induced increases on the lesioned side (ML) were significantly higher than those observed in the MNL and MC groups (Figure 4D).We found no significant changes in Syt10 expression on the side of the lesion in animals treated with saline (Figure 4E).In addition, injection of METH caused no changes in striatum of controls animals (MC group).In contrast, the drug caused significant increases in Syt10 expression in both the intact (1.68-fold) and DAdenervated (2.13-fold) sides of 6-OHDA-treated rats in comparison to the SC group and to the SNL side of hemiparkinsonian rats.Figure 4F shows that 6-OHDA did not cause any significant changes in the expression of Synj2bp in the saline-treated animals (SL).However, METH induced small but significant increases (1.35-fold) in Synj2bp in the DA depleted side (ML), but not on the

Discussion
Injections of 6-OHDA in the MFB caused complete loss of DA and significant decreases in 5-HT content on the lesioned side of the striatum.The 6-OHDA lesion was also associated with METH-induced increases in 5-HIAA/5-HT ratios.6-OHDAinduced striatal DA depletion resulted in substantial changes in transcript levels on the lesioned side.We also found that METH dose that caused changes in mRNA levels on the intact side elicited alterations in the expression of, mostly, different transcripts on the DA-depleted side.METH injection also failed to impact the expression of several genes whose mRNA levels were modulated by DA depletion in the saline-treated hemiparkinsonian rats.

Effects of unilateral 6-OHDA-induced MFB lesions on striatal gene expression
We found significant increases in the expression of Nts, Nmu, and Tac2 and decreases in Tac1 expression in the DA-depleted striata of rats treated with saline.These results are consistent with previous reports of lesion-induced increases in Nmu [44], Nts [45,46] and Tac2 [47] as well as decreases in Tac1 [18,48].Interestingly, although METH injection did not further influence the changes observed in Tac1 and Tac2 mRNA levels on the DAdepleted side, the drug did cause further increases in Nmu expression on the lesioned side, suggesting that Nmu expression is regulated by additional neurotransmitters in the absence of DA innervation.The observations that the small dose of METH caused significant increases in Nts mRNA levels in the control animals (MC) that were of lesser magnitude than those induced by the drug on the lesioned striatum suggest that METH might enhance Nts gene expression in association with or via increased 5- HT release, given that METH caused increased 5-HIAA/5-HT ratios on the 6-OHDA lesioned side.When taken together, our results suggest that the expression of Tac2, Nmu, and Nts might be under tonic inhibitory control by DA.The findings that this dose of METH did not cause any significant increases in Tac2 and Nmu transcripts in the intact striatum suggest a potential DA-mediated floor effect under these conditions.The situation appears more complex for Nts mRNA which showed METH-mediated increases in both the presence and absence of DA.Increases in Nts mRNA after METH have been reported previously and are thought to be related to stimulation of DA D1 receptors [49].However, the fact that METH caused greater increases in Nts transcript levels on the lesioned side suggests that both DA and 5-HT might be involved in regulating Nts transcript levels after injection of METH since methylenedioxymethamphetamine (MDMA) which causes 5-HT release can also increase Nts mRNA levels [50].Other potential regulators of gene expression after 6-OHDA lesions might include changes in glutamatergic functions because hemiparkinsonian rats exhibit increases in extracellular glutamate [51][52][53][54] and changes in glutamate synapses [52] within the lesioned striatum.It is also possible to suggest the DA depletion-mediated changes in transcript levels might be related to alterations in the levels of Nur77 (Nr4a1) [55], the only transcription factor that showed 2fold increases on the DA-denervated side (see Table 6).Therefore, there might exist a certain correspondence between DA depletion, Nur77 expression, and up-regulated transcripts of some of the identified genes in the hemiparkinsonian brain.
It is also of interest to discuss the substantial increases in Nmu mRNA observed after DA depletion reported in the present study and in a previous microarray paper [44] because NMU is a neuropeptide which is found ubiquitously in the body, with very high levels reported in the gastrointestinal tract and the pituitary gland [56].NMU was first isolated from porcine spinal cord [57].
Subsequently, NMU was located in the rodent brain and in other tissues [58][59][60].Using radioimmunoassay, Domin et al. [59] reported high concentrations of NMU in the nucleus accumbens, septum, and hypothalamus but moderate levels in the substantia nigra and the globus pallidus and much lower levels in the striatum.Fujii et al. [61] used RT-PCR and reported that Nmu mRNA levels were very high in the pituitary.Other structures such as the medulla oblongata, hypothalamus and striatum showed moderate Nmu mRNA levels whereas the cerebral cortex, hippocampus, and the cerebellum had very low levels.NMU has also been reported to participate in several physiological functions including release of corticotrophin releasing hormone, excitation of nociceptive neurons, and regulation of body weight [62,63].NMU exerts its physiological effects by stimulation of the NMUR1 [61,64] and NMUR2 [65] receptors.In the brain, the effects of this peptide might occur through NMUR2 because qPCR analysis failed to detect much Nmur1 in the brain whereas moderate-tohigh levels of Nmur2 were found in the substantia nigra, the nucleus accumbens, and the striatum [66].Because Nmu mRNA is expressed in the striatum and because its level is substantially regulated after DA depletion [44] and by METH administration (present study), it is possible to suggest that this neuropeptide might play important roles in striatal functions.This idea is supported by the report that intra-cerebroventricular NMU injections cause increased motor activity in rats [66].
The DA depletion-induced increases in Fst mRNA levels are consistent with those reported in a recent microarray study [44].FST, a monomeric glycoprotein which is co-expressed with the activins, irreversibly binds with activins and prevents their interactions with their receptors [67][68][69].Interestingly, we also found increases in Inhba transcript which encodes a subunit of inhibins and activins; both of which are members of a family of polypetides that also include transforming growth factor-beta and bone morphogenic proteins [70,71].Activins are disulfidelinked homodimers of INHBA (activin A) and of INHBB (activin B) and can form the heterodimer, activin AB [72].Activin A is up-regulated in the brain consequent to kainate- [73] and ischemia- [74] induced insults.Because activins are also thought to have neurotrophic properties [75], the METH-induced increases in Inhba expression, and by extension of activins (or inhibins), suggest the possibility that this family of proteins might  participate in METH-induced neuroplastic changes in the brain [37].

Effects of METH injections on the expression of transcription factors in the rat striatum
Previous studies have also shown that the expression of several transcription factors is affected by injections of larger doses of METH in normal animals [29][30][31]76].These genes are similar to those reported by Berke et al. [16] who had used the DA D1 agonist, SKF38393, to report changes in gene expression mainly on the side of the 6-OHDA-induced DA depletion.In the present study, we used the indirect agonist, METH, which is dependent on DA release to exert its actions on DA receptors [21,22,[24][25][26].Interestingly, we found that the METH injection caused increases in c-fos mRNA levels on both sides of the hemiparkinsonian rats, with more prominent increases on the intact side (2.7-fold) than on the lesioned (1.75-fold) side (Table 6).The observations of increased c-fos mRNA levels on the lesioned side are consistent with previous reports that METH caused increase in c-Fos protein levels in the DA-depleted striatum [77].Together, these observations suggest that METH-induced changes in c-fos mRNA levels might dependent on both DA and non-DA, presumably 5-HT, systems in the rodent striatum [78,79].The role of glutamate in these responses needs to be also considered since amphetamine-induced changes in striatal responses involves stimulation of NMDA receptors [80].
As reported above, we found that Egr2 and Egr3 transcripts were more induced by the METH injection on the intact side of the lesioned rats than in the control animals, whereas those in Egr1 mRNA levels were similar in these two METH-treated groups.These results suggest that Egr2 and Egr3 transcription might be coregulated in the rodent striatum in a manner comparable to observations in T cells [81].The differential responses between the control animals and the hemiparkinsonian rats might be due, in part, to higher levels of baseline DA released in dialysates collected on the intact side of unilaterally 6-OHDA-lesioned rats in comparison to DA levels measured in control animals [24,82].Thus, the steady-state increases in DA released in the synaptic cleft of the 6-OHDA-treated animals might cause potentiated transcriptional sensitivity to METH-mediated DA-induced changes in Egr2 and Egr3 expression in the intact striatum.

Differential effects of METH on gene expression between the DA-innervated and -depleted striatum of hemiparkinsonian rats
METH administration also caused substantial changes in the expression of several genes on the DA-depleted side of hemiparkinsonian rats (Figure 4).The Pdgf-d transcript is of interest because the PDGF-D protein belongs to a family of trophic factors that are involved in the growth and survival of mesenchymal cells [83,84].These factors include PDGF-A, PDGF-B, PDGF-C, and PDGF-D [85].Although the effects of the PDGF-A and PDGF-B have been studied on diverse cell types of the nervous system [86,87], little is known about the functions of PDGF-D in the brain.A previous study had found early and prolonged increases in neuronal Pdgf-a and Pdgf-b mRNA levels on the side ipsilateral to a unilateral 6-OHDA injection in the rat median forebrain bundle [88].There were also more gradual increases in these transcripts on the contralateral side of the lesion [88].Those findings differ from our present data because we found increases in Pdgf-d mRNA on both sides of the METH-treated hemiparkinsonian rats.Our observations that METH can cause increases in Pdgf-d mRNA levels on the lesioned side of hemiparkinsonian rats suggest that the PDGF-D protein might play a role in 5-HT-mediated changes after striatal DA depletion because of METH-induced increases in 5-HT turnover on the lesioned side.Because PDGF proteins act as proliferative factors for glial cells [89,90], it will be of interest to examine to what extent PDGF-D might participate in METHinduced reactive astrocytosis or microgliosis [91].Of related interest, we found that METH caused increases in the Cox-2 transcript on both sides of the brains in hemiparkinsonian rats.These data are consistent with reports by other authors who have previously tabulated METH-induced changes in COX-2 protein expression [92,93].These METH-induced increases in COX-2 might be related, in part, to the increases in PDGF-D because Cox-2 mRNA is induced during activation of microglial cells [94] and because PDGFs are proliferative factors for glial cells [89,90].
Although we have focused the discussion on the known biochemical effects of METH on neurotransmitter release, it could also be argued that some of the changes in gene expression might be secondary to rotational behaviors, independent of drug effects.This is an important issue because immobilization of animals after apomorphine injection prevented priming responses to the direct DA D1 agonist, SKF38393, thus suggesting movement-mediated induction of these behavioral changes [95].However, the fact that the pattern of changes in gene expression is similar in the striatum of the METH-treated control rats and the nonlesioned striatal side of the METH-treated hemiparkinsonian rats suggests that the changes in mRNA levels are secondary to drug effects not to intense turning behaviors since control rats exhibited no METH-induced rotation.The latter argument is consistent with the report that DA D1-induced behavioral responses are dissociated from changes in c-Fos protein expression [96].This contention is also supported by the fact that administration of the indirect DA agonist, amphetamine, whose actions like those of METH [21][22][23][24][25] are also dependent on DA release from nerve terminals, did not cause priming to DA D1 receptor stimulation [96].Nevertheless, the veracity of this claim will have to await results of time-dependent experiments in which animals are euthanized during the time of peak METH-induced rotation which occurred at around 30-45 min after the METH injection (see Figure 1B).The results of the proposed studies will be need to be contrasted to the present findings which were obtained in animals euthanized at 2 hours after the METH injection at a time when METH-induced rotation had subsided.

Conclusion
In summary, we have reported that a single injection of a relatively low METH dose that caused substantial changes in gene expression in the intact striatum also triggered alterations in the expression of a different set of genes in the striatum that was completely lacking of DA.Because the 6-OHDA-induced lesion completely depleted dopamine in the striatum, these METHinduced changes on the lesioned are probably not related to disuse-induced supersensitivity of striatal DA receptors because METH is an indirect agonist that depends on intact DA terminals for its action on DA receptors.Thus, these observations implicate DA-independent phenomena in the METH-mediated regulation of these transcripts and point to the possibility that the absence of DA might cause plastic changes that render the striatum differentially responsive to the effects of METH on transcript levels in intrinsic striatal cells.This idea is consistent with the results of some studies that have documented differential physiological responses in the striatum of hemiparkinsonian rats [97][98][99].It remains to be determined to what extent these changes might serve as substrates for synaptic plasticity observed in the lesioned striatum.Finally, our observations of METH-mediated DA-independent alterations in transcript levels suggest that other molecular pathways should be taken into consideration when discussing therapeutic approaches to METH abusers who show pathological changes, including DA depletion, in their brains.

Figure 2 .
Figure 2. Differential profiles of transcript levels in saline-and METH-treated 6-OHDA-induced hemiparkinsonian rats.The Venn diagram shows the distinctiveness and the overlap of genes identified by the four separate comparisons shown in the figure.The rats were treated and euthanized as described in the text.RNA was extracted from striatal tissues obtained from both sides of the brains of saline-and METH-injected hemiparkinsonian rats.Microarray analyses were performed as described in the text.Genes were identified as differently expressed if they showed greater than 61.7-fold changes at p,0.025.doi:10.1371/journal.pone.0015643.g002

Figure 3 .
Figure 3. Effects of DA depletion and METH on striatal expression of Tac1, Tac2, Pdyn, Nts, Nmu and Fst.The values represent means 6 SEM (fold changes in comparison to the SC group) of 6-10 measurements.Statistical analyses were done by ANOVA followed by PLSD.Key to statistics: Key to statistics: *p,0.05, in comparison to the SC group; #p,0.05, in comparison to the SNL group; !p,0.05, in comparison to the MC group; ^p,0.05, in comparison to the SL group; $p,0.05, in comparison to the MNL group.doi:10.1371/journal.pone.0015643.g003

Figure 4 .
Figure 4. Effects of METH on the expression of Inhba, Acvr1, Pdgf-d, Cox-2, Syt and Synj2bp in DA-depleted striatum.The values represent means 6 SEM (fold changes in comparison to the SC group) of 6-10 measurements.Statistical analyses were done by ANOVA followed by PLSD.Key to statistics: *p,0.05,incomparison to the SC group; #p,0.05, in comparison to the SNL group; !p,0.05, in comparison to the MC group; ^p,0.05, in comparison to the SL group; $p,0.05, in comparison to the MNL group.doi:10.1371/journal.pone.0015643.g004

Table 1 .
Effects of 6-OHDA and METH on striatal monoamine levels.The values represent means 6 SEM (percentage changes with respect to the SC group) of 8-10 measurements per group.The groups are as described in the text.Statistical analyses were done by ANOVA followed by PLSD.Key to statistics:

Table 2 .
Partial list of genes showing changes in the DA-depleted striatum of hemiparkinsonian rats.

Table 3 .
Partial list of METH-induced changes in the levels of gene transcripts on the intact striatum of hemiparkinsonian rats.

Table 4 .
Partial list of genes showing METH-induced changes in the DA-depleted striatum.

Table 5 .
METH-induced changes in levels of gene transcripts on the lesioned side of hemiparkinsonian rats.The data were generated from the ML vs SL comparison.The genes are listed in descending order according to fold changes in comparison to expression in the SL group. doi:10.1371/journal.pone.0015643.t005

Table 6 .
Effects of METH on striatal expression of transcription factors in hemiparkinsonian rats.The values represent means 6 SEM (fold changes with respect to the SC group) of 8-10 measurements per group.The groups are as described in the text.Statistical analyses were done by ANOVA followed by PLSD.Key to statistics: