Upregulation of neuronal astrocyte elevated gene-1 protects nigral dopaminergic neurons in vivo

Upregulation of neuronal astrocyte elevated gene-1 protects nigral dopaminergic neurons in vivo The role of astrocyte elevated gene-1 (AEG-1) in nigral dopaminergic (DA) neurons has not been studied. Here we report that the expression of AEG-1 was significantly lower in DA neurons in the postmortem substantia nigra of patients with Parkinson’s disease (PD) compared to age-matched controls. Similarly, decreased AEG-1 levels were found in the 6-hydroxydopamine (6-OHDA) mouse model of PD. An adeno-associated virus-induced increase in the expression of AEG-1 attenuated the 6-OHDA-triggered apoptotic death of nigral DA neurons. Moreover, the neuroprotection conferred by the AEG-1 upregulation significantly intensified the neurorestorative effects of the constitutively active ras homolog enriched in the brain [Rheb(S16H)]. Collectively, these results demonstrated that the sustained level of AEG-1 as an important anti-apoptotic factor in nigral DA neurons might potentiate the therapeutic effects of treatments, such as Rheb(S16H) administration, on the degeneration of the DA pathway that characterizes PD. Introduction amyotrophic lateral sclerosis (ALS) by activating apopto- Astrocyte elevated gene-1 (AEG-1), also known as tic signaling pathways via inhibition of the phosphatidy- metadherin, was originally identified as a human immu- linositol-4,5-bisphosphate 3-kinase/protein kinase B nodeficiency virus-1- and tumor necrosis factor-alpha- (PI3K/Akt) signaling pathway . inducible gene in human fetal astrocytes, and its upre- The aberrant activation of apoptotic signaling pathways gulation is a well-established important oncogenic event in the adult brain is a well-known neurotoxic event that is 1–4 in various types of human cancer . The downregulation associated with neuronal loss, such as that observed in of neuronal AEG-1 has recently been shown to reduce the neurodegenerative diseases, including Parkinson’s disease 6–8 viability of motor neurons in a mouse model of (PD) and Alzheimer’s disease (AD) , and the PI3K/Akt/ mammalian target of rapamycin complex 1 (mTORC1) signaling pathway has been shown to elicit neuroprotec- Correspondence: Seok-Geun Lee (seokgeun@khu.ac.kr) or Sang tive effects on the survival and growth of neurons in the Ryong Kim (srk75@knu.ac.kr) 1 9–11 School of Life Sciences, BK21 Plus KNU Creative BioResearch Group, Institute nigrostriatal dopaminergic (DA) system . However, of Life Science & Biotechnology, Kyungpook National University, Daegu 41566, little is known about the neuroprotective role of AEG-1 in Republic of Korea PD. Department of Neural Development and Disease, Department of Structure & Function of Neural Network, Korea Brain Research Institute, Daegu 41062, Here we found that the loss of DA neurons in post- Republic of Korea mortem substantia nigra (SN) tissue from patients with Full list of author information is available at the end of the article These authors contributed equally: Eunju Leem, Hyung-Jun Kim, Minji Choi. Edited by G. Raschellà. © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; Leem et al. Cell Death and Disease (2018) 9:449 Page 2 of 15 PD were associated with significant decreases in the levels neurons, we evaluated the in vivo effects of AEG-1 over- of expression of AEG-1 in nigral DA neurons of patients expression on the basal levels of apoptotic markers, such with PD compared to age-matched controls. These find- as cleaved caspase-3 and cleaved poly (ADP-ribose) ings suggested that the relationship between AEG-1 polymerase 1 (PARP-1) in nigral DA neurons. Mice were downregulation and the pathogenesis of PD are clini- sacrificed 4 weeks after intranigral injections of AAV- cally relevant. To investigate the role of AEG-1 as a sur- AEG-1 or the control vector AAV-green fluorescent protein (GFP), and the transduction of DA neurons was vival factor in nigral DA neurons in the adult brain, we examined the effects of the adeno-associated virus (AAV)- confirmed by the patterns of GFP expression and the mediated overexpression of AEG-1 on these neurons in immunoperoxidase staining of the hemagglutinin (HA) the 6-hydroxydopamine (6-OHDA)-treated animal model epitope in the AAV-AEG-1 vector, respectively (Fig. 2a). 9,10,12 of PD . Additionally, we examined whether the neu- HA- and GFP-positive cells were clearly colocalized with roprotection conferred by AEG-1 overexpression, which TH-positive DA neurons (Fig. 2b) but not with glial might be a therapeutic intervention, contributed to the fibrillary acidic protein (GFAP)-positive astrocytes or neurorestorative effects on the in vivo nigrostriatal DA ionized calcium binding adaptor molecule 1 (Iba1)-posi- system of treatment strategies, such as the administration tive microglia in the SN (Fig. 2c). Upregulation of AEG-1, of constitutively active ras homolog enriched in brain which showed no neurotoxicity (Fig. 2d–f), resulted in a (with a S16H mutation) [Rheb(S16H)], which induces significant decrease in the basal levels of cleaved caspase-3 9,10 axonal regrowth in damaged DA neurons . and cleaved PARP-1 in the SN compared to noninjected and GFP controls (Fig. 2g; *p = 0.005 vs. CON). Results Similar to previous reports that have implicated apop- 6,8 Decreased levels of AEG-1 expression in the SN of patients tosis in the loss of DA neurons in patients with PD , with PD and a neurotoxin-based model of PD western blot analyses revealed significant increases in the To investigate the alterations in the levels of AEG-1 levels of caspase-3, cleaved caspase-3, and cleaved PARP- expression in the SN of patients with PD (Fig. 1a), we 1 were observed in the SN of patients with PD compared & && performed immunohistochemical staining of the expression to age-matched controls (Fig. 3a; p = 0.014, p = 0.019, &&& patterns (Fig. 1b) and quantified the changes using western and p = 0.009 vs. CON, respectively). As shown in the blotting (Fig. 1c). AEG-1-positive immunoreactivity (blue) experimental schematic (Fig. 3b), the double immuno- was clearly reduced in neuromelanin-positive DA neurons fluorescence staining of TH (red) and cleaved caspase-3 (brown) in the SN of patients with PD compared to age- (green) and of TH and cleaved PARP-1 (green) (Fig. 3c), matched controls (Fig. 1b). Western blot analyses revealed and western blot analyses (Fig. 3d) showed that the levels significant decreases in the levels of AEG-1 and tyrosine of both cleaved caspase-3 and cleaved PARP-1 sig- hydroxylase (TH, a marker of DA neurons) in the SN of the nificantly increased 2 days post-lesion in the TH-positive patients with PD compared to age-matched controls DA neurons in the SN of mice injected with 6-OHDA # 15,16 (Fig. 1c; p= 0.033 and *p= 0.022 for AEG-1 and TH, only . However, the upregulation of AEG-1 sig- respectively, vs. CON). However, decreased AEG-1 nificantly inhibited the cleavage of both caspase-3 and expression was not observed in the hippocampus of PARP-1 in the nigral DA neurons following the 6-OHDA patients with AD compared to age-matched controls, even injections compared to injections of 6-OHDA alone ### though there was a significant loss of neuronal nuclei (Fig. 3c, d; ***p = 0.009 and p = 0.002, respectively, vs. (NeuN, a marker of neurons) in that region in the patients 6-OHDA alone). The anti-apoptotic effects of AEG-1 on compared to controls (Fig. 1f, g; p= 0.001 vs.CON). The the 6-OHDA-induced neurotoxicity in DA neurons were reduction of AEG-1 (brown) was specifictothe SN pars confirmed with western blot analyses of the B-cell lym- compacta of 1 day post-lesion 6-OHDA-treated mice phoma 2/Bcl-2-associated X protein (Bcl-2/Bax) ratio (Fig. 1d), which is a well-known neurotoxin-based model of (Supplementary Figure S1). 9,10,12–14 PD . Western blot analyses similarly showed a sig- nificant decrease in the levels of AEG-1 expression in the Neuroprotective effects of AEG-1 upregulation against SN after 6-OHDA administration, compared to untreated 6-OHDA neurotoxicity controls, 1 day post-lesion (Fig. 1e; *p= 0.024 vs.CON), We evaluated the neuroprotective effects of AAV-AEG- even though the levels of TH were not significantly 1 in the 6-OHDA-treated mouse model of PD (Fig. 4a). decreased in the SN (Fig. 1e). 6-OHDA administration clearly caused neurotoxicity in the nigrostriatal DA projections (Fig. 4b, c), and the Decreased levels of apoptotic signaling molecules by AEG- transduction of DA neurons with AEG-1 but not GFP 1 transduction of mature neurons in the SN effectively mitigated the 6-OHDA-induced neurotoxicity To determine whether the upregulation of AEG-1 in the SN compared to the effects of treatment with affected the apoptotic signaling pathways in nigral DA 6-OHDA alone (Fig. 4b; p = 0.026 vs. 6-OHDA alone). Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 3 of 15 Fig. 1 (See legend on next page.) Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 4 of 15 (see figure on previous page) Fig. 1 Decreased levels of astrocyte elevated gene-1 (AEG-1) in the postmortem substantia nigra (SN) of patients with Parkinson’s disease (PD) and the SN of 6-hydroxydopamine (6-OHDA)-treated mice. a Description of the human postmortem SN tissue. b Immunohistochemistry for AEG-1 in the human SN. Scale bar, 50 μm. The square insets in the left panels contain magnifications of the photomicrographs in the right panels. Scale bar, 20 μm. The blue arrows indicate AEG-1 immunoreactivity, and the brown arrows indicate neuromelanin immunoreactivity. c Western blot analysis of the levels of tyrosine hydroxylase (TH) and AEG-1 in the human SN. *p= 0.022 and p = 0.033 vs. age-matched controls (CON) (t-test; n = 4 for each group). d Representative sections showing AEG-1 expression (with Nissl counterstaining) in the mouse SN pars compacta (SNpc), which is outlined by the dotted lines. Scale bar, 20 μm. The square insets in the left contain magnifications of the photomicrographs in the right panels. Scale bar, 20 μm. e Western blot analyses of the levels of AEG-1 and TH in the mouse SN. AA ascorbic acid. *p = 0.024 and **p= 0.001 vs. intact CON; p < 0.001 for TH, significantly different from CON [one-way analysis of variance (ANOVA) with Tukey’s post hoc test; n = 4 for each group]. f Description of the human postmortem hippocampal tissue. g Western blot analyses of the levels of AEG-1 and neuronal nuclei (NeuN) in the hippocampus of patients with Alzheimer’s disease (AD) and CON. Please note that of the levels of AEG-1 are not decreased in the postmortem hippocampus of patients with AD compared with CON. p = 0.001 vs. CON (t-test; n = 5 for each group) Western blot analyses also showed that AEG-1 trans- sustained activation of the Akt/mTORC1 signaling path- duction significantly preserved the levels of TH expres- way induces axonal regeneration in damaged neu- 9,10,20,21 sion following 6-OHDA-induced neurotoxicity compared rons . However, western blot analyses showed that to treatment with 6-OHDA alone (Fig. 4d; p = 0.002 vs. the overexpression of AEG-1 alone did not alter the 6-OHDA alone) in the SN but not in the striatum (STR). phosphorylation statuses of the Thr37/46 residues of 4E- Similar to the limited neuroprotective effects observed BP1, which are indicative of mTORC1 activity (Supple- using immunostaining and western blotting, the results mentary Figure S4). Similarly, AEG-1 overexpression did obtained with reversed-phase high-performance liquid not induce significant increases in the levels of p-Akt, chromatography (HPLC) analyses indicated that the levels which is associated with activation of mTORC1, com- of striatal dopamine and its metabolites, including 3,4- pared to the levels in noninjected controls (Supplemen- dihydroxyphenylacetic acid (DOPAC) and homovanillic tary Figure S4). Additionally, no significant changes were acid (HVA), did not significantly differ between the mice observed in the levels of microtubule-associated protein treated with 6-OHDA following AEG-1 transduction and 1A/1B-light chain 3 (LC3)-I and II, which are used as an 12,19,22 those treated with 6-OHDA alone (Supplementary indicator of autophagosome formation , following Figure S2). injections of AAV-AEG-1 in the SN of healthy mice To evaluate the effects of glial AEG-1 in the nigrostriatal compared to those in noninjected controls (Supplemen- DA system, we unilaterally injected adenovirus (Ad)-AEG- tary Figure S4). 1 (Ad-AEG-1) or Ad-null, which was a control vector, in In the SN of patients with PD, we observed significant the SN of healthy mice and examined whether any neu- increases in the levels of LC3-II and p62, which are well- 12,19,22,23 roprotective effects were observed against 6-OHDA neu- known markers of ALP , compared to those in rotoxicity. Four weeks after intranigral injections of Ad- age-matched controls (Supplementary Figure S5a). Con- AEG-1, immunohistochemical staining of the HA epitope sistent with the increase in the levels of LC3-II and p62, a tag for Ad-AEG-1 indicated the site-specific transduction significant decrease in the levels of p-4E-BP1 expression of SN microglia with Ad-AEG-1 (Supplementary Fig- was also observed in the SN of patients with PD (Sup- ures S3a to c), and no neurotoxicity was observed in the plementary Figure S5a). These observations suggested nigrostriatal DA projections in the brains of the healthy that the suppression of aberrant ALP through the acti- mice (Supplementary Figures S3d and e). However, we did vation of the mTORC1 signaling pathway might also be not observe neuroprotective effects of the overexpression associated with neuroprotection of the nigrostriatal DA of microglial AEG-1 against 6-OHDA-induced neuro- system. However, AEG-1 overexpression in DA neurons toxicity at 1 week post-lesion (Fig. 4e, f). did not suppress the aberrant accumulation of autophagic components, such as LC3-II and p62, and the decrease in Lack of activation of the Akt/mTORC1 signaling pathway in mTORC1 activity following 6-OHDA neurotoxicity nigral DA neurons by AEG-1 upregulation (Supplementary Figure S5b). The results of previous studies suggested that the Akt/ mTORC1 signaling pathway could be regulated by Application of AEG-1-induced neuroprotection to the 3–5,17,18 changes in AEG-1 expression , and the activation functional recovery of the disrupted nigrostriatal DA of Akt/mTORC1 signaling pathway could regulate the system autophagy–lysosomal pathway (ALP), that is associated To determine the importance of sustaining the 12,19 with axonal degeneration in PD . Moreover, the increased levels of neuronal AEG-1 in the adult Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 5 of 15 Fig. 2 Adeno-associated virus (AAV)-AEG-1 transduction of dopaminergic (DA) neurons in the in vivo SN of healthy mice. a Experimental schematic and the immunostaining for green fluorescent protein (GFP; green) and hemagglutinin (HA; brown) in the SNpc, which is outlined by the dotted elliptical shape, which was conducted following each viral injection. Scale bar, 200 μm. b Representative double immunofluorescent labeling of TH (red) and GFP (green) or TH and HA (green) in the SNpc. Scale bar, 20 μm. c Representative double immunofluorescent labeling for glial fibrillary acidic protein (GFAP)/ionized calcium binding adaptor molecule 1 (Iba1; red), which are markers of astrocytes and microglia, respectively, and GFP/ HA (green) in the SNpc of healthy mice. Scale bar, 20 μm. d Immunostaining for TH in the SN and striatum (STR). Scale bars, 200 μm (black) and 50 μm (white) for the SN, and 1000 μm for the STR. e, f The number and optical density of the nigral TH-positive neurons and striatal TH-positive fibers, respectively (one-way ANOVA with Tukey’s post hoc test; n = 4 for each group). CON contralateral side, IPSI ipsilateral side. g Western blot analyses of the levels of cleaved caspase-3 (c-caspase-3) and cleaved poly (ADP-ribose) polymerase 1 (c-PARP-1) following AEG-1 transduction in the SN of healthy brains. *p = 0.005 vs. CON (one-way ANOVA with Tukey’s post hoc test; n = 4 for each group) 9,10,12 nigrostriatal DA system and, consequently, the potential axonal regeneration in damaged DA neurons .As of AEG-1 overexpression as a therapeutic approach for shown in the experimental schematic (Fig. 5a), treatment PD, we examined the effects of AEG-1 overexpression with 6-OHDA alone induced significant reductions in the following post treatment with AAV-Rheb(S16H) on the motor performance, which was measured using the open- functional recovery of nigral DA neurons and induction of field test (Fig. 5b, c) and rotarod test (Fig. 5d), compared Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 6 of 15 Fig. 3 Anti-apoptotic effects of AEG-1 transduction in DA neurons on 6-OHDA neurotoxicity. a Western blot analyses show a significant increase in the levels of caspase-3, c-caspase-3, and c-PARP1 in the postmortem tissues of patients with PD compared with CON. p = 0.014 for && &&& caspase-3, p = 0.019 for c-caspase-3, and p = 0.009 for c-PARP-1, vs. CON (t-test; n = 4 for each group). b Experimental schematic for Fig. 3c, d. c Representative double immunofluorescence labeling for TH (red) and c-caspase-3 (green) or TH and c-PARP-1 (green) in the mouse SN. AEG-1 upregulation induces reductions in the levels of expression of both c-caspase-3 and c-PARP-1 in TH-positive DA neurons in the SN with 6-OHDA neurotoxicity. Scale bar, 20 μm. d Western blot analyses of the levels of c-caspase-3 and c-PARP-1 in the SN 2 days after 6-OHDA treatment. *p = # ## ### 0.029, **p = 0.025, p = 0.032, and p = 0.016 vs. CON; ***p = 0.009 and p = 0.002 vs. 6-OHDA alone (one-way ANOVA with Tukey’s post hoc test; n = 4 for each group) to that in noninjected controls (*p < 0.001 vs. intact con- AAV-Rheb(S16H)]. Consistent with these results, the trols). Similar to the results described in our previous immunohistochemical staining of TH demonstrated that report , Rheb(S16H) overexpression rescued the motor Rheb(S16H) overexpression following 6-OHDA adminis- impairments induced by 6-OHDA neurotoxicity com- tration induced axonal regeneration in damaged DA # # 9,10 pared to mice injected with 6-OHDA alone ( p = 0.008 neurons (Fig. 5f, h; p < 0.001 vs. 6-OHDA alone) , and ## and p = 0.01 vs. 6-OHDA alone). In particular, we found Rheb(S16H) overexpression in the presence of increased that AEG-1-overexpressing mice that were injected with levels of AEG-1 significantly restored the density of DA AAV-Rheb(S16H) following 6-OHDA injections demon- fibers in the STR compared to the effects in the absence of strated significant reversals of the motor impairments that AEG-1 [Fig. 5f, h; p = 0.003 vs. 6-OHDA + AAV-Rheb were caused by 6-OHDA neurotoxicity compared with (S16H)]. Moreover, the depleted levels of striatal dopa- Rheb(S16H)-treated mice without virally overexpressed mine following 6-OHDA administration, which were § §§ AEG-1 [Fig. 5b–d; p = 0.007 and p = 0.007 for open- measured using HPLC, were greatly restored following field test and rotarod test, respectively, vs. 6-OHDA + Rheb(S16H) overexpression in the presence of virally Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 7 of 15 Fig. 4 (See legend on next page.) Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 8 of 15 (see figure on previous page) Fig. 4 Upregulation of neuronal AEG-1 protects DA neurons from 6-OHDA-induced neurotoxicity. a Experimental schematic for b–d. b Representative coronal sections of the SN stained with anti-TH at 7 days post-lesion. Representative high-power micrographs are shown in the inset to aid visualization. Scale bars, 200 μm (black) and 50 μm (white). The quantitative analysis shows a population of preserved TH-positive neurons in the SN. *p < 0.001 and **p = 0.023 vs. CON; p= 0.026 vs. 6-OHDA (n = 4 for each group). c Neuroprotective effects of AEG-1 are not observed on striatal TH-positive fibers. Scale bar, 1000 μm. The histogram shows the optical densities of the striatal TH-positive fibers. *p < 0.001 vs. CON (n = 4 for each group). d Western blot analysis of the levels of TH with 6-OHDA-induced neurotoxicity in the nigrostriatal DA system. *p < 0.001 for SN and **p < 0.001 for STR vs. CON; p = 0.002 vs. 6-OHDA alone (n = 4 for each group). e, f Experimental schematic and representative coronal sections show that glial AEG-1 upregulation by Ad transduction does not protect the nigrostriatal DA system in the 6-OHDA-treated mouse model of PD. Scale bars, 200 μm for SN and 1000 μm for STR. *p < 0.001 vs. CON (n = 4 for each group). One-way ANOVA with Tukey’s post hoc test was used in b–f overexpressed AEG-1 compared to the levels in its signaling pathway in in vivo and in vitro models of ALS . absence [Fig. 5i; p = 0.024 vs. 6-OHDA + AAV-Rheb Although the role of AEG-1 in the pathogenesis of (S16H)]. Similarly, the levels of the metabolites of dopa- PD was unknown, these results suggest that AEG-1 mine, including DOPAC and HVA, were restored by Rheb might be critical for the survival of DA neurons in the SN (S16H) overexpression, and the effects were more obvious of patients with PD. Here we observed decreased in the presence of overexpressed AEG-1 than in its expression of AEG-1 in damaged DA neurons in absence (Supplementary Figure S6). the adult brain in an immunohistochemical analysis Rheb(S16H) administration did not affect the levels of of the SN of patients with PD and 6-OHDA-treated expression of AEG-1 in the SN of the mice (Supplemen- mice (Fig. 1b, d). However, no significant reductions in tary Figure S7), which suggested that the neurorestorative AEG-1 were observed in the hippocampus of patients effects from this administration might be independent of with AD compared to age-matched controls (Fig. 1g). AEG-1 expression and that its upregulation might activate Thus, these observations suggested that the decreased the Akt/mTORC1 signaling pathway as a supplementary levels of AEG-1 were a specific event that occurred in mechanism in the presence of AEG-1. These data sug- damaged DA neurons and AEG-1 downregulation and the gested that AEG-1, which is significantly reduced in the loss of nigral DA neurons in PD might be clinically SN of patients with PD, is an important endogenous correlated. factor that protects nigral DA neurons from neurotoxicity Here, AAV-mediated overexpression of AEG-1 in DA and that this protection by the anti-apoptotic effects of neurons decreased the levels of apoptotic markers, neuronal AEG-1 enhances the restoration of the disrupted including cleaved caspase-3 and cleaved PARP-1, follow- 15,16 nigrostriatal DA system (Fig. 6), as shown by the effects of ing 6-OHDA administration (Fig. 3c, d) , resulting in Rheb(S16H) administration in the neurotoxin model of neuroprotection in the SN (Fig. 4b, d). The AEG-1- PD (Fig. 5). induced anti-apoptotic effects were confirmed by western blot analyses of the Bcl-2/Bax ratio (Supplementary Fig- Discussion ure S1). An increase in the levels of apoptotic signaling Under physiological conditions, apoptosis is an essential molecules, which was similar to the results obtained in the homeostatic mechanism that maintains the cell popula- mouse model (Fig. 3c, d), were also observed in the SN of 24,25 tion in healthy tissue and protects cells from injury . patients with PD (Fig. 3a). These results suggested that However, aberrant apoptosis, which is one of the neuro- AEG-1 plays a role as a negative regulator of apoptosis in toxic events in the adult brain, may inevitably be related to adult DA neurons and the sustained levels of neuronal 6,8,26 neurodegenerative diseases, such as PD and AD . AEG-1 following neurotoxic events may confer neuro- Moreover, consistent with the results of studies on protection to nigral DA neurons in vivo. Moreover, the 6,8 patients with PD , increases in the levels of apoptotic overexpression of microglial AEG-1 in the 6-OHDA- markers, such as cleaved caspase-3 and cleaved PARP-1, based mouse model of PD did not confer neuroprotection 15,16 are observed in in vivo and in vitro models of PD .As (Fig. 4e, f). Therefore, consistent with the effects of reported previously, the inhibition of apoptotic pathways reduced AEG-1 expression in motor neurons in an ALS 7,27–30 5 protects DA neurons against neurotoxin treatment . mouse model , our results showed that the sustained Thus, these results suggest that the expression and increased levels of neuronal AEG-1 were important for maintenance of endogenous anti-apoptotic factors are attenuating the vulnerability of nigral DA neurons in the beneficial for the survival of nigral DA neurons in the SN of adult brain. adult brain. Despite the significant anti-apoptotic effects of AEG-1, The downregulation of AEG-1 contributes to the its overexpression was not sufficient to protect the whole apoptosis of motor neurons by inhibiting the PI3K/Akt nigrostriatal DA projection against 6-OHDA-induced Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 9 of 15 Fig. 5 (See legend on next page.) Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 10 of 15 (see figure on previous page) Fig. 5 Synergistic effects of AEG-1 and Rheb(S16H) in the disrupted nigrostriatal DA system in vivo.a Experimental schematic. b, c Total ## ### § distance traveled for 5 min and velocity in the open-field test. *p < 0.001 vs. intact controls; p = 0.010 and p < 0.001 vs. 6-OHDA alone; p = 0.007 vs. 6-OHDA + AAV-Rheb(S16H) group (one-way ANOVA with Tukey’s post hoc test; n = 5 for each group). p = 0.008 vs. 6-OHDA alone (t-test). d # ## § Motor deficits measured by using the rotarod test. *p < 0.001 and **p = 0.002 vs. intact controls; p < 0.001 and p = 0.002 vs. 6-OHDA alone; p = 0.006 vs. 6-OHDA+ AAV-Rheb(S16H) group (n = 5 for each group). e, f Representative coronal SN (scale bar, 200 μm) and STR (scale bar, 1000 μm) sections stained with anti-TH at 11 weeks post-lesion. g Quantitative analysis showing the population of preserved TH-positive neurons in the SN. *p < 0.001 vs. intact controls; p = 0.006 vs. 6-OHDA alone (n = 3 for each group). h The histogram shows the optical densities of the striatal TH-positive # ## § fibers. *p < 0.001 vs. intact controls; p < 0.001 and p = 0.001 vs. 6-OHDA alone; p = 0.003 vs. 6-OHDA+ AAV-Rheb(S16H) group (n = 3 for each group). i The levels of striatal dopamine, which were measured with high-performance liquid chromatography, were quantitatively expressed as a # ## ### § percentage of intact control. *p < 0.001 vs. intact controls; p < 0.001, p = 0.024 and p = 0.028 vs. 6-OHDA alone; p = 0.024 vs. 6-OHDA+ AAV- Rheb(S16H) group [n = 4 for AAV-GFP+ 6-OHDA + AAV-Rheb(S16H) group; n = 5 for the other groups]. One-way ANOVA with Tukey’s post hoc test was used in d and g–i One explanation for the observation that AEG-1 over- expression in DA neurons was insufficient to protect the DA system seems related to the mTORC1 pathway. The mTOR kinase, which plays a central role in the integration of responses to various environmental conditions, exists 10,31,32 in two complexes, mTORC1 and mTORC2 . mTORC1 is an important mediator of Akt. mTORC2 can activate Akt, which in turn can act on mTORC1. Acti- vation of the Akt/mTOR signaling pathway enhances the activity of cell survival pathways under various conditions, including trophic factor withdrawal, ischemic shock, and 9–12,32 oxidative stress . Moreover, recent reports have showed that the activation of neuronal mTORC1, which is a key biomolecule for neurotrophic support, by either the delivery of a specific gene or the direct administration of trophic factors induces protective effects against neuro- 11,33 degeneration in animal models of PD . We recently demonstrated that the Rheb(S16H) delivery-induced activation of mTORC1 in DA neurons protects and reconstructs the damaged nigrostriatal DA projections in a mouse model of PD, which suggested that the activation of the Akt/mTORC1 signaling pathway might be a pro- mising therapeutic strategy for the functional recovery of 9,10 DA neurons . Here, however, our results indicated that AEG-1 upregulation in nigral DA neurons did not enhance Akt/mTORC1 signaling in healthy mice (Sup- Fig. 6 Schematic of the importance of neuronal AEG-1 preservation in the nigrostriatal DA system. Following neurotoxic plementary Figure S4). Moreover, the activation events in the nigrostriatal DA system, AEG-1 transduction in DA of the Akt/mTORC1 signaling pathway suppressed the neurons prevents the apoptotic cell death of the DA neurons (but not initiation of autophagy and prevented the aberrant accu- their functional properties). Meanwhile, decreases in the levels of mulation of autophagic components, which might inhibit neuronal AEG-1 and the consequent loss of DA neurons were normal lysosomal degradation, such as the removal of observed. Moreover, AEG-1 transduction in DA neurons can intensify the neurorestorative effects by synergizing with the administration of expended macromolecules and organelles, in the nigros- 12,19 therapeutic agents for axonal regeneration and the inhibition of triatal DA system . Although autophagy is physiolo- abnormal activation of the autophagy–lysosomal pathway. Thus, these gically important for preserving cellular homeostasis and results suggest the importance of AEG-1 preservation in potential inducing protective effects, such as the suppression of therapeutic strategies against PD 34–37 apoptosis and axonal degeneration , its aberrant activity promotes neurodegeneration in the SN of patients neurotoxicity (Fig. 4c, d and Supplementary Figure S2). with PD . Therefore, these observations suggest that These observations indicated limitations in the protective autophagic stress that is caused by the aberrant accu- effects of AEG-1 against 6-OHDA-induced neurotoxicity. mulation of autophagic components might be one of the Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 11 of 15 critical mechanisms that induces the loss of DA neurons it did not affect the aberrant accumulation of autophagic in PD. components and Akt/mTORC1 activity following AEG-1 is known to induce autophagy, which results in 6-OHDA administration, which could contribute to the the survival of cancerous cells under metabolic stress and neurotoxic effects on the nigrostriatal DA system. To apoptosis resistance, and these results may underlie its overcome this limitation of AEG-1, we further transduced considerable cancer-promoting properties . However, the Akt/mTORC1 activator Rheb(S16H) into AEG-1- overexpressing DA neurons. Surprisingly, the synergistic our results indicated that AEG-1 overexpression in DA neurons had no effect on the levels of LC3-II in healthy effects of the two factors restored the nigrostriatal DA brains (Supplementary Figure S4), and on the aberrant system that was disrupted by 6-OHDA administration, accumulation of LC3 and p62, which are critical for ALP and the effects were more obvious in the presence of 19,22,40 activation following 6-OHDA administration AEG-1 than in its absence (Figs. 5 and 6). Therefore, we (Supplementary Figure S5b). Moreover, the decrease in concluded that AEG-1 was an important endogenous the activity of mTORC1 following 6-OHDA neurotoxicity factor for protecting nigral DA neurons from aberrant was not inhibited by the presence of AEG-1 in DA neu- apoptotic signaling pathway in the adult brain and that rons in vivo (Supplementary Figure S5b). Therefore, to the maintenance of increased levels of AEG-1 in nigral potentiate the beneficial effects of AEG-1 in nigral DA DA neurons in patients with PD, in combination with neurons, it may be necessary to activate the Akt/ therapeutic agents, such as an Akt/mTORC1 signaling mTORC1 signaling pathway, which suppresses the aber- activator, may be a highly promising therapeutic strategy rant accumulation of autophagic components as a sup- to maximize the functional recovery of the damaged plementary protective mechanism in the presence of nigrostriatal DA system (Fig. 6). AEG-1 and consequently results in enhanced neuropro- tection of the nigrostriatal DA projection in the adult Materials and methods brain. Ethics statement The overexpression of Rheb(S16H), which is a con- All animal experiments were performed in accordance 9,10 stitutively active form of Rheb and activates mTORC1 , with the approved animal protocols and guidelines suppresses the induction of the abnormal autophagy sig- established by the Animal Care Committee at Kyungpook naling pathway by 6-OHDA treatment . To further National University (Number: KNU 2016-42). Experi- examine if AEG-1 overexpression strengthened the neu- ments involving human tissue were approved by the roprotective and neurorestorative effects of therapeutic Bioethics Committee, Institutional Review Board Kyung- agents, such as Rheb(S16H), AAV-Rheb(S16H) was pook National University Industry Foundation (IRB injected into the SN 3 weeks post-lesion, which is when Number: KNU 2014-0007 and 2016-0011). the maximum neurodegeneration is observed after 6- 9,10 OHDA treatment , in the absence or presence of virally Human brain tissue overexpressed AEG-1 (Fig. 5a). Our results demonstrated Frozen and paraffin-fixed brain tissues were obtained that Rheb(S16H) upregulation in the presence of from the Victorian Brain Bank Network (VBBN), sup- increased levels of AEG-1 induced synergistic neuror- ported by the Florey Institute of Neuroscience and Mental estorative effects, such as restored motor functions, in the Health, The Alfred, and the Victorian Forensic Institute of nigrostriatal DA system disrupted by 6-OHDA neuro- Medicine, and funded by Australia’s National Health & toxicity (Fig. 5b–i). As shown in Supplementary Figure S7, Medical Research Council and Parkinson’s Victoria. Fro- the Rheb(S16H) transduction of DA neurons did not zen and paraffin-fixed brain tissues were used in quanti- affect the levels of AEG-1 in the SN of the mice, sug- tatively analyzing the level of proteins and in observing gesting that the Rheb(S16H)-induced neurorestoration the expression pattern of target molecule, respectively. was independent of AEG-1 upregulation and that the AEG-1-induced neuroprotection against 6-OHDA neu- Materials rotoxicity augmented the beneficial effects of Rheb(S16H) Materials were purchased from the following compa- in the lesioned nigrostriatal DA system in vivo (Fig. 5). nies: 6-OHDA (Sigma, St Louis, MO), desipramine In conclusion, our findings suggested that AEG-1 (Sigma), L-ascorbic acid (Sigma), rabbit anti-TH (Pel- functioned as an anti-apoptotic factor in nigral DA neu- Freez, Brown Deer, WI), mouse anti-TH (R&D Systems, rons of the adult brain and the decrease in AEG-1 might Minneapolis, MN), rabbit anti-Iba1 (Wako Pure Chemical be involved in the loss of DA neurons, which is one of the Industries, Osaka, Japan), rabbit anti-GFAP (Millipore, key pathological features in PD. However, the over- Billerica, MA), rabbit anti-AEG-1 (Invitrogen, Camarillo, expression of AEG-1 in DA neurons was not sufficient to CA), rabbit anti-GFP (Millipore), mouse anti-HA (Cell protect the whole nigrostriatal DA projection in the ani- Signaling, Beverly, MA), rabbit anti-HA (Cell Signaling), mal model of PD owing to its limited protective effects as rabbit anti-FLAG (Sigma), rabbit anti-caspase-3 (Cell Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 12 of 15 Signaling), rabbit anti-cleaved caspase-3 (Cell Signaling), −0.35 cm, ML: −0.11 cm, DV: −0.37 cm, relative to rabbit anti-PARP-1 (Cell Signaling), rabbit anti-cleaved bregma) using a 30-gauge Hamilton syringe attached to PARP-1 (Cell Signaling), rabbit anti-LC3B (Cell Signal- an automated microinjector . Viral vector suspension in ing), rabbit anti-4E-BP1 (Cell Signaling), rabbit anti-p-4E- a volume of 2.0 μl was injected at a rate of 0.1 μl/min over BP1 (Cell Signaling), mouse anti-NeuN (Millipore), rabbit 20 min. After injection, the needle was left in place for an anti-Akt (Cell Signaling), rabbit anti-p-Akt (Cell Signal- additional 5 min before being slowly retracted. ing), rabbit anti-β-actin (Cell Signaling), rabbit anti-α- tubulin (Cell Signaling), mouse anti-Bcl-2 (Santa Cruz Intrastriatal 6-OHDA injection Biotechnology, Santa Cruz, CA), mouse anti-Bax (Santa The intrastriatal 6-OHDA model was induced as pre- 9,10 Cruz Biotechnology), rabbit anti-p62/SQSTM1 (Sigma), viously described . Mice were intraperitoneally injected biotinylated anti-rabbit IgG (Vector laboratories, Burlin- with desipramine (25 mg/kg in 0.9% NaCl), and then game, CA), Texas Red-conjugated anti-rabbit/mouse IgG anesthetized with chloral hydrate. Anesthetized mice were (Vector Laboratories), fluorescein (FITC)-conjugated placed in a stereotaxic frame, and a solution of 6-OHDA anti-mouse IgG (Vector Laboratories), FITC-conjugated (5 mg/ml in 0.9% NaCl/0.02% ascorbic acid), with a final anti-rabbit IgG (Jackson ImmunoResearch Laboratories, volume of 3.0 μl was injected by Hamilton syringe at a rate Bar Harbor, ME), horseradish peroxidase (HRP)-con- of 0.5 μl/min. The injection was performed into the right jugated anti-rabbit IgG (Enzo Life Sciences, Farmingdale, STR at coordinates (AP: +0.09 cm; ML: −0.22 cm; DV: NY) and HRP-conjugated anti-mouse IgG (Thermo −0.25 cm, relative to bregma) . The needle was with- Fisher Scientific, Rockford, IL). drawn slowly after 5 min. Animals were sacrificed and analyzed at the indicated time points for each 9,10 Production of viral vectors experiment . The two types of viral vectors used in the present study were AAV serotype 1 and Ad serotype 5. Viral vectors Behavioral tests were produced as described previously, with some mod- Open-field test 10,17,41,42 ifications . Ad viral vectors were supplied by S.G. The open-field test was performed as described pre- Lee. Briefly, for the production of AAV viral vectors viously, with some modifications . Briefly, 11 weeks after carrying AEG-1 with a HA-encoding sequence at the the 6-OHDA injection, mice were placed individually in 3ʹ-end (AEG-1-HA), AEG-1 cDNA tagged with HA was the corner of a test chamber (40 × 40 × 40 cm) enclosed amplified from the mammalian expression plasmid of with white acrylic walls. After a 1 min adaptation period, AEG-1 (pcDNA3.1-AEG-1-HA), as previously descri- animal behaviors such as the total distance traveled (in 17,42 bed . AEG-1-HA obtained from pcDNA3.1-AEG-1- cm) and velocity (in cm/sec) were recorded for 5 min HA was cloned into an AAV packaging construct that using a video camera. The change in locomotor activity utilizes the chicken β-actin promotor and contains a 3′ was analyzed offline by video-tracking software (SMART, WPRE . Constitutively activated Rheb was also cloned Panlab, Barcelona, Spain). The test chamber was cleaned into the same AAV packaging construct . All nucleotide between trials with 70% ethyl alcohol. To minimize stress sequences in the AAV packaging construct were con- levels, tests were performed under conditions of low firmed before viral vector production. AAVs were pro- illumination. duced at the University of North Carolina Vector Core. The genomic titer of AAV-AEG-1 and AAV-Rheb(S16H) Rotarod test 12 12 were 9.4 × 10 viral genomes/ml and 3.6 × 10 viral The rotarod test was performed at 11 weeks post-lesion, genomes/ml, respectively. Enhanced GFP, used as a con- using a previously described procedure with some mod- trol, was subcloned into the same AAV viral backbone, ifications . Before 6-OHDA treatment, all mice were pre- and viral stock was produced at a titer of 2.0 × 10 viral trained on the rotarod apparatus (3 cm rod diameter; genomes/ml. Genomic titers of both Ad-AEG-1 and Ad- Scitech Inc., Seoul, Korea) at 10 revolutions per min (rpm) null viral stocks were 2.0 × 10 infectious unit/ml. for 10 min, and the training was performed for 3 con- secutive days. Eleven weeks after the 6-OHDA injection, Intranigral AAV and Ad injection performance on the rod was evaluated at a constant Adult (8- to 10-week-old) male C57BL/6 mice were acceleration rate of 4–40 rpm in 300 sec. Two consecutive obtained from Daehan Biolink (Eumseong, Korea). As trials were performed at 60 min intervals. 9,10 previously described, with some modifications , mice were anesthetized with chloral hydrate solution and Immunohistochemical staining placed in a stereotaxic frame (Kopf Instruments, Tujunga, Postmortem brain tissues were processed for immuno- CA) with a mouse adapter. Each mouse received a uni- histochemistry as described previously . Briefly, the lateral injection of AAV or Ad into the right SN (AP: human SN sections were deparaffinized and subjected to Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 13 of 15 citrate-based antigen retrieval, and then washed in cold at 4 °C with the following primary antibodies: rabbit anti- PBS and blocked with blocking solution. The sections TH (1:2000), rabbit anti-AEG-1 (1:1000), mouse anti- were incubated with primary antibodies against AEG-1 NeuN (1:500), rabbit anti-GFP (1:500), rabbit anti-HA (1:500) at 4 °C overnight, and then incubated with bioti- (1:1000), rabbit anti-FLAG (1:3000), rabbit anti-caspase-3 nylated secondary antibodies for 1 h at room temperature, (1:1000), rabbit anti-cleaved caspase-3 (1:1000), rabbit followed by addition of the avidin-biotin reagent (Vec- anti-PARP-1 (1:1000), rabbit anti-cleaved PARP-1 (1:1000), mouse anti-Bcl-2 (1:1000), mouse anti-Bax tastain ABC kit, Vector Laboratories) for 1 h at room temperature. The SN sections were visualized using a (1:1000), rabbit anti-LC3B (1:1000), rabbit anti-p62/ 3,3′-diaminobenzidine (DAB; Sigma) peroxidase substrate SQSTM1 (1:1000), rabbit anti-Akt (1:1000), rabbit anti-p- solution [0.05% DAB, 0.05% Cobalt Chloride (Sigma), Akt (1:2000), rabbit anti-4E-BP1 (1:1000), and rabbit anti- 0.05% Nickel Ammonium Sulfate (Sigma) and 0.015% p-4E-BP1 (1:1000). Subsequently, the membranes were H O in PBS, pH 7.2]. Each section was covered with a incubated with secondary antibodies for 1 h at room 2 2 thin glass coverslip and analyzed under a bright-field temperature, and the bands were finally detected using microscope (Carl Zeiss, Oberkochen, Germany). Western-blot detection reagents (Thermo Fisher Scien- 9,10,44 As previously described , mice were transcardially tific, Rockford, IL). For quantitative analyses, the density perfused and fixed, and the brains were dissected out, of each band was measured using a Computer Imaging frozen, and cut into 30-μm-thick coronal sections using a Device and accompanying software (Fuji Film, Tokyo, cryostat microtome (Thermo Fisher Scientific). Briefly, Japan), and the levels were quantitatively expressed as the the brain sections were washed in PBS and blocked with density normalized to the housekeeping protein band for blocking buffer, and then incubated at 4 °C for 2 days with each sample. the following primary antibodies: rabbit anti-TH (1:2000), rabbit anti-AEG-1 (1:500), rabbit anti-Iba1 (1:2000), rab- Stereological estimation 10,44 bit anti-GFAP (1:1000), mouse anti-HA (1:1000), rabbit As previously described , the total number of TH- anti-cleaved caspase-3 (1:400), rabbit anti-cleaved PARP-1 positive neurons was counted in the various animal (1:400) and mouse anti-TH (1:500). After incubation, the groups using the optical fractionator method. Counting of brain sections were incubated with biotinylated secondary TH-positive neurons in the SN was performed on a antibodies, followed by addition of the avidin-biotin bright-field microscope (Olympus Optical, BX51, Tokyo, reagent (Vectastain ABC kit) for 1 h at room tempera- Japan) using Stereo Investigator software (MBF ture, or incubated with fluorescence-conjugated second- Bioscience, Williston, VT). This unbiased stereological ary antibodies for 1 h. The signal following treatment with method of cell counting is not affected by either the avidin-biotin reagent was detected by incubating the reference volume (SN pars compacta) or the size of the sections in 0.5 mg/ml DAB (Sigma) in 0.1 M PB con- counted elements (neurons). taining 0.003% H O . The sections incubated with 2 2 fluorescence-conjugated secondary antibodies were Quantitative determination of striatal TH washed in 0.1 M PBS. The stained sections were mounted immunoperoxidase staining on gelatin-coated slides and analyzed under a light or Densitometric analysis of the mouse STR was carried 10,44 fluorescence microscope (Carl Zeiss). out as previously described . Briefly, an average of 6 coronal sections of the STR that gathered according to the Western blot analysis bregma of the brain atlas was imaged at a ×1.25 mag- Animal and human tissue samples were prepared for nification. The density of striatal TH-positive fibers was 10,44 western blot analysis as previously described . Briefly, measured using the Science Lab 2001 Image Gauge animal SN tissues were removed and sliced using a brain (Fujifilm, Tokyo, Japan). To control for variations in matrix (Roboz Surgical Instrument Co., Gaithersburg, background illumination, the density of the corpus cal- MD). Animal or human tissue samples were homogenized losum was subtracted from the density of STR for each and centrifuged at 4 °C for 20 min at 14,000 × g; the section. The density in both the contralateral and ipsi- supernatant was transferred to a fresh tube and the con- lateral sides was expressed by comparing with the average centration was determined using a bicinchoninic acid density of TH-positive fiber innervating the contralateral assay (BCA) kit (Bio-Rad Laboratories, Hercules, CA). side. Aliquots containing 50 μg of protein were electrophoresed on a sodium dodecyl sulfate (SDS)/polyacrylamide gel Measurement of dopamine and its metabolites in the STR (Bio-Rad Laboratories) and transferred to polyvinylidene As previously described , the levels of striatal dopa- difluoride (PVDF) membranes (Millipore, Billerica, MA) mine and its metabolites were measured by high perfor- using an electrophoretic transfer system (Bio-Rad mance liquid chromatography (HPLC, 1260 Infinity Laboratories) The membranes were incubated overnight system, Agilent Technologies, Santa Clara, CA) using an Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 14 of 15 ESA Coulochem III electrochemical detector. Briefly, the designed the viral constructs; S.R.K., S.-G.L., P.B.F., and N.K. contributed the viral vectors; C.M. contributed the human brain samples; Y.-S.O., K.J.L., Y.S.C., K.S.A., B. brain tissues were homogenized and centrifuged, and the K.J., D.W.K., J.M.L., S.G.L., and S.R.K. supervised the analysis of the data obtained supernatants were injected using an autosampler at 4 °C with the human brain samples; S.R.K. supervised the whole project and wrote (Waters 717 plus autosampler) and eluted through a μ the paper. All of the authors contributed to the data analysis and preparation of the manuscript. Sunfire C18 column (4.6 × 100 mm × 5 μm; Waters Cor- poration, Milford, MA) with a mobile phase of MDTM/ Conflict of interest acetonitrile (90:10). The peaks of dopamine and its The authors declare that they have no conflict of interest. metabolites were analyzed and integrated using a Chem- Station software (Agilent Technologies, Santa Clara, CA), Publisher's note and all samples were normalized for protein content as Springer Nature remains neutral with regard to jurisdictional claims in spectrophotometrically determined using the Pierce BCA published maps and institutional affiliations. protein assay kit (Thermo Scientific, Waltham, MA). Supplementary Information accompanies this paper at (https://doi.org/ 10.1038/s41419-018-0491-3). Statistical analysis Differences between two groups were analyzed using t- Received: 29 January 2018 Accepted: 14 February 2018 test. Multiple comparisons between groups were per- formed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Behavioral test was analyzed using one-way ANOVA and t-tests. All values References represent the mean ± standard error of the mean (SEM), 1. Su,Z.Z.etal. Identification and cloning of human astrocyte genes displaying and statistical analyses were performed using SigmaStat elevated expression after infection with HIV-1 or exposure to HIV-1 envelope glycoprotein by rapid subtraction hybridization, RaSH. Oncogene 21, software (Systat Software, San Leandro, CA). 3592–3602 (2002). 2. Su,Z.Z.etal. Customized rapidsubtraction hybridization (RaSH) gene Acknowledgements microarrays identify overlapping expression changes in human fetal astrocytes This study was supported by grants from the Korea Healthcare Technology resulting from human immunodeficiency virus-1 infection or tumor necrosis R&D Project, Ministry of Health & Welfare (HI15C1928), and the National factor-alpha treatment. Gene 306,67–78 (2003). Research Foundation of Korea (NRF-2013R1A2A2A01069099, NRF- 3. Lee, S. G.,Kang, D. C.,DeSalle,R., Sarkar,D.&Fisher,P.B.AEG-1/MTDH/LYRIC, 2015R1A4A1042399, NRF-2016M3C7A1905074, and NRF-2017R1A2B4002675). the beginning: initial cloning, structure, expression profile, and regulation of expression. Adv. Cancer Res. 120,1–38 (2013). Author details 4. Emdad, L. et al. AEG-1/MTDH/LYRIC: a promiscuous protein partner critical in School of Life Sciences, BK21 Plus KNU Creative BioResearch Group, Institute cancer, obesity, and CNS diseases. Adv. Cancer Res. 131,97–132 (2016). of Life Science & Biotechnology, Kyungpook National University, Daegu 41566, 5. Yin, X. et al. Downregulated AEG-1 together with inhibited PI3K/Akt pathway Republic of Korea. Department of Neural Development and Disease, is associated with reduced viability of motor neurons in an ALS model. Mol. Department of Structure & Function of Neural Network, Korea Brain Research Cell. Neurosci. 68, 303–313 (2015). Institute, Daegu 41062, Republic of Korea. Department of Science in Korean 6. Hartmann, A. et al. Caspase-3: a vulnerability factor and final effector in Medicine, Graduate School, Kyung Hee University, Seoul 02447, Republic of apoptotic death of dopaminergic neurons in Parkinson’sdisease. Proc. Natl Korea. Department of Brain-Cognitive Science, Daegu-Gyeongbuk Institute of Acad. Sci. USA 97,2875–2880 (2000). Science and Technology, Daegu 42988, Republic of Korea. Predictive Model 7. Kanthasamy, A. G. et al. A novel peptide inhibitor targeted to caspase-3 Research Center, Korea Institute of Toxicology, Korea Research Institute of cleavage site of a proapoptotic kinase protein kinase C delta (PKCdelta) Chemical Technology, Daejeon 34114, Republic of Korea. Department of protects against dopaminergic neuronal degeneration in Parkinson’sdisease Biochemisry and Molecular Biology, Department of Neuroscience Graduate models. Free. Radic. Biol. Med. 41, 1578–1589 (2006). School, School of Medicine, Kyung Hee University, Seoul 02447, Republic of 8. Radi,E., Formichi, P., Battisti,C.& Federico, A.Apoptosis andoxidative stress in Korea. Department of Anatomy, Brain Research Institute, School of Medicine, neurodegenerative diseases. J. Alzheimers Dis. 42,S125–S152 (2014). Chungnam National University, Daejeon 34134, Republic of Korea. Victorian 9. Kim, S. R. et al. Dopaminergic pathway reconstruction by Akt/Rheb-induced Brain Bank Network, Florey Institute of Neuroscience and Mental Health, The axon regeneration. Ann. Neurol. 70, 110–120 (2011). University of Melbourne, Melbourne, VIC 3004, Australia. Department of 10. Kim,S. R., Kareva, T., Yarygina, O., Kholodilov, N. &Burke,R.E.AAV transduction Anatomical Pathology, Alfred Hospital, Melbourne, VIC 3004, Australia. of dopamine neurons with constitutively active Rheb protects from Department of Human and Molecular Genetics, VCU Institute of Molecular neurodegeneration and mediates axon regrowth. Mol. Ther. 20,275–286 Medicine, VCU Massey Cancer Center, School of Medicine, Virginia (2012). Commonwealth University, Richmond, VA 23298, USA. Department of 11. Nam, J. H. et al. Induction of GDNF and BDNF by hRheb(S16H) transduction of Neurology, Columbia University, New York, NY 10032, USA. Department of SNpc neurons: neuroprotective mechanisms of hRheb(S16H) in a model of Biochemistry and Cell Biology, Cell and Matrix Research Institute, BK21 Plus Parkinson’sdisease. Mol. Neurobiol. 51,487–499 (2015). KNU Biomedical Convergence Program, School of Medicine, Kyungpook 12. Cheng, H. C. et al. Akt suppresses retrograde degeneration of dopaminergic National University, Daegu 41944, Republic of Korea. Department of Food axons by inhibition of macroautophagy. J. Neurosci. 31,2125–2135 (2011). Science and Nutrition, Pukyong National University, Busan 48513, Republic of 13. Kim, B. W. et al. Pathogenic upregulation of glial lipocalin-2 in the Parkinsonian Korea. KHU-KIST Department of Converging Science and Technology, Kyung dopaminergic system. J. Neurosci. 36,5608–5622 (2016). Hee University, Seoul 02447, Republic of Korea. Brain Science and 14. Kim,H.D., Jeong, K. H.,Jung, U. J. & Kim,S.R.Naringintreatment induces Engineering Institute, Kyungpook National University, Daegu 41944, Republic neuroprotective effects in a mouse model of Parkinson’s disease in vivo, but of Korea not enough to restore the lesioned dopaminergic system. J. Nutr. Biochem. 28, 140–146 (2016). Author contributions 15. Lev, N., Melamed, E. & Offen, D. Apoptosis and Parkinson’sdisease. Prog. E.L., H.-J.K., M.C., S.-G.L., and S.R.K. conceived and designed the experiments; E. Neuropsychopharmacol. Biol. Psychiatry 27,245–250 (2003). L., H.-J.K., M.C., U.J.J., S.K., J.-Y.U., W.-H.S., J.Y.J., S.G.L., and S.R.K. conducted the 16. Hernandez-Baltazar, D., Mendoza-Garrido,M.E.& Martinez-Fong, D. Activation experiments, analyzed the data, and generated the figures; S.R.K. and S.-G.L. of GSK-3beta and caspase-3 occurs in Nigral dopamine neurons during the Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 15 of 15 development of apoptosis activated by a striatal injection of 6- 31. Kim,S.R.Mammaliantarget ofrapamycincomplex 1asaninducer of neu- hydroxydopamine. PLoS ONE 8, e70951 (2013). rotrophic factors in dopaminergic neurons. Neural Regen. Res. 9,2036–2037 17. Lee, S.G., Su,Z. Z., Emdad, L.,Sarkar, D. &Fisher, P. B. Astrocyteelevatedgene-1 (2014). (AEG-1) is a target gene of oncogenic Ha-ras requiring phosphatidylinositol 3- 32. Jeon, M. T. et al.InvivoAAV1 transductionwithhRheb(S16H) protects hip- kinase and c-Myc. Proc. Natl Acad. Sci. USA 103, 17390–17395 (2006). pocampal neurons by BDNF production. Mol. Ther. 23,445–455 (2015). 18. Lee,S.G.etal. Astrocyteelevatedgene-1 activates cell survival pathways 33. Allen, S. J.,Watson, J. J.,Shoemark, D. K.,Barua,N.U.& Patel, N. K. GDNF,NGF through PI3K-Akt signaling. Oncogene 27, 1114–1121 (2008). and BDNF as therapeutic options for neurodegeneration. Pharmacol. Ther. 19. Burke,R.E.&O’Malley, K. Axon degeneration in Parkinson’sdisease. Exp. Neurol. 138,155–175 (2013). 246,72–83 (2013). 34. Boya, P. et al. Inhibition of macroautophagy triggers apoptosis. Mol. Cell. Biol. 20. Liu, K. et al. PTEN deletion enhances the regenerative ability of adult corti- 25,1025–1040 (2005). cospinal neurons. Nat. Neurosci. 13, 1075–1081 (2010). 35. Komatsu, M. et al. Essential role for autophagy protein Atg7 in the main- 21. Miao, L. et al. mTORC1 is necessary but mTORC2 and GSK3beta are inhibitory tenance of axonal homeostasis and the prevention of axonal degeneration. for AKT3-induced axon regeneration in the central nervous system. eLife 5, Proc. Natl Acad. Sci. USA 104, 14489–14494 (2007). e14908 (2016). 36. Yue,Z., Wang,Q.J. & Komatsu, M. Neuronal autophagy: goingthe distance to 22. Pankiv, S. et al. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degra- the axon. Autophagy 4,94–96 (2008). dation of ubiquitinated protein aggregates by autophagy. J. Biol. Chem. 282, 37. Petibone, D. M., Majeed, W. & Casciano, D. A. Autophagy function and its 24131–24145 (2007). relationship to pathology, clinical applications, drug metabolism and toxicity. J. 23. Alves, S. et al. The autophagy/lysosome pathway is impaired in SCA7 patients Appl. Toxicol. 37,23–37 (2017). and SCA7 knock-in mice. Acta Neuropathol. 128, 705–722 (2014). 38. Anglade, P. et al. Apoptosis and autophagy in nigral neurons of patients with 24. Elmore, S. Apoptosis: a review of programmed cell death. Toxicol. Pathol. 35, Parkinson’sdisease. Histol. Histopathol. 12,25–31 (1997). 495–516 (2007). 39. Bhutia, S. K. et al. Astrocyte elevated gene-1 induces protective autophagy. 25. Norbury, C. J. & Hickson, I. D. Cellular responses to DNA damage. Annu. Rev. Proc. Natl Acad. Sci. USA 107, 22243–22248 (2010). Pharmacol. Toxicol. 41,367–401 (2001). 40. Marin, C. & Aguilar, E. In vivo 6-OHDA-induced neurodegeneration and nigral 26. Zhu,X.et al. Neuroprotectiveproperties of Bcl-w in Alzheimer disease. J. autophagic markers expression. Neurochem. Int. 58,521–526 (2011). Neurochem. 89,1233–1240 (2004). 41. Emdad, L. et al. Activation of the nuclear factor kappaB pathway by astrocyte 27. Sredni, B. et al. Multifunctional tellurium molecule protects and restores elevated gene-1: implications for tumor progression and metastasis. Cancer dopaminergic neurons in Parkinson’s disease models. FASEB J. 21,1870–1883 Res. 66, 1509–1516 (2006). (2007). 42. Lee, S. G. et al. Oncogene AEG-1 promotes glioma-induced neurodegenera- 28. Yamada, M., Kida, K., Amutuhaire, W., Ichinose, F. & Kaneki, M. Gene disruption tion by increasing glutamate excitotoxicity. Cancer Res. 71, 6514–6523 (2011). of caspase-3 prevents MPTP-induced Parkinson’s disease in mice. Biochem. 43. Paxinos,G.& Franklin,K.B.J. The Mouse Brain in Stereotaxic Coordinates, Biophys. Res. Commun. 402,312–318 (2010). Compact 2nd edn (Elsevier Academic Press, Amsterdam; Boston, 2004). 29. Latchoumycandane, C., Anantharam, V., Jin, H. & Kanthasamy, A. Dopaminergic 44. Shin, W. H. et al. Induction of microglial Toll-like receptor 4 by prothrombin neurotoxicant 6-OHDA induces oxidative damage through proteolytic acti- kringle-2: a potential pathogenic mechanism in Parkinson’sdisease. Sci. Rep. 5, vation of PKCdelta in cell culture and animal models of Parkinson’sdisease. 14764 (2015). Toxicol. Appl. Pharmacol. 256,314–323 (2011). 30. Abdelkader,N.F., Safar, M. M. &Salem,H.A.Ursodeoxycholic acid ameliorates apoptotic cascade in the Rotenone Model of Parkinson’s disease: modulation of mitochondrial perturbations. Mol. Neurobiol. 53,810–817 (2016). Official journal of the Cell Death Differentiation Association http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Cell Death & Disease Springer Journals

Loading next page...
 
/lp/springer_journal/upregulation-of-neuronal-astrocyte-elevated-gene-1-protects-nigral-yyfC053s2j
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Life Sciences; Life Sciences, general; Biochemistry, general; Cell Biology; Immunology; Cell Culture; Antibodies
eISSN
2041-4889
D.O.I.
10.1038/s41419-018-0491-3
Publisher site
See Article on Publisher Site

Abstract

The role of astrocyte elevated gene-1 (AEG-1) in nigral dopaminergic (DA) neurons has not been studied. Here we report that the expression of AEG-1 was significantly lower in DA neurons in the postmortem substantia nigra of patients with Parkinson’s disease (PD) compared to age-matched controls. Similarly, decreased AEG-1 levels were found in the 6-hydroxydopamine (6-OHDA) mouse model of PD. An adeno-associated virus-induced increase in the expression of AEG-1 attenuated the 6-OHDA-triggered apoptotic death of nigral DA neurons. Moreover, the neuroprotection conferred by the AEG-1 upregulation significantly intensified the neurorestorative effects of the constitutively active ras homolog enriched in the brain [Rheb(S16H)]. Collectively, these results demonstrated that the sustained level of AEG-1 as an important anti-apoptotic factor in nigral DA neurons might potentiate the therapeutic effects of treatments, such as Rheb(S16H) administration, on the degeneration of the DA pathway that characterizes PD. Introduction amyotrophic lateral sclerosis (ALS) by activating apopto- Astrocyte elevated gene-1 (AEG-1), also known as tic signaling pathways via inhibition of the phosphatidy- metadherin, was originally identified as a human immu- linositol-4,5-bisphosphate 3-kinase/protein kinase B nodeficiency virus-1- and tumor necrosis factor-alpha- (PI3K/Akt) signaling pathway . inducible gene in human fetal astrocytes, and its upre- The aberrant activation of apoptotic signaling pathways gulation is a well-established important oncogenic event in the adult brain is a well-known neurotoxic event that is 1–4 in various types of human cancer . The downregulation associated with neuronal loss, such as that observed in of neuronal AEG-1 has recently been shown to reduce the neurodegenerative diseases, including Parkinson’s disease 6–8 viability of motor neurons in a mouse model of (PD) and Alzheimer’s disease (AD) , and the PI3K/Akt/ mammalian target of rapamycin complex 1 (mTORC1) signaling pathway has been shown to elicit neuroprotec- Correspondence: Seok-Geun Lee (seokgeun@khu.ac.kr) or Sang tive effects on the survival and growth of neurons in the Ryong Kim (srk75@knu.ac.kr) 1 9–11 School of Life Sciences, BK21 Plus KNU Creative BioResearch Group, Institute nigrostriatal dopaminergic (DA) system . However, of Life Science & Biotechnology, Kyungpook National University, Daegu 41566, little is known about the neuroprotective role of AEG-1 in Republic of Korea PD. Department of Neural Development and Disease, Department of Structure & Function of Neural Network, Korea Brain Research Institute, Daegu 41062, Here we found that the loss of DA neurons in post- Republic of Korea mortem substantia nigra (SN) tissue from patients with Full list of author information is available at the end of the article These authors contributed equally: Eunju Leem, Hyung-Jun Kim, Minji Choi. Edited by G. Raschellà. © The Author(s) 2018 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to theCreativeCommons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Official journal of the Cell Death Differentiation Association 1234567890():,; 1234567890():,; Leem et al. Cell Death and Disease (2018) 9:449 Page 2 of 15 PD were associated with significant decreases in the levels neurons, we evaluated the in vivo effects of AEG-1 over- of expression of AEG-1 in nigral DA neurons of patients expression on the basal levels of apoptotic markers, such with PD compared to age-matched controls. These find- as cleaved caspase-3 and cleaved poly (ADP-ribose) ings suggested that the relationship between AEG-1 polymerase 1 (PARP-1) in nigral DA neurons. Mice were downregulation and the pathogenesis of PD are clini- sacrificed 4 weeks after intranigral injections of AAV- cally relevant. To investigate the role of AEG-1 as a sur- AEG-1 or the control vector AAV-green fluorescent protein (GFP), and the transduction of DA neurons was vival factor in nigral DA neurons in the adult brain, we examined the effects of the adeno-associated virus (AAV)- confirmed by the patterns of GFP expression and the mediated overexpression of AEG-1 on these neurons in immunoperoxidase staining of the hemagglutinin (HA) the 6-hydroxydopamine (6-OHDA)-treated animal model epitope in the AAV-AEG-1 vector, respectively (Fig. 2a). 9,10,12 of PD . Additionally, we examined whether the neu- HA- and GFP-positive cells were clearly colocalized with roprotection conferred by AEG-1 overexpression, which TH-positive DA neurons (Fig. 2b) but not with glial might be a therapeutic intervention, contributed to the fibrillary acidic protein (GFAP)-positive astrocytes or neurorestorative effects on the in vivo nigrostriatal DA ionized calcium binding adaptor molecule 1 (Iba1)-posi- system of treatment strategies, such as the administration tive microglia in the SN (Fig. 2c). Upregulation of AEG-1, of constitutively active ras homolog enriched in brain which showed no neurotoxicity (Fig. 2d–f), resulted in a (with a S16H mutation) [Rheb(S16H)], which induces significant decrease in the basal levels of cleaved caspase-3 9,10 axonal regrowth in damaged DA neurons . and cleaved PARP-1 in the SN compared to noninjected and GFP controls (Fig. 2g; *p = 0.005 vs. CON). Results Similar to previous reports that have implicated apop- 6,8 Decreased levels of AEG-1 expression in the SN of patients tosis in the loss of DA neurons in patients with PD , with PD and a neurotoxin-based model of PD western blot analyses revealed significant increases in the To investigate the alterations in the levels of AEG-1 levels of caspase-3, cleaved caspase-3, and cleaved PARP- expression in the SN of patients with PD (Fig. 1a), we 1 were observed in the SN of patients with PD compared & && performed immunohistochemical staining of the expression to age-matched controls (Fig. 3a; p = 0.014, p = 0.019, &&& patterns (Fig. 1b) and quantified the changes using western and p = 0.009 vs. CON, respectively). As shown in the blotting (Fig. 1c). AEG-1-positive immunoreactivity (blue) experimental schematic (Fig. 3b), the double immuno- was clearly reduced in neuromelanin-positive DA neurons fluorescence staining of TH (red) and cleaved caspase-3 (brown) in the SN of patients with PD compared to age- (green) and of TH and cleaved PARP-1 (green) (Fig. 3c), matched controls (Fig. 1b). Western blot analyses revealed and western blot analyses (Fig. 3d) showed that the levels significant decreases in the levels of AEG-1 and tyrosine of both cleaved caspase-3 and cleaved PARP-1 sig- hydroxylase (TH, a marker of DA neurons) in the SN of the nificantly increased 2 days post-lesion in the TH-positive patients with PD compared to age-matched controls DA neurons in the SN of mice injected with 6-OHDA # 15,16 (Fig. 1c; p= 0.033 and *p= 0.022 for AEG-1 and TH, only . However, the upregulation of AEG-1 sig- respectively, vs. CON). However, decreased AEG-1 nificantly inhibited the cleavage of both caspase-3 and expression was not observed in the hippocampus of PARP-1 in the nigral DA neurons following the 6-OHDA patients with AD compared to age-matched controls, even injections compared to injections of 6-OHDA alone ### though there was a significant loss of neuronal nuclei (Fig. 3c, d; ***p = 0.009 and p = 0.002, respectively, vs. (NeuN, a marker of neurons) in that region in the patients 6-OHDA alone). The anti-apoptotic effects of AEG-1 on compared to controls (Fig. 1f, g; p= 0.001 vs.CON). The the 6-OHDA-induced neurotoxicity in DA neurons were reduction of AEG-1 (brown) was specifictothe SN pars confirmed with western blot analyses of the B-cell lym- compacta of 1 day post-lesion 6-OHDA-treated mice phoma 2/Bcl-2-associated X protein (Bcl-2/Bax) ratio (Fig. 1d), which is a well-known neurotoxin-based model of (Supplementary Figure S1). 9,10,12–14 PD . Western blot analyses similarly showed a sig- nificant decrease in the levels of AEG-1 expression in the Neuroprotective effects of AEG-1 upregulation against SN after 6-OHDA administration, compared to untreated 6-OHDA neurotoxicity controls, 1 day post-lesion (Fig. 1e; *p= 0.024 vs.CON), We evaluated the neuroprotective effects of AAV-AEG- even though the levels of TH were not significantly 1 in the 6-OHDA-treated mouse model of PD (Fig. 4a). decreased in the SN (Fig. 1e). 6-OHDA administration clearly caused neurotoxicity in the nigrostriatal DA projections (Fig. 4b, c), and the Decreased levels of apoptotic signaling molecules by AEG- transduction of DA neurons with AEG-1 but not GFP 1 transduction of mature neurons in the SN effectively mitigated the 6-OHDA-induced neurotoxicity To determine whether the upregulation of AEG-1 in the SN compared to the effects of treatment with affected the apoptotic signaling pathways in nigral DA 6-OHDA alone (Fig. 4b; p = 0.026 vs. 6-OHDA alone). Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 3 of 15 Fig. 1 (See legend on next page.) Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 4 of 15 (see figure on previous page) Fig. 1 Decreased levels of astrocyte elevated gene-1 (AEG-1) in the postmortem substantia nigra (SN) of patients with Parkinson’s disease (PD) and the SN of 6-hydroxydopamine (6-OHDA)-treated mice. a Description of the human postmortem SN tissue. b Immunohistochemistry for AEG-1 in the human SN. Scale bar, 50 μm. The square insets in the left panels contain magnifications of the photomicrographs in the right panels. Scale bar, 20 μm. The blue arrows indicate AEG-1 immunoreactivity, and the brown arrows indicate neuromelanin immunoreactivity. c Western blot analysis of the levels of tyrosine hydroxylase (TH) and AEG-1 in the human SN. *p= 0.022 and p = 0.033 vs. age-matched controls (CON) (t-test; n = 4 for each group). d Representative sections showing AEG-1 expression (with Nissl counterstaining) in the mouse SN pars compacta (SNpc), which is outlined by the dotted lines. Scale bar, 20 μm. The square insets in the left contain magnifications of the photomicrographs in the right panels. Scale bar, 20 μm. e Western blot analyses of the levels of AEG-1 and TH in the mouse SN. AA ascorbic acid. *p = 0.024 and **p= 0.001 vs. intact CON; p < 0.001 for TH, significantly different from CON [one-way analysis of variance (ANOVA) with Tukey’s post hoc test; n = 4 for each group]. f Description of the human postmortem hippocampal tissue. g Western blot analyses of the levels of AEG-1 and neuronal nuclei (NeuN) in the hippocampus of patients with Alzheimer’s disease (AD) and CON. Please note that of the levels of AEG-1 are not decreased in the postmortem hippocampus of patients with AD compared with CON. p = 0.001 vs. CON (t-test; n = 5 for each group) Western blot analyses also showed that AEG-1 trans- sustained activation of the Akt/mTORC1 signaling path- duction significantly preserved the levels of TH expres- way induces axonal regeneration in damaged neu- 9,10,20,21 sion following 6-OHDA-induced neurotoxicity compared rons . However, western blot analyses showed that to treatment with 6-OHDA alone (Fig. 4d; p = 0.002 vs. the overexpression of AEG-1 alone did not alter the 6-OHDA alone) in the SN but not in the striatum (STR). phosphorylation statuses of the Thr37/46 residues of 4E- Similar to the limited neuroprotective effects observed BP1, which are indicative of mTORC1 activity (Supple- using immunostaining and western blotting, the results mentary Figure S4). Similarly, AEG-1 overexpression did obtained with reversed-phase high-performance liquid not induce significant increases in the levels of p-Akt, chromatography (HPLC) analyses indicated that the levels which is associated with activation of mTORC1, com- of striatal dopamine and its metabolites, including 3,4- pared to the levels in noninjected controls (Supplemen- dihydroxyphenylacetic acid (DOPAC) and homovanillic tary Figure S4). Additionally, no significant changes were acid (HVA), did not significantly differ between the mice observed in the levels of microtubule-associated protein treated with 6-OHDA following AEG-1 transduction and 1A/1B-light chain 3 (LC3)-I and II, which are used as an 12,19,22 those treated with 6-OHDA alone (Supplementary indicator of autophagosome formation , following Figure S2). injections of AAV-AEG-1 in the SN of healthy mice To evaluate the effects of glial AEG-1 in the nigrostriatal compared to those in noninjected controls (Supplemen- DA system, we unilaterally injected adenovirus (Ad)-AEG- tary Figure S4). 1 (Ad-AEG-1) or Ad-null, which was a control vector, in In the SN of patients with PD, we observed significant the SN of healthy mice and examined whether any neu- increases in the levels of LC3-II and p62, which are well- 12,19,22,23 roprotective effects were observed against 6-OHDA neu- known markers of ALP , compared to those in rotoxicity. Four weeks after intranigral injections of Ad- age-matched controls (Supplementary Figure S5a). Con- AEG-1, immunohistochemical staining of the HA epitope sistent with the increase in the levels of LC3-II and p62, a tag for Ad-AEG-1 indicated the site-specific transduction significant decrease in the levels of p-4E-BP1 expression of SN microglia with Ad-AEG-1 (Supplementary Fig- was also observed in the SN of patients with PD (Sup- ures S3a to c), and no neurotoxicity was observed in the plementary Figure S5a). These observations suggested nigrostriatal DA projections in the brains of the healthy that the suppression of aberrant ALP through the acti- mice (Supplementary Figures S3d and e). However, we did vation of the mTORC1 signaling pathway might also be not observe neuroprotective effects of the overexpression associated with neuroprotection of the nigrostriatal DA of microglial AEG-1 against 6-OHDA-induced neuro- system. However, AEG-1 overexpression in DA neurons toxicity at 1 week post-lesion (Fig. 4e, f). did not suppress the aberrant accumulation of autophagic components, such as LC3-II and p62, and the decrease in Lack of activation of the Akt/mTORC1 signaling pathway in mTORC1 activity following 6-OHDA neurotoxicity nigral DA neurons by AEG-1 upregulation (Supplementary Figure S5b). The results of previous studies suggested that the Akt/ mTORC1 signaling pathway could be regulated by Application of AEG-1-induced neuroprotection to the 3–5,17,18 changes in AEG-1 expression , and the activation functional recovery of the disrupted nigrostriatal DA of Akt/mTORC1 signaling pathway could regulate the system autophagy–lysosomal pathway (ALP), that is associated To determine the importance of sustaining the 12,19 with axonal degeneration in PD . Moreover, the increased levels of neuronal AEG-1 in the adult Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 5 of 15 Fig. 2 Adeno-associated virus (AAV)-AEG-1 transduction of dopaminergic (DA) neurons in the in vivo SN of healthy mice. a Experimental schematic and the immunostaining for green fluorescent protein (GFP; green) and hemagglutinin (HA; brown) in the SNpc, which is outlined by the dotted elliptical shape, which was conducted following each viral injection. Scale bar, 200 μm. b Representative double immunofluorescent labeling of TH (red) and GFP (green) or TH and HA (green) in the SNpc. Scale bar, 20 μm. c Representative double immunofluorescent labeling for glial fibrillary acidic protein (GFAP)/ionized calcium binding adaptor molecule 1 (Iba1; red), which are markers of astrocytes and microglia, respectively, and GFP/ HA (green) in the SNpc of healthy mice. Scale bar, 20 μm. d Immunostaining for TH in the SN and striatum (STR). Scale bars, 200 μm (black) and 50 μm (white) for the SN, and 1000 μm for the STR. e, f The number and optical density of the nigral TH-positive neurons and striatal TH-positive fibers, respectively (one-way ANOVA with Tukey’s post hoc test; n = 4 for each group). CON contralateral side, IPSI ipsilateral side. g Western blot analyses of the levels of cleaved caspase-3 (c-caspase-3) and cleaved poly (ADP-ribose) polymerase 1 (c-PARP-1) following AEG-1 transduction in the SN of healthy brains. *p = 0.005 vs. CON (one-way ANOVA with Tukey’s post hoc test; n = 4 for each group) 9,10,12 nigrostriatal DA system and, consequently, the potential axonal regeneration in damaged DA neurons .As of AEG-1 overexpression as a therapeutic approach for shown in the experimental schematic (Fig. 5a), treatment PD, we examined the effects of AEG-1 overexpression with 6-OHDA alone induced significant reductions in the following post treatment with AAV-Rheb(S16H) on the motor performance, which was measured using the open- functional recovery of nigral DA neurons and induction of field test (Fig. 5b, c) and rotarod test (Fig. 5d), compared Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 6 of 15 Fig. 3 Anti-apoptotic effects of AEG-1 transduction in DA neurons on 6-OHDA neurotoxicity. a Western blot analyses show a significant increase in the levels of caspase-3, c-caspase-3, and c-PARP1 in the postmortem tissues of patients with PD compared with CON. p = 0.014 for && &&& caspase-3, p = 0.019 for c-caspase-3, and p = 0.009 for c-PARP-1, vs. CON (t-test; n = 4 for each group). b Experimental schematic for Fig. 3c, d. c Representative double immunofluorescence labeling for TH (red) and c-caspase-3 (green) or TH and c-PARP-1 (green) in the mouse SN. AEG-1 upregulation induces reductions in the levels of expression of both c-caspase-3 and c-PARP-1 in TH-positive DA neurons in the SN with 6-OHDA neurotoxicity. Scale bar, 20 μm. d Western blot analyses of the levels of c-caspase-3 and c-PARP-1 in the SN 2 days after 6-OHDA treatment. *p = # ## ### 0.029, **p = 0.025, p = 0.032, and p = 0.016 vs. CON; ***p = 0.009 and p = 0.002 vs. 6-OHDA alone (one-way ANOVA with Tukey’s post hoc test; n = 4 for each group) to that in noninjected controls (*p < 0.001 vs. intact con- AAV-Rheb(S16H)]. Consistent with these results, the trols). Similar to the results described in our previous immunohistochemical staining of TH demonstrated that report , Rheb(S16H) overexpression rescued the motor Rheb(S16H) overexpression following 6-OHDA adminis- impairments induced by 6-OHDA neurotoxicity com- tration induced axonal regeneration in damaged DA # # 9,10 pared to mice injected with 6-OHDA alone ( p = 0.008 neurons (Fig. 5f, h; p < 0.001 vs. 6-OHDA alone) , and ## and p = 0.01 vs. 6-OHDA alone). In particular, we found Rheb(S16H) overexpression in the presence of increased that AEG-1-overexpressing mice that were injected with levels of AEG-1 significantly restored the density of DA AAV-Rheb(S16H) following 6-OHDA injections demon- fibers in the STR compared to the effects in the absence of strated significant reversals of the motor impairments that AEG-1 [Fig. 5f, h; p = 0.003 vs. 6-OHDA + AAV-Rheb were caused by 6-OHDA neurotoxicity compared with (S16H)]. Moreover, the depleted levels of striatal dopa- Rheb(S16H)-treated mice without virally overexpressed mine following 6-OHDA administration, which were § §§ AEG-1 [Fig. 5b–d; p = 0.007 and p = 0.007 for open- measured using HPLC, were greatly restored following field test and rotarod test, respectively, vs. 6-OHDA + Rheb(S16H) overexpression in the presence of virally Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 7 of 15 Fig. 4 (See legend on next page.) Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 8 of 15 (see figure on previous page) Fig. 4 Upregulation of neuronal AEG-1 protects DA neurons from 6-OHDA-induced neurotoxicity. a Experimental schematic for b–d. b Representative coronal sections of the SN stained with anti-TH at 7 days post-lesion. Representative high-power micrographs are shown in the inset to aid visualization. Scale bars, 200 μm (black) and 50 μm (white). The quantitative analysis shows a population of preserved TH-positive neurons in the SN. *p < 0.001 and **p = 0.023 vs. CON; p= 0.026 vs. 6-OHDA (n = 4 for each group). c Neuroprotective effects of AEG-1 are not observed on striatal TH-positive fibers. Scale bar, 1000 μm. The histogram shows the optical densities of the striatal TH-positive fibers. *p < 0.001 vs. CON (n = 4 for each group). d Western blot analysis of the levels of TH with 6-OHDA-induced neurotoxicity in the nigrostriatal DA system. *p < 0.001 for SN and **p < 0.001 for STR vs. CON; p = 0.002 vs. 6-OHDA alone (n = 4 for each group). e, f Experimental schematic and representative coronal sections show that glial AEG-1 upregulation by Ad transduction does not protect the nigrostriatal DA system in the 6-OHDA-treated mouse model of PD. Scale bars, 200 μm for SN and 1000 μm for STR. *p < 0.001 vs. CON (n = 4 for each group). One-way ANOVA with Tukey’s post hoc test was used in b–f overexpressed AEG-1 compared to the levels in its signaling pathway in in vivo and in vitro models of ALS . absence [Fig. 5i; p = 0.024 vs. 6-OHDA + AAV-Rheb Although the role of AEG-1 in the pathogenesis of (S16H)]. Similarly, the levels of the metabolites of dopa- PD was unknown, these results suggest that AEG-1 mine, including DOPAC and HVA, were restored by Rheb might be critical for the survival of DA neurons in the SN (S16H) overexpression, and the effects were more obvious of patients with PD. Here we observed decreased in the presence of overexpressed AEG-1 than in its expression of AEG-1 in damaged DA neurons in absence (Supplementary Figure S6). the adult brain in an immunohistochemical analysis Rheb(S16H) administration did not affect the levels of of the SN of patients with PD and 6-OHDA-treated expression of AEG-1 in the SN of the mice (Supplemen- mice (Fig. 1b, d). However, no significant reductions in tary Figure S7), which suggested that the neurorestorative AEG-1 were observed in the hippocampus of patients effects from this administration might be independent of with AD compared to age-matched controls (Fig. 1g). AEG-1 expression and that its upregulation might activate Thus, these observations suggested that the decreased the Akt/mTORC1 signaling pathway as a supplementary levels of AEG-1 were a specific event that occurred in mechanism in the presence of AEG-1. These data sug- damaged DA neurons and AEG-1 downregulation and the gested that AEG-1, which is significantly reduced in the loss of nigral DA neurons in PD might be clinically SN of patients with PD, is an important endogenous correlated. factor that protects nigral DA neurons from neurotoxicity Here, AAV-mediated overexpression of AEG-1 in DA and that this protection by the anti-apoptotic effects of neurons decreased the levels of apoptotic markers, neuronal AEG-1 enhances the restoration of the disrupted including cleaved caspase-3 and cleaved PARP-1, follow- 15,16 nigrostriatal DA system (Fig. 6), as shown by the effects of ing 6-OHDA administration (Fig. 3c, d) , resulting in Rheb(S16H) administration in the neurotoxin model of neuroprotection in the SN (Fig. 4b, d). The AEG-1- PD (Fig. 5). induced anti-apoptotic effects were confirmed by western blot analyses of the Bcl-2/Bax ratio (Supplementary Fig- Discussion ure S1). An increase in the levels of apoptotic signaling Under physiological conditions, apoptosis is an essential molecules, which was similar to the results obtained in the homeostatic mechanism that maintains the cell popula- mouse model (Fig. 3c, d), were also observed in the SN of 24,25 tion in healthy tissue and protects cells from injury . patients with PD (Fig. 3a). These results suggested that However, aberrant apoptosis, which is one of the neuro- AEG-1 plays a role as a negative regulator of apoptosis in toxic events in the adult brain, may inevitably be related to adult DA neurons and the sustained levels of neuronal 6,8,26 neurodegenerative diseases, such as PD and AD . AEG-1 following neurotoxic events may confer neuro- Moreover, consistent with the results of studies on protection to nigral DA neurons in vivo. Moreover, the 6,8 patients with PD , increases in the levels of apoptotic overexpression of microglial AEG-1 in the 6-OHDA- markers, such as cleaved caspase-3 and cleaved PARP-1, based mouse model of PD did not confer neuroprotection 15,16 are observed in in vivo and in vitro models of PD .As (Fig. 4e, f). Therefore, consistent with the effects of reported previously, the inhibition of apoptotic pathways reduced AEG-1 expression in motor neurons in an ALS 7,27–30 5 protects DA neurons against neurotoxin treatment . mouse model , our results showed that the sustained Thus, these results suggest that the expression and increased levels of neuronal AEG-1 were important for maintenance of endogenous anti-apoptotic factors are attenuating the vulnerability of nigral DA neurons in the beneficial for the survival of nigral DA neurons in the SN of adult brain. adult brain. Despite the significant anti-apoptotic effects of AEG-1, The downregulation of AEG-1 contributes to the its overexpression was not sufficient to protect the whole apoptosis of motor neurons by inhibiting the PI3K/Akt nigrostriatal DA projection against 6-OHDA-induced Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 9 of 15 Fig. 5 (See legend on next page.) Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 10 of 15 (see figure on previous page) Fig. 5 Synergistic effects of AEG-1 and Rheb(S16H) in the disrupted nigrostriatal DA system in vivo.a Experimental schematic. b, c Total ## ### § distance traveled for 5 min and velocity in the open-field test. *p < 0.001 vs. intact controls; p = 0.010 and p < 0.001 vs. 6-OHDA alone; p = 0.007 vs. 6-OHDA + AAV-Rheb(S16H) group (one-way ANOVA with Tukey’s post hoc test; n = 5 for each group). p = 0.008 vs. 6-OHDA alone (t-test). d # ## § Motor deficits measured by using the rotarod test. *p < 0.001 and **p = 0.002 vs. intact controls; p < 0.001 and p = 0.002 vs. 6-OHDA alone; p = 0.006 vs. 6-OHDA+ AAV-Rheb(S16H) group (n = 5 for each group). e, f Representative coronal SN (scale bar, 200 μm) and STR (scale bar, 1000 μm) sections stained with anti-TH at 11 weeks post-lesion. g Quantitative analysis showing the population of preserved TH-positive neurons in the SN. *p < 0.001 vs. intact controls; p = 0.006 vs. 6-OHDA alone (n = 3 for each group). h The histogram shows the optical densities of the striatal TH-positive # ## § fibers. *p < 0.001 vs. intact controls; p < 0.001 and p = 0.001 vs. 6-OHDA alone; p = 0.003 vs. 6-OHDA+ AAV-Rheb(S16H) group (n = 3 for each group). i The levels of striatal dopamine, which were measured with high-performance liquid chromatography, were quantitatively expressed as a # ## ### § percentage of intact control. *p < 0.001 vs. intact controls; p < 0.001, p = 0.024 and p = 0.028 vs. 6-OHDA alone; p = 0.024 vs. 6-OHDA+ AAV- Rheb(S16H) group [n = 4 for AAV-GFP+ 6-OHDA + AAV-Rheb(S16H) group; n = 5 for the other groups]. One-way ANOVA with Tukey’s post hoc test was used in d and g–i One explanation for the observation that AEG-1 over- expression in DA neurons was insufficient to protect the DA system seems related to the mTORC1 pathway. The mTOR kinase, which plays a central role in the integration of responses to various environmental conditions, exists 10,31,32 in two complexes, mTORC1 and mTORC2 . mTORC1 is an important mediator of Akt. mTORC2 can activate Akt, which in turn can act on mTORC1. Acti- vation of the Akt/mTOR signaling pathway enhances the activity of cell survival pathways under various conditions, including trophic factor withdrawal, ischemic shock, and 9–12,32 oxidative stress . Moreover, recent reports have showed that the activation of neuronal mTORC1, which is a key biomolecule for neurotrophic support, by either the delivery of a specific gene or the direct administration of trophic factors induces protective effects against neuro- 11,33 degeneration in animal models of PD . We recently demonstrated that the Rheb(S16H) delivery-induced activation of mTORC1 in DA neurons protects and reconstructs the damaged nigrostriatal DA projections in a mouse model of PD, which suggested that the activation of the Akt/mTORC1 signaling pathway might be a pro- mising therapeutic strategy for the functional recovery of 9,10 DA neurons . Here, however, our results indicated that AEG-1 upregulation in nigral DA neurons did not enhance Akt/mTORC1 signaling in healthy mice (Sup- Fig. 6 Schematic of the importance of neuronal AEG-1 preservation in the nigrostriatal DA system. Following neurotoxic plementary Figure S4). Moreover, the activation events in the nigrostriatal DA system, AEG-1 transduction in DA of the Akt/mTORC1 signaling pathway suppressed the neurons prevents the apoptotic cell death of the DA neurons (but not initiation of autophagy and prevented the aberrant accu- their functional properties). Meanwhile, decreases in the levels of mulation of autophagic components, which might inhibit neuronal AEG-1 and the consequent loss of DA neurons were normal lysosomal degradation, such as the removal of observed. Moreover, AEG-1 transduction in DA neurons can intensify the neurorestorative effects by synergizing with the administration of expended macromolecules and organelles, in the nigros- 12,19 therapeutic agents for axonal regeneration and the inhibition of triatal DA system . Although autophagy is physiolo- abnormal activation of the autophagy–lysosomal pathway. Thus, these gically important for preserving cellular homeostasis and results suggest the importance of AEG-1 preservation in potential inducing protective effects, such as the suppression of therapeutic strategies against PD 34–37 apoptosis and axonal degeneration , its aberrant activity promotes neurodegeneration in the SN of patients neurotoxicity (Fig. 4c, d and Supplementary Figure S2). with PD . Therefore, these observations suggest that These observations indicated limitations in the protective autophagic stress that is caused by the aberrant accu- effects of AEG-1 against 6-OHDA-induced neurotoxicity. mulation of autophagic components might be one of the Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 11 of 15 critical mechanisms that induces the loss of DA neurons it did not affect the aberrant accumulation of autophagic in PD. components and Akt/mTORC1 activity following AEG-1 is known to induce autophagy, which results in 6-OHDA administration, which could contribute to the the survival of cancerous cells under metabolic stress and neurotoxic effects on the nigrostriatal DA system. To apoptosis resistance, and these results may underlie its overcome this limitation of AEG-1, we further transduced considerable cancer-promoting properties . However, the Akt/mTORC1 activator Rheb(S16H) into AEG-1- overexpressing DA neurons. Surprisingly, the synergistic our results indicated that AEG-1 overexpression in DA neurons had no effect on the levels of LC3-II in healthy effects of the two factors restored the nigrostriatal DA brains (Supplementary Figure S4), and on the aberrant system that was disrupted by 6-OHDA administration, accumulation of LC3 and p62, which are critical for ALP and the effects were more obvious in the presence of 19,22,40 activation following 6-OHDA administration AEG-1 than in its absence (Figs. 5 and 6). Therefore, we (Supplementary Figure S5b). Moreover, the decrease in concluded that AEG-1 was an important endogenous the activity of mTORC1 following 6-OHDA neurotoxicity factor for protecting nigral DA neurons from aberrant was not inhibited by the presence of AEG-1 in DA neu- apoptotic signaling pathway in the adult brain and that rons in vivo (Supplementary Figure S5b). Therefore, to the maintenance of increased levels of AEG-1 in nigral potentiate the beneficial effects of AEG-1 in nigral DA DA neurons in patients with PD, in combination with neurons, it may be necessary to activate the Akt/ therapeutic agents, such as an Akt/mTORC1 signaling mTORC1 signaling pathway, which suppresses the aber- activator, may be a highly promising therapeutic strategy rant accumulation of autophagic components as a sup- to maximize the functional recovery of the damaged plementary protective mechanism in the presence of nigrostriatal DA system (Fig. 6). AEG-1 and consequently results in enhanced neuropro- tection of the nigrostriatal DA projection in the adult Materials and methods brain. Ethics statement The overexpression of Rheb(S16H), which is a con- All animal experiments were performed in accordance 9,10 stitutively active form of Rheb and activates mTORC1 , with the approved animal protocols and guidelines suppresses the induction of the abnormal autophagy sig- established by the Animal Care Committee at Kyungpook naling pathway by 6-OHDA treatment . To further National University (Number: KNU 2016-42). Experi- examine if AEG-1 overexpression strengthened the neu- ments involving human tissue were approved by the roprotective and neurorestorative effects of therapeutic Bioethics Committee, Institutional Review Board Kyung- agents, such as Rheb(S16H), AAV-Rheb(S16H) was pook National University Industry Foundation (IRB injected into the SN 3 weeks post-lesion, which is when Number: KNU 2014-0007 and 2016-0011). the maximum neurodegeneration is observed after 6- 9,10 OHDA treatment , in the absence or presence of virally Human brain tissue overexpressed AEG-1 (Fig. 5a). Our results demonstrated Frozen and paraffin-fixed brain tissues were obtained that Rheb(S16H) upregulation in the presence of from the Victorian Brain Bank Network (VBBN), sup- increased levels of AEG-1 induced synergistic neuror- ported by the Florey Institute of Neuroscience and Mental estorative effects, such as restored motor functions, in the Health, The Alfred, and the Victorian Forensic Institute of nigrostriatal DA system disrupted by 6-OHDA neuro- Medicine, and funded by Australia’s National Health & toxicity (Fig. 5b–i). As shown in Supplementary Figure S7, Medical Research Council and Parkinson’s Victoria. Fro- the Rheb(S16H) transduction of DA neurons did not zen and paraffin-fixed brain tissues were used in quanti- affect the levels of AEG-1 in the SN of the mice, sug- tatively analyzing the level of proteins and in observing gesting that the Rheb(S16H)-induced neurorestoration the expression pattern of target molecule, respectively. was independent of AEG-1 upregulation and that the AEG-1-induced neuroprotection against 6-OHDA neu- Materials rotoxicity augmented the beneficial effects of Rheb(S16H) Materials were purchased from the following compa- in the lesioned nigrostriatal DA system in vivo (Fig. 5). nies: 6-OHDA (Sigma, St Louis, MO), desipramine In conclusion, our findings suggested that AEG-1 (Sigma), L-ascorbic acid (Sigma), rabbit anti-TH (Pel- functioned as an anti-apoptotic factor in nigral DA neu- Freez, Brown Deer, WI), mouse anti-TH (R&D Systems, rons of the adult brain and the decrease in AEG-1 might Minneapolis, MN), rabbit anti-Iba1 (Wako Pure Chemical be involved in the loss of DA neurons, which is one of the Industries, Osaka, Japan), rabbit anti-GFAP (Millipore, key pathological features in PD. However, the over- Billerica, MA), rabbit anti-AEG-1 (Invitrogen, Camarillo, expression of AEG-1 in DA neurons was not sufficient to CA), rabbit anti-GFP (Millipore), mouse anti-HA (Cell protect the whole nigrostriatal DA projection in the ani- Signaling, Beverly, MA), rabbit anti-HA (Cell Signaling), mal model of PD owing to its limited protective effects as rabbit anti-FLAG (Sigma), rabbit anti-caspase-3 (Cell Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 12 of 15 Signaling), rabbit anti-cleaved caspase-3 (Cell Signaling), −0.35 cm, ML: −0.11 cm, DV: −0.37 cm, relative to rabbit anti-PARP-1 (Cell Signaling), rabbit anti-cleaved bregma) using a 30-gauge Hamilton syringe attached to PARP-1 (Cell Signaling), rabbit anti-LC3B (Cell Signal- an automated microinjector . Viral vector suspension in ing), rabbit anti-4E-BP1 (Cell Signaling), rabbit anti-p-4E- a volume of 2.0 μl was injected at a rate of 0.1 μl/min over BP1 (Cell Signaling), mouse anti-NeuN (Millipore), rabbit 20 min. After injection, the needle was left in place for an anti-Akt (Cell Signaling), rabbit anti-p-Akt (Cell Signal- additional 5 min before being slowly retracted. ing), rabbit anti-β-actin (Cell Signaling), rabbit anti-α- tubulin (Cell Signaling), mouse anti-Bcl-2 (Santa Cruz Intrastriatal 6-OHDA injection Biotechnology, Santa Cruz, CA), mouse anti-Bax (Santa The intrastriatal 6-OHDA model was induced as pre- 9,10 Cruz Biotechnology), rabbit anti-p62/SQSTM1 (Sigma), viously described . Mice were intraperitoneally injected biotinylated anti-rabbit IgG (Vector laboratories, Burlin- with desipramine (25 mg/kg in 0.9% NaCl), and then game, CA), Texas Red-conjugated anti-rabbit/mouse IgG anesthetized with chloral hydrate. Anesthetized mice were (Vector Laboratories), fluorescein (FITC)-conjugated placed in a stereotaxic frame, and a solution of 6-OHDA anti-mouse IgG (Vector Laboratories), FITC-conjugated (5 mg/ml in 0.9% NaCl/0.02% ascorbic acid), with a final anti-rabbit IgG (Jackson ImmunoResearch Laboratories, volume of 3.0 μl was injected by Hamilton syringe at a rate Bar Harbor, ME), horseradish peroxidase (HRP)-con- of 0.5 μl/min. The injection was performed into the right jugated anti-rabbit IgG (Enzo Life Sciences, Farmingdale, STR at coordinates (AP: +0.09 cm; ML: −0.22 cm; DV: NY) and HRP-conjugated anti-mouse IgG (Thermo −0.25 cm, relative to bregma) . The needle was with- Fisher Scientific, Rockford, IL). drawn slowly after 5 min. Animals were sacrificed and analyzed at the indicated time points for each 9,10 Production of viral vectors experiment . The two types of viral vectors used in the present study were AAV serotype 1 and Ad serotype 5. Viral vectors Behavioral tests were produced as described previously, with some mod- Open-field test 10,17,41,42 ifications . Ad viral vectors were supplied by S.G. The open-field test was performed as described pre- Lee. Briefly, for the production of AAV viral vectors viously, with some modifications . Briefly, 11 weeks after carrying AEG-1 with a HA-encoding sequence at the the 6-OHDA injection, mice were placed individually in 3ʹ-end (AEG-1-HA), AEG-1 cDNA tagged with HA was the corner of a test chamber (40 × 40 × 40 cm) enclosed amplified from the mammalian expression plasmid of with white acrylic walls. After a 1 min adaptation period, AEG-1 (pcDNA3.1-AEG-1-HA), as previously descri- animal behaviors such as the total distance traveled (in 17,42 bed . AEG-1-HA obtained from pcDNA3.1-AEG-1- cm) and velocity (in cm/sec) were recorded for 5 min HA was cloned into an AAV packaging construct that using a video camera. The change in locomotor activity utilizes the chicken β-actin promotor and contains a 3′ was analyzed offline by video-tracking software (SMART, WPRE . Constitutively activated Rheb was also cloned Panlab, Barcelona, Spain). The test chamber was cleaned into the same AAV packaging construct . All nucleotide between trials with 70% ethyl alcohol. To minimize stress sequences in the AAV packaging construct were con- levels, tests were performed under conditions of low firmed before viral vector production. AAVs were pro- illumination. duced at the University of North Carolina Vector Core. The genomic titer of AAV-AEG-1 and AAV-Rheb(S16H) Rotarod test 12 12 were 9.4 × 10 viral genomes/ml and 3.6 × 10 viral The rotarod test was performed at 11 weeks post-lesion, genomes/ml, respectively. Enhanced GFP, used as a con- using a previously described procedure with some mod- trol, was subcloned into the same AAV viral backbone, ifications . Before 6-OHDA treatment, all mice were pre- and viral stock was produced at a titer of 2.0 × 10 viral trained on the rotarod apparatus (3 cm rod diameter; genomes/ml. Genomic titers of both Ad-AEG-1 and Ad- Scitech Inc., Seoul, Korea) at 10 revolutions per min (rpm) null viral stocks were 2.0 × 10 infectious unit/ml. for 10 min, and the training was performed for 3 con- secutive days. Eleven weeks after the 6-OHDA injection, Intranigral AAV and Ad injection performance on the rod was evaluated at a constant Adult (8- to 10-week-old) male C57BL/6 mice were acceleration rate of 4–40 rpm in 300 sec. Two consecutive obtained from Daehan Biolink (Eumseong, Korea). As trials were performed at 60 min intervals. 9,10 previously described, with some modifications , mice were anesthetized with chloral hydrate solution and Immunohistochemical staining placed in a stereotaxic frame (Kopf Instruments, Tujunga, Postmortem brain tissues were processed for immuno- CA) with a mouse adapter. Each mouse received a uni- histochemistry as described previously . Briefly, the lateral injection of AAV or Ad into the right SN (AP: human SN sections were deparaffinized and subjected to Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 13 of 15 citrate-based antigen retrieval, and then washed in cold at 4 °C with the following primary antibodies: rabbit anti- PBS and blocked with blocking solution. The sections TH (1:2000), rabbit anti-AEG-1 (1:1000), mouse anti- were incubated with primary antibodies against AEG-1 NeuN (1:500), rabbit anti-GFP (1:500), rabbit anti-HA (1:500) at 4 °C overnight, and then incubated with bioti- (1:1000), rabbit anti-FLAG (1:3000), rabbit anti-caspase-3 nylated secondary antibodies for 1 h at room temperature, (1:1000), rabbit anti-cleaved caspase-3 (1:1000), rabbit followed by addition of the avidin-biotin reagent (Vec- anti-PARP-1 (1:1000), rabbit anti-cleaved PARP-1 (1:1000), mouse anti-Bcl-2 (1:1000), mouse anti-Bax tastain ABC kit, Vector Laboratories) for 1 h at room temperature. The SN sections were visualized using a (1:1000), rabbit anti-LC3B (1:1000), rabbit anti-p62/ 3,3′-diaminobenzidine (DAB; Sigma) peroxidase substrate SQSTM1 (1:1000), rabbit anti-Akt (1:1000), rabbit anti-p- solution [0.05% DAB, 0.05% Cobalt Chloride (Sigma), Akt (1:2000), rabbit anti-4E-BP1 (1:1000), and rabbit anti- 0.05% Nickel Ammonium Sulfate (Sigma) and 0.015% p-4E-BP1 (1:1000). Subsequently, the membranes were H O in PBS, pH 7.2]. Each section was covered with a incubated with secondary antibodies for 1 h at room 2 2 thin glass coverslip and analyzed under a bright-field temperature, and the bands were finally detected using microscope (Carl Zeiss, Oberkochen, Germany). Western-blot detection reagents (Thermo Fisher Scien- 9,10,44 As previously described , mice were transcardially tific, Rockford, IL). For quantitative analyses, the density perfused and fixed, and the brains were dissected out, of each band was measured using a Computer Imaging frozen, and cut into 30-μm-thick coronal sections using a Device and accompanying software (Fuji Film, Tokyo, cryostat microtome (Thermo Fisher Scientific). Briefly, Japan), and the levels were quantitatively expressed as the the brain sections were washed in PBS and blocked with density normalized to the housekeeping protein band for blocking buffer, and then incubated at 4 °C for 2 days with each sample. the following primary antibodies: rabbit anti-TH (1:2000), rabbit anti-AEG-1 (1:500), rabbit anti-Iba1 (1:2000), rab- Stereological estimation 10,44 bit anti-GFAP (1:1000), mouse anti-HA (1:1000), rabbit As previously described , the total number of TH- anti-cleaved caspase-3 (1:400), rabbit anti-cleaved PARP-1 positive neurons was counted in the various animal (1:400) and mouse anti-TH (1:500). After incubation, the groups using the optical fractionator method. Counting of brain sections were incubated with biotinylated secondary TH-positive neurons in the SN was performed on a antibodies, followed by addition of the avidin-biotin bright-field microscope (Olympus Optical, BX51, Tokyo, reagent (Vectastain ABC kit) for 1 h at room tempera- Japan) using Stereo Investigator software (MBF ture, or incubated with fluorescence-conjugated second- Bioscience, Williston, VT). This unbiased stereological ary antibodies for 1 h. The signal following treatment with method of cell counting is not affected by either the avidin-biotin reagent was detected by incubating the reference volume (SN pars compacta) or the size of the sections in 0.5 mg/ml DAB (Sigma) in 0.1 M PB con- counted elements (neurons). taining 0.003% H O . The sections incubated with 2 2 fluorescence-conjugated secondary antibodies were Quantitative determination of striatal TH washed in 0.1 M PBS. The stained sections were mounted immunoperoxidase staining on gelatin-coated slides and analyzed under a light or Densitometric analysis of the mouse STR was carried 10,44 fluorescence microscope (Carl Zeiss). out as previously described . Briefly, an average of 6 coronal sections of the STR that gathered according to the Western blot analysis bregma of the brain atlas was imaged at a ×1.25 mag- Animal and human tissue samples were prepared for nification. The density of striatal TH-positive fibers was 10,44 western blot analysis as previously described . Briefly, measured using the Science Lab 2001 Image Gauge animal SN tissues were removed and sliced using a brain (Fujifilm, Tokyo, Japan). To control for variations in matrix (Roboz Surgical Instrument Co., Gaithersburg, background illumination, the density of the corpus cal- MD). Animal or human tissue samples were homogenized losum was subtracted from the density of STR for each and centrifuged at 4 °C for 20 min at 14,000 × g; the section. The density in both the contralateral and ipsi- supernatant was transferred to a fresh tube and the con- lateral sides was expressed by comparing with the average centration was determined using a bicinchoninic acid density of TH-positive fiber innervating the contralateral assay (BCA) kit (Bio-Rad Laboratories, Hercules, CA). side. Aliquots containing 50 μg of protein were electrophoresed on a sodium dodecyl sulfate (SDS)/polyacrylamide gel Measurement of dopamine and its metabolites in the STR (Bio-Rad Laboratories) and transferred to polyvinylidene As previously described , the levels of striatal dopa- difluoride (PVDF) membranes (Millipore, Billerica, MA) mine and its metabolites were measured by high perfor- using an electrophoretic transfer system (Bio-Rad mance liquid chromatography (HPLC, 1260 Infinity Laboratories) The membranes were incubated overnight system, Agilent Technologies, Santa Clara, CA) using an Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 14 of 15 ESA Coulochem III electrochemical detector. Briefly, the designed the viral constructs; S.R.K., S.-G.L., P.B.F., and N.K. contributed the viral vectors; C.M. contributed the human brain samples; Y.-S.O., K.J.L., Y.S.C., K.S.A., B. brain tissues were homogenized and centrifuged, and the K.J., D.W.K., J.M.L., S.G.L., and S.R.K. supervised the analysis of the data obtained supernatants were injected using an autosampler at 4 °C with the human brain samples; S.R.K. supervised the whole project and wrote (Waters 717 plus autosampler) and eluted through a μ the paper. All of the authors contributed to the data analysis and preparation of the manuscript. Sunfire C18 column (4.6 × 100 mm × 5 μm; Waters Cor- poration, Milford, MA) with a mobile phase of MDTM/ Conflict of interest acetonitrile (90:10). The peaks of dopamine and its The authors declare that they have no conflict of interest. metabolites were analyzed and integrated using a Chem- Station software (Agilent Technologies, Santa Clara, CA), Publisher's note and all samples were normalized for protein content as Springer Nature remains neutral with regard to jurisdictional claims in spectrophotometrically determined using the Pierce BCA published maps and institutional affiliations. protein assay kit (Thermo Scientific, Waltham, MA). Supplementary Information accompanies this paper at (https://doi.org/ 10.1038/s41419-018-0491-3). Statistical analysis Differences between two groups were analyzed using t- Received: 29 January 2018 Accepted: 14 February 2018 test. Multiple comparisons between groups were per- formed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Behavioral test was analyzed using one-way ANOVA and t-tests. All values References represent the mean ± standard error of the mean (SEM), 1. Su,Z.Z.etal. Identification and cloning of human astrocyte genes displaying and statistical analyses were performed using SigmaStat elevated expression after infection with HIV-1 or exposure to HIV-1 envelope glycoprotein by rapid subtraction hybridization, RaSH. Oncogene 21, software (Systat Software, San Leandro, CA). 3592–3602 (2002). 2. Su,Z.Z.etal. Customized rapidsubtraction hybridization (RaSH) gene Acknowledgements microarrays identify overlapping expression changes in human fetal astrocytes This study was supported by grants from the Korea Healthcare Technology resulting from human immunodeficiency virus-1 infection or tumor necrosis R&D Project, Ministry of Health & Welfare (HI15C1928), and the National factor-alpha treatment. Gene 306,67–78 (2003). Research Foundation of Korea (NRF-2013R1A2A2A01069099, NRF- 3. Lee, S. G.,Kang, D. C.,DeSalle,R., Sarkar,D.&Fisher,P.B.AEG-1/MTDH/LYRIC, 2015R1A4A1042399, NRF-2016M3C7A1905074, and NRF-2017R1A2B4002675). the beginning: initial cloning, structure, expression profile, and regulation of expression. Adv. Cancer Res. 120,1–38 (2013). Author details 4. Emdad, L. et al. AEG-1/MTDH/LYRIC: a promiscuous protein partner critical in School of Life Sciences, BK21 Plus KNU Creative BioResearch Group, Institute cancer, obesity, and CNS diseases. Adv. Cancer Res. 131,97–132 (2016). of Life Science & Biotechnology, Kyungpook National University, Daegu 41566, 5. Yin, X. et al. Downregulated AEG-1 together with inhibited PI3K/Akt pathway Republic of Korea. Department of Neural Development and Disease, is associated with reduced viability of motor neurons in an ALS model. Mol. Department of Structure & Function of Neural Network, Korea Brain Research Cell. Neurosci. 68, 303–313 (2015). Institute, Daegu 41062, Republic of Korea. Department of Science in Korean 6. Hartmann, A. et al. Caspase-3: a vulnerability factor and final effector in Medicine, Graduate School, Kyung Hee University, Seoul 02447, Republic of apoptotic death of dopaminergic neurons in Parkinson’sdisease. Proc. Natl Korea. Department of Brain-Cognitive Science, Daegu-Gyeongbuk Institute of Acad. Sci. USA 97,2875–2880 (2000). Science and Technology, Daegu 42988, Republic of Korea. Predictive Model 7. Kanthasamy, A. G. et al. A novel peptide inhibitor targeted to caspase-3 Research Center, Korea Institute of Toxicology, Korea Research Institute of cleavage site of a proapoptotic kinase protein kinase C delta (PKCdelta) Chemical Technology, Daejeon 34114, Republic of Korea. Department of protects against dopaminergic neuronal degeneration in Parkinson’sdisease Biochemisry and Molecular Biology, Department of Neuroscience Graduate models. Free. Radic. Biol. Med. 41, 1578–1589 (2006). School, School of Medicine, Kyung Hee University, Seoul 02447, Republic of 8. Radi,E., Formichi, P., Battisti,C.& Federico, A.Apoptosis andoxidative stress in Korea. Department of Anatomy, Brain Research Institute, School of Medicine, neurodegenerative diseases. J. Alzheimers Dis. 42,S125–S152 (2014). Chungnam National University, Daejeon 34134, Republic of Korea. Victorian 9. Kim, S. R. et al. Dopaminergic pathway reconstruction by Akt/Rheb-induced Brain Bank Network, Florey Institute of Neuroscience and Mental Health, The axon regeneration. Ann. Neurol. 70, 110–120 (2011). University of Melbourne, Melbourne, VIC 3004, Australia. Department of 10. Kim,S. R., Kareva, T., Yarygina, O., Kholodilov, N. &Burke,R.E.AAV transduction Anatomical Pathology, Alfred Hospital, Melbourne, VIC 3004, Australia. of dopamine neurons with constitutively active Rheb protects from Department of Human and Molecular Genetics, VCU Institute of Molecular neurodegeneration and mediates axon regrowth. Mol. Ther. 20,275–286 Medicine, VCU Massey Cancer Center, School of Medicine, Virginia (2012). Commonwealth University, Richmond, VA 23298, USA. Department of 11. Nam, J. H. et al. Induction of GDNF and BDNF by hRheb(S16H) transduction of Neurology, Columbia University, New York, NY 10032, USA. Department of SNpc neurons: neuroprotective mechanisms of hRheb(S16H) in a model of Biochemistry and Cell Biology, Cell and Matrix Research Institute, BK21 Plus Parkinson’sdisease. Mol. Neurobiol. 51,487–499 (2015). KNU Biomedical Convergence Program, School of Medicine, Kyungpook 12. Cheng, H. C. et al. Akt suppresses retrograde degeneration of dopaminergic National University, Daegu 41944, Republic of Korea. Department of Food axons by inhibition of macroautophagy. J. Neurosci. 31,2125–2135 (2011). Science and Nutrition, Pukyong National University, Busan 48513, Republic of 13. Kim, B. W. et al. Pathogenic upregulation of glial lipocalin-2 in the Parkinsonian Korea. KHU-KIST Department of Converging Science and Technology, Kyung dopaminergic system. J. Neurosci. 36,5608–5622 (2016). Hee University, Seoul 02447, Republic of Korea. Brain Science and 14. Kim,H.D., Jeong, K. H.,Jung, U. J. & Kim,S.R.Naringintreatment induces Engineering Institute, Kyungpook National University, Daegu 41944, Republic neuroprotective effects in a mouse model of Parkinson’s disease in vivo, but of Korea not enough to restore the lesioned dopaminergic system. J. Nutr. Biochem. 28, 140–146 (2016). Author contributions 15. Lev, N., Melamed, E. & Offen, D. Apoptosis and Parkinson’sdisease. Prog. E.L., H.-J.K., M.C., S.-G.L., and S.R.K. conceived and designed the experiments; E. Neuropsychopharmacol. Biol. Psychiatry 27,245–250 (2003). L., H.-J.K., M.C., U.J.J., S.K., J.-Y.U., W.-H.S., J.Y.J., S.G.L., and S.R.K. conducted the 16. Hernandez-Baltazar, D., Mendoza-Garrido,M.E.& Martinez-Fong, D. Activation experiments, analyzed the data, and generated the figures; S.R.K. and S.-G.L. of GSK-3beta and caspase-3 occurs in Nigral dopamine neurons during the Official journal of the Cell Death Differentiation Association Leem et al. Cell Death and Disease (2018) 9:449 Page 15 of 15 development of apoptosis activated by a striatal injection of 6- 31. Kim,S.R.Mammaliantarget ofrapamycincomplex 1asaninducer of neu- hydroxydopamine. PLoS ONE 8, e70951 (2013). rotrophic factors in dopaminergic neurons. Neural Regen. Res. 9,2036–2037 17. Lee, S.G., Su,Z. Z., Emdad, L.,Sarkar, D. &Fisher, P. B. Astrocyteelevatedgene-1 (2014). (AEG-1) is a target gene of oncogenic Ha-ras requiring phosphatidylinositol 3- 32. Jeon, M. T. et al.InvivoAAV1 transductionwithhRheb(S16H) protects hip- kinase and c-Myc. Proc. Natl Acad. Sci. USA 103, 17390–17395 (2006). pocampal neurons by BDNF production. Mol. Ther. 23,445–455 (2015). 18. Lee,S.G.etal. Astrocyteelevatedgene-1 activates cell survival pathways 33. Allen, S. J.,Watson, J. J.,Shoemark, D. K.,Barua,N.U.& Patel, N. K. GDNF,NGF through PI3K-Akt signaling. Oncogene 27, 1114–1121 (2008). and BDNF as therapeutic options for neurodegeneration. Pharmacol. Ther. 19. Burke,R.E.&O’Malley, K. Axon degeneration in Parkinson’sdisease. Exp. Neurol. 138,155–175 (2013). 246,72–83 (2013). 34. Boya, P. et al. Inhibition of macroautophagy triggers apoptosis. Mol. Cell. Biol. 20. Liu, K. et al. PTEN deletion enhances the regenerative ability of adult corti- 25,1025–1040 (2005). cospinal neurons. Nat. Neurosci. 13, 1075–1081 (2010). 35. Komatsu, M. et al. Essential role for autophagy protein Atg7 in the main- 21. Miao, L. et al. mTORC1 is necessary but mTORC2 and GSK3beta are inhibitory tenance of axonal homeostasis and the prevention of axonal degeneration. for AKT3-induced axon regeneration in the central nervous system. eLife 5, Proc. Natl Acad. Sci. USA 104, 14489–14494 (2007). e14908 (2016). 36. Yue,Z., Wang,Q.J. & Komatsu, M. Neuronal autophagy: goingthe distance to 22. Pankiv, S. et al. p62/SQSTM1 binds directly to Atg8/LC3 to facilitate degra- the axon. Autophagy 4,94–96 (2008). dation of ubiquitinated protein aggregates by autophagy. J. Biol. Chem. 282, 37. Petibone, D. M., Majeed, W. & Casciano, D. A. Autophagy function and its 24131–24145 (2007). relationship to pathology, clinical applications, drug metabolism and toxicity. J. 23. Alves, S. et al. The autophagy/lysosome pathway is impaired in SCA7 patients Appl. Toxicol. 37,23–37 (2017). and SCA7 knock-in mice. Acta Neuropathol. 128, 705–722 (2014). 38. Anglade, P. et al. Apoptosis and autophagy in nigral neurons of patients with 24. Elmore, S. Apoptosis: a review of programmed cell death. Toxicol. Pathol. 35, Parkinson’sdisease. Histol. Histopathol. 12,25–31 (1997). 495–516 (2007). 39. Bhutia, S. K. et al. Astrocyte elevated gene-1 induces protective autophagy. 25. Norbury, C. J. & Hickson, I. D. Cellular responses to DNA damage. Annu. Rev. Proc. Natl Acad. Sci. USA 107, 22243–22248 (2010). Pharmacol. Toxicol. 41,367–401 (2001). 40. Marin, C. & Aguilar, E. In vivo 6-OHDA-induced neurodegeneration and nigral 26. Zhu,X.et al. Neuroprotectiveproperties of Bcl-w in Alzheimer disease. J. autophagic markers expression. Neurochem. Int. 58,521–526 (2011). Neurochem. 89,1233–1240 (2004). 41. Emdad, L. et al. Activation of the nuclear factor kappaB pathway by astrocyte 27. Sredni, B. et al. Multifunctional tellurium molecule protects and restores elevated gene-1: implications for tumor progression and metastasis. Cancer dopaminergic neurons in Parkinson’s disease models. FASEB J. 21,1870–1883 Res. 66, 1509–1516 (2006). (2007). 42. Lee, S. G. et al. Oncogene AEG-1 promotes glioma-induced neurodegenera- 28. Yamada, M., Kida, K., Amutuhaire, W., Ichinose, F. & Kaneki, M. Gene disruption tion by increasing glutamate excitotoxicity. Cancer Res. 71, 6514–6523 (2011). of caspase-3 prevents MPTP-induced Parkinson’s disease in mice. Biochem. 43. Paxinos,G.& Franklin,K.B.J. The Mouse Brain in Stereotaxic Coordinates, Biophys. Res. Commun. 402,312–318 (2010). Compact 2nd edn (Elsevier Academic Press, Amsterdam; Boston, 2004). 29. Latchoumycandane, C., Anantharam, V., Jin, H. & Kanthasamy, A. Dopaminergic 44. Shin, W. H. et al. Induction of microglial Toll-like receptor 4 by prothrombin neurotoxicant 6-OHDA induces oxidative damage through proteolytic acti- kringle-2: a potential pathogenic mechanism in Parkinson’sdisease. Sci. Rep. 5, vation of PKCdelta in cell culture and animal models of Parkinson’sdisease. 14764 (2015). Toxicol. Appl. Pharmacol. 256,314–323 (2011). 30. Abdelkader,N.F., Safar, M. M. &Salem,H.A.Ursodeoxycholic acid ameliorates apoptotic cascade in the Rotenone Model of Parkinson’s disease: modulation of mitochondrial perturbations. Mol. Neurobiol. 53,810–817 (2016). Official journal of the Cell Death Differentiation Association

Journal

Cell Death & DiseaseSpringer Journals

Published: Apr 18, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off