Background
Poly(ADP-ribose) polymerase-1 (PARP-1) serves as a double-edged sword in cellular processes, playing pivotal roles in both cell survival and death. As a key facilitator of DNA damage repair via polyADP-ribosylation (PARylation), PARP-1 contributes to cellular survival [
18]. Nevertheless, its overactivation results in cytosolic NAD+ depletion, poly-ADP-ribose (PAR) formation, mitochondrial apoptosis-inducing factor (AIF) release, nuclear AIF translocation, and ultimately, parthanatos-mediated cell death[
26]. Thus, while PARP-1 inhibition may promote cell lethality through excessive DNA damage in cancer therapy, prudent PARP-1 suppression could offer a therapeutic avenue to regulate abnormal neuronal death in conditions such as Parkinson's disease (PD) and ischemia [
56].
PD, a neurodegenerative disorder, is characterized by dopaminergic neuronal loss in the substantia nigra pars compacta (SNpc) [
52]. With aging, PD prevalence increases, and affected individuals' brain exhibits abnormal protein aggregates called Lewy bodies, composed of parkin (an E3 ubiquitin ligase component) and α-synuclein. Although the etiology of PD remains enigmatic, parkin mutations have been linked to autosomal recessive juvenile parkinsonism (AR-JP) [
32,
49]. In sporadic PD, the more common form, parkin often becomes inactivated via S-nitrosylation, oxidative and dopaminergic stress, and c-Abl phosphorylation [
17].
Given parkin's role as an SCF-like ubiquitin ligase complex component, potential substrates degraded by parkin, such as aminoacyl-tRNA synthetase interacting multifunctional protein 2 (AIMP2, also known as p38 or JTV-1) [
15,
52], have emerged as promising therapeutic targets for PD. Elevated AIMP2 expression in post-mortem PD patient’s brain and parkin knockout mice suggests its critical involvement in PD pathophysiology [
33]. AIMP2 is known to respond to genetic damage through direct interaction with p53 and to promote TNF-α-dependent cell death via TNF receptor-associated factor 2 (TRAF2), possibly acting as a multidirectional apoptotic factor [
12]. However, the exact mechanism underlying AIMP2's influence on PD remains elusive. Recent findings by Lee et al. revealed that AIMP2 overexpression directly activates PARP-1 and triggers dopaminergic neuronal cell death in AIMP2 transgenic mice [
38]. Consequently, AIMP2-induced PD implicates PARP-1-mediated parthanatos as a central driver of disease progression.
Although AIMP2 serves as the upstream regulator of PARP-1 activation, its undruggability—owing to its non-enzymatic and housekeeping scaffold properties [
45]—precludes it as a viable PD therapeutic target. Instead, PARP inhibitors hold promise as potential PD therapeutics by suppressing AIMP2-aberrantly activated PARP-1. However, the ability of PARP inhibitors to cross the blood–brain barrier remains uncertain [
22]. Furthermore, considering PARP-1's crucial role in DNA damage sensing and repair, long-term treatment with PARP inhibitors may provoke significant adverse effects in PD patients [
7].
To identify a safe and specific target for treating PD, we assessed AIMP2 as a potential candidate for mitigating dopaminergic cell death. In our previous research, we discovered that the AIMP2 splice variant, DX2, which lacks exon 2 of the full-length AIMP2, competes with AIMP2 for p53 binding [
11]. Remarkably, we observed distinct functional characteristics between full-length AIMP2 and its splice variant DX2 in TNF-α signaling. In ovarian epithelial cells, such as SKOV3 and A2780, AIMP2 elicited apoptosis by degrading TRAF2 and inhibiting NF-κB. In contrast, DX2 facilitated cell survival by stabilizing TRAF2 in response to TNF-α treatment [
13]. These findings propose that DX2 might function as an endogenous antagonist of proapoptotic full-length AIMP2, maintaining a delicate equilibrium between cellular survival and death. If DX2 can effectively impede AIMP2’s function in PD, it may mitigate dopaminergic neuronal cell death. Our study demonstrates that DX2 proficiently suppresses AIMP2 function by modulating PARP-1 activation.
Central nervous system (CNS) disorders display convoluted pathophysiology and elaborate molecular mechanisms, making drug delivery to target cells arduous due to the CNS's unique anatomical features [
4]. As a result, the pursuit of more efficacious treatment strategies to supersede existing drugs persists. The current study endeavors to evaluate the potency of a recombinant adeno-associated virus (AAV) vector system in delivering DX2, consequently inhibiting AIMP2 function in PARP-1 overactivation both in vitro and in vivo.
Materials and method
Cell lines and reagents
SK-SH5Y, SK-N-SH, and N2A cells were obtained from the Korean Cell Line Bank (KCLB, Seoul, Korea). Cells were grown in RPMI-1640, and supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin (HyClone, PA, USA). The transient transfection of myc-tagged or GFP-tagged parkin, AIMP2, DX2, HA-tagged ubiquitin, and empty vector was performed using lipofectamine 2000 (Invitrogen, CA, USA). Propidium iodide solution (PI), 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetra-zolium bromide (MTT), rotenone, and 6-OHDA were obtained from Sigma-Aldrich (St. MO, USA). siRNA targeting DX2 (CACGUGCAGGAUUACGGGGC) was purchased from Bioneer (Seoul, Korea).
Primary neuronal cell isolation
To prepare primary neurons, embryonic brain cortices at E18-E19 were removed and transferred to new conical tubes and gently washed three times with PBS. The washed cortices were incubated in papain solution containing 3.5 mg/ml papain (0.5 units/g; Wako, VA, USA), 0.5 mg/ml EDTA-disodium salt (Wako, VA, USA), and deoxyribonuclease I (DNase I, 5 units/mL; Takara, Shiga, Japan) for 15 min. The digested cortices were dissociated by pipetting 12 times with a glass Pasteur pipette and filtered with a cell strainer (40-mm mesh; BD Biosciences, NJ, USA). The harvested neuronal cells were cultured in MEM containing 20% FBS and N2 supplement (Thermo Fisher Scientific, MA, USA).
Vialble cell counting
Cells were detached using Trypsin/EDTA solution and subjected to centrifugation at 300 g for 5 min. A portion of the collected cells were then sampled and suspended in a 1:1 mixture of media and trypan blue solution (Cat no. 15250061, Thermo Fisher, CA, USA). The cell suspension was subsequently loaded into a cell counting device (Cat no. AMQAF2001, Countess 3 Autometated cell counter, Thermo Fisher, CA, USA) following the device process. All cell counting experiments were conducted in triplicate for quantification.
Pulse-chase assay
HEK293 cells transfected with myc-tagged AIMP2 or DX2 were then incubated in a methionine-free medium for 1 h. Then, [35S] methionine (50 μCi/ml) was added and incubated for 1 h. After the radioactive methionine was washed off with fresh medium, targeted proteins AIMP2 and DX2 were immunoprecipitated with the myc antibody (SC, CA; USA Cat. #sc-40), separated by 12% SDS-PAGE, and subjected to autoradiography using a BAS scanner (FLA-3000; Tokyo, Japan).
Primary hippocampal neuron culture and transfection
Hippocampal CA3-CA1 regions from (0–1 day-old) Sprague Dawley rats (DBL; Strain code: NTac:SD) were dissected, dissociated, and plated onto poly-ornithine-coated coverslips inside a 6-mm-diameter cylinder. For live imaging of synaptic physiology, the plasmids (vGlut1-pHluorin with strep-DX2 or strep-AIMP2) were transfected 8 days after plating. Experiments were conducted 14–21 days after plating. All animal experiments were performed with the approval of the Institutional Animal Care and Use Committee of Kyung Hee University. The plasmids were incubated with 2 mM Ca2+, 2 × HeBS (273 mM NaCl, 9.5 mM KCl, 1.4 mM Na2HPO4.7H2O, 15 mM D-glucose, 42 mM HEPES), and the mixture was transfected to hippocampal neurons at DIV 8 using the Ca2+ phosphate precipitation method.
Optical live imaging for synaptic function
For synaptic function imaging, we utilized a pHluorin-based assay. Coverslips with neurons were mounted in a laminar-flow-perfusion system of a metal chamber on the stage of a custom-built, laser-illuminated epifluorescence microscope. Live-cell images were acquired using an Andor iXon Ultra 897 (Model #DU-897U-CS0-#BV) back-illuminated EM CCD camera with a diode-pumped OBIS 488 laser (Coherent), shuttered by TTL on/off signal from the EMCCD camera during imaging. Fluorescence excitation/emission and collection were achieved using a 40× Fluor Zeiss objective lens (1.3 NA) with 500–550 nm emission and 498 nm dichroic filters (Chroma) for pHluorin. Action potentials were caused by passing a 1 ms current pulse through platinum-iridium electrodes using an isolated current stimulator (World Precision Instruments). Neurons were perfused in saline-based Tyrode’s buffer consisting of 119 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 25 mM HEPES, 30 mM glucose, 10 μM 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX), and 50 μM D,L-2-amino-5-phosphonovaleric acid (AP5), adjusted to pH 7.4. All live presynaptic terminal imaging was conducted at 30 °C; all images were acquired at 2 Hz with a 50-ms exposure.
Immunoprecipitation assay
After washing the prepared cells with cold PBS, cell lysate was collected with 1% Trition × 100 lysis buffer diluted in PBS after incubation at 4 °C for 20 min. The 500 μg of protein in the supernatant was incubated with 1 μg antibody for pull down for 2 h and then incubated with 25 μl protein A/G plus-agarose beads (sc-2003, SC; CA, USA) at 4 °C overnight. After washing the sinked bead with PBS-T (0.05% tween in PBS), 2 × sample buffer was mixed with the bead. For denaturation, the samples were boiled for 10 min. Pull down was performed with PARP-1 antibody, c-Myc antibody for Parkin, and Flag antibody for AIMP2 and DX2.
Western blot
The prepared protein supernatant was loaded into a 10% acrylamide gel for Western blotting electrophoresis, before transferring the protein from the gel to a PVDF membrane; a 5% skimmed milk solution was then used for blocking. The membrane was incubated in primary antibody diluted with Tris-buffered saline (TBS) with 0.05% Tween (TBS-T) for 2 h. After three washes with TBS-T, the diluted secondary antibody was incubated for 1 h. Detection was performed after luminol (SC; CA, USA). Antibodies against c-myc (sc-40), Ha-tag (sc-7392), PARP-1 (sc-8007), GFP (sc-101525), PCNA (sc-56), and α-tubulin (sc-5286) were purchased from Santa Cruz (SC, CA, USA) and tyrosine hydroxylase antibody (p40101-150) was purchased from Pel-Freez (AR, USA). Poly-(ADP-ribose) (83732s), lamin A/C (2032s), AIF (4642s) and YY1 (46365s) antibodies were purchased from Cell Signalling (MA, USA). For detecting basal AIMP2 and DX2 expression, NMS-01-0011 (curebio, Seoul) was used. All experiments were 3 times repeated independently.
Ethics statement
All mice experiments were performed under the Kyung Hee University Institutional Animal Care & Use Committee guidelines (KHUASP(SE)-18-101).
Rotenone-induced PD mouse model
To confirm the anti-neurodegenerative role of DX2 in vivo, chicken-β-actin promoter induced whole body DX2 transgenic animals were provided by Dr. Sunghoon Kim (Seoul National University, South Korea). Male C57Bl/6n-based DX2 transgenic mice and wild type mice were used. Since 8 weeks of age, fresh rotenone solution was orally administrated at a concentration of 30 mg/kg by gavage once a day for 4 weeks to wild type and DX2-TG mice. Rotenone (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in 4% carboxymethylcellulose (CMC, Sigma-Aldrich) with 1.25% chloroform (Beijing Chemical Works, Beijing, China). During rotenone treatment, the rotarod behaviour test was performed once a week, and the latency to fall was measured. At 4 weeks after chemical administration, pole test was performed to measure the turning time above top pole and descending time from top to bottom. To confirm this dopaminergic neuron survival in the substantial nigra, immunohistochemistry staining was performed using tyrosine hydroxylase antibody with cryo-sectioned brain.
Production and purification of scAAV2
To produce AAV-GFP and AAV-DX2, AAV-GFP and AAV-DX2 were generated by ligating oligonucleotides encoding the GFP or DX2 sequence into the pSF-AAV-ITR-CMV-ITR-KanR (OXGENE, Oxford, UK) vector. AAV viral vectors were prepared and transfected with AAV2 helper (OXGENE, Oxford, UK), AAV2 Rep-Cap (OXGENE, Oxford, UK), and pSF AAV-ITR-DX2 or pSF AAV-ITR-GFP in HEK293 cells. Media and cells were then collected and lysed.
To produce recombinant AAV serotype 2 (rAAV2) encoding DX2 or GFP at 40 L (20 L × 2) scales, one day before transfection of plasmid vectors, HEK 293 cells were counted (8 × 105 cells/ml) and centrifuged at 300 × g for 7 min. The cell pellet was resuspended in HEK293 medium supplemented with 4 mM ultraglutamine and incubated for 24 h. Then, the DNA-PEIproHQ (transfection reagent) complexes (transfection mix) were generated according to the manufacturer’s recommendations. Briefly, two mixtures were generated in culture medium: (1) the plasmid DNA mix (scAAV2 DX2 or GFP–4 mg, rep/cap plasmid–7.2 mg, helper plasmid–8.8 mg, total 20 mg of DNA per 1 L of HEK293 medium with final 4 mM ultraglutamine) and (2) the PEIproHQ mix (60 mg of PEI per 1 L of HEK293 medium with final 4 mM ultraglutamine). These were combined and incubated for 10–15 min at room temperature. Finally, a volume equating to 10% of the production volume was added to each of the wave bags to be transfected (2 L of DNA-PEIproHQ complexes was added to 18 L of culture). Three days post-transfection, the cells were harvested, lysed with lysis buffer containing 300 mM NaCl, 50 mM Tris–Cl, pH8.0, and 0.1% PF68. To remove the residual DNA, the lysate was incubated for 1 h at 37 °C after adding a nuclease. Then, we purified the AAV product using tangential flow filtration (TFF) and affinity chromatography, and the neutralized affinity chromatography eluate material was solubilized in Tris-based buffer solution. In consideration of the effect of the AAV formulation on safety, the Tris-based buffer was changed to PBS-based buffer (PBS with 300 mM NaCl and 0.001% PF-68) by the 2nd TFF diafiltration to produce high-titer virus vector stock. Part of the formulated (diafiltrated) viral particles were concentrated using 100 kDa Amicon Centricon™ and the concentrated products were sterilized by filtration using a 0.22 μm syringe filter (Satorius). A final product (post purification) viral genome (VG) titer of 4.05 × 1013 VG (1.62 × 1013 VG/mL, by qPCR) was obtained with this process.
RNA library preparation, sequencing, and data analysis
AAV-GFP or -DX2 was administrated to N2A and SK-N-SH cells. After 3 days, RNA was isolated using TRIzol Reagent (Thermo Fisher Scientific Inc., MA, USA). For control and test RNAs, the construction of the library was performed using QuantSeq 3′ mRNA-Seq Library Prep Kit (Lexogen Inc., VIE, Austria) according to the manufacturer’s instructions. QuantSeq 3′ mRNA-Seq reads were aligned using Bowtie2 [
37]. Bowtie2 indices were either generated from genome assembly sequence or the representative transcript sequences for aligning to the genome and transcriptome. The alignment file was used for assembling transcripts, estimating their abundances, and detecting differential expression of genes. Differentially expressed genes were determined based on counts from unique and multiple alignments using coverage in Bedtools [
44]. The RC (Read Count) data were processed based on the quantile normalization method using EdgeR within R using Bioconductor [
21]. Gene category was based on searches done in DAVID (
http://david.abcc.ncifcrf.gov/) and Medline databases.
Surgical procedures
All experiments in mice were performed under the Kyung Hee University Institutional Animal Care & Use Committee guidelines (KHUASP(SE)-18-101). In the 6-OHDA group, male c57bl/6n mice aged 8 weeks were injected with 4 μg/μl of 6-OHDA in the right striatum. Each mouse was ethically anesthetized using ketamine and a muscle relaxant (Rompun). The anesthetized mice were then placed on the stereotaxic device with tooth and ear bars. The total volume of 6-OHDA per mouse was 3 μl. The injection was performed using a Hamilton syringe at the following coordinates: AP: + 0.5 mm, ML: + 1.8 mm, DV: − 3.7 mm. To evaluate the efficacy of DX2 in the 6-OHDA mice model, AAV-DX2 and its control virus AAV-GFP were injected on both sides of the lesion. The injection was performed using a Hamilton syringe at the following coordinates: coordinates 1: AP: + 1.2 mm, ML: + 1.5 mm, DV: − 3.5 mm; coordinates 2: AP: − 0.1 mm, ML: + 2.1 mm, DV: − 3.2 mm. All stereotaxic injections were performed at a rate of 0.5 μl/min and the needle was removed out of place for another 10 min after the injection before slowly being removed.
Animal behaviour test in the cylinder
Every mouse in the 6-OHDA model group was behaviourally tested by using a cylinder with a total diameter of 11 cm and height of 20 cm. The blind test method was used through all experiments and analysis. Behaviour tests were conducted at different points after the injection of AAV.
-
Apomorphine-induced rotation test: Apomorphine-induced rotation test was conducted at 2, 6, and 10 weeks post-injection of 6-OHDA. To avoid possible hypersensitization, an interval of a minimum of 3 to 4 weeks was chosen. Each mouse received 0.5 mg/kg of apomorphine hydrochloride via subcutaneous injection (Sigma-Aldrich) and then was placed in the individual cylinder described above. Mice were allowed to adjust to their environment for 5–10 min before any turns ipsilateral or contralateral to the lesion site. All procedures were recorded for 35 min. Results were organized by net contralateral rotations per minute.
-
Cylinder test: Forelimb movement was analyzed by the cylinder test using a modified mouse version [
47]. Every mouse in the 6-OHDA group was tested using a cylinder (11 cm in diameter and 20 cm in length). All analyses were performed by an experimenter blinded to the group. Behaviour tests were conducted at different points after the AAV injection. The cylinder test was conducted at 2, 6, and 10 weeks post-injection of 6-OHDA at the onset of the dark cycle. Each mouse was placed in the plastic cylinder described above.
-
Elevated body swing (EBS) test: The EBS test was performed after the cylinder test in the 6-OHDA model. Each mouse was held 3 cm vertically above the ground for 60 s. The swinging pattern of each mouse was recorded and marked whenever the mouse turned its head over 30 degrees from the axis. Each test was conducted with two examiners: one was holding the mouse while the other recorded the movement. Each test was recorded as a ratio of turns to the total number of swings.
Pole test
The mouse was placed on top of a pole, and the duration for which the mouse stayed on the pole was measured. Since mice with deficits in motor activity will likely fall off the pole and mice with improved motor activity will descend the pole, an increase in the time in pole descent indicates an improvement in motor activity
Rotarod test
To examine the motility of the rotenone mouse models, a rotarod test was conducted at 3 and 15 days post lesion. All mice were trained on the same machine with equal rotating speed to show a stable performance. Training procedures consisted of three sessions on 2 days in which each session included three individual trials, lasting at least 200 s each. All mice were trained at 5, 10, and 15 rpm (revolutions per minute). The final test was conducted at 15 rpm (three sessions, 300 s each) on the third day. To reduce fatigue, each mouse was given at least 5 min of rest between each measurement.
Tissue preparation
At the end of the behavioural assessment, the total duration was 12 weeks for the 6-OHDA model. Mice were sacrificed after being anesthetized with isoflurane and perfused with 0.9% saline mixed with 4% paraformaldehyde (PFA; Yakuri Pure Chemicals) in phosphate buffer (PBS) at a pH of 7.4. Organs including the lungs, heart, liver, kidney, ovary, spleen, bladder, spinal cord, and brain were collected. The brains were fixed overnight in 4% PFA followed by a dehydration step with 30% sucrose in PBS. Brain samples were then frozen with optimum cutting temperature compound (O.C.T. compound; Tissue-Tek) and cut on a cryotome (Leica CM1850; Leica Biosystems) into 30 μm thick coronal sections through the entire substantia nigra and striatum. All sections were made from the posterior end of the SN to the anterior end of the striatum. The sections were stored at − 20 °C in a cryoprotectant solution made with ethylene glycol, glycerol, and 0.2 M phosphate buffer.
Immunofluorescence staining
For the staining, the floating brain tissues in cryoprotectant solution or attached cells on a coverslip were moved to a slide glass. After washing with PBS three times for 5 min, blocking with 5% BSA in PBS was performed for 30 min followed by permeabilization with 0.2% triton × 100 in PBS. The primary tyrosine hydroxylase antibody was 1:500 diluted with 1% BSA in PBS, applied to the section/cells, and incubated at room temperature for 1 h. After three washes with PBS-T (0.1% tween in PBS) for 5 min, the secondary antibody diluted with 1% BSA in PBS was incubated in a dark room for 1 h. Mounting with VECTASHILD antifade mounting medium with DAPI was used for nucleus counterstaining.
Image analysis
Images were analyzed with Image J (
http://rsb.info.nih.gov/ij) using the Time Series Analyzer plugin, accessible at
https://imagej.nih.gov/ij/plugins/time-series.html. Synaptic boutons of neurons were appointed as oval regions of interest (diameter, 10 pixels), and the amplitude of pHluorin-based fluorescence at synapses was counted using Origin Pro 2020. The kinetics of endocytosis was measured using a single exponential decay function.
Statistics analyses
The p values are represented for mean ± SEM and data were statistically analyzed using Student's t-test or ANOVA, where appropriate. A p-value of less than 0.05 was considered to be statistically significant.
Discussion
The most important finding of this study was that DX2, a splice variant of AIMP2 [
11], can effectively turn off PARP-1 activation. The DX2 transgene encoded by rAAV2 vector system, competed with AIMP2, bound to PARP-1 more strongly than AIMP2, but did not induce its overactivation, resulting in neuronal cell survival. Over the past decade, PARP-1 inhibitors have been explored in a few clinical trials to treat ischemia [
51]. However, it still has a technical limitation in effectively accessing the central nervous system. Since PARP-1 activation is critical for sensing and recovery of DNA damage [
56], systemic or long-term treatment with PARP-1 inhibitors may cause serious side effects, like teratogenicity and anaemia [
36,
57].Thus, spatiotemporal or conditional inhibition would be preferred in clinical interests. DX2, a competitive antagonist of AIMP2, could be an alternative for inhibiting the PARP overactivation pathway, since AIMP2 is an upstream regulator of PARP-1 in the cell death regulation and AIMP2 accumulation results in PARP-1 overactivation and cell death, even without DNA damage [
38].
Thus, using DX2 for selective inhibition of PARP-1 overactivation, only when AIMP2 accumulation is present, would be a smart strategy for treating PD effectively while avoiding unnecessary adverse effects due to the use of PARP inhibitors.
Chronic diseases, such as PD, need long-term intervention. Many allopathic medications are applicable for patients with PD, but all of them have considerable disadvantages in controlling the symptoms. Levodopa, one of the representative drugs for the treatment of PD [
41,
43], has many side effects, such as the end-of-dose deterioration of function or on/off oscillations, especially in patients on chronic levodopa therapy [
8,
16]. To overcome the side effects of levodopa, many other drugs, including dopamine agonists, MAO-B inhibitors [
25,
28], and COMT inhibitors [
6,
54] are used; however, they also have other side effects. To treat patients with PD more efficiently, novel drug targets with fewer side effects than conventional drugs need to be developed; ideally, the new drug should assist the survival of dopaminergic neurons [
24], besides dopamine production or secretion. In this study, DX2 was shown to improve motor activity (Figs.
4, 7 and 8) and rescue dopaminergic neuronal cell death (Figs. 7e and 8e).
Parkin is frequently mutated or inactivated in patients with sporadic and familial PD [
1,
32]. In previous reports, AIMP2 had been reported as a pro-apoptotic protein ubiquitinated by Parkin E3 ubiquitin ligase, and to be highly expressed in patients with PD [
38]. Malfunctional PD can cause AIMP2 protein to accumulate in cell, leading to aberrant cell death. A portion of the alleviated AIMP2 seems to be located in the nucleus to bind to PARP-1 [
38], hence inducing parthanatos. DX2, lacking exon 2, functions as a competitive antagonist of AIMP2 [
12,
23]. However, since DX2 seems to act only when extra amount of AIMP2 is present, induced in a specific stress condition (Figs.
1,
2 and
3), logically its overexpression is non-oncogenic and it barely disturbs the tumour suppressive function of AIMP2 in normal condition (Fig.
6b and Additional file
1: Fig. S5) [
10].
Although majority of our study confirmed the preventive effect of DX2 on 6-OHDA-induced mouse model, the concept of prevention and therapy in the PD-induced mouse can be used interchangeably. Even though we tested a couple of preventive concept animal models, the results indicated remarkable therapeutic application prospects. Neurodegeneration is an event caused by imbalance in the neuronal population undergoing survival and death. If, with AAV-DX2 therapy, we can increase the surviving cell population and suppress cell death, the rate of natural neuronal regeneration may be increased [
35]; delay of neurodegeneration may occur via natural neuronal regeneration [
31]. Therefore, improved motor activity and reduced dopaminergic neuronal cell death due to DX2 overexpression in the substantia nigra of PD mouse model are important observations that could be exploited as therapeutic targets. In the 6-OHDA-induced Parkinson’s disease model, since 6-OHDA does not cross the blood–brain barrier, it needs to be injected directly into the substantia nigra, medial forebrain bundle, or striatum to inhibit the nigrostriatal dopaminergic pathway [
29]. It has been reported that 6-OHDA acts specifically on monoaminergic neurons through increased ROS and quinone [
20,
46], and the formation of Lewy body is not induced in such mice [
55]. Chronic administration of rotenone causes nigrostriatal dopaminergic degeneration and Parkinson's symptoms while increasing the α-synuclein and significantly increasing Lewy body inclusions [
5]. In particular, inhibition of systemic complex I by rotenone is known to induce selective effects of oxidative stress on the nigrostriatal dopamine system compared to that in the MPTP model [
48]. Since none of the mouse models of Parkinsonism, induced by a neurotoxin, could explain PD symptoms completely, we confirmed that the chemicals effectively induced aberrant activation of PARP-1. DX2 seemed to be an effective therapeutic tool for promoting the survival of dopaminergic neuronal cells in all the three mouse models discussed above. Further, we showed the expression level of DX2 to be important for the prevention of progressive motor deficits after neuronal damage. Collectively, we found the regulation of DX2 expression to play a critical role in improving PD symptoms, strongly suggesting that DX2 is an effective drug target for PD.
Gene therapy is a therapeutic method of directly injecting a gene to treat a genetic defect, and has been widely developed as a therapeutic option for the treatment of degenerative and intractable diseases [
50]. Adeno-associated viruses are currently used for gene therapy to treat PD. Clinical trials using AAV targeting PD-inducing genes, such as AAV2-hGDNF [
30,
53] and AAV2-hAADC [
3], are currently in progress. Despite concerns about the safety of gene therapy, direct brain administration of AAV has been approved in phase I/II clinical trials of PD [
9,
14,
19].
Gene therapy may be a promising strategy for accessing CNS and treating diseases, such as Parkinson’s disease, with minimal invasiveness of intracranial micro-injection. Among the various viral vectors, adeno-associated virus serotype (AAV) is accepted as a suitable gene delivery tool, owing to the following advantages: (1) long-term and stable transgene expression in non-dividing cells following the administration of a single dose and (2) relative safety of use due to the lack of pathogenicity. More than 100 AAV serotypes are known to exist, of which, AAV2 is the most-selected for CNS gene therapy application due to its specific tropism, defined by CNS tissue-specific distribution and low-level systemic distribution [
40].
The current study is the first to report that overexpression of DX2 using AAV in the substantia nigra rescues motor activity and neuronal cell death in PD-induced mouse models. Considering the proven safety of AAV in clinical trials, we suggested that AAV-DX2 could be used as a therapeutic drug for patients with PD.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.