Neuroprotection of Indole-Derivative Compound NC001-8 by the Regulation of the NRF2 Pathway in Parkinson's Disease Cell Models

Parkinson's disease (PD) is a common neurodegenerative disease accompanied by a loss of dopaminergic (DAergic) neurons. The development of therapies to prevent disease progression is the main goal of drug discovery. There is increasing evidence that oxidative stress and antioxidants may contribute to the pathogenesis and treatment of PD, respectively. In the present study, we investigated the antioxidative protective effects of the indole-derivative compound NC001-8 in DAergic neurons derived from SH-SY5Y cells and PD-specific induced pluripotent stem cells (PD-iPSCs) carrying a PARKIN ex5del mutation. In SH-SY5Y-differentiated DAergic neurons under 1-methyl-4-phenylpyridinium (MPP+) treatment, NC001-8 remarkably reduced the levels of reactive oxygen species (ROS) and cleaved caspase 3; upregulated nuclear factor erythroid 2-related factor 2 (NRF2) and NAD(P)H dehydrogenase, quinone 1 (NQO1); and promoted neuronal viability. In contrast, NRF2 knockdown abolished the effect of NC001-8 on the reduction of ROS and improvement of neuronal viability. In H2O2-treated DAergic neurons differentiated from PD-iPSCs, NC001-8 rescued the aberrant increase in ROS and cleaved caspase 3 by upregulating NRF2 and NQO1. Our results demonstrated the protective effect of NC001-8 in DAergic neurons via promoting the NRF2 antioxidative pathway and reducing ROS levels. We anticipate that our present in vitro assays may be a starting point for more sophisticated in vivo models or clinical trials that evaluate the potential of NC001-8 as a disease modifier for PD.

Indole and its derivative compounds with an aromatic heterocyclic structure have been recently used in studies on neurodegenerative diseases [17][18][19][20][21][22][23]. Endogenous melatoninrelated indole-3-propionic acid exhibited an effect against ROS damage and cell death resulting from the addition of Aβ peptide on SK-N-SH human neuroblastoma cells and primary rat hippocampal neurons [20]. Another endogenous indole derivative, indolepropionamide, was also shown to prolong rotifer lifespan and recover the mitochondrial metabolic function in rodents by decreasing the generation of free radicals [21]. We have previously demonstrated that the indole derivative NC001-8 is able to mitigate oxidative stress and polyglutamine (polyQ) aggregation by upregulated chaperones and/or autophagy in spinocerebellar ataxia-type (SCA) cell models [22,23]. In the present study, we examined the effect of NC001-8 in MPP + -treated DAergic neurons derived from SH-SY5Y cells, focusing particularly on the maintenance of neuronal survival and reduction of ROS by increasing NRF2 and NQO1 expression. Furthermore, NC001-8 neuroprotective effects were also explored in DAergic neurons derived from induced pluripotent stem cells (iPSCs) carrying a PARKIN ex5del mutation (PD-iPSC) [24]. Our findings demonstrated that NC001-8 reduced oxidative stress via the enhancement of the NRF2 antioxidative pathway, suggesting the potential of NC001-8 in treating PD.

Extraction of RNA and Profiling of Relevant ROS
Genes by the q-PCR Array. Total RNA was isolated using the Trizol reagent (Invitrogen). Reverse transcription (RT) was performed by superscript III (Invitrogen) with an initial concentration of 5 μg total RNA. We established an in-house human panel for ROS profiling analysis with real-time qPCR using SYBR green reagents (Applied Biosystems). The gene entities of the q-PCR array are listed in the supplementary table (available here). The thermocyclic conditions were as follows: 50°C for 2 min, 95°C for 10 min, 95°C for 15 s, and 60°C for 1 min for 40 cycles, which were through the ABI 7900 HT RT-PCR system (Applied Biosysple was assessed in triplicate. Relative expression values were normalized to β-actin. Relative gene expressions were calculated using the 2 -ΔΔCT method, ΔC T = C T ðtarget geneÞ -C T ðβ − actinÞ, in which C T indicates cycle threshold (the fractional cycle number where the fluorescent signal reaches detection threshold). Primer sequences of 4 validated genes and the endogenous control are as follows: NRF2 "CATGCCCTCACCTGCT ACTT (forward)" and "TGTTCTGGTGATGCCACACT (reverse)," NQO1 "TTACTATGGGATGGGGTCCA (forward)" and "TTTCAATGCACCACAAGAGG (reverse)," SOD2 "GATGTGCATCAAGCCTGGTA (forward)" and "TGCAGACCTCTTTGATGGTTG (reverse)," GCLM "TC CTACCTGCACCCTCAACT (forward)" and "TGTGAA CATCAGCCTGGAAA (reverse)," and β-actin "TCCCTG GAGAAGAGCTACGA (forward)" and "AGCACTGTGTT GGCGTACAG (reverse)." 2.4. Small Interfering RNA Transfection. The SH-SY5Y cells were transfected with siRNA for NRF2 (Sigma-Aldrich) using the Lipofectamine® RNAiMAX Transfection Reagent (Invitrogen) according to the manufacturer's instructions. Furthermore, the expression of NRF2 was confirmed by qRT-PCR and Western blotting after 48 hours of transfection.

Immunofluorescent Staining and Measurement of Neurite
Outgrowth. The cells cultured on coverslips were washed with phosphate-buffered saline (PBS, Invitrogen) and fixed with 4% paraformaldehyde (Sigma-Aldrich) for 10 minutes at room temperature. After three rinses (10 minutes each) with PBS containing Tween 20 (Invitrogen) (PBST), the cells were incubated in blocking solution consisting of PBST and 10% BSA (Sigma-Aldrich) for 30 minutes at room temperature. Subsequently, the samples were hybridized with the primary anti-TH (1 : 500, Millipore) and TUBB3 (1 : 500, Biolegend) antibodies in blocking solution overnight at 4°C. After three rinses (10 minutes each) with PBST, the cells were incubated with the diluted secondary antibody conjugated with Alexa 594 or Alexa 647 (Thermo Fisher Scientific) in blocking solution in the dark for 1 hour. The cells were counterstained with 4 ′ -6-diamidino-2-phenylindole (DAPI, 1 : 1 000; Thermo Fisher Scientific) for nuclear detection. Subsequently, the coverslips were mounted with DAKO mounting solution onto microscopic slides. The cells were observed using a Leica TCS confocal microscope. The neurite outgrowth features (TH-positive) including total outgrowth, processes, and branches were counted randomly with more than 500 cells and assessed by MetaMorph microscopy automation and using the image analysis software (Molecular Devices).

Trypan
Blue Cell Viability Assay. The cell viability was determined by light microscopy. Cells that were not stained with trypan blue were considered viable.
2.9. LDH Assay. On day 14 after neuronal differentiation, cells were cultured with 100 μl of LDH reaction mixture based on the manufacturer's instructions (Roche) and incubated at room temperature for 20 minutes. The samples' absorbance was set at 490 nm, and reading was performed by using a microplate spectrophotometer. Sigma-Aldrich), and 5 mM dithiothreitol (DTT; Sigma-Aldrich) on ice for 20 minutes. After centrifugation, the supernatants containing proteins were quantified to evaluate the caspase 3 activity using Ac-DEVD-AMC as a fluorogenic substrate. The fluorescence wavelength of the AMC reading was set at 360 nm excitation in conjunction with a 460 nm emission filter.
2.11. Assessment of ROS. The cells were incubated at 37°C for 60 minutes in the Fluorogenic CellROX™ Deep Green Reagent (5 μM, Molecular Probes), which is designed to measure ROS reliably in live cells. Subsequently, the cells were washed with PBS and the ROS levels were evaluated by measuring green fluorescence using a Leica TCS confocal microscope, with excitation/emission wavelengths of 488/520 nm. The cells were costained with TH, and only TH-positive cells were counted and analyzed.

Statistical
Analyses. Data were presented as the means ± SD from three different passages and analyzed using Student's t-test or one-way analysis of variance (ANOVA) with Bonferroni's post hoc test. All statistical analyses were performed using the SPSS statistical software (version 18.0, IBM). Hierarchical clustering analysis was performed using the free academic software Cluster 3.0 (http://bonsai.hgc.jp/ mdehoon/software/cluster/software.htm#ctv). Significant statistical differences were considered at p < 0:05.

Derivative NC001-8 Demonstrates No Neurotoxicity in
DAergic Neurons. We applied 12-O-tetradecanoyl-phorbol-13-acetate (TPA) to induce the differentiation of SH-SY5Y cells into DAergic neurons [26]. The SH-SY5Y cells were cultured in a medium containing 120 nM TPA for 2 weeks (Figure 1(a)). On day 14, the cells had obviously changed morphologically and developed long neuritic processes. They also expressed the neuronal marker TUBB3 and the DAergic neuronal marker tyrosine hydroxylase (TH) (Figure 1(b)).
As an indole-derivative compound with an aromatic structure (Figure 1(a)), NC001-8, at a concentration of 100 nM, has demonstrated its neuroprotective potential in  SCA cell models [22,23]. Before examining the neuroprotective potential of NC001-8 in PD, we first evaluated its neurotoxicity in DAergic neurons. As shown in Figure 1(c), NC001-8 (concentration needed to be less cytotoxic for maintaining 90% of growth: 100 nM) was added to the medium during DAergic differentiation. After 14 days of differentiation, cells treated with NC001-8 also displayed the properties of DAergic neurons, including profound outgrowth of neurites and expression of TH and TUBB3 (Figure 1(d)). Treatment with NC001-8 did not affect the proportion of TH-positive cells (Figure 1(e)) and the level of neurite outgrowth (Figure 1(f)). These results suggested that NC001-8, at 100 nM, is not toxic to neurons and did not affect DAergic differentiation.
3.3. Pretreatment with NC001-8 Potentiates the NRF2 Antioxidative Pathway. To identify further key molecular targets of NC001-8 in ROS reduction and neuroprotection, we examined the expression alterations using an in-house quantitative polymerase chain reaction (q-PCR) array that carried 83 candidate genes involved in antioxidative and chaperon pathways (supplementary table) using SH-SY5Yderived DAergic neurons treated with MPP + and/or NC001-8 (Figure 3(a)). By this array, we found four genes (NRF2, NQO1, GCLM (glutamate-cysteine ligase modifier   These results indicated that NC001-8 demonstrate an antioxidative effect by enhancement of the NRF2 pathway through translocation of NRF2 into the nucleus. Although the q-PCR array showed the upregulation of NRF2 and NQO1 in the MPP + -untreated/NC001-8-treated neurons, the validation results did not demonstrate significant upregulations of NRF2 and NQO1 by pretreatment with NC001-8 in MPP +untreated neurons (Figures 3(c)-3(e)). The dose and timeresponse experiments also showed that pretreatment with NC001-8 did not alter the expression of NRF2 and NQO1 in MPP + -untreated neurons (supplementary figure).

Discussion
Currently, effective treatments that modify neurodegeneration in PD are still lacking. By using our DAergic neuronal models derived from SH-SY5Y cells and PD-iPSCs, we demonstrated for the first time the role of the indole derivative NC001-8 as an antioxidant to protect DAergic neurons. In SH-SY5Y-derived DAergic neurons, NC001-8 significantly attenuated the toxic effects of MPP + by mitigating the ROS overproduction, neuronal apoptosis, and impairment of neurite outgrowth. Similar neuroprotective effects of NC001-8 were also observed in H 2 O 2 -treated DAergic neurons derived from PD-iPSCs. These results suggested therapeutic potentials of NC001-8 in PD.
One of the critical challenges in the research on PD is the lack of live DAergic neurons from patients for mechanistic studies and new drug discovery. The iPSCs derived from patients with PD can recapitulate disease phenotypes to serve as a platform for testing new potential therapeutic strategies [24,[36][37][38]. The PD-iPSC-derived DAergic neurons demonstrated low expression levels of NRF2 and NQO1, in addition to a higher susceptibility to the environmental stressor such as H 2 O 2 , thus serving as a good model to evaluate the effects of NC001-8. Treatment with NC001-8 reduced the overproduction of ROS and expression of cleaved caspase 3 by H 2 O 2 toxicity and upregulated NRF2 and NQO1 in PD-iPSCderived DAergic neurons, suggesting its neuroprotective and antioxidative potentials in DAergic neurons from patients with PD. Similarly, we have previously demonstrated that genipin improved the abnormal susceptibility of PD-iPSC-derived DAergic neurons to H 2 O 2 treatment by activating the NRF2 pathway [24]. Thus, strategies to induce the expression of antioxidative genes such as NRF2 or NQO1 could be a viable approach to develop neuroprotective therapies for PD.
Although we have shown that NC001-8 exerted neuroprotection via the enhancement of the NRF2 antioxidative pathway, pleiotropic effects of this compound may also lead to its neuroprotective effects. For example, NC001-8 also attenuated oxidative stress and polyQ-mediated neurotoxicity by upregulating chaperones including HSF1, HSPA1A, and HSP70 [22,23]. However, our q-PCR array did not reveal these expression alterations in MPP + -treated DAergic neurons. The molecular mechanism to control NRF2 expression is still not well understood, although KRAS, BRAF, and MYC have been reported to upregulate NRF2 expression via binding to the NRF2 promoter site [39]. Future genome-wide expression studies are warranted to explore whether there is any other underlying mechanism contributing to the neuroprotective effects of NC001-8 in PD.
Our results also indicated that NC001-8 exerts its neuroprotective effects only when applied before MPP + treatment Above these results were presented as the means ± SD from three different passages; * p < 0:05. NC-iPSC: induced pluripotent stem cells derived from a healthy volunteer; PD-iPSC: induced pluripotent stem cells carrying a PARKIN ex5del mutation; GAPDH: glyceraldehyde 3-phosphate dehydrogenase; NRF2: nuclear factor erythroid 2-related factor 2; NQO1: NAD(P)H dehydrogenase, quinone 1; ROS: reactive oxygen species.
in SH-SY5Y-derived DAergic neurons (pretreatment), suggesting a concept for an early window of therapeutic intervention in PD before exposure to environmental hazards. However, the clinical presentations of PD appear when approximately 50% of the DAergic neurons are lost [40]. Identification of diagnostic markers for preclinical PD is crucial to ensure effective intervention during a limited period early in the course of the disease for preventing subsequent disease progression.
In conclusion, our in vitro study provided evidence that the indole derivative NC001-8 could be a novel compound for PD treatment through the activation of the NRF2 antioxidative pathway. However, the NC001-8 neuroprotective effects and disease-modifying potentials should be further validated in vivo in PD animal models. The heterogeneous nature of PD may also limit the generalization of this neuroprotective strategy to all patients with PD. In the future, development of personalized medicine may be valuable for identifying possible patient candidates responsive to this therapeutic strategy.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.