MicroRNA-6862 inhibition elevates sphingosine kinase 1 and protects neuronal cells from MPP+-induced apoptosis

MPP+ (1-methyl-4-phenylpyridinium)-induced dopaminergic neuronal cell apoptosis is associated with sphingosine kinase 1 (SphK1) inhibition. We here tested the potential effect of microRNA-6862 (miR-6862), a novel SphK1-targeting miRNA, on MPP+-induced cytotoxicity in neuronal cells. MiR-6862 locates in the cytoplasm of SH-SY5Y neuronal cells. It directly binds to SphK1 mRNA. In SH-SY5Y cells and HCN-2 cells, ectopic overexpression of miR-6862 decreased SphK13’-untranslated region luciferase reporter activity and downregulated its expression. miR-6862 inhibition exerted opposite activity and elevated SphK1 expression. In neuronal cells, MPP+-induced cell death was significantly inhibited through miR-6862 inhibition. Conversely, ectopic overexpression of miR-6862 or CRISPR/Cas9-induced SphK1 knockout augmented MPP+-induced apoptosis in the neuronal cells. Importantly, antagomiR-6862 failed to inhibit MPP+-induced apoptosis in SphK1-knockout SH-SY5Y cells. These results suggest that inhibition of miR-6862 induces SphK1 elevation and protects neuronal cells from MPP+-induced cell death.


INTRODUCTION
Currently, Parkinson's disease (PD) is the second most common neurodegenerative disease [1,2]. It affects over seven million people globally [2]. PD is characterized by a progressive loss of dopaminergic (DA) neurons in the midbrain [3][4][5][6]. Sustained and unsolved oxidative stress could be one primary cause of DA neuronal cell death. It is also known as the primary pathogenesis mechanism of PD [7].
Studies have shown that MPP + serve as a main contributor of subsequent neuronal cell apoptosis as it downregulated SphK1 expression and inhibited its activity [20,21]. Conversely, SphK1 activation by exogenously adding S1P protected neuronal cells from MPP + -induced cell apoptosis [22]. Therefore, restoring SphK1expression/activity could protect neuronal cells from MPP + [20,21]. microRNAs (miRNAs) is a large family of small noncoding RNA molecules containing ~22 nucleotides [23,24]. miRNAs are capable of silencing targeted genes through a post-transcriptional mechanism [23,24]. Specifically, miRNAs function via base-pairing with the complementary sequences in the 3'-untranslated region (UTR) of targeted mRNAs/genes. This will cause the degradation and/or translation inhibition of targeted genes [23,24]. Several SphK1-targeting miRNAs have been identified, and many of them are in cancer cells [25][26][27].
Here we identifiedmicroRNA-6862 (miR-6862) as a novel SphK1-targeting miRNA. The current study is to determine whether miR-6862 would affect SphK1 expression in neuronal cells, and if so, whether miR-6862 inhibition would protect neuronal cells from MPP+induced damage.
Stable cells were established with puromycin selection. In stable cells with lv-antagomiR-6862, the mature miR-6862 expression was depleted (over 95% reduction of control cells, Figure 1K). As a result, SphK1 3'-UTR luciferase reporter activity ( Figure 1L) and SphK1 mRNA expression ( Figure 1M) were boosted. SphK1 protein elevation was detected as well ( Figure 1N) and SphK2 expression was unchanged ( Figure 1N). The lentiviral construct encoding anti-sense control sequence, lv-antagomiRC, did not alter miR-6862 and SphK1 expression ( Figure 1K-1N). These results showed that miR-6862 directly binds to and silences SphK1 in SH-SY5Y neuronal cells.

AGING
# P < 0.05 vs. MPP + treatment in "Pare" cells or "miRiC" cells. Experiments in this figure were repeated five times, with the similar results obtained.

DISCUSSION
The loss of DA neurons in the midbrain is a key pathological characteristics of PD progression, and it could be due to mitochondrial dysfunction, free radicals (ROS) accumulation, sustained oxidative injury, as well as abnormal protein aggregation and inflammatory lesions [5,29]. Of particular relevance, mitochondrial dysfunction, causing ROS accumulation and oxidative injury, is one primary cause of DA neuronal cell death [5,29].
The postmortem brain analyses and animal model studies have revealed that miRNA dysregulation is associated with PD [30,31] and other neurodegenerative disorders such as Alzheimer's disease and Huntington's disease [32,33]. Understanding miRNAs and their targeted genes in neurodegenerative diseases should help to identify novel genes for neuronal functions and to reveal the underlying molecular mechanisms underlying these diseases [32,33]. Since miRNAs regulating genes and pathways that are associated with PD are altered in PD, it is useful in the diagnose of PD and even different PD subtypes [30,31]. Furthermore, miRNAs are important in regulating neuronal cell death by MPP + .
Importantly, CRISPR/Cas9-induced SphK1 KO augmented MPP + -induced neuronal cell death, which KO construct ("koSphK1" cells) were further transduced with or without the lentiviral construct encoding the anti-sense of premiR-6862 (lv-antagomiR-686), control cells were transduced with lentiCRISPR-GFP empty vector ("Cas9-C"); Cells were treated with or without MPP + (3 mM) and then cultured for applied time periods, SphK1 mRNA and protein expression was tested (A, B); Cell viability and death were tested by CCK-8 (C) and LDH release (D) assays, respectively; Cell apoptosis was examined by nuclear TUNEL staining assay (E). miR-6862 expression was shown (F). Bars stand for mean ± standard deviation (SD, n=5). * P < 0.05 vs. "Ctrl" treatment in "Cas9-C" cells. # P < 0.05 vs. MPP + treatment in "Cas9-C" cells. "n.s." stands for non-statistical difference. Experiments in this figure were repeated five times, with the similar results obtained. AGING mimicked miR-6862 overexpression-induced activity. Significantly, in koSphK1 SH-SY5Y neuronal cells, exogenously miR-6862 inhibition by lv-antagomiR-6862 failed to alleviate MPP + -induced apoptosis. These results indicated that SphK1 should be the primary target of miR-6862 in neuronal cells. Moreover, miR-6862 inhibition-induced neuronal cell protection against MPP + should be through elevating SphK1 expression.
However, the findings of antagomiR-6862-induced neuronal cell protection in vitro cannot be simply translated in vivo. The in vivo experiments are regarded as the gold standard for PD studies [39]. The potential effect of antagomiR-6862 remains to be evaluated in MPTP-induced PD animal model and possibly in genetic-based PD animal models. Also, only established neuronal cell lines, SH-SY5Y and HCN-2, were tested in this study. Our findings have to be further verified in primary DA neurons. In addition, miR-6862 expression was unchanged in MPP + -treated cells. Further exploring the expression of miR-6862 in PD postmortem brain tissues and PD animal models would be necessary.
In summary, we identified miR-6862 as a novel SphK1targeting miRNA. miR-6862 inhibition upregulated SphK1 and protected neuronal cells from MPP + -induced damage. Therefore, SphK1 elevation induced bymiR-6862 inhibition could be a novel genetic strategy to protect DA neurons against oxidative injury.

Reagents and antibodies
MPP + , puromycin, polybrene and CCK-8 assay kit were obtained from Sigma-Aldrich (St. Louis, Mo). The antibodies were all provided by Cell Signaling Tech (Danvers, MA) and Abcam (Shanghai, China). Lipofectamine 2000 and other transfection reagents were provided by Thermo-Fisher Invitrogen (Shanghai, China). All the primers, sequences, and viral constructs were provided by Shanghai Genechem Co. (Shanghai, China) unless otherwise mentioned.

Cell culture
Culturing the differentiated SH-SY5Y neuronal cells was reported before [10,40]. The HCN-2 neuronal cell line was purchased from the Cell Bank of Shanghai Institute of Biological Science (Shanghai, China). HCN-2 cells were cultured in DMEM with 10% FBS.MPP + was dissolved in PBS to obtain 30 mM MPP + stock solution, and was then filtrated and stored in dark at -20° C. Neuronal cells with applied genetic modifications were treated with MPP + (3 mM). This concentration was chosen based on our previous studies [10,41].

Transfection of miR mimic
SH-SY5Y and HCN-2 neuronal cells were cultured into six well-tissue plates (at 2 × 10 5 cells per well) and transfected with 500 nM of the applied miR mimic or miR inhibitor through a Lipofectamine 2000 protocol [42].

RNA-Pull down assay
RNA-Pull down was analyzed by a Pierce Magnetic RNA Pull-Down Kit [43,44]. Briefly, SH-SY5Y neuronal cells were cultured into six well-tissue plates (at 2 × 10 5 cells per well) and transfected with biotinylated miR-6862 mimic (100 nmol/L) for 48h [44]. The streptavidin-coated magnetic beads were added to total cell lysates to pulldown biotin-captured miR-6862-bound RNA complex [43]. SphK1 mRNA expression was tested by qPCR with level normalized (% of input controls).

Cell viability
Neuronal cells were plated into 96 well-tissue plates (at 3, 000 cells per well). Following the applied MPP + treatment, a CCK-8 assay kit was utilized to test cell viability with the attached protocol. CCK-8's optical density (OD) was tested at the test-wavelength of 550 nm.
AGING Quantitative real-time reverse transcriptase polymerase chain reaction (qPCR) The detailed protocols of qPCR were described in our previous study [10]. Briefly, TRIzol reagents (Sigma) were utilized to extract total cellular RNA, which was then reversely transcribed to cDNA. A TOYOBO ReverTra Ace qPCR kit (Tokyo, Japan) was applied for qPCR under the ABI Prism 7500H fast Real-Time PCR system (Foster City, CA). The methods for mRNA data quantification and normalization to GAPDH were described before [10]. A TransStartTM SYBR Green qPCR Supermix (TransGen Biotech) was utilized to examine miR-6862 expression, and the results were normalized to U6 RNA. The mRNA primers were listed in Table 2.

Lactate dehydrogenase (LDH) assay
Neuronal cells were plated in six well-tissue plates. Following the applied MPP + treatment, LDH contents were tested by a simple two-step LDH detection kit (Promega, Shanghai, China). LDH contents in the medium were always normalized to total LDH contents.

Western blotting
Neuronal cells were plated into six well-tissue plates (at 2 × 10 5 cells per well). Following the applied treatment, cell lysates were obtained. Detailed protocols of Western blotting were described before [10]. Data quantification was performed through ImageJ software (NIH).

Caspase activity
Neuronal cells were plated into six well-tissue plates (at 2 × 10 5 cells per well). Following the MPP + treatment, cell lysates were obtained. The relative activities of caspase-3 and caspase-9 were tested by fluorometric caspase assay kits (Beyotime Biotechnology, Wuxi, China) [45] with 20 μg lysates per sample. The caspase-3 or the caspase-9 p-nitroaniline (pNA) absorbance was detected at the wavelength of 405 nm.
Single strand DNA (ssDNA) ELISA assay. Neuronal cells were seeded into six-well plates. Following the applied MPP + treatment, ssDNA contents were tested using a ApoStrandTM ELISA kit (BIOMOL International, Plymouth Meeting, PA).

Mitochondrial depolarization assay
Neuronal cells were plated into 96 well-tissue plates (at 3, 000 cells per well). Following the applied MPP + treatment, cells were stained with the mito-dye JC-1. JC-1 green fluorescence intensity (at 488 nm) was recorded. The representative JC-1 images were presented as well.

SphK1 knockout (KO)
A lentiCRISPR-GFP construct encoding the small guide RNA (sgRNA) against human SphK1 was provided by Dr. Yao at Nanjing Medical University [25]. SH-SY5Y neuronal cells were cultured into six well-tissue plates and transfected with SphK1-KO construct. FACS-mediated GFP sorting were utilized to select transfected cells. Cells were then distributed into 96-well plates. Single stable cells with SphK1 KO were further screened by qPCR. Control SH-SY5Y cells were transfected with the empty vector [25].

Statistics
Data were presented as mean ±standard deviation (SD). Statistical differences were analyzed by one-way analysis of variance (ANOVA) followed by multiple comparisons performed with post hoc Bonferroni test (SPSS, 23.0, SPSS Co. Chicago, CA). A two-tailed unpaired T test was utilized (Excel 2007) when comparing the difference between two treatment groups. Values with P < 0.05 were considered statistically significant.

AUTHOR CONTRIBUTIONS
GX performed all neuronal culture and signaling studies. YL helped to established and verified constructs virus and sequences. JC performed all cell death studies. ZZ, JZ and WD: Study conception and design, and data analysis, Figure organization, involved in drafting the article and revising it critically for important intellectual content, and with final approval of the version submitted to the journal.

CONFLICTS OF INTEREST
None of the authors have any conflicts of interest.