Advertisement

Molecular Neurobiology

, Volume 54, Issue 2, pp 1577–1586 | Cite as

Altered Expression of the Long Noncoding RNA NEAT1 in Huntington’s Disease

  • Jun-Sang Sunwoo
  • Soon-Tae Lee
  • Wooseok Im
  • Mijung Lee
  • Jung-Ick Byun
  • Keun-Hwa Jung
  • Kyung-Il Park
  • Ki-Young Jung
  • Sang Kun Lee
  • Kon Chu
  • Manho Kim
Article

Abstract

Huntington’s disease (HD) is a devastating neurodegenerative disease caused by cytosine-adenine-guanine trinucleotide repeat expansion in the huntingtin gene. Growing evidence supports the regulatory functions of long noncoding RNAs (lncRNAs) in the disease process, but little is known about the association between lncRNAs and neuronal death in HD. Here, we evaluated the altered expression profiles of lncRNA in HD by using microarrays. Among dysregulated lncRNAs, we focused on the upregulation of nuclear paraspeckle assembly transcript 1 (NEAT1). Quantitative PCR analysis validated increased NEAT1 levels in the R6/2 mouse brain as well as the human HD postmortem brain. To determine the biological effects of NEAT1 on neuronal survival, neuro2A cells were transfected with the NEAT1 short isoform vector and were subjected to H2O2-induced injury. Subsequently, NEAT1-transfected cells showed increased viability under oxidative stress. Our observations support the notion that NEAT1 upregulation in HD contributes to the neuroprotective mechanism against neuronal injury rather than the pathological process underlying neurodegeneration in HD.

Keywords

Huntington’s disease Long noncoding RNA NEAT1 Microarray Neuroprotection 

Introduction

Huntington’s disease (HD) is a hereditary neurodegenerative disease characterized by autosomal dominant inheritance, cognitive dysfunction, choreoathetosis, and psychiatric disturbances. A selective loss of medium spiny neurons in the caudate nucleus and putamen are pathological features typical of HD [1]. The abnormal expansion of cytosine-adenine-guanine (CAG) trinucleotide repeats located in the huntingtin gene causes the formation of mutant huntingtin proteins containing expanded polyglutamine tracts [2]. Mutant huntingtin was reported to induce neurodegeneration via several mechanisms, such as transcriptional dysregulation, impaired clearance of misfolded proteins, toxic N-terminal fragments, mitochondrial dysfunction, and oxidative stress [3, 4, 5, 6, 7]. Nevertheless, HD pathogenesis is still not completely understood, and thus, disease-modifying treatments have not yet been established in human clinical trials.

Long noncoding RNAs (lncRNAs) are transcripts that are longer than 200 nucleotides and lack the ability to code proteins [8]. Recent advances in high-throughput genomic technologies have helped determine that lncRNAs outnumber messenger RNAs (mRNAs) encoding proteins and that they are involved in various biological processes, including epigenetic regulation of gene expression, genomic imprinting, embryonic development, and cellular differentiation [9, 10]. LncRNAs employ several epigenetic mechanisms to influence gene expression, such as chromatin modification, molecular decoys, allosteric modulation of RNA-binding proteins, and transcriptional repression of sense mRNAs in a cis manner [8, 11]. Considering the regulatory and pathobiological functions of lncRNAs, aberrant expression of lncRNAs in disease states has been the subject of intense investigation in recent years. Beta-secretase-1 antisense transcript (BACE1-AS) is an lncRNA that is known to mediate the pathophysiology of Alzheimer’s disease. In vitro and in vivo experiments showed that BACE1-AS enhances BACE1 mRNA stability and amyloid beta peptide accumulation through a feed-forward regulatory mechanism [12]. In addition, altered expression of lncRNA has been reported in animal models of epilepsy as well as Alzheimer’s disease [13, 14].

Regarding HD, to date, only a few studies have addressed the implications of lncRNA dysregulation. One study found that the human accelerated region 1 (HAR1) transcript was downregulated in the striatum of HD patients and that the HAR1 locus was targeted by RE1-silencing transcription factor (REST), which contributes to the widespread transcriptional suppression of neurons in close association with HD pathogenesis [15]. However, the lack of knowledge about HAR1 transcript functions has led to limited understanding of its mechanistic relevance to the neurodegeneration seen in HD. Another study identified the antisense transcript of huntingtin (HTTAS) and showed that HTTAS negatively regulated huntingtin transcript levels depending on trinucleotide repeat length [16]. Moreover, mouse lncRNA Abhd11os (abhydrolase domain containing 11, opposite strand) was recently reported to be downregulated in the striatum of R6/2 mice and to be neuroprotective against mutant huntingtin in vivo [17]. However, little information has been obtained concerning genome-wide profiles of lncRNA expression in HD. Although a previous study explored the microarray data which formerly misannotated many lncRNAs as protein-coding transcripts [18], merely 1127 probes targeting lncRNAs were employed to identify seven differentially expressed lncRNAs in HD.

Nuclear paraspeckle assembly transcript 1 (NEAT1) is a nuclear-enriched noncoding RNA essential for the formation and maintenance of paraspeckles, which are subnuclear bodies found in mammalian cells [19, 20, 21]. NEAT1 is transcribed from the multiple endocrine neoplasia type 1 (MEN1) gene located on human chromosome 11 and has two variant transcripts of 3.7 kb (NEAT1_1) and 23 kb (NEAT1_2) in length [22]. NEAT1 was identified to be evolutionarily conserved as opposed to other lncRNAs that commonly lack sequence conservation with other species, supporting a potential biological function of NEAT1 [23]. Of note, NEAT1 gained the attention of cancer researchers because of the finding that NEAT1 dysregulation is involved in a series of cancers, such as prostate cancer, squamous cell carcinoma, high grade glioma, and colorectal cancers [24, 25, 26, 27]. NEAT1 levels were found to be significantly increased in tumor tissues and to independently predict overall survival of patients, supporting its biological function with regard to oncogenesis and tumor progression. NEAT1 was observed to modulate the epigenetic landscape for active transcription in prostate cancer [24]. In addition, in vitro experiments demonstrated that NEAT1 promotes tumor cell proliferation, migration, and invasion [25]. Growing evidence indicates the pivotal role of NEAT1 in multiple cellular processes, but little is known about NEAT1 dysregulation in HD and its functional implication in neurodegeneration.

In the present study, we hypothesized that lncRNAs are differentially expressed in HD and have harmful or protective effects on neuronal survival. To solve this question, we profiled the changes in lncRNA levels in HD using microarrays and analyzed the biological effect, while focusing on NEAT1.

Materials and Methods

Preparation of Brain Samples

Postmortem human brain samples were obtained from the NICHD Brain and Tissue Bank for Developmental Disorders at the University of Maryland (Baltimore, MD, USA). We selected five HD brain samples with a postmortem interval of not more than 10 h. Controls were matched for age, sex, and postmortem intervals (Table 1). The brain tissue used in this study (500 mg per sample) was derived from the caudate nucleus. This protocol was exempted from review by an Institutional Review Board at Seoul National University Hospital in accordance with the exemption criteria (E-1401-048-547).
Table 1

Demographic characteristics of human postmortem brain tissue

Subject

Age (years)

Sex

Ethnicity

PMI (h)

Cause of death

HD

     

1

43

Male

Caucasian

10

Complication of HD

2

55

Male

Caucasian

10

Complication of HD

3

78

Female

Caucasian

2

Complication of HD

4

82

Female

Caucasian

6

Complication of HD

5

69

Female

Caucasian

10

Chronic obstructive pulmonary disease

Control

     

1

43

Male

Caucasian

9

Drug intoxication

2

56

Male

Caucasian

10

Atherosclerotic cardiovascular disease

3

77

Female

Caucasian

8

Drowning

4

80

Female

Caucasian

10

Pneumonia

5

71

Female

Caucasian

9

Multiple medical disorders

PMI postmortem interval, HD Huntington’s disease

For the animal model of HD, we used transgenic HD mice of the R6/2 line (B6CBA-Tg (HDexon1) 62Gpb/3 J, 111 CAGs) and their wild-type littermates (The Jackson Laboratory, Bar Harbor, ME, USA). R6/2 mice were obtained by crossbreeding ovarian transplant hemizygote females with B6CBAF1/J males and were shown to express the human mutant huntingtin carrying CAG repeat expansions [28]. The mice were housed with ad libitum access to food and water under a 12 h light/12 h dark cycle. R6/2 mouse brains were isolated under deep anesthesia at 12 weeks of age, as previously described [6, 29]. The animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of Seoul National University Hospital.

Microarray Analysis

Total RNA was isolated from the caudate nucleus using TRIzol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s protocol. The synthesis of target complementary RNA (cRNA) probes and hybridization was conducted using Agilent’s Low RNA Input Linear Amplification kit (Agilent Technologies, Santa Clara, CA, USA). The transcription master mix was prepared as per the manufacturer’s protocol, using 4X Transcription buffer, 0.1 M DTT, NTP mix, 50 % PEG, RNase-Out, inorganic pyrophosphatase, T7-RNA polymerase, and cyanine 3-CTP. For the transcription of double-stranded DNA (dsDNA), we added the transcription master mix to the dsDNA reaction samples and incubated it at 40 °C for 2 h. The cRNA Cleanup Module (Agilent Technologies) was used to purify amplified and labeled cRNA. The labeled cRNA target was measured by using an ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). After checking labeling efficiency, cRNA was fragmented by adding 10X blocking agent and 25X fragmentation buffer and incubating at 60 °C for 30 min. Then, the fragmented cRNA was resuspended with 2X hybridization buffer and the sample was directly transferred onto an assembled Agilent Sureprint G3 Human GE 8 × 60 K v2 Microarray. Hybridization was performed at 65 °C for 17 h using an Agilent Hybridization oven (Agilent Technologies). The hybridized microarrays were washed as per the manufacturer’s washing protocol (Agilent Technologies). Sureprint G3 Human GE 8 × 60 K v2 Microarray contained 17,001 probes targeting lncRNAs among a total of 50,599 probes. The microarray analysis was conducted by eBiogen (Seoul, South Korea).

Data Acquisition and Analysis

We scanned the hybridized images using Agilent’s DNA microarray scanner and quantified them with Feature Extraction Software (Agilent Technologies). All data normalization and selection of genes with fold-change in expression were conducted by using GeneSpringGX 7.3 (Agilent Technologies). The averages of normalized ratios were calculated by dividing the average of normalized signal channel intensity by the average of normalized control channel intensity. The Student’s t test was used to identify significantly dysregulated genes between groups. Multiple comparisons were corrected by the Benjamini-Hochberg method with a false discovery rate (FDR) of less than 0.05. Hierarchical clustering was analyzed using MultiExperiment Viewer version 4.8.1 (http://www.tm4.org) with Pearson correlation and average linkage clustering methods.

Quantitative Reverse Transcription PCR

Extracted RNA was reverse transcribed into complementary DNA (cDNA) using Superscript III Reverse Transcriptase (Invitrogen Life Technologies, Gaithersburg, MD, USA) and random hexamer primers (Ambion-Applied Biosystems, Foster City, CA, USA). All real-time reactions were performed in triplicate on an ABI PRISM 7000 sequence detection system (Ambion-Applied Biosystems). The expression levels were normalized to those of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and calculated using the comparative threshold cycle. Sequences for the primers used in this study were as follows: human NEAT1 forward, 5′-CTTCCTCCCTTTAACTTATCCATTCAC-3′; human NEAT1 reverse, 5′-CTCTTCCTCCACCATTACCAACAATAC-3′; mouse NEAT1 forward, 5′-TTGGGACAGTGGACGTGTGG-3′; mouse NEAT1 reverse, 5′-TCAAGTGCCAGCAGACAGCA-3′; human GAPDH forward; 5′-CCATGAGAAGTATGACAACAGCC-3′; human GAPDH reverse, 5′-CCTTCCACGATACCAAAGTTG-3′;mouse GAPDH forward, 5′-GTCGTGGAGTCTACTGGTGT-3′; mouse GAPDH reverse, 5′-TGCTGACAATCTTGAGTGAG-3′. The PCR primers for NEAT1 recognize common sequences between two NEAT1 isoforms [30], and therefore, NEAT1 levels measured in this study include both NEAT1_1 and NEAT1_2 isoforms.

In Vitro HD Models and Cell Culture

Transfection of mutant huntingtin into neuro2A cells was performed as previously described [31]. Briefly, the cDNA transcripts containing the 5′ one third segment (3221 bp) of human huntingtin (a gift from Marian DiFiglia, Harvard Medical School, MA, USA) were constructed with 18 and 100 CAG repeats for wild-type and mutant huntingtin, respectively. Neuro2A cells (American Type Culture Collection, Manassas, VA, USA) were cultured in 24-well plates to reach 60–70 % confluence and then transfected with Lipofectamine 2000 (Life Technologies, Carlsbad, CA, USA) and pcDNA3-wild-type huntingtin or pcDNA3-mutant huntingtin for 3 h in serum-free OptiMEM (Invitrogen). Afterwards, the transfected cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Thermo Scientific, MA, USA) supplemented with 10 % fetal bovine serum (FBS), 100 units/mL penicillin, and 0.1 mg/mL streptomycin.

Mouse knock-in striatal neuronal cell lines were also used as a cellular model of HD. STHdh Q7/7 and STHdh Q111/7 cells were derived from the striatum of wild-type mice and knock-in HD mice, respectively (a kind gift from M. E. MacDonald, Harvard Medical School, MA, USA) [32]. STHdh cells were grown in DMEM supplemented with 10 % FBS, 2-mM glutamine, and 400 μg/mL G418.

NEAT1 Overexpression

The pCMV6-AC-GFP-NEAT1 plasmid (which encodes the open reading frame of mouse NEAT1 with a carboxy-terminal turboGFP tag; 9.6 kb, GFP-NEAT1) was manufactured by Enzynomics (Hanam-si, South Korea). Briefly, full-length mouse NEAT1_1 (3177 bp short isoform; NCBI Gene ID 66961) was amplified with the following primers: forward, 5′ GGCGGCCGGGAATTCAGGAGTTAGTGACAAGGAGGGCT-3′; reverse, 5′-CGGCCGCGTACGCGTGAAGCTTCAATCTCAAACCTTTATT-3′. Then, it was cloned into the EcoRI and MluI sites of the pCMV-AC-GFP vector. The vector was designed to express NEAT1 lncRNA and GFP reporter under the same CMV promoter. Full sequences of the vector are presented in the supplementary file. The DNA construct was quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and confirmed by sequencing. Empty pCMV-AC-GFP vector was used as a transfection control in NEAT1 overexpression experiments. Neuro2A cells (American Type Culture Collection) were transfected with pCMV6-AC-GFP-NEAT1 or control vector using Lipofectamine 2000 (Life Technologies). Transfection efficiency was measured after 48 h by detecting GFP fluorescence. Transfected neuro2A cells were collected using FACSAria II (Becton Dickinson Biosciences, San Jose, CA, USA).

H2O2 Cell Injury and WST-1 Assay

Neuro2A cells transfected with NEAT1 or the empty vector were seeded at 5 × 103 cells per well on 96-well plates with DMEM containing 10 % fetal bovine serum. Next day, cells were treated for 24 h with H2O2 used at different concentrations (0, 500, 1000, and 2000 μM). Then, cell viability was measured using the WST-1 assay (Roche, Basel, Switzerland) according to the manufacturer’s protocols. The relative viability of each group was normalized to the viability of the neuro2A cells that were not treated with H2O2.

Statistical Analysis

The relative expression levels and cell viability were analyzed between groups by the Mann–Whitney U test using SPSS Statistics version 22 (IBM corp., Armonk, NY, USA). Two-tailed p values less than 0.05 were considered statistically significant. For microarray data, p values were adjusted with FDR of less than 0.05 by using the Benjamini-Hochberg method. Upregulation of lncRNAs was defined as a >1.5-fold increase, while downregulation was defined as a <0.67-fold decrease compared with controls.

Results

Differentially Expressed lncRNAs in HD

We measured expression profiles of lncRNAs in the human HD brain compared to the control brain by using microarrays. After removing bad spots, we identified 181 differentially expressed lncRNAs: 35 were upregulated (fold change > 1.5) and 146 were downregulated (fold change < 0.67). Hierarchical clustering analysis of dysregulated lncRNAs demonstrated separate clustering in HD and control samples (Fig. 1a). XLOC_l2_010636 (probe ID: A_21_P0012722) was the lncRNA showing the most increase in HD, with a fold change of 8.45, while LOC100507043 (A_21_P0000821) showed the most decrease, with a fold change of 0.069. Table 2 summarizes the top 20 lncRNAs showing the greatest upregulation or downregulation in HD.
Fig. 1

Increased NEAT1 expression in the brain of individuals with Huntington’s disease (HD). a Microarray analysis shows differentially expressed long noncoding RNAs in human HD brain samples (absolute fold change >1.5 or <0.67, adjusted p value <0.05, n = 4 per group). The colors correspond to a normalized expression level of each transcript, where red indicates upregulation and green indicates downregulation. b Quantitative reverse transcription PCR (qRT-PCR) validated upregulated NEAT1 expression in the human HD brain (n = 5 per group). c Increased expression of NEAT1 in the R6/2 mouse brain was confirmed by qRT-PCR (n = 4 per group). Human and mouse GAPDH were used as an internal control. Bars show mean ± SEM. *p < 0.05. Abbreviations: HD Huntington’s disease

Table 2

The 20 most upregulated and downregulated lncRNAs in the human Huntington’s disease brain

Upregulated

Downregulated

Gene symbol

FC

p value

Gene symbol

FC

p value

XLOC_l2_010636

8.454

3.31E-02

LOC100507043

0.069

2.52E-02

XLOC_l2_010636

7.653

4.85E-02

ANKRD20A5P

0.117

2.84E-02

XLOC_001246

6.731

4.87E-02

XLOC_013980

0.130

2.86E-02

XLOC_004512

4.157

4.90E-02

XLOC_l2_003881

0.148

3.74E-02

LOC221442

3.349

4.01E-02

C21orf81

0.154

2.30E-02

XLOC_002192

3.134

2.50E-02

DGCR9

0.154

2.97E-04

XLOC_004361

3.067

1.76E-02

XLOC_l2_009292

0.165

4.83E-02

XLOC_012025

2.986

2.84E-02

LOC100507278

0.181

4.90E-02

XLOC_l2_002767

2.939

4.00E-02

XLOC_000107

0.192

4.28E-02

IFITM4P

2.759

2.36E-02

LOC646627

0.192

3.39E-02

LOC100652997

2.706

2.51E-02

LOC283070

0.202

2.83E-02

LOC100131096

2.538

3.12E-03

XLOC_l2_009136

0.210

2.83E-02

XLOC_009509

2.445

3.32E-02

LOC100506502

0.215

3.36E-02

LOC100131138

2.333

3.60E-02

LOC440934

0.221

3.73E-02

XLOC_007102

2.315

4.46E-03

FLJ30838

0.225

4.88E-02

LOC100506516

2.191

1.64E-02

XLOC_l2_015172

0.239

3.70E-02

LOC100170939

1.879

1.12E-02

XLOC_l2_015202

0.247

2.93E-02

XLOC_007433

1.862

2.86E-02

XLOC_008013

0.258

1.75E-02

LOC100505495

1.802

1.06E-02

XLOC_008390

0.260

1.76E-02

MGC70870

1.726

3.31E-02

LOC100507588

0.275

1.74E-02

p values were adjusted using the Benjamini-Hochberg procedure with FDR < 0.05

FC fold change

Increased NEAT1 Expression in HD

Among dysregulated lncRNAs, we focused on NEAT1 upregulation in HD. The microarray kit used in this study contained four probes targeting NEAT1: A_19_P00318409; A_19_P00321333; A_21_P0011024; A_19_P00321332. All these probes hybridize to both NEAT1_1 (MENε) and NEAT1_2 (MENβ) transcripts and measure expression levels of both NEAT1 isoforms because the long isoform NEAT1_2 is a continuation of the short isoform NEAT1_1 [33]. Microarray data revealed that NEAT1 was consistently upregulated across all NEAT1-targeting probes with fold changes ranging between 2.46 and 4.04 (Supplementary Table 1). However, NEAT1 upregulation in microarray failed to reach significance levels after multiple testing corrections at the FDR level of 0.05. Therefore, to confirm increased NEAT1 expression in HD, we performed quantitative reverse transcription PCR (qRT-PCR) for the human HD and control brain tissues. NEAT1 levels measured in this study refer to expression of both NEAT1 isoforms because NEAT1 primers were designed to hybridize to common sequences between NEAT1_1 and NEAT1_2 transcripts as described previously [30]. NEAT1 expression significantly increased by 2.24-fold in the postmortem brain of HD patients (p = 0.047; Fig. 1b). In addition, we assessed the R6/2 mouse model of HD to support the results obtained for the human HD brain. Brain tissue was obtained from R6/2 transgenic mice (111 CAG repeats) and their wild-type littermates at 12 weeks of age. Real-time PCR showed that R6/2 transgenic mice expressed 21.5-fold higher levels of NEAT1 than the control mice (p = 0.034; Fig. 1c), which was consistent with the human HD data.

Effect of NEAT1 Overexpression on Cell Survival

To evaluate the effect of NEAT1 overexpression in HD, we measured NEAT1 levels from in vitro HD models: mutant huntingtin-transfected neuro2A cells and mouse striatal neuron-derived cell lines (STHdh). Neuro2A cells were transfected with either mutant huntingtin (100 CAGs) or wild-type huntingtin (18 CAGs). Mutant STHdh cells expressed heterozygous mutant huntingtin (111/7 CAGs), while wild-type STHdh cells expressed homozygous wild-type huntingtin (7/7 CAGs). However, both in vitro models containing mutant huntingtin failed to replicate upregulation of NEAT1 in HD (data not shown). It is presumed that cellular models of HD mutant huntingtin were unable to entirely recapitulate the complex pathophysiology of HD seen in humans or R6/2 mice. Accordingly, we employed a NEAT1 overexpression model by directly transfecting the NEAT1 vector into neuro2A cells. More specifically, the NEAT1 overexpressing model was designed to induce overexpression of NEAT1_1, not NEAT1_2 isoform, because the vector contained only the genomic sequences of NEAT1_1 shorter isoform. The overall transfection efficiency measured by flow cytometry ranged between 11.3 and 16.4 %. (Fig. 2a–d). NEAT1-transfected neuro2A cells were isolated by using FACS and, as expected, showed a marked (174-fold) increase in NEAT1 expression compared to cells transfected with vehicle or GFP vector when measured by real-time PCR (p = 0.029; Fig. 2e).
Fig. 2

NEAT1 overexpression and H2O2-induced cell death. a DAPI (blue, left), GFP (green, middle), and DAPI merged with GFP (right) images were taken by confocal microscopy after GFP-NEAT1 vector transfection. bd Flow cytometry analysis to determine transfection efficiency by detecting GFP expression. Dot plots (left panel) show a live neuro2A cell subpopulation (P1) based on forward scatter (FSC) and side scatter (SSC). Neuro2A cells transfected with a vehicle (b), GFP vector (c), and NEAT1 vector (d) demonstrated a comparable percentage of live neuro2A cells ranging between 95.6 and 98.5 %. Histograms (right panel) show that 16.4 % of GFP-transfected cells and 11.3 % of NEAT1-transfected cells were GFP-positive, compared with 0 % of vehicle-transfected cells, as measured by FACS. e After harvesting transfected cells by FACS, NEAT1 expression was measured by real-time PCR. NEAT1-transfected neuro2A cells showed a 174-fold increase in the NEAT1 transcript level compared with vehicle- or GFP-transfected cells (p = 0.029, n = 4 per group). f WST-1 assay after H2O2 injury demonstrated significantly reduced cell death in NEAT1-transfected cells compared to GFP-transfected cells at 500, 1000, and 2000 μM concentrations of H2O2 (p = 0.009 at all concentrations, n = 5 per group). Bars show mean ± SEM. *p < 0.05; **p < 0.01

To rule out the possibility that NEAT1 signals resulted from the plasmid DNA contamination, we performed qRT-PCR including no-reverse transcription (no-RT) controls. NEAT1 levels were measured in unsorted neuro2A cells using the same PCR protocol except that the reverse transcriptase was omitted in the cDNA synthesis step for no-RT controls. NEAT1-transfected cells showed significantly increased NEAT1 levels in the cytoplasm (205-fold) as well as the nucleus (4091-fold) compared to GFP-transfected cells (p = 0.008 for each compartment). However, NEAT1 signals were not detected in the PCR for NEAT1-transfected no-RT controls as expected (Supplementary Fig. 1a). These findings suggested that NEAT1 levels measured by PCR did not originate from contaminating DNA. Moreover, it is evident that induced NEAT1 transcripts were enriched in the nucleus of neuro2A cells.

To validate that induced NEAT1 transcripts localized to the nucleus, we additionally measured nuclear-cytoplasmic ratio of NEAT1 expression in FACS-sorted cells. In line with the results from unsorted cells, NEAT1-transfected cells demonstrated higher nuclear accumulation of NEAT1 than GFP-transfected controls (41.5 ± 2.72 vs. 2.7 ± 0.71, p = 0.008; Supplementary Fig. 1b). The results supported the specific nuclear localization of NEAT1 lncRNA in the overexpression model.

We therefore examined the influence of NEAT1 short isoform overexpression on neuronal cell death. NEAT1-overexpressed neuro2A cells demonstrated significantly increased survival against H2O2-induced cell injury, compared with GFP-transfected cells (p = 0.009 in each condition; Fig. 2f). Viability of NEAT1-transfected cells increased by 31.3, 19.5, and 39.8 % at H2O2 concentrations of 500, 1000, and 2000 μM, respectively. These results indicate the protective effect of NEAT1 overexpression against oxidative injury in neuronal cell lines.

Discussion

In this study, we detected dysregulated lncRNA expression, particularly upregulation of NEAT1, in the human HD postmortem brain and R6/2 mouse brain. When we transfected the NEAT1 short isoform into neuro2A cells, cell death due to oxidative stress induced by H2O2 significantly decreased. Our results reveal the functional relevance of lncRNA NEAT1, especially as a neuroprotective mechanism in HD pathogenesis.

NEAT1 is an evolutionarily conserved noncoding RNA transcript, which is enriched in the nucleus and serves as a scaffold for the formation of nuclear paraspeckles [20, 23]. NEAT1 contributes to the regulation of gene expression via nuclear retention of mRNAs containing inverted repeats, while also affecting the nucleocytoplasmic relocation of paraspeckle proteins, which can act as transcription factors [34, 35, 36]. Upregulation of NEAT1 has been implicated in the cellular response to infections with a variety of viruses, including Japanese encephalitis virus, human immunodeficiency virus, and rabies virus [30, 37]. Following viral infection or poly I:C stimulation, NEAT1 was activated through the Toll-like receptor 3-mediated pathway, consequently leading to the expression of antiviral inflammatory cytokine interleukin 8 (IL-8). Expression occurred following the relocation of paraspeckle proteins, which under normal conditions repress IL-8 transcription [35]. These results indicate that NEAT1 acts as a transcriptional regulator in the innate immune response, rather than acting through nonspecific transcripts that induce bystander effects.

Relatively few studies have evaluated NEAT1 expression in neurological diseases. Previous lncRNA data derived from the human nucleus accumbens showed that NEAT1 was upregulated in heroin users [38]. Additionally, the NEAT1_2 transcript was reported to be upregulated in conjunction with increased paraspeckle formation in spinal motor neurons of patients with amyotrophic lateral sclerosis, a devastating neurodegenerative disease [39]. Previous research using microarray data mining methods detected NEAT1 overexpression in HD, in accordance with our study [18]. Differential expression of MEG3, LINC00341, LINC00342, and DGCR5 was also consistent with our microarray data, although the difference of the latter three transcripts did not reach statistically significant levels after correction for multiple testing. However, the TUG1 level did not change in contrast to that in the previous study, while RPS20P22 expression could not be evaluated because the microarray kit used in the present study did not include the corresponding probes.

Among a number of lncRNAs dysregulated in HD, we focused on NEAT1 based on its biological functions and potential association with HD pathogenesis. Hirose et al. demonstrated that proteasome inhibition by MG132 or bortezomib treatment induced NEAT1 lncRNA synthesis with paraspeckle enlargement [36]. It should be noted that ubiquitin-proteasome system impairment and subsequent aggregation of abnormal proteins within neurons are well-known pathomechanisms for neurodegeneration in HD [40]. In this regard, the link between proteasome inhibition and NEAT1 induction could partially explain our results that NEAT1 levels were elevated in the human HD postmortem brain and R6/2 mouse brain. Although proteasome inhibition activated de novo proteasome synthesis by a feedback mechanism [41], it is not understood whether induced NEAT1 transcripts affected proteasome homeostasis or how NEAT1 expression is regulated upon proteasome inhibition.

Previous research suggested the nuclear translocation of transcription factors through the cooperative actions of NEAT1 and paraspeckles in response to cellular stimuli [35, 36]. Paraspeckle proteins are not merely structural components of paraspeckles but are also transcription factors that control gene expression [42, 43]. Furthermore, it has been reported that alterations in expression or subcellular location of several transcription factors are associated with widespread transcriptional changes in HD [4, 44, 45]. Nevertheless, the mechanistic link between mutant huntingtin and transcriptional dysregulation in HD has not been fully elucidated. Our demonstration of NEAT1 upregulation in HD provides novel insights into the disease mechanism that sequestration of various paraspeckle proteins resulting from NEAT1 upregulation might contribute to the pathogenic alteration of transcriptional status in HD. In this context, further research will be needed to define transcription factors directly affected by NEAT1 dysregulation in the disease state and subsequent downstream processes contributing to neuronal dysfunction and death.

NEAT1 overexpression has been reported to engage in the cell survival pathway under stress conditions such as prolonged proteasome inhibition and hypoxia [36, 46]. In both conditions, knockdown of NEAT1 caused a significant increase in cell death, which supports the anti-apoptotic function of NEAT1. In agreement with this, NEAT1_1 transfection into neuro2A cells showed protective effects against H2O2-induced cell death in our study. Thus, it is likely that the NEAT1 overexpression in HD does not demonstrate a pathological process leading to neuronal death but instead represents a defensive mechanism against cell injury. However, it should be noted that the cellular effects shown in this study are limited to the short isoform of NEAT1. The lack of examination for the role of the long isoform is the limitation of the present study, because there might be differential effects between two NEAT1 isoforms upon the neuronal survival.

Regarding the constructs of NEAT1 overexpression vector, GFP located downstream of NEAT1 sequences was used as a reporter gene. Despite the presence of noncoding RNA regulated by the common promoter, GFP protein expression was confirmed by confocal microscopic analysis. In agreement with our results, a GFP reporter was inserted after the transcription ending region of Glt2 lncRNA, and GFP protein was successfully expressed without interfering with lncRNA expression [47]. Moreover, RNCR2 lncRNA, which is implicated in regulation of retinal cell differentiation, was also fused to GFP sequences and GFP fluorescence indicated that GFP gene were indeed translated [48]. However, GFP tagging was shown to inhibit nuclear retention of RNCR2 lncRNA and recapitulate the effect of RNCR2 knockdown. Despite conflicting data on lncRNA subcellular localization, previous results supported that GFP fused to lncRNAs undergo translation.

Although additional experiments are needed to elucidate the precise pro-survival mechanisms related with NEAT1 and the temporal relationship between NEAT1 induction and disease progression, our observations provide potential targets for developing lncRNA-based novel therapeutics in HD. Further research is warranted to identify NEAT1-regulating pathways in connection with mutant huntingtin and determine NEAT1-binding paraspeckle proteins contributing to transcriptional dysregulation in HD.

Notes

Acknowledgments

This study was supported by grants from the Korean Health Technology R&D Project (HI14C2348), Ministry of Health and Welfare, Republic of Korea, and the National Research Foundation (NRF) of Korea (2011–0012728). S-T.L was supported by the Seoul National University Hospital Research Fund (04-2014-0730) and Korean Health Technology R&D Project (HI12C1773).

Compliance with Ethical Standards

Conflict of Interest

The authors declare that they have no conflict of interest.

Funding

This study was supported by grants from the Korean Health Technology R&D Project (HI14C2348), Ministry of Health and Welfare, Republic of Korea, and the National Research Foundation (NRF) of Korea (2011–0012728). S-T.L was supported by the Seoul National University Hospital Research Fund (04-2014-0730) and Korean Health Technology R&D Project (HI12C1773).

Supplementary material

12035_2016_9928_MOESM1_ESM.docx (32 kb)
ESM 1 (DOCX 32 kb)

References

  1. 1.
    Labbadia J, Morimoto RI (2013) Huntington’s disease: underlying molecular mechanisms and emerging concepts. Trends Biochem Sci 38:378–385. doi: 10.1016/j.tibs.2013.05.003 CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    MacDonald ME, Ambrose CM, Duyao MP, Myers RH, Lin C, Srinidhi L, Barnes G, Taylor SA et al (1993) A novel gene containing a trinucleotide repeat that is expanded and unstable on Huntington’s disease chromosomes. Cell 72:971–983. doi: 10.1016/0092-8674(93)90585-E CrossRefGoogle Scholar
  3. 3.
    Graham RK, Deng Y, Slow EJ, Haigh B, Bissada N, Lu G, Pearson J, Shehadeh J et al (2006) Cleavage at the caspase-6 site is required for neuronal dysfunction and degeneration due to mutant huntingtin. Cell 125:1179–1191. doi: 10.1016/j.cell.2006.04.026 CrossRefPubMedGoogle Scholar
  4. 4.
    Sugars KL, Rubinsztein DC (2003) Transcriptional abnormalities in Huntington disease. Trends Genet 19:233–238. doi: 10.1016/S0168-9525(03)00074-X CrossRefPubMedGoogle Scholar
  5. 5.
    Trushina E, McMurray CT (2007) Oxidative stress and mitochondrial dysfunction in neurodegenerative diseases. Neuroscience 145:1233–1248. doi: 10.1016/j.neuroscience.2006.10.056 CrossRefPubMedGoogle Scholar
  6. 6.
    Lee ST, Chu K, Jung KH, Im WS, Park JE, Lim HC, Won CH, Shin SH et al (2009) Slowed progression in models of Huntington disease by adipose stem cell transplantation. Ann Neurol 66:671–681. doi: 10.1002/ana.21788 CrossRefPubMedGoogle Scholar
  7. 7.
    Ross CA, Tabrizi SJ (2011) Huntington’s disease: from molecular pathogenesis to clinical treatment. Lancet Neurol 10:83–98. doi: 10.1016/S1474-4422(10)70245-3 CrossRefPubMedGoogle Scholar
  8. 8.
    Rinn JL, Chang HY (2012) Genome regulation by long noncoding RNAs. Annu Rev Biochem 81:145–166. doi: 10.1146/annurev-biochem-051410-092902 CrossRefPubMedGoogle Scholar
  9. 9.
    Iyer MK, Niknafs YS, Malik R, Singhal U, Sahu A, Hosono Y, Barrette TR, Prensner JR et al (2015) The landscape of long noncoding RNAs in the human transcriptome. Nat Genet 47:199–208. doi: 10.1038/ng.3192 CrossRefPubMedPubMedCentralGoogle Scholar
  10. 10.
    Kung JTY, Colognori D, Lee JT (2013) Long noncoding RNAs: past, present, and future. Genetics 193:651–669. doi: 10.1534/genetics.112.146704 CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Wahlestedt C (2013) Targeting long non-coding RNA to therapeutically upregulate gene expression. Nat Rev Drug Discov 12:433–446. doi: 10.1038/nrd4018 CrossRefPubMedGoogle Scholar
  12. 12.
    Faghihi MA, Modarresi F, Khalil AM, Wood DE, Sahagan BG, Morgan TE, Finch CE, St Laurent G 3rd et al (2008) Expression of a noncoding RNA is elevated in Alzheimer’s disease and drives rapid feed-forward regulation of beta-secretase. Nat Med 14:723–730. doi: 10.1038/nm1784 CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Lee DY, Moon J, Lee ST, Jung KH, Park DK, Yoo JS, Sunwoo JS, Byun JI et al (2015) Distinct expression of long non-coding RNAs in an Alzheimer’s disease model. J Alzheimers Dis 45:837–849. doi: 10.3233/jad-142919 PubMedGoogle Scholar
  14. 14.
    Lee DY, Moon J, Lee ST, Jung KH, Park DK, Yoo JS, Sunwoo JS, Byun JI et al (2015) Dysregulation of long non-coding RNAs in mouse models of localization-related epilepsy. Biochem Biophys Res Commun 462:433–440. doi: 10.1016/j.bbrc.2015.04.149 CrossRefPubMedGoogle Scholar
  15. 15.
    Johnson R, Richter N, Jauch R, Gaughwin PM, Zuccato C, Cattaneo E, Stanton LW (2010) Human accelerated region 1 noncoding RNA is repressed by REST in Huntington’s disease. Physiol Genomics 41:269–274. doi: 10.1152/physiolgenomics.00019.2010 CrossRefPubMedGoogle Scholar
  16. 16.
    Chung DW, Rudnicki DD, Yu L, Margolis RL (2011) A natural antisense transcript at the Huntington’s disease repeat locus regulates HTT expression. Hum Mol Genet 20:3467–3477. doi: 10.1093/hmg/ddr263 CrossRefPubMedPubMedCentralGoogle Scholar
  17. 17.
    Francelle L, Galvan L, Gaillard MC, Petit F, Bernay B, Guillermier M, Bonvento G, Dufour N, Elalouf JM, Hantraye P, Deglon N, de Chaldee M, Brouillet E (2015) Striatal long noncoding RNA Abhd11os is neuroprotective against an N-terminal fragment of mutant huntingtin in vivo. Neurobiol Aging 36:1601.e1607-1616. doi: 10.1016/j.neurobiolaging.2014.11.014
  18. 18.
    Johnson R (2012) Long non-coding RNAs in Huntington’s disease neurodegeneration. Neurobiol Dis 46:245–254. doi: 10.1016/j.nbd.2011.12.006 CrossRefPubMedGoogle Scholar
  19. 19.
    Sasaki YT, Ideue T, Sano M, Mituyama T, Hirose T (2009) MENepsilon/beta noncoding RNAs are essential for structural integrity of nuclear paraspeckles. Proc Natl Acad Sci U S A 106:2525–2530. doi: 10.1073/pnas.0807899106 CrossRefPubMedPubMedCentralGoogle Scholar
  20. 20.
    Clemson CM, Hutchinson JN, Sara SA, Ensminger AW, Fox AH, Chess A, Lawrence JB (2009) An architectural role for a nuclear noncoding RNA: NEAT1 RNA is essential for the structure of paraspeckles. Mol Cell 33:717–726. doi: 10.1016/j.molcel.2009.01.026 CrossRefPubMedPubMedCentralGoogle Scholar
  21. 21.
    Sunwoo H, Dinger ME, Wilusz JE, Amaral PP, Mattick JS, Spector DL (2009) MEN epsilon/beta nuclear-retained non-coding RNAs are up-regulated upon muscle differentiation and are essential components of paraspeckles. Genome Res 19:347–359. doi: 10.1101/gr.087775.108 CrossRefPubMedPubMedCentralGoogle Scholar
  22. 22.
    Fox AH, Lamond AI (2010) Paraspeckles. Cold Spring Harb Perspect Biol 2:a000687. doi: 10.1101/cshperspect.a000687 PubMedPubMedCentralGoogle Scholar
  23. 23.
    Hutchinson JN, Ensminger AW, Clemson CM, Lynch CR, Lawrence JB, Chess A (2007) A screen for nuclear transcripts identifies two linked noncoding RNAs associated with SC35 splicing domains. BMC Genomics 8:39. doi: 10.1186/1471-2164-8-39 CrossRefPubMedPubMedCentralGoogle Scholar
  24. 24.
    Chakravarty D, Sboner A, Nair SS, Giannopoulou E, Li R, Hennig S, Mosquera JM, Pauwels J et al (2014) The oestrogen receptor alpha-regulated lncRNA NEAT1 is a critical modulator of prostate cancer. Nat Commun 5:5383. doi: 10.1038/ncomms6383 CrossRefPubMedPubMedCentralGoogle Scholar
  25. 25.
    Chen X, Kong J, Ma Z, Gao S, Feng X (2015) Up regulation of the long non-coding RNA NEAT1 promotes esophageal squamous cell carcinoma cell progression and correlates with poor prognosis. Am J Cancer Res 5:2808–2815CrossRefPubMedPubMedCentralGoogle Scholar
  26. 26.
    He C, Jiang B, Ma J, Li Q (2015) Aberrant NEAT1 expression is associated with clinical outcome in high grade glioma patients. APMIS. doi: 10.1111/apm.12480 Google Scholar
  27. 27.
    Wu Y, Yang L, Zhao J, Li C, Nie J, Liu F, Zhuo C, Zheng Y et al (2015) Nuclear-enriched abundant transcript 1 as a diagnostic and prognostic biomarker in colorectal cancer. Mol Cancer 14:191. doi: 10.1186/s12943-015-0455-5 CrossRefPubMedPubMedCentralGoogle Scholar
  28. 28.
    Mangiarini L, Sathasivam K, Seller M, Cozens B, Harper A, Hetherington C, Lawton M, Trottier Y et al (1996) Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87:493–506CrossRefPubMedGoogle Scholar
  29. 29.
    Lee ST, Chu K, Im WS, Yoon HJ, Im JY, Park JE, Park KH, Jung KH et al (2011) Altered microRNA regulation in Huntington’s disease models. Exp Neurol 227:172–179. doi: 10.1016/j.expneurol.2010.10.012 CrossRefPubMedGoogle Scholar
  30. 30.
    Zhang Q, Chen CY, Yedavalli VS, Jeang KT (2013) NEAT1 long noncoding RNA and paraspeckle bodies modulate HIV-1 posttranscriptional expression. MBio 4:e00596–00512. doi: 10.1128/mBio.00596-12 CrossRefPubMedPubMedCentralGoogle Scholar
  31. 31.
    Kim M, Lee HS, LaForet G, McIntyre C, Martin EJ, Chang P, Kim TW, Williams M et al (1999) Mutant huntingtin expression in clonal striatal cells: dissociation of inclusion formation and neuronal survival by caspase inhibition. J Neurosci 19:964–973PubMedGoogle Scholar
  32. 32.
    Trettel F, Rigamonti D, Hilditch-Maguire P, Wheeler VC, Sharp AH, Persichetti F, Cattaneo E, MacDonald ME (2000) Dominant phenotypes produced by the HD mutation in STHdh(Q111) striatal cells. Hum Mol Genet 9:2799–2809. doi: 10.1093/hmg/9.19.2799 CrossRefPubMedGoogle Scholar
  33. 33.
    Naganuma T, Hirose T (2013) Paraspeckle formation during the biogenesis of long non-coding RNAs. RNA Biol 10:456–461. doi: 10.4161/rna.23547 CrossRefPubMedPubMedCentralGoogle Scholar
  34. 34.
    Chen LL, Carmichael GG (2009) Altered nuclear retention of mRNAs containing inverted repeats in human embryonic stem cells: functional role of a nuclear noncoding RNA. Mol Cell 35:467–478. doi: 10.1016/j.molcel.2009.06.027 CrossRefPubMedPubMedCentralGoogle Scholar
  35. 35.
    Imamura K, Imamachi N, Akizuki G, Kumakura M, Kawaguchi A, Nagata K, Kato A, Kawaguchi Y et al (2014) Long noncoding RNA NEAT1-dependent SFPQ relocation from promoter region to paraspeckle mediates IL8 expression upon immune stimuli. Mol Cell 53:393–406. doi: 10.1016/j.molcel.2014.01.009 CrossRefPubMedGoogle Scholar
  36. 36.
    Hirose T, Virnicchi G, Tanigawa A, Naganuma T, Li R, Kimura H, Yokoi T, Nakagawa S et al (2014) NEAT1 long noncoding RNA regulates transcription via protein sequestration within subnuclear bodies. Mol Biol Cell 25:169–183. doi: 10.1091/mbc.E13-09-0558 CrossRefPubMedPubMedCentralGoogle Scholar
  37. 37.
    Saha S, Murthy S, Rangarajan PN (2006) Identification and characterization of a virus-inducible non-coding RNA in mouse brain. J Gen Virol 87:1991–1995. doi: 10.1099/vir.0.81768-0 CrossRefPubMedGoogle Scholar
  38. 38.
    Michelhaugh SK, Lipovich L, Blythe J, Jia H, Kapatos G, Bannon MJ (2011) Mining Affymetrix microarray data for long non-coding RNAs: altered expression in the nucleus accumbens of heroin abusers. J Neurochem 116:459–466. doi: 10.1111/j.1471-4159.2010.07126.x CrossRefPubMedGoogle Scholar
  39. 39.
    Nishimoto Y, Nakagawa S, Hirose T, Okano HJ, Takao M, Shibata S, Suyama S, Kuwako K et al (2013) The long non-coding RNA nuclear-enriched abundant transcript 1_2 induces paraspeckle formation in the motor neuron during the early phase of amyotrophic lateral sclerosis. Mol Brain 6:31. doi: 10.1186/1756-6606-6-31 CrossRefPubMedPubMedCentralGoogle Scholar
  40. 40.
    Bennett EJ, Shaler TA, Woodman B, Ryu KY, Zaitseva TS, Becker CH, Bates GP, Schulman H et al (2007) Global changes to the ubiquitin system in Huntington’s disease. Nature 448:704–708. doi: 10.1038/nature06022 CrossRefPubMedGoogle Scholar
  41. 41.
    Meiners S, Heyken D, Weller A, Ludwig A, Stangl K, Kloetzel PM, Kruger E (2003) Inhibition of proteasome activity induces concerted expression of proteasome genes and de novo formation of Mammalian proteasomes. J Biol Chem 278:21517–21525. doi: 10.1074/jbc.M301032200 CrossRefPubMedGoogle Scholar
  42. 42.
    Shav-Tal Y, Zipori D (2002) PSF and p54(nrb)/NonO—multi-functional nuclear proteins. FEBS Lett 531:109–114. doi: 10.1016/S0014-5793(02)03447-6 CrossRefPubMedGoogle Scholar
  43. 43.
    Dong X, Sweet J, Challis JR, Brown T, Lye SJ (2007) Transcriptional activity of androgen receptor is modulated by two RNA splicing factors, PSF and p54nrb. Mol Cell Biol 27:4863–4875. doi: 10.1128/MCB.02144-06 CrossRefPubMedPubMedCentralGoogle Scholar
  44. 44.
    Bae BI, Xu H, Igarashi S, Fujimuro M, Agrawal N, Taya Y, Hayward SD, Moran TH et al (2005) p53 mediates cellular dysfunction and behavioral abnormalities in Huntington’s disease. Neuron 47:29–41. doi: 10.1016/j.neuron.2005.06.005 CrossRefPubMedGoogle Scholar
  45. 45.
    Johnson R, Buckley NJ (2009) Gene dysregulation in Huntington’s disease: REST, microRNAs and beyond. Neuromolecular Med 11:183–199. doi: 10.1007/s12017-009-8063-4 CrossRefPubMedGoogle Scholar
  46. 46.
    Choudhry H, Albukhari A, Morotti M, Haider S, Moralli D, Smythies J, Schodel J, Green CM et al (2015) Tumor hypoxia induces nuclear paraspeckle formation through HIF-2alpha dependent transcriptional activation of NEAT1 leading to cancer cell survival. Oncogene 34:4482–4490. doi: 10.1038/onc.2014.378 CrossRefPubMedGoogle Scholar
  47. 47.
    Li Z, Wang L, Wang Y, Liu L, Wang L, Li W, Zhou Q (2015) Generation of an LncRNA Gtl2-GFP reporter for rapid assessment of pluripotency in mouse induced pluripotent stem cells. J Genet Genomics 42:125–128. doi: 10.1016/j.jgg.2015.02.004 CrossRefPubMedGoogle Scholar
  48. 48.
    Rapicavoli NA, Poth EM, Blackshaw S (2010) The long noncoding RNA RNCR2 directs mouse retinal cell specification. BMC Dev Biol 10:49. doi: 10.1186/1471-213x-10-49 CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2016

Authors and Affiliations

  • Jun-Sang Sunwoo
    • 1
    • 2
    • 3
  • Soon-Tae Lee
    • 1
    • 2
  • Wooseok Im
    • 1
    • 2
  • Mijung Lee
    • 1
    • 2
  • Jung-Ick Byun
    • 1
    • 2
    • 4
  • Keun-Hwa Jung
    • 1
    • 2
  • Kyung-Il Park
    • 5
  • Ki-Young Jung
    • 1
    • 2
  • Sang Kun Lee
    • 1
    • 2
  • Kon Chu
    • 1
    • 2
  • Manho Kim
    • 1
    • 2
    • 6
  1. 1.Department of Neurology, Laboratory for Neurotherapeutics, Biomedical Research InstituteSeoul National University HospitalSeoulSouth Korea
  2. 2.Program in Neuroscience, College of MedicineSeoul National UniversitySeoulSouth Korea
  3. 3.Department of Neurology, School of MedicineSoonchunhyang UniversitySeoulSouth Korea
  4. 4.Department of NeurologyKyung Hee University Hospital at GangdongSeoulSouth Korea
  5. 5.Department of Neurology, Healthcare System Gangnam CenterSeoul National University HospitalSeoulSouth Korea
  6. 6.Protein Metabolism Medical Research Center, College of MedicineSeoul National UniversitySeoulSouth Korea

Personalised recommendations