1471-2199-10-9.fm

BMC Molecular Biology
Research article
Spinocerebellar ataxia type 8 larger triplet expansion alters histone
modification and induces RNA foci
I-Cheng Chen†1, Hsuan-Yuan Lin†1, Ghin-Chueh Lee1, Shih-Huan Kao1,
Chiung-Mei Chen2, Yih-Ru Wu2, Hsiu-Mei Hsieh-Li1, Ming-Tsan Su1 and
Guey-Jen Lee-Chen*1
Address: 1Department of Life Science, National Taiwan Normal University, Taipei 116, Taiwan and 2Department of Neurology, Chang Gung Memorial Hospital, Chang-Gung University College of Medicine, Taipei 105, Taiwan Email: I-Cheng Chen - [email protected]; Hsuan-Yuan Lin - [email protected]; Ghin-Chueh Lee - [email protected]; Shih-Huan Kao - [email protected]; Chiung-Mei Chen - [email protected]; Yih-Ru Wu - [email protected]; Hsiu-Mei Hsieh-Li - [email protected]; Ming-Tsan Su - [email protected]; Guey-Jen Lee-Chen* - [email protected] * Corresponding author Published: 10 February 2009 Received: 30 July 2008Accepted: 10 February 2009 BMC Molecular Biology 2009, 10:9
2009 Chen et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Background: Spinocerebellar ataxia type 8 (SCA8) involves the expression of an expanded CTG/
CAG combined repeats (CR) from opposite strands producing CUG expansion transcripts (ataxin
8 opposite strand, ATXN8OS) and a polyglutamine expansion protein (ataxin 8, ATXN8). The
pathogenesis of SCA8 is complex and the spectrum of clinical presentations is broad.
Results: Using stably induced cell models expressing 0, 23, 88 and 157 CR, we study the role of
ATXN8OS transcripts in SCA8 pathogenesis. In the absence of doxycycline, the stable ATXN8OS
CR cell lines exhibit low levels of ATXN8OS expression and a repeat length-related increase in
staurosporine sensitivity and in the number of annexin positive cells. A repeat length-dependent
repression of ATXN8OS expression was also notable. Addition of doxycycline leads to 25 50
times more ATXN8OS RNA expression with a repeat length-dependent increase in fold of
ATXN8OS RNA induction. ChIP-PCR assay using anti-dimethyl-histone H3-K9 and anti-acetyl-
histone H3-K14 antibodies revealed increased H3-K9 dimethylation and reduced H3-K14
acetylation around the ATXN8OS cDNA gene in 157 CR line. The repeat length-dependent
increase in induction fold is probably due to the increased RNA stability as demonstrated by
monitoring ATXN8OS RNA decay in cells treated with the transcriptional inhibitor, actinomycin
D. In cells stably expressing ATXN8OS, RNA FISH experiments further revealed ribonuclear foci
formation in cells carrying expanded 88 and 157 CR.
Conclusion: The present study demonstrates that the expanded CUG-repeat tracts are toxic to
human cells and may affect ATXN8OS RNA expression and stability through epigenetic and post-
transcriptional mechanisms.
(page number not for citation purposes) BMC Molecular Biology 2009, 10:9
on other repeat expansion diseases with similar neuronal Spinocerebellar ataxia type 8 (SCA8) is a dominantly pathology using this cell line has been reporte].
inherited, slowly progressive neurodegenerative disorder The derived ATXN8OS cell lines are isogenic except for the caused by the expansion of CTA/CTG combined repeats number of CTA/CTG combined repeats. The repeat (CR) in the ataxin 8 opposite strand (ATXN8OS) gene number in these cell lines was stable (data not shown).
located on chromosome 13reported repeat ATXN8OS RNA levels were measured by real-time PCR lengths associated with ataxia vary dramatically, ranging quantification using ATXN8OS-specific probe and prim- from 68 [1000 repeats [e general popula- ers. The expression of the endogenous ATXN8OS RNA in tion more than 99% of the alleles have 16 37 CR ].
vector only cell line was too low to be efficiently detected.
Nevertheless the penetrance of the SCA8 repeat expansion In the absence of doxycycline, all ATXN8OS CR cell lines and ataxia is not complete, as expansions do not always expressed low level of ATXN8OS RNA, ranging from segregate with ataxia in families and they are present in 0.017 to 0.042 compared with endogenous HPRT1 (Fig.
rare instances in normal and non-ataxic diseased popula- A repeat length-dependent repression of ATXN8OS expression is notable. ATXN8OS 88 and 157 CR cells weremore sensitive to staurosporine (30 and 50 nM), an exter- The pathogenesis of SCA8 is complex. In addition to a nal apoptotic stimulus ). A repeat length related CTG repeat expansion in the ATXN8OS gene, it also increase in the number of annexin V-positive cells was involves a CAG repeat expansion in another overlapping also observed when the viability of these ATXN8OS CR gene, ataxin 8 (ATXN8 the CTG direction, cell lines was examined. Annexin V binds phosphatidyl ATXN8OS expresses non-coding transcripts containing serine (PS) located in the plasma membrane. PS is only the CUG expansion which overlap with the 5' region of accessible to annexin V during apoptosis when the PS the Kelch-like 1 (KLHL1) transcripts, and in the CAG moves from the inner to the outer plasma membrane, or direction ATXN8 expresses transcripts encoding a nearly during necrosis when membrane integrity is lost. In CR pure polyglutamine expansion protein. As a consequence, cells grown without doxycycline, while the absolute level three plausible mechanisms were proposed for SCA8: of cell death was relatively modest, in 88 and 157 CR the RNA gain-of function], partial loss of KLHL1 function number of dying cells was statistically significant amount- polyglutamine expansion protein in the CAG ing to 3 times that seen in the cells with 23 CR, die present study, we focus on the RNA – Dox). In CR cells grown with doxycycline, cell death was gain-of function mechanism.
also significantly increased, amounting to 2 times thatseen in the cells with 23 CR (Fig. ).
The causative agent for myotonic dystrophy (DM1) is alsoknown to be a CTG expansion in the 3'-UTR of the DMPK Repeat length-related change in ATXN8OS expression The geThe expanded CUG repeat in the DMPK RNA induction of ATXN8OS RNA levels were further examined impaired nuclear cytoplasmic transport, resulting in After induction with doxycycline nuclear retention and ribonuclear foci formati, for 1 and 2 days, the amount of ATXN8OS RNA in 0, 23, In addition, expanded CTG repeats in DM1 alter the adja- 88 and 157 CR cells increased significantly. When the cent chromatin structure [] and several proteins bind to amount of ATXN8OS RNA present at the time of doxycy- CUG repeat-containing RN]. Using PC12 neuro- cline addition was set to 100% for each CR cell line, the nal cells expressing the CUG repeat-bearing mRNA, cis- fold of induction increased with repeat length, with effects through the reporter gene and neuronal death after (CR)n/fold of induction being (CR)0/29 32, (CR)23/ cell differentiation in vitro were repo]. Expression 25 27, (CR)88/41 42, and (CR)157/47 50.
of a Huntington's disease-like 2 JPH3 transcript with anexpanded CUG repeat also resulted in the formation of Involvement of reduced ATXN8OS expression with histone
RNA foci and cell toxicity d on these previous studies, we established ATXN8OS stably induced HEK- The expression of the ATXN8OS in 0, 23, 88 and 157 CR 293 cell lines carrying 0, 23, 88 and 157 CR to investigate lines was driven by the same hybrid CMV/TetO2 pro- the possible epigenetic and post-transcriptional regula- moter. As CUG triplet repeat expansion in DM1 may alter tions of the ATXN8OS expression.
the adjacent chromatin structure], the observed repeatlength-dependent repression of ATXN8OS expression may be due to chromatin remodeling. Modifications of ATXN8OS CR cell lines
the H3 have been shown to induce a change in chromatin The pcDNA5/FRT/TO vector and ATXN8OS constructs activity . To examine the possibility that site-specific containing 0, 23, 88 and 157 CR were used to generate H3 modifications might regulate ATXN8OS expression ATXN8OS CR cell lines. These cell lines were originated with CUG repeat expansions, ChIP-PCR analysis was per- from human embryonic kidney 293 cells, which express formed using primary antibodies anti-dimethyl-H3-K9 many neuron-specific mR. A large body of work (associated with transcriptionally repressed chromatin) (page number not for citation purposes) BMC Molecular Biology 2009, 10:9
ATXN8OS 0 157 CR cells. (A) The ATXN8OS cDNA containing exons D, C2, C1, B and A. The combined repeats (CR)n
inside exon A are indicated. The thick black line below cDNA represents a 213-bp fragment spanning exons B to A for ChIP-
PCR. (B) The features of pcDNA5/FRT/TO-ATXN8OS plasmids. ATXN8OS cDNA was cloned into the NotI site of the
pcDNA5/FRT/TO vector and its expression driven by hybrid CMV/TetO2 promoter (PCMV 2X TetO2). Also shown is the frag-
ment containing CMV promoter, HaloTag open reading frame and SV40 late poly(A) signal placed at the PvuII site between
bovine growth hormone poly(A) signal and Flp recombination target (FRT) site. (C) Real-time PCR quantification of ATXN8OS
RNA level relative to endogenous HPRT1 RNA in ATXN8OS CR cells grown without doxycycline. (D) The effect of 0 nM
(white bar), 15 nM (gray bar), 30 nM (charcoal gray bar) and 50 nM (black bar) staurosporine on the survival of ATXN8OS 23,
88 and 157 CR cells grown without doxycycline. (E) Annexin V binding in ATXN8OS 0, 23, 88 and 157 CR cells grown with
(right) or without (left) doxycycline. Data are represented as the mean ± SD of three independent experiments, each per-
formed in duplicate. The * indicates the difference between the indicated samples are statistically significant (P < 0.05).
and anti-acetyl-H3-K14 (associated with transcriptionally the ATXN8OS cDNA gene. If chromatin structure was active chromatin). The methylation of H3-K9 in 157 CR affected in expanded CR lines, reduced expression of cells was evident, but not in 23 or 88 CR cell HaloTag would be expected. To examine this, HaloTag Also hypoacetylation of H3-K14 was observed in 157 CR RNA level relative to endogenous HPRT1 RNA was first cells, but not in 23 or 88 CR quantified by real-time PCR. As shown in Fig. agRNA expression in expanded 88 and 157 CR lines was sig- Repeat length-dependent repression of HaloTag gene
nificantly reduced to 70 71% when compared to the nor- located next to ATXN8OS cDNA gene
mal 23 CR line (P < 0.05). To confirm the expression The cloning vector used to generate ATXN8OS CR lines change, proteins were collected and subjected to western was modified by placing a HaloTag gene downstream of blotting with HaloTag and β-actin antibodies. Consistent (page number not for citation purposes) BMC Molecular Biology 2009, 10:9
TXN8OS RNA in 0 157 CR and vector only lines Expression of ATXN8OS RNA in 0 157 CR and vector only lines. Real-time PCR quantification was performed after
addition of doxycycline for 0 2 days. To calculate the fold of induction, the relative ATXN8OS RNA levels in each CR line at
the time of doxycycline addition (0 d) are set as 1.0. Data are represented as the mean ± SD of two independent experiments,
each performed in duplicate.
s ChIP-PCR assay using anti-dimethyl H3 (K9) and anti-acetyl-histone H3 (K14) antibodies ATXN8OS CR lines ChIP-PCR assay using anti-dimethyl H3 (K9) and anti-acetyl-histone H3 (K14) antibodies.
ATXN8OS CR cells were sonicated to shear DNA. DNA was then isolated, precipitated with anti-dimethyl H3 (K9) (A) or
anti-acetyl-histone H3 (K14) (B) antibody, and used as template for PCR amplification of 213-bp exon B to A-containing frag-
ment. Lanes P represents positive control.
(page number not for citation purposes) BMC Molecular Biology 2009, 10:9
with the results of real-time PCR quantification, expres- for 48 h and actinomycin D (1 μg/ml) was added to block sion levels of HaloTag protein were significantly transcription of new RNA molecules. The stability of the decreased in ATXN8OS 88 and 157 CR lines as compared ATXN8OS RNA was then determined by real-time PCR to that of 23 CR line (Fig. 79 81%, P < 0.05). Fluores- quantification of the amount of ATXN8OS RNA present in cence microscope examination after immunocytochemi- cells harvested at different time points after actinomycin cal staining using HaloTag antibody also revealed the D addition. The amount of ATXN8OS RNA present at the reduced expression of HaloTag protein (Fig. start of the experiment immediately before actinomycin Daddition was set to 100%. As shown g Increased ATXN8OS transcript stability and ribonuclear
HPRT1 mRNA as an internal control, the levels of foci formation with CUG repeat expansion
ATXN8OS mRNA at 12 h after addition of actinomycin D To examine the observed repeat length-dependent in 23, 88 and 157 CR cells were 7.2%, 12.1% and 22.0%, increase of fold of induction, the effect of repeat length on respectively. Therefore, expanded CR causes stabilization the stability of ATXN8OS RNA was investigated. The of ATXN8OS mRNA and subsequently reduces RNA ATXN8OS cells were grown in the presence of doxycycline aloTag gene located next to ATXN8OS cDNA gene Expression of HaloTag gene located next to ATXN8OS cDNA gene. (A) Real-time PCR quantification of HaloTag
RNA level relative to endogenous HPRT1 RNA. To normalize, the relative HaloTag RNA in 23 CR cells is set as 1.0. (B) Rep-
resentative western blot image of CR lines using HaloTag and β-actin antibodies. Levels of HaloTag were normalized with an
internal control (β-actin). The relative immunoreactivity of HaloTag is shown in the right panel. For both (A) and (B), data are
expressed as the mean ± SEM values from three independent experiments. * indicates P < 0.05. (C) Fluorescence microscopy
examination after immunostaining using HaloTag antibody (green). Nuclei were counterstained with DAPI (blue).
(page number not for citation purposes) BMC Molecular Biology 2009, 10:9
Figure 5 S transcript stability and ribonuclear foci formation with CUG repeat expansion
ATXN8OS transcript stability and ribonuclear foci formation with CUG repeat expansion. (A) Real-time PCR
quantification of ATXN8OS RNA level relative to endogenous HPRT1 RNA following addition of doxycycline for 2 days and
actinomycin D treatment for 0, 3, 6, 9 or 12 hours. The relative ATXN8OS RNA in 23, 88 and 157 CR cells at the time of
actinomycine D addition (0 h) is set as 100%. (B) Ribonuclear foci formation with ATXN8OS CUG expansion. ATXN8OS 23,
88 and 157 CR cells were grown with doxycycline for 2 days and analyzed by RNA-FISH using a Cy3 labeled (CAG) (green)
or Cy5 labeled ATXN8OS unique sequence (red) oligonucleotide probe. Nuclei were counterstained with DAPI (blue).
Since the mutant DMPK and JPH3 transcripts accumu- formed two days later. As shown in Fi ribonuclear lated in the nuclei of patient cells and aggregated to form foci were seen in cells expressing ATXN8OS 23 CR. How- distinct foci , we investigated whether the ever, distinct ribonuclear foci, mostly perinuclear, were expanded ATXN8OS CUG repeats form ribonuclear foci.
observed in cells expressing expanded 88 and 157 CR.
The ATXN8OS 23, 88 and 157 CR cells were grown in the Similar results were obtained using an oligonucleotide presence of doxycycline and FISH experiments using a probe specific to ATXN8OS. Since this probe can only Cy3-labeled (CAG)10 oligonucleotide probe was per- bind to the ATXN8OS RNA in single copy, the inability to (page number not for citation purposes) BMC Molecular Biology 2009, 10:9
detect ribonuclear foci with 23 CR is not an artifact of the responsible for the observed repression of ATXN8OS RNA copy number of the repeats in the 88 or 157 CR cells.
in the 88 CR cells. Nevertheless the epigenetic changeinduced by large SCA8 repeat expansions may be one of the genetic factors that suppress the disease symptoms in Clinical and genetic studies have shown that SCA8 is a control individuals carrying large SCA8 repeat expan- slowly progressive inherited disorder with highly incom- sions. Work is now in progress to examine the roles of plete penetrance, in addition to its rare occurrence of phe- HP1 and its associated regulators on the observed repeat- notype in individuals carrying a much larger repeat dependent repression of ATXN8OS expression.
expansion []. Thus the pathogenic mechanismsunderlying SCA8 are expected to be complicated. In this Increased ATXN8OS transcript stability and ribonuclear
study we developed a number of otherwise isogenic human cell lines expressing transcripts with 0 157 CTA/ Previously it was reported that expanded CUG repeat tran- CTG CR. The repeat number in these cell lines was stable scripts form stable hairpins d muscleblind and and expression of ATXN8OS RNA containing expanded MBNL1 increase steady state levels of CUG repeat RNA 88 and 157 CR causes an increase in the likelihood of cell ]. The possibility of stable hairpin formation and pro- de cells are also sensitive to stau- teins binding may explain the observation that expanded rosporine treatment which can induce apoptosis (Fig.
CR causes stabilization of ATXN8OS mRand servation coincides with the reported length- in turn leads to repeat length-dependent increase in fold dependent toxicity of untranslated CUG repeats in cell of ATXN8OS induction (Fig. d DM2, the and Caenorhabditis elegans experim expanded repeat RNA forms discrete ribonuclear foci to cells were used to investigate the possible epigenetic and sequester CUG binding proteins and subsequently jeop- post-transcriptional controls of the ATXN8OS expression.
ardize the normal cellular functions of these proteins, The implications of the findings in the pathogenesis of which would then lead to abnormal RNA splicing of sev- SCA8 are discussed as the following.
eral genes []. Although the induced expression levels ofATXN8OS RNA in these CR cells were low (ranging from Epigenetic changes of ATXN8OS expression
0.659 1.346 compared with endogenous HPRT1), ribo- Previously expansions of CTG repeat in myotonic dystro- nuclear foci were detected in our ATXN8OS 88 or 157 CR phy and GAA repeat in Friedreich's ataxia conferred varie- cells (Fig. ). Most of the RNA foci formed are located gation of expression on a linked transgene in mice near nuclear membrane, which may be compatible with Silencing was correlated with a decrease in promoter the observation by Koch and colleague that the hairpin accessibility and was enhanced by the classical position structure formed by long CUG repeats (> 44) cannot pass effect variegation modifier heterochromatin protein 1 through nucleic pores [. The ribonuclear foci observed (HP1), which is able to bind to methylated histone H3-K9 in the nucleus may also result in transcriptional dysfunc- ed levels of histone H3 dimethylated on K9 tion to lead to the disease, as indicated by transcription were also seen in Friedreich's ataxia cells consistent with a factors leaching from chromatin by mutant RNA in myot- repressive chromatin organizatio]. The "Histone Code" hypothesis proposed that amino-terminal exten-sions of histones are subject to a variety of posttransla- tional modifica. Histone methylation and In summary, our data provide evidence of epigenetic and acetylation are among the best characterized modifica- post-transcriptional regulations of the ATXN8OS expres- tions to regulate gene expression through alterations in sion. Although the in vitro cell culture study may not truly the chromatin structure. In chromatin domains that are reflect the pathological events in vivo, our study may shed transcriptionally repressed, high levels of histone H3-K9 insights into the pathogenesis of this disease.
methylation and H3-K14 hypoacetylation were observed.
Therefore, it is possible to predict the transcriptional com- petence of a particular genomic region by examining the ATXN8OS cDNA constructs
H3 methylation and acetylation patterns. Using ChIP Human cerebellum polyadenylated RNA (200 ng) (Clon- assay, we provided direct evidence of H3-K9 dimethyla- tech) was reverse transcribed using the SuperScript™ III tion and H3-K14 hypoacetylation and repression of reverse transcriptase (Invitrogen). Sense and antisense ATXN8OS RNA in the 157 CR cells (Fig. ). As reduced primers used for amplification of ATXN8OS cDNA were expression of adjacent HaloTag gene was also seen in the 5'-ATCCTTCACCTGTTGCCT-3' and 5'-GCTTGTGAG- 88ndent of H3-K9 dimethylation GACTGAGAATG-3', respectively. The 1.3-kb full-length, and H3-K14 hypoaceDNA methylation 23 CR [(CTA)11(CTG)12] containing cDNA (including or other histone modifications such as arginine methyla- exons D, C2, C1, B, anA cloned into tion and serine/threonine phosphoray be pGEM-T Easy vector (Promega) and sequenced. The (page number not for citation purposes) BMC Molecular Biology 2009, 10:9
cloned ATXN8OS cDNA containing 88 CR was made by The amount of HaloTag mRNA was determined by cus- replacing a 178 bp NlaIII-AflII fragment with a 373 bp tomized Assays-by-Design probe (Forward primer: fragment from the PCR clone of a PD patient CCGACGTGGGACGAATGG, Reverse primer: CGGAAG- [(CTA)8CCACTACTGCTACTGCTA(CTG)74]. The 88 GCCTGGAAGGT, TaqMan® probe: GAATTCGCCCGTGA) CR was further expanded to 157 CR (4331348, Applied Biosystems). Fold change was calcu- lated using the formula 2ΔCt, ΔCT = CT (control) - CT (tar- TA(CTG)9] by ligating Fnu4HI partially digested frag- get), in which CT indicates cycle threshold.
ments. The interruption of the CTG repeat tract by CCAand CTA is similar to that reported seen in SCA8 patients Cell viability assay
construct the clone without combined repeats (0 To quantify the cell viability, mitochondrial dehydroge- CR), a DraI site was added to the 5' end of CTA/CTG nase activity was evaluated by measuring its cleaving activ- repeats by site-directed mutagenesis using primer 5'- ity of the tetrazolium salt WST-1 (Takara). The cultured cell suspension was seeded into a 96-well plate. After stau- TAAAA(CTA)8C-3' and a 273-bp DraI fragment contain- rosporine (0 50 nM) treatment for one day, PreMix WST- ing 88 CTA/CTG combined repeats was removed. Then 1 was added to cells in a final 1:10 dilution and incubated the ATXN8OS cDNA was cloned into the NotI site of for 2 hours. The absorbance of the red colored formazan pcDNA5/FRT/TO vector (Invitrogen) for establishing sta- dye cleaved from WST-1 was measured at 450 nm by bly induced ATXN8OS CR cell lines. The pcDNA5/FRT/ Microplate Autoreader EL311 (Bio-Tek Instruments Inc.).
TO vector used was modified by inserting a 2.3 kb BglII Reactions were performed in triplicate. Annexin V-posi- (blunted)-FspI fragment containing CMV enhancer/pro- tive cells were determined using Annexin V-FITC Apopto- moter, HaloTag open reading frame and SV40 late sis Detection Kit (Strong Biotech Corp.) according to the poly(A) signal from pHT2 (Promega) at the PvuII site supplier's instructions. Each cell line was tested at least 3 between bovine growth hormone poly(A) signal and Flp times and apoptotic cells quantitatively determined by recombination target (FRT) site flow cytometry.
Cell culture and ATXN8OS CR cell lines
HEK-293-derived Flp-In™-293 (Invitrogen) cells were cul- Cells on five 10 cm dishes were incubated with 1% for- tivated in Dulbecco's modified Eagle's medium contain- maldehyde for 10 min to cross-link histones to DNA.
ing 10% fetal bovine serum in a 37°C humidified Cells were then washed twice using cold PBS containing incubator with a 5% CO2 atmosphere. The cloned protease inhibitors (1 mM phenylmethylsulfonyl fluo- pcDNA5/FRT/TO-ATXN8OS plasmids and vector were ride, 1 μg/ml aprotinin and 1 μg/ml pepstain A). The used to generate the ATXN8OS CR and vector only cell washed cells were scraped and resuspended in lysis buffer lines by targeting insertion into Flp-In™-293 cells, accord- (1% SDS, 10 mM EDTA, 50 mM Tris pH 8.1) (Upstate ing to the supplier's instructions. The repeats in these Biotechnology). The resulting lysate was subjected to son- ATXN8OS cell lines were examined by PCR and sequenc- ication. The sample was centrifuged to remove cell debris.
ing. These cell lines were grown in medium containing 5 The volume of the chromatin supernatant was divided μg/ml blasticidin and 100 μg/ml hygromycin. Doxycy- into several parts. The first part was used as input (posi- cline (1 μg/ml) was added to induce ATXN8OS expres- tive) control, and other parts were diluted with ChIP dilu- sion. To evaluate the stability of ATXN8OS transcripts, tion buffer containing protease inhibitors to bring up the actinomycin D (1 μg/ml) (A1410, Sigma) was added 48 volume to 2 ml. The chromatin solution was incubated hr after induction for 0, 3, 6, 9 and 12 hr for total RNA with 75 μl of a mixture of protein A-agarose/salmon sperm DNA slurry (Upstate Biotechnology) for 30 min at4°C with agitation to reduce nonspecific background. Fol- Real-time PCR quantification of mRNA
lowing this preclearing procedure, the solution was centri- Total RNA was extracted from ATXN8OS CR and vector fuged and the supernatant was collected. Five micrograms only cells using the Trizol (Invitrogen). The RNA was of anti-dimethyl H3-K9 or anti-acetylated-H3-K14 anti- DNase (Stratagene) treated, quantified, and reverse-tran- body (Upstate Biotechnology) were added to the chroma- scribed to cDNA using High Capacity cDNA Reverse Tran- tin solution and incubated overnight at 4°C with scription Kit (Applied Biosystems) with random primers.
rotation. The resulting immune complexes were collected Using ABI PRISM® 7000 Sequence Detection System by addition of 60 μl of protein A-agarose/salmon sperm (Applied Biosystems), real-time quantitative PCR was per- DNA slurry and incubated at 4°C with rotation for 1 h.
formed on a cDNA amount equivalent to 250 ng total The beads were washed several times and the attached RNA with TaqMan fluorogenic probes Hs01382089-m1 immune complexes were eluted with 250 μl elution buffer (exon C2 and C1 boundary) for ATXN8OS and 4326321E containing 1% SDS and 0.1 M NaHCO3. Cross-links were for HPRT1 (endogenous control) (Applied Biosystems).
reversed by the addition of 8 μl 5 M NaCl and incubating (page number not for citation purposes) BMC Molecular Biology 2009, 10:9
the samples at 65°C for 4 hr. The DNA was purified by ton X-100 treatment for 10 min, a Cy3-(CAG)10 (Operon) phenol/chloroform extraction and analyzed by PCR using the primer pair specific to the exons B to A region (5'- (unique toATXN8OS) oligonucleotide probe was added CAAACTTCAGAGAGAGAGGG-3' and 5'-CAGAGT- at 37°C for 2 hr for FISH exper TAATCTCTCCGTGG-3', 213 bp fragment) of ATXN8OS ed using DAPI (4'-6- cDNA gene (Fig. ).
diamidino-2-phenylindole). Fluorescent signals are visu-alized using a Leica TCS confocal laser scanning micro- Western blot analysis
scope optimized for simultaneous dual fluorescent ATXN8OS CR cells were lysed in RIPA buffer (10 mM Tris pH 7.5, 150 mM NaCl, 5 mM EDTA pH 8.0, 0.1% sodiumdodecyl sulphate (SDS), 1% deoxycholate, 1% NP-40) containing the protease inhibitor mixture (Sigma). After ICC and HYL carried out the experiment including sonication and sitting on ice for 30 min, the lysates were ATXN8OS CR construct engineering, stable lines estab- centrifuged at 13 000 g for 30 min at 4°C. Protein concen- lishment and expression studies. GCL carried out the trations were determined with the Bio-Rad protein assay experiment of ChIP-PCR. SHK performed cell viability kit, using albumin as standards. Laemmli sample buffer assay. GJLC supervised the study design. All authors par- was then added to 30 μg of protein and heated in a boiling ticipated in the revising of the manuscript.
water bath for 10 min. Equal amounts of protein fromeach sample were fractionated in a 12% SDS-polyacryla- mide gel electrophoresis (PAGE). The fractionated protein We thank Taiwan Foundation for Rare Disorders for providing thesis samples were transferred onto a nitrocellulose membrane scholarship and Confocal Microscope Lab, Instrumentation Center, NTU (Schleicher and Schuell), and non-specific binding was for technical support. This work was supported by grants NSC-94-3112-B- blocked in 5% non-fat dry milk for overnight at 4°C. After 003-001, NSC-95-3112-B-003-002 and NSC-97-2311-B-003-010-MY3 from the National Science Council, Executive Yuan, Taiwan; and 96TOP001 washing with Tris-buffered saline (TBS), the blots were from National Taiwan Normal University, Taipei, Taiwan.
probed with a 1:1000 dilution of HaloTag antibody(Promega) or a 1:10000 dilution of actin antibody (Chemicon) in TBS/1% bovine serum albumin/0.1% Koob MD, Moseley ML, Schut LJ, Benzow KA, Bird TD, Day JW, Tween 20 for 1 hr at room temperature with gentle shak- ing. After extensive washing, the blots were probed with a Nature genetics 1999,
21(4):379-384.
1:10000 dilution of goat anti-mouse or goat anti-rabbit Stevanin G, Herman A, Durr A, Jodice C, Frontali M, Agid Y, Brice IgG conjugated to horseradish peroxidase (Jackson Nature genetics 2000, 24(3):213. author reply 215.
ImmunoResearch). The blots were washed extensively Ikeda Y, Dalton JC, Moseley ML, Gardner KL, Bird TD, Ashizawa T, and the protein detected using Immobilon™ western Seltzer WK, Pandolfo M, Milunsky A, Potter NT, et al.: chemiluminescent HRP substrate (Millipore).
American journal of
human genetics 2004, 75(1):3-16.
Worth PF, Houlden H, Giunti P, Davis MB, Wood NW: ATXN8OS CR cells on coverslips were washed with PBS Nature genetics 2000, 24(3):214-215.
and fixed in 4% paraformaldehyde in PBS for 10 min, fol- Sobrido MJ, Cholfin JA, Perlman S, Pulst SM, Geschwind DH: lowed by 20 min incubation with 0.1% Triton X-100 in Neu-
rology 2001, 57(7):1310-1312.
PBS to permeabilize cells, overnight incubation with 0.5% Mosemiller AK, Dalton JC, Day JW, Ranum LP: Molecular genetics
bovine serum albumin in PBS to block non-specific bind- of spinocerebellar ataxia type 8 (SCA8). Cytogenetic and genome
ing. The primary anti-HaloTag antibody, diluted 1:500 in Schols L, Bauer I, Zuhlke C, Schulte T, Kolmel C, Burk K, Topka H, 1% BSA in TBS, was used to stain cells at 4°C overnight.
Bauer P, Przuntek H, After washing, cells were incubated for 2 hr at room tem- Annals of neurology 2003, 54(1):110-115.
Ikeda Y, Daughters RS, Ranum LP: Bidirectional expression of the
perature in FITC-conjugated secondary antibody diluted SCA8 expansion mutation: One mutation, two genes. Cere-
to 1:500 in TBS containing 1% BSA, and washed in PBS.
bellum (London, England); 2008. Nuclei were detected using 4'-6-diamidino-2-phenylin- Ranum LP, Day JW: Trends Genet 2004, 20(10):506-512.
dole (DAPI). The stained cells were examined after He Y, Zu T, Benzow KA, Orr HT, Clark HB, Koob MD mounted in Vectashield (Vector Laboratories Inc.) using a Leica TCS confocal laser scanning microscope.
J Neurosci 2006, Fluorescent in situ hybridization (FISH)
Moseley ML, Zu T, Ikeda Y, Gao W, Mosemiller AK, Daughters RS,Chen G, Weatherspoon MR, Clark HB, Ebner TJ, et al.: To examine the ribonuclear foci, cells were grown on cov- erslips, washed, and fixed for 15 min at room temperature in 4% formaldehyde and 10% acetic acid. After 0.1% Tri- Nature genetics 2006, 38(7):758-769.
(page number not for citation purposes) BMC Molecular Biology 2009, 10:9
Fu YH, Pizzuti A, Fenwick RG Jr, King J, Rajnarayan S, Dunne PW, Ranum LP, Day JW: Myotonic dystrophy: clinical and molecular
Dubel J, Nasser GA, Ashizawa T, de Jong P, et al.: parallels between myotonic dystrophy type 1 and type 2. Cur-
rent neurology and neuroscience reports 2002, 2(5):465-470.
Science 1992, 255(5049):1256-1258.
Koch KS, Leffert HL: Taneja KL, McCurrach M, Schalling M, Housman D, Singer RH Journal of theoretical biology The Journal of cell biology 1995, Ebralidze A, Wang Y, Petkova V, Ebralidse K, Junghans RP Davis BM, McCurrach ME, Taneja KL, Singer RH, Housman DE Nemes JP, Benzow KA, Moseley ML, Ranum LP, Koob MD: Proceedings of the National Academy of Sciences of the United States of America 1997, Human molecular genetics 2000, 9(10):1543-1551.
Otten AD, Tapscott SJ: Wu YR, Lin HY, Chen CM, Gwinn-Hardy K, Ro LS, Wang YC, Li SH, Hwang JC, Fang K, Hsieh-Li HM, et al.: ings of the National Academy of Sciences of the United States of America Timchenko LT, Miller JW, Timchenko NA, DeVore DR, Datar KV, Lin Clinical genetics 2004, 65(3):209-214.
L, Roberts R, Caskey CT, Swanson MS: Moseley ML, Schut LJ, Bird TD, Koob MD, Day JW, Ranum LP: Nucleic acids research 1996, Human molecular genetics 2000, 9(14):2125-2130.
Kino Y, Mori D, Oma Y, Takeshita Y, Sasagawa N, Ishiura S:Human molecular genetics 2004, 13(5):495-507.
Furuya H, Shinnoh N, Ohyagi Y, Ikezoe K, Kikuchi H, Osoegawa M,
Fukumaki Y, Nakabeppu Y, Hayashi T, Kira J: Biochemical pharmacology 2005,
69(3):503-516.
Rudnicki DD, Holmes SE, Lin MW, Thornton CA, Ross CA, Margolis
RL: Annals of neurology 2007,
61(3):272-282.
Shaw G, Morse S, Ararat M, GrahFaseb J 2002, 16(8):869-871.
Anborgh PH, Godin C, Pampillo M, Dhami GK, Dale LB, Cregan SP,Truant R, Fergus The Journal of biological chemistry 2005, Handa V, Goldwater D, Stiles D, Cam M, Poy G, Kumari D, Usdin K FEBS letters
2005, 579(12):2702-2708.
Strahl BD, Allis CD: Nature 2000, 403(6765):41-45.
Chen KY, Pan H, Lin MJ, Li YY, Wang LC, Wu YC, Hsiao KM:
Length-dependent toxicity of untranslated CUG repeats on
Caenorhabditis elegans. Biochemical and biophysical research com-
munications 2007, 352(3):774-779.
Saveliev A, Everett C, Sharpe T, Webster Z, Festenstein R: Nature 2003, 422(6934):909-913.
Bannister AJ, Zegerman P, Partridge JF, Miska EA, Thomas JO, Allshire
RC, Kouzarides T: Nature 2001,
410(6824):120-124.
Greene E, Mahishi L, Entezam A, Kumari D, Usdin K: scientist can read your work free of charge Nucleic acids research
2007, 35(10):3383-3390.
"BioMed Central will be the most significant development for Kim JK, Samaranayake M, Pradhan S: disseminating the results of biomedical researc h in our lifetime." Cell Mol Life Sci 2008 in press.
Sir Paul Nurse, Cancer Research UK Tian B, White RJ, Xia T, Welle S, Turner DH, Mathews MB, ThorntonCA: Your research papers will be: available free of charge to the entire biomedical community RNA 2000, 6(1):79-87.
Houseley JM, Wang Z, Brock GJ, Soloway J, Artero R, Perez-Alonso peer reviewed and published immediately upon acceptance M, O'Dell KM, Monckton DG: cited in PubMed and archived on PubMed Central Human molecular genetics 2005, yours — you keep the copyright Submit your manuscript here: (page number not for citation purposes)

Source: http://www.biol.ntnu.edu.tw/download.php?filename=97_5da856da.pdf&dir=writing_journal/8&title=%E6%AA%94%E6%A1%88%E4%B8%8B%E8%BC%89