Portal de Periódicos da CAPES

Protein Tyrosine Phosphatases Are Up-regulated and Participate in Cell Death Induced by Polyglutamine Expansion

2002; Elsevier BV; Volume: 277; Issue: 46 Linguagem: Inglês

10.1074/jbc.m206890200

1083-351X

Zhiliang Wu, Teresa M. O’Kane, Richard Scott, Mary J. Savage, Donna Bozyczko‐Coyne,

Cardiomyopathy and Myosin Studies

Polyglutamine expansion is the cause of several neurodegenerative diseases. An in vitro model of polyglutamine-induced neuronal cell death was developed using truncated mutant huntingtin (Htt) and PC12 cells. Cell death was specifically observed in cells expressing a truncated mutant huntingtin-green fluorescence protein (GFP) fusion protein with 118 glutamine repeats (Gln118), as demonstrated by the release of lactate dehydrogenase (LDH). To gain further insights into the mechanisms of polyglutamine expansion-induced cell death, the Affymetrix rat genome array U34A was used to investigate gene expression changes associated with polyglutamine-mediated protein aggregation and cell death. Among the up-regulated genes, the increase of four protein tyrosine phosphatases (PTPs) was further confirmed by real-time quantitative reverse transcription PCR. Protein expression of mitogen activated protein (MAP) kinase phosphatase 1 (MKP1) was also increased as demonstrated by Western blot. Furthermore, phosphorylation of MAP kinase extracellular signal-regulated kinase 1/2 (ERK1/2) was substantially reduced in association with protein aggregation, and two general PTP inhibitors, sodium orthovanadate and bpV(pic), dramatically rescued the cells from polyglutamine-induced cell death. These results suggest that one or more of the PTPs are involved in the polyglutamine-induced cell death. Polyglutamine expansion is the cause of several neurodegenerative diseases. An in vitro model of polyglutamine-induced neuronal cell death was developed using truncated mutant huntingtin (Htt) and PC12 cells. Cell death was specifically observed in cells expressing a truncated mutant huntingtin-green fluorescence protein (GFP) fusion protein with 118 glutamine repeats (Gln118), as demonstrated by the release of lactate dehydrogenase (LDH). To gain further insights into the mechanisms of polyglutamine expansion-induced cell death, the Affymetrix rat genome array U34A was used to investigate gene expression changes associated with polyglutamine-mediated protein aggregation and cell death. Among the up-regulated genes, the increase of four protein tyrosine phosphatases (PTPs) was further confirmed by real-time quantitative reverse transcription PCR. Protein expression of mitogen activated protein (MAP) kinase phosphatase 1 (MKP1) was also increased as demonstrated by Western blot. Furthermore, phosphorylation of MAP kinase extracellular signal-regulated kinase 1/2 (ERK1/2) was substantially reduced in association with protein aggregation, and two general PTP inhibitors, sodium orthovanadate and bpV(pic), dramatically rescued the cells from polyglutamine-induced cell death. These results suggest that one or more of the PTPs are involved in the polyglutamine-induced cell death. A growing number of neurodegenerative diseases result from the expansion of unstable trinucleotide CAG repeats, including Huntington's disease (HD), 1The abbreviations used are: HD, Huntington's disease; Htt, huntingtin; GFP, green fluorescent protein; 118Q, Htt·GFP protein with 118 Gln repeats (Gln118); 17Q, Htt·GFP protein with Gln17; bpV(pic), dipotassium bisperoxo(picolinato)oxovanadate (V); LDH, lactate dehydrogenase; PTP, protein tyrosine kinase; RT, reverse transcription; MAP, mitogen-activated protein; MKP, MAP kinase phosphatase; ERK, extracellular signal-regulated kinase; NLS, nuclear localization signal; DMEM, Dulbecco's modified Eagle's medium; NGF, nerve growth factor; Dox, doxycycline 1The abbreviations used are: HD, Huntington's disease; Htt, huntingtin; GFP, green fluorescent protein; 118Q, Htt·GFP protein with 118 Gln repeats (Gln118); 17Q, Htt·GFP protein with Gln17; bpV(pic), dipotassium bisperoxo(picolinato)oxovanadate (V); LDH, lactate dehydrogenase; PTP, protein tyrosine kinase; RT, reverse transcription; MAP, mitogen-activated protein; MKP, MAP kinase phosphatase; ERK, extracellular signal-regulated kinase; NLS, nuclear localization signal; DMEM, Dulbecco's modified Eagle's medium; NGF, nerve growth factor; Dox, doxycyclinespinobulbar muscular atrophy, spinocerebellar ataxia types 1, 2, 6, and 7, Machado-Joseph Disease, and dentatorubral pallidoluysian atrophy. Because the CAG trinucleotide encodes the amino acid glutamine, these disorders are often referred to as polyglutamine diseases, although the polyglutamine expansions occur in different proteins (1Zoghbi H.Y. Orr H.T. Annu. Rev. Neurosci. 2000; 23: 217-247Crossref PubMed Scopus (1082) Google Scholar). One of the well studied examples is HD, an inherited progressive neurological disorder caused by CAG expansion in the Htt gene (2The Huntington's Disease Collaborative Research Group Cell. 1993; 72: 971-983Abstract Full Text PDF PubMed Scopus (6893) Google Scholar). Although research in the past several years has revealed many important characteristics of the mutant Htt, both the biological function of normal Htt and the pathogenesis of HD are still not fully understood. Normal Htt is a ∼350 kDa cytoplasmic protein with a short polyglutamine tract at the N terminus. It has multiple predicted caspase cleavage sites and is cleaved by caspase 1 and caspase 3in vitro (3Goldberg Y.P. Nicholson D.W. Rasper D.M. Kalchman M.A. Koide H.B. Graham R.K. Bromm M. Kazemi-Esfarjani P. Thornberry N.A. Vaillancourt J.P. Hayden M.R. Nat. Genet. 1996; 13: 442-449Crossref PubMed Scopus (498) Google Scholar, 4Wellington C.L. Ellerby L.M. Hackman A.S. Margolis R.L. Trifiro M.A. Singaraja R. McCutcheon K. Salvesen G.S. Propp S.S. Bromm M. Rowland K.J. Zhang T. Rasper D. Roy S. Thornberry N. Pinsky L. Kakizuka A. Ross C.A. Nicholson D.W. Bredesen D.E. Hayden M.R. J. Biol. Chem. 1998; 273: 9158-9167Abstract Full Text Full Text PDF PubMed Scopus (491) Google Scholar) as well as by calpain (5Gafni J. Ellerby L.M. J. Neurosci. 2002; 22: 4842-4849Crossref PubMed Google Scholar). Studies in both cell cultures and transgenic mice indicate that truncated forms of mutant Htt are more toxic than the full-length mutant protein and that the toxicity is dependent on the degree of polyglutamine expansion (6Cooper J.K. Schilling G. Peters M.F. Herring W.J. Sharp A.H. Kaminsky Z. Masone J. Khan F.A. Delanoy M. Borchelt D.R. Dawson V.L. Dawson T.M. Ross C.A. Hum. Mol. Genet. 1998; 7: 783-790Crossref PubMed Scopus (305) Google Scholar, 7Lunkes A. Mandel J.L. Hum. Mol. Genet. 1998; 7: 1355-1361Crossref PubMed Scopus (173) Google Scholar, 8Hackam A.S. Singaraja R. Wellington C.L. Metzler M. McCutcheon K. Zhang T. Kalchman M. Hayden M.R. J. Cell Biol. 1998; 141: 1097-1105Crossref PubMed Scopus (280) Google Scholar, 9Martindale D. Hackam A. Wieczorek A. Ellerby L. Wellington C. McCutcheon K. Singaraja R. Kazemi-Esfarjani P. Devon R. Kim S.U. Bredesen D.E. Tufaro F. Hayden M.R. Nat. Genet. 1998; 18: 150-154Crossref PubMed Scopus (418) Google Scholar, 10Moulder K.L. Onodera O. Burke J.R. Strittmatter W.J. Johnson Jr., E.M. J. Neurosci. 1999; 19: 705-715Crossref PubMed Google Scholar). Transgenic mice expressing exon 1 of the Htt gene with 100–150 CAG repeats develop a number of features reminiscent of HD, including a late onset neurodegeneration within the anterior cingulate cortex, dorsal striatum, and cerebellum (11Turmaine M. Raza A. Mahal A. Mangiarini L. Bates G.P. Davies S.W. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 8093-8097Crossref PubMed Scopus (379) Google Scholar) as well as the development of brain nuclear inclusions and a behavioral phenotype (12Mangiarini L. Sathasivam K. Seller M. Cozens B. Harper A. Hetherington C. Lawton M. Trottier Y. Lehrach H. Davies S.W. Bates G.P. Cell. 1996; 87: 493-506Abstract Full Text Full Text PDF PubMed Scopus (2536) Google Scholar,13Davies S.W. Turmaine M. Cozens B.A. DiFiglia M. Sharp A.H. Ross C.A. Scherzinger E. Wanker E.E. Mangiarini L. Bates G.P. Cell. 1997; 90: 537-548Abstract Full Text Full Text PDF PubMed Scopus (1876) Google Scholar, 14Carter R.J. Lione L.A. Humby T. Mangiarini L. Mahal A. Bates G.P. Dunnett S.B. Morton A.J. J. Neurosci. 1999; 19: 3248-3257Crossref PubMed Google Scholar). Although the use of truncated Htt may not fully recapture the disease process initiated by full-length mutant Htt, it is still a valuable approach to dissecting biochemical mechanisms associated with polyglutamine toxicity. The mechanisms of neurodegeneration in polyglutamine diseases are not fully known. Genome-wide gene expression has been evaluated in bothin vivo (15Luthi-Carter R. Strand A. Peters N.L. Solano S.M. Hollingsworth Z.R. Menon A.S. Frey A.S. Spektor B.S. Penney E.B. Schilling G. Ross C.A. Borchelt D.R. Tapscott S.J. Young A.B. Cha J.-H. Olson J.M. Hum. Mol. Genet. 2000; 9: 1259-1271Crossref PubMed Scopus (608) Google Scholar, 16Iannicola C. Moreno S. Oliverio S. Nardacci R. Ciofi-Luzzatto A. Piacentini M. J. Neurochem. 2000; 75: 830-839Crossref PubMed Scopus (67) Google Scholar) and in vitro models (17Wyttenbach A. Swartz J. Kita H. Thykjaer T. Carmichael J. Bradley J. Brown R. Maxwell M. Schapira A. Orntoft T.F. Kato K. Rubinsztein D.C. Hum. Mol. Genet. 2001; 15: 1829-1845Crossref Google Scholar) of HD. Diminished expression of neurotransmitter, calcium, and retinoid signaling pathways was observed in the R6/2 mouse, a transgenic HD model (15Luthi-Carter R. Strand A. Peters N.L. Solano S.M. Hollingsworth Z.R. Menon A.S. Frey A.S. Spektor B.S. Penney E.B. Schilling G. Ross C.A. Borchelt D.R. Tapscott S.J. Young A.B. Cha J.-H. Olson J.M. Hum. Mol. Genet. 2000; 9: 1259-1271Crossref PubMed Scopus (608) Google Scholar). A study using a small panel cDNA array showed that some genes were down-regulated and that others were significantly up-regulated in the affected brain (16Iannicola C. Moreno S. Oliverio S. Nardacci R. Ciofi-Luzzatto A. Piacentini M. J. Neurochem. 2000; 75: 830-839Crossref PubMed Scopus (67) Google Scholar). DNA microarray analysis on a PC12 cellular model of polyglutamine toxicity (17Wyttenbach A. Swartz J. Kita H. Thykjaer T. Carmichael J. Bradley J. Brown R. Maxwell M. Schapira A. Orntoft T.F. Kato K. Rubinsztein D.C. Hum. Mol. Genet. 2001; 15: 1829-1845Crossref Google Scholar) identified 56 genes with significant changes in mRNA expression by comparing a mutant cell line with a control cell line at one time point. Yet, the role of each gene in polyglutamine toxicity was not further investigated in any reported study. Here, stable PC12 cell lines with inducible expression of Htt·GFP·NLS were developed and characterized. In mutant cells with Gln118, designated here as 118Q cells, protein aggregates developed predominantly in the cytoplasm and perinuclear area from 16 h after induction. Cell death followed expression of Gln118 Htt·GFP but not of the Gln17 Htt·GFP as measured by LDH release. Gene expression was examined at 8, 16, and 48 h after induction in both 17Q and 118Q cells using the Affymetrix U34A rat genome array. Among the up-regulated candidates were four PTPs. Their increased expressions were confirmed by real-time quantitative RT-PCR, and the protein levels of one, MKP1, was also increased. The functional importance of these phosphatases was demonstrated using pharmacological inhibitors that increased both the levels of activated extracellular signal-regulated kinase and cell survival. The DNA of full-length huntingtin with Gln17 or Gln118repeats (kindly supplied by Dr. Jean-Louis Mandel) was digested withAvrII and BamHI (New England Biolabs, Beverly, MA). The DNA fragments containing the truncated huntingtingene (corresponding to the 304- and 405-amino acid N-terminal fragment of the control and the mutant huntingtin, respectively) were purified and ligated to PCR products of GFP·NLS (nuclear localization signal), resulting in expression constructs of Htt·GFP·NLS. The NLS primer sequences are 5′-GATCTCCAAAAAAGAAGAGAAAGGTAG-3′ and 5′-TCGACTACCTTTCTCTTCTTTTTTGGA-3′. The NLS was first added into the GFP gene between the SalI andBglII sites. To amplify the GFP·NLSgene, a primer with the BamHI site (5′-TCCGGTGGATCCCTAGCCCGCGGTAC-3′) and a primer with theAvrII site (5′-CGCCACCCTAGGTAGCAAGGG-3′) were used. The PCR program was 94 °C (30 s), 55 °C (10 s), and 72 °C (1 min) for 25 cycles. All the clones of thehuntingtin gene were transformed into Escherichia coli SURE (Stratagene, La Jolla, CA) competent cells, and the transformed strains were cultured at 30 °C to reduce the instability of CAG repeats. DNA was isolated using a Qiagen (Valencia, CA) plasmid DNA purification kit. PC12 Tet-Off cells (Clontech) were co-transfected with both the pTK-Hy DNA and the expression plasmid DNA (1:10, 1 μg/well of 6-well plates) using the Effectene transfection reagent (Qiagen). The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 7.5% defined fetal bovine serum, 7.5% horse serum, 10 mm HEPES, and 1 μg/ml doxycycline for 4 days and then subjected to G418 (500 μg/ml) and hygromycin (200 μg/ml) selection after splitting the cells. The cells were fed with fresh media every 4 days for 3–4 weeks when single colonies appeared. Each clone was tested in low serum (0.18% fetal bovine serum/horse serum, 1:1) DMEM containing 50 ng/ml NGF (Sigma) for Htt·GFP·NLS induction and neuronal differentiation in the absence of doxycycline. Clones with a higher induction of GFP fusion proteins were selected and characterized further. All images were taken on a Nikon ECLIPSE TE300 inverted microscope with an ELWD 20x/0.45 lens using an Optronics Magnafire digital camera. PC12 stable cell lines were routinely cultured in DMEM growth media (7.5% defined fetal bovine serum, 7.5% horse serum, 100 μg/ml G418, and 100 μg/ml hygromycin) with 10 ng/ml doxycycline. For cell survival assays, the cells were seeded in doxycycline-free low serum (0.18%) DMEM with NGF (50 ng/ml) at low density (5 × 103 cells/well) in 96-well plates coated with laminin and poly-l-ornithine (Sigma). The media were collected at different time points after induction, and released LDH was measured according to the manufacturer's instruction (Roche Molecular Biochemicals). Sodium orthovanadate or bpV(pic) (dipotassium bisperoxo(picolinato)oxovanadate (V)) were purchased from Alexis Biochemicals (San Diego, CA). Stock solutions were prepared in water and stored at −20 °C. Affymetrix rat genome array U34A contains probe sets representing all known rat genes and some expressed sequence tags. For gene expression analysis on an array, the procedure recommended by Affymetrix was followed. Total RNA was isolated using Trizol (Invitrogen) and purified by RNeasy spin column (Qiagen) according to the manufacturer's protocols. RNA was eluted in nuclease-free water and stored at −80 °C. The first strand of cDNA was synthesized from T7-(dT)24 primer with Superscript II reverse transcriptase (Invitrogen). Each reverse transcription reaction included 8 μg of RNA and 200 units of transcriptase in a total volume of 20 μl. After the second strand synthesis, the cDNA was phenol/chloroform extracted and ethanol precipitated. The final cDNA pellet was dissolved in 12 μl of water. Biotinylated cRNA was synthesized using the BioArray high yield RNA transcript labeling kit (Enzo Diagnostics, Farmingdale, NY) and fragmented according to Affymetrix protocol. Rat U34A array was hybridized, washed, stained, and scanned using a GeneChip fluidics station 400 (Affymetrix) and a GeneArray scanner (Agilent Technologies, Palo Alto, CA). The data of each array were collected and initially analyzed using Microarray Suite 4.0 software. Affymetrix empirical algorithms were used at default settings for both absolute and comparison analysis. For comparison of multiple arrays, the average signal intensity of each array was scaled to 250 target intensity. For further analysis of multiple comparisons, Affymetrix MicroDB and data mining tool (version 2.0) were used. Duplicate total RNA samples were prepared from both 17Q and 118Q cells cultured for 8, 16, and 48 h in doxycycline-free medium, respectively. Each RNA sample was converted to cDNA followed by cRNA synthesis, hybridization, and chip scanning. Four-pair comparison analysis was performed to identify differentially regulated genes at 16 and 48 h, using the expression at 8 h as a baseline by Affymetrix MS4.0 software. Genes scored as increased or decreased more than three times in all four-pair comparisons were considered high confidence changes. Other genes were considered unchanged. The -fold changes under "Results" are the mean of the four-pair comparisons. Real-time quantitative RT-PCR was performed on the ABI PRISM 7000 sequence detection system (Applied Biosystems, Foster City, CA) by monitoring the increase of fluorescence by the binding of SYBR Green to double-stranded DNA. Dissociation analysis was performed at the end of each PCR reaction to ensure there was only specific product. The PCR primers were designed with PRIMER EXPRESS software (Applied Systems) at the default setting for a small amplicon of the PCR product. The first strand cDNA template was synthesized from 5 μg of total RNA using oligo(dT) and diluted into 500 μl of water. For a 50-μl PCR, a 5-μl cDNA template was mixed with forward and reverse primers (250 nm each primer at final concentration) and 2× SYBR Green PCR Master Mix (Applied Biosystems). The reaction was run at the default setting program (95 °C (15 s), 60 °C (1 min), 40 cycles). Gene-specific PCR was performed in triplicate on each cDNA sample. The gene-specific primers, with GenBankTM accession numbers in parentheses, were as follows: MKP1 (U02553), 5′-TGAGGACTAACCGAGTGAAG-3′, 5′-TGCGGTCAAGTCATTGTTATG-3′; MKP3 (X94185), 5′-ATCACTGGAGCCAAAACCTG-3′, 5′-GAGAAACTGAAGGAATGAGAC-3′; MKP cpg21 (AF013144), 5′-GCTTACCTCATGAAGACCAAG-3′, 5′-TTGCACGGGTATGAAATGTC-3′; and TD14 (AF077000), 5′-GAGCAAGCAGAACTTTCTGC-3′, 5′-GTCAGGTTCCGCAAAGGAAG-3′. For quantification of gene expression changes, the ΔΔCt method was used to calculate relative -fold changes normalized against the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene as described in user bulletin number 2 (Applied Biosystems). After rinsing with phosphate-buffered saline, cells were lysed in 150 μl of FRAK buffer (1% Triton X-100, 50 mm NaCl, 30 μm sodium pyrophosphate, 50 mm sodium fluoride, 1 mmsodium orthovanadate, 10 mm Tris-HCl, pH 7.6) containing protease inhibitors (1 mm Pefabloc, 10 μg/ml pepstatin, 10 μg/ml aprotinin, and 1 mm E-64) on ice for 30 min. The lysates were then collected and homogenized by passing through a 21-gauge needle attached to a syringe. After centrifugation (20 min at 14,000 rpm), the supernatants were collected, and an aliquot was taken for protein determination by BCA assay (Pierce). Specific phospho-p44/42 MAP kinase (Thr-202/Tyr-204) antibody (catalog no. 9101) and p44/42 MAP kinase antibody (catalog no. 9102) were purchased from Cell Signaling(Beverly, MA). The MKP1 antibody was purchased from Sigma and used at 1:1000 dilution. The previous protocol (18Bozyczko-Coyne D. O'Kane T.M., Wu, Z.-L. Dobrzanski P. Murthy S. Vaught J.L. Scott R.W. J. Neurochem. 2001; 77: 849-863Crossref PubMed Scopus (154) Google Scholar) was followed with minor modifications. In brief, samples were mixed with 5× Laemmli buffer and heated for 2 min at 95 °C. Samples of equivalent total protein (usually 10 μg) were run in NuPage Bis-Tris 10% gels (Invitrogen) and transferred to PVDF membrane (Bio-Rad). The membranes were blocked in 5% nonfat milk, 0.05% Tween 20, and Tris-buffered saline and then incubated with antibody diluted in 0.05% Tween 20 and Tris-buffered saline for 1 h at room temperature. Following three 15-min washes, the blots were incubated with horseradish peroxidase-conjugated secondary goat antibody (Southern Biotech) diluted 1:10,000 in 0.05% Tween 20/Tris-buffered saline. The blots were then washed three times. Enhanced chemiluminescence was used to detect signals developed on Hyperfilm MP (Amersham Biosciences). The images were analyzed using the FOTO/ANALYST® electronic documentation system (Fotodyne, Hartland, WI) and the Gel-Pro® analyzer (Media Cybernetics, Silver Spring, MD). Full-length huntingtin cDNA (6Cooper J.K. Schilling G. Peters M.F. Herring W.J. Sharp A.H. Kaminsky Z. Masone J. Khan F.A. Delanoy M. Borchelt D.R. Dawson V.L. Dawson T.M. Ross C.A. Hum. Mol. Genet. 1998; 7: 783-790Crossref PubMed Scopus (305) Google Scholar) was truncated at theAvrII site and fused to the coding sequence of the GFP. Constructs were generated for tetracycline-regulated expression of Htt·GFP fusion proteins with an SV40 NLS. The NLS was engineered into the 3′-end of the GFP gene before the TGA stop codon. The NLS was added to increase cellular toxicity of the Htt·GFP protein according to published reports (19Peters M.F. Nucifora Jr., F.C. Kushi J. Seaman H.C. Cooper J.K. Herring W.J. Dawson V.L. Dawson T.M. Ross C.A. Mol. Cell. Neurosci. 1999; 14: 121-128Crossref PubMed Scopus (167) Google Scholar, 20Saudou F. Finkbeiner S. Devys D. Greenberg M.E. Cell. 1998; 95: 55-66Abstract Full Text Full Text PDF PubMed Scopus (1351) Google Scholar). The constructs were confirmed by DNA sequencing (Fig. 1 A). PC12 Tet-Off cells were transfected, and stable cell lines were selected. Each cell line was examined by fluorescence microscopy for the presence of the Htt·GFP protein following induction in the absence of doxycycline (Dox). Clones with a higher induction of the fusion proteins were selected for further characterization. Among the cell lines expressing the Htt·GFP·NLS, a control cell line, 17Q, and a mutant cell line, 118Q, were used in this study. Each of these cell lines had the highest expression in its group. Expression and localization of the Htt·GFP·NLS fusion proteins was examined in live cells at different time points after induction by fluorescence microscopy. Both the 17Q and 118Q stable lines exhibited evenly distributed proteins throughout the cytoplasm and nucleus at 8 h after induction (Fig.1 B). Several small protein aggregates often appeared in the cytoplasm and/or the perinuclear region in the 118Q cells and became a single large aggregate at later time points following the Htt fusion protein induction. The aggregates were present in roughly 40 and 90% of the cells at 16 and 48 h, respectively (Fig.1 B). The cellular toxicity of the mutant proteins was assessed by release of LDH at various time points after induction. Across a 4-day time course (Fig. 1 C) there was a significant increase of released LDH by 2 days after induction in the 118Q cells. Compared with the 118Q cells cultured with Dox, the LDH release in the absence of Dox was increased by more than 5- and 10-fold at 3 and 4 days, respectively. In contrast, there was no significant difference of LDH release in the 17Q cells either in the presence or absence of Dox. Furthermore, the 118Q cells had 3–4-fold increase of LDH release at the equivalent level of Htt·GFP protein expression compared with the 17Q cells (data not shown). Thus, the cell death is caused by the mutant protein and not purely by protein overexpression. Based on the time course of the mutant Htt·GFP protein localization and aggregation (see Fig. 1 B), the 118Q cells at 8, 16, and 48 h were defined as early, middle, and late stages of the polyglutamine-induced cell death, respectively. In the early stage, the cells looked similar to normal cells; therefore, the gene expression at 8 h was used as the baseline to identify gene expression changes associated with protein aggregation and cell death at the middle and late stages. The mutant cells could be rescued by adding Dox up to 24 but not 48 h after the mutant protein expression (data not shown). By including a late time point (48 h) when the cells were committed to die, patterns of gene expression important for the execution of cell death can be revealed. Among the genes with high confidence changes, there were 111 up-regulated genes at 16 h and 253 at 48 h. Among them, 86 were found at both 16 and 48 h. The number of down-regulated genes was 210 at 16h and 480 at 48h. The down-regulated genes outnumbered the up-regulated genes at a ratio of ∼2:1 at each time point. The up-regulated PTP genes in the Affymetrix U34A array analysis are summarized in Table I. There are seven GenBankTM en-tries corresponding to four PTPs (MKP1, MKP3, CPG21, and TD14). Three of them are dual specificity protein phosphatases, also known as MAP kinase phosphatases or MKPs. In the 17Q cells, only MKP1 expression was increased. In the 118Q cells, MKP1, MKP3, and MKP cpg21 were increased at both 16 and 48 h. TD14 was up-regulated at 48 h only. Compared with the 17Q cells, MKP1 was induced more than 2-fold in the 118Q cells. The -fold changes at 16 and 48 h in the 118Q cells suggested that MKP1 and MKP3 were progressively increased from 8 to 48 h. This was the period of transition from a relatively normal state to aggregate formation and cell death. MKP cpg21 demonstrated a 9-fold increase at 16 and 48 h. The increase of TD14 by 34-fold was not confirmed by real-time RT-PCR (Fig. 2 A). This large increase was likely due to inappropriate calculation by the Affymetrix software, because the signal intensity at both 8 and 16 h was below zero (data not shown).Table IUp-regulation of protein tyrosine phosphatases in 17Q control and 118Q mutant cells by Affymetrix U34A array analysisGeneAccession17Q118Q16h/8h*48h/8h*16h/8h*48h/8h*MKP1S743511.72.33.55.1S81478NC1.83.34.5U02553NC3.14.98.3MKP3X94185NCNC9.716.0U42627NCNC2.64.4CPG21AF013144NCNC8.99.1TD14AF077000NCNCNC34.0Asterisk denotes -fold change. NC, no change. Open table in a new tab Asterisk denotes -fold change. NC, no change. The up-regulation of the four PTPs was further investigated by real-time quantitative RT-PCR (Fig. 2 A). In the 17Q cells, expressions of MKP1 and TD14 were increased 4- and 2-fold, respectively, at both 16 h and 48 h. The -fold changes of MKP1 were about 11- and 18-fold at 16 and 48 h in the 118Q cells, much greater than in the 17Q cells. This PCR result is consistent with the Affymetrix analysis, which also shows higher MKP1 induction at 48 than that at 16 h. This pattern of increase is also shown with MKP3 and TD14, but not with MKP cpg21. Both MKP3 and MKP cpg21 were increased in the 118Q cells, but not in the 17Q cells. This is also consistent with the data obtained by Affymetrix microarray analysis. To eliminate any effect of NGF on PTP induction and ensure that up-regulation of PTPs was caused by the mutant Htt·GFP protein and not clonal variation, the two independent cell lines, 118Q and 118Q-G, were cultured in growth media without NGF before real-time quantitative RT-PCR. Following gene induction in the absence of Dox, both cell lines showed a similar increase of MKP1 at 20 and 48 h against an 8-h baseline. The absence of the Gln118 Htt·GFP protein in both of the cell lines cultured in the presence of Dox had effect on MKP1 expression (Fig. 2 B). Protein expression of MKP1 was further analyzed by Western blot. There was no detectable MKP1 in the 17Q cells following the control Htt·GFP induction up to 48 h (data not shown). In contrast, an increase of the MKP1 protein was observed in the 118Q cells from 8 to 48 h following the mutant Htt·GFP induction (Fig.3). The lack of MKP1 protein in the 17Q cells on Western blot might be due to the low level of MKP1 expression and/or the low sensitivity of the antibody. Protein expressions of MKP3, cpg21, and TD14 were not investigated because of the lack of antibody reagents. A consequence of MKP enhancement would be the suppression of MAP kinase ERK1/2 activation. The activation of ERK1/2 was examined by Western blot using a specific anti-phospho-ERK1/2 antibody. At 24 and 36 h, respectively, the 118Q cells demonstrated a 60 and 80% decrease of ERK1 phosphorylation and a 40 and 70% decrease of ERK2 phosphorylation compared with that observed at 14 h (Fig. 4 A). This decrease of ERK1/2 activation from 14 to 36 h was coincident with the increase of protein aggregates in the 118Q cells (Fig.1 B). In contrast, such a decrease of ERK1/2 phosphorylation was not demonstrated in the 17Q cells (Fig. 4 B). This suggests that polyglutamine expansion leads to suppression of ERK1/2 phosphorylation. Furthermore, addition of PTP inhibitors enhanced ERK1/2 phosphorylation at 2 μm (Fig. 4 C). The PTP inhibitors were tested for their protective effects in the mutant 118Q cells. A significant decrease in LDH release was obtained with bpV(pic) or orthovanadate at concentrations equal to and greater than 2 μm (Fig. 5). After subtraction of the basal LDH release, there was 49 and 43% protection with bpV(pic) and orthovanadate at 2 μm, respectively. A greater protection of 81 and 71% was observed with 10 μmof bpV(pic) and orthovanadate, respectively. A general serine/threonine protein phosphatase inhibitor, okadaic acid, did not show protection in the same survival assay (data not shown). In this study, an in vitro model of polyglutamine-mediated cell death associated with truncated mutant Htt was developed in a stable PC12 cell line. The induced cell death was highly significant and easily detected by LDH assay at 3 days after induction. To gain new insights regarding molecular mechanisms possibly contributing to the polyglutamine toxicity, gene expressions were evaluated using Affymetrix rat genome array U34A. As a result, 86 genes were found to be up-regulated at 16 and 48 h following induction, coincident with mutant Htt protein aggregation. Among the 86 up-regulated genes, four PTPs were especially interesting based on their role in the regulation of the MAP kinase signaling pathway (21Keyse S.M. Curr. Opin. Cell Biol. 2000; 12: 186-192Crossref PubMed Scopus (700) Google Scholar). The up-regulation of the four PTPs was further confirmed by real-time quantitative RT-PCR. The increase of MKP1 was demonstrated both at the levels of mRNA and protein. A recently published study (22Seta K.A. Kim R. Kim H.W. Millhorn D.E. Beitner-Johnson D. J. Biol. Chem. 2001; 276: 44405-44412Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar) using a similar approach showed that MKP1 is also induced at both the mRNA and protein levels by hypoxia in PC12 cells. MKP1, MKP3, and MKP cpg21 can be induced by oxidative stress, heat shock, and growth factors. In fact, MKP1 was first identified as an immediate early gene induced rapidly in cultured cells within 30–120 min of exposure to these agents (23Keyse S.M. Emslie E.A. Nature. 1992; 359: 644-646Crossref PubMed Scopus (568) Google Scholar). MKP1 expression in PC12 cells is slightly increased by agents promoting neuronal differentiation such as NGF (22Seta K.A. Kim R. Kim H.W. Millhorn D.E. Beitner-Johnson D. J. Biol. Chem. 2001; 276: 44405-44412Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar). In this study, the induction of MKP1 in the 118Q cells is certainly much higher in magnitude compared with the 17Q cells by both Affymetrix rat U34A array analysis and real-time PCR. Therefore, the induction of MKP1 is significantly related to the mutant Htt·GFP expression. Furthermore, the induction of all four PTPs is sustained in the progression of protein aggregation in 118Q cells. It is unclear how the polyglutamine-expanded protein affects the transcriptional control of these PTPs. Interestingly, MAP kinase p38 was suggested to play a role in the induction of MKP1 in response to stress (22Seta K.A. Kim R. Kim H.W. Millhorn D.E. Beitner-Johnson D. J. Biol. Chem. 2001; 276: 44405-44412Abstract Full Text Full Text PDF PubMed Scopus (69) Google Scholar, 24Li J. Gorospe M. Hutter D. Barnes J. Keyse S.M. Liu Y. Mol. Cell. Biol. 2002; 21: 8213-8224Crossref Scopus (169) Google Scholar). Perhaps the p38 kinase is also responsible in part for the induction of MKP1 in this PC12 model of polyglutamine toxicity. PTPs play an important role in the regulation of MAP kinase signaling pathways, which relay, amplify, and integrate signals from a diverse range of stimuli and elicit appropriate physiological responses including proliferation, differentiation, development, inflammation, and apoptosis (21Keyse S.M. Curr. Opin. Cell Biol. 2000; 12: 186-192Crossref PubMed Scopus (700) Google Scholar). Among the four PTPs identified by Affymetrix genome array analysis, three of them (MKP1, MKP3, and MKP cpg21) are dual-specificity (threonine/tyrosine) enzymes. They are known to dephosphorylate MAP kinase ERK1/2 both in vitro and in vivo (23Keyse S.M. Emslie E.A. Nature. 1992; 359: 644-646Crossref PubMed Scopus (568) Google Scholar, 25Zheng C.F. Guan K.L. J. Biol. Chem. 1993; 268: 16116-16119Abstract Full Text PDF PubMed Google Scholar, 26Groom L.A. Sneddon A.A. Alessi D.R. Dowd S. Keyse S.M. EMBO J. 1996; 15: 3621-3632Crossref PubMed Scopus (365) Google Scholar, 27Ishibashi T. Bottaro D.P. Michieli P. Kelley C.A. Aaronson S.A. J. Biol. Chem. 1994; 269: 29897-29902Abstract Full Text PDF PubMed Google Scholar). TD14 is not well studied but may also be involved in the regulation of MAPK signaling based on its homology to yeast protein BRO1, which is involved in the MAP kinase signaling pathway (28Cao L. Zhang L. Ruiz-Lozano P. Yang Q. Chien K.R. Graham R.M. Zhou M. J. Biol. Chem. 1998; 273: 21077-21083Abstract Full Text Full Text PDF PubMed Scopus (41) Google Scholar). Because MKPs were up-regulated in our model of polyglutamine toxicity, it was reasonable to think that the MAP kinase ERK1/2 signaling pathway would be suppressed. Decreased ERK1/2 phosphorylation was observed in the 118Q cells but not the 17Q cells from 14 to 36 h after induction of the Htt·GFP. Along this line, it was recently shown that MAP kinase ERK1/2 activation was attenuated in a stable PC12 cell line expressing mutant huntingtin (29Song C. Perides G. Liu Y.F. J. Biol. Chem. 2002; 277: 6703-6707Abstract Full Text Full Text PDF PubMed Scopus (62) Google Scholar) in response to NGF treatment. Furthermore, in our study, inhibitors of PTP significantly reversed both the cell death and the reduction in phospho-ERK 1/2 associated with the truncated mutant Htt·GFP expression. Therefore, inhibition of the MAP kinase ERK1/2 pathway mediated by increased MKPs is likely an important part of polyglutamine toxicity. Further studies will be of interest to determine the role of individual MKPs in polyglutamine toxicity. In conclusion, a cellular model of polyglutamine-induced toxicity has been developed that recapitulates some of the pathological features of protein aggregation and neuronal cell loss found in the HD brain. The presented evidence shows that MKPs may be involved in the polyglutamine-induced cell death and may serve as therapeutic targets for polyglutamine diseases. We thank Dr. Jean-Louis Mandel (Institute de Génétique et de Biologie Moléculaire et Cellulaire, CNRS/LGME, INSERM Unité 184, ULP, Parc d'Innovation, BP 163, 67404 Illkirch cedex, France) for providing mouse huntingtin cDNA.