Inhibition of mixed lineage kinase 3 attenuates MPP+-induced neurotoxicity in SH-SY5Y cells

Brain Research 1003 (2004) 86 – 97 www.elsevier.com/locate/brainres

Research report

Inhibition of mixed lineage kinase 3 attenuates MPP+-induced neurotoxicity in SH-SY5Y cells Joanne R. Mathiasen a,*, Beth Ann W. McKenna a, Michael S. Saporito a, Ghanashyam D. Ghadge b, Raymond P. Roos b, Beverly P. Holskin a, Zhi-Liang Wu a, Stephen P. Trusko a, Thomas C. Connors a, Anna C. Maroney a, Beth Ann Thomas a, Jeffrey C. Thomas a, Donna Bozyczko-Coyne a a

Neurobiology, Cephalon, Inc., 145 Brandywine Parkway, West Chester, PA 19380, USA b Department of Neurology, The University of Chicago, Chicago, IL 60637, USA Accepted 3 November 2003

Abstract The neuropathology of Parkinson’s Disease has been modeled in experimental animals following MPTP treatment and in dopaminergic cells in culture treated with the MPTP neurotoxic metabolite, MPP+. MPTP through MPP+ activates the stress-activated c-Jun N-terminal kinase (JNK) pathway in mice and SH-SY5Y neuroblastoma cells. Recently, it was demonstrated that CEP-1347/KT7515 attenuated MPTPinduced nigrostriatal dopaminergic neuron degeneration in mice, as well as MPTP-induced JNK phosphorylation. Presumably, CEP-1347 acts through inhibition of at least one upstream kinase within the mixed lineage kinase (MLK) family since it has been shown to inhibit MLK 1, 2 and 3 in vitro. Activation of the MLK family leads to JNK activation. In this study, the potential role of MLK and the JNK pathway was examined in MPP+-induced cell death of differentiated SH-SY5Y cells using CEP-1347 as a pharmacological probe and dominant negative adenoviral constructs to MLKs. CEP-1347 inhibited MPP+-induced cell death and the morphological features of apoptosis. CEP-1347 also prevented MPP+-induced JNK activation in SH-SY5Y cells. Endogenous MLK 3 expression was demonstrated in SH-SY5Y cells through protein levels and RT-PCR. Adenoviral infection of SH-SY5Y cells with a dominant negative MLK 3 construct attenuated the MPP+mediated increase in activated JNK levels and inhibited neuronal death following MPP+ addition compared to cultures infected with a control construct. Adenoviral dominant negative constructs of two other MLK family members (MLK 2 and DLK) did not protect against MPP+induced cell death. These studies show that inhibition of the MLK 3/JNK pathway attenuates MPP+-mediated SH-SY5Y cell death in culture and supports the mechanism of action of CEP-1347 as an MLK family inhibitor. D 2004 Elsevier B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Degenerative disease: Parkinson’s Keywords: Adenovirus; CEP-1347; c-Jun N-terminal kinase; Mixed lineage kinase; MPP+; SH-SY5Y

1. Introduction Degeneration of nigro-striatal dopaminergic (DA) neurons is a neuropathological hallmark of Parkinson’s Disease (PD). Part of the underlying degenerative process is thought to be due to a selective vulnerability of DA neurons to mitochondrial dysfunction. This mitochondrial dysfunction * Corresponding author. Tel.: +1-610-738-6634; fax: +1-610-3440065. E-mail address: [email protected] (J.R. Mathiasen). 0006-8993/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2003.11.073

is widespread throughout many cell types of the PD patient [54,56]. Thus, great interest lies in understanding the key biochemical events that are triggered following blockade of mitochondrial respiration and are causal to neuronal death. Mechanistic studies of DA neuron death akin to PD are routinely conducted using toxins that interfere with electron transport at the site of mitochondrial complex I and/or complex II. Classically, MPTP neurotoxicity has gained broad acceptance as a model of PD since it is a potent and selective nigro-striatal DA neurotoxin that produces PD-like symptoms in humans, non-human primates and

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

mice [1,4,17,25,26,34]. Moreover, MPP+, the neurotoxic byproduct of MPTP metabolic oxidation [43,50,62,65], is selectively taken up into DA neurons and inhibits complex I of mitochondrial electron transport. Several intraneuronal signaling pathways have been implicated in MPTP-induced neurotoxicity in vivo. For example, mice overexpressing a mitochondrial membrane signaling protein, Bcl-2, or mice that are deficient in another signaling peptide, p53, are resistant to MPTP-induced neurotoxicity [44,64,69]. MPTP also activates mitogen-activated protein kinase kinase 4 (MKK4) and a downstream substrate, c-Jun N-terminal kinase (JNK) in both the striatum and substantia nigra of mice [52]. Overexpression of all members of the mixed lineage kinase (MLK) family leads to an activation of the JNK pathway [5,9,13,21,27,49,58]. Downstream of the MLKs are the dual-specificity mitogen activated protein kinase kinases (MKK4 and MKK7) which phosphorylate JNKs on serine and threonine residues [14]. Gene transfer of JNK-interacting protein-1 (JIP) in SHSY5Y cells and mice has implicated the JNK pathway in MPTP (MPP+)-induced DA cell death [67] and adenoviral expression of dominant negative c-Jun in an axotomyinduced rat model of DA cell death has demonstrated inhibition of c-Jun activation and cell death [8]. Of note, in a variety of neuronal cell types, signaling through these pathways manifests in morphological and biochemical features of apoptosis, including nuclear chromatin condensation, membrane blebbing, DNA laddering and activation of caspase(s) [11,12,23,41,59]. Additionally, an inhibitor of the JNK pathway, the indolocarbazole CEP-1347/KT7515 [36], attenuates MPTPmediated nigrostriatal DA neuron loss in mice [51]. Recent studies exemplify that CEP-1347 likely inhibits JNK activation indirectly through the MLK family [37]. To establish a direct role of MLK in DA neuron death elicited by MPP+, studies were conducted in neuronally differentiated SHSY5Y cells, evaluating CEP-1347 protection against and interception of a variety of events associated with apoptotic death. MPP+-treated SH-SY5Y cells were chosen as a model system to investigate signaling pathways causative of cell death because these cells exhibit (1) DA neuron characteristics including dopamine synthesis, (2) expression of dopamine receptors, (3) specific uptake and sequestration of dopamine consistent with expression of the dopamine transporter [3,14,57] and (4) they can be differentiated into a neuronal phenotype by incubation with retinoic acid [32]. Furthermore, MPP+ induces apoptotic cell death in SHSY5Y cells and activates JNK and the early transcription factor nuclear factor nB [7,30,33,46,55]. The current studies investigated the ability of CEP-1347 to (1) inhibit the neurotoxic effects of MPP+ in retinoic acid (RA) differentiated SH-SY5Y cells; and (2) inhibit MPP+-induced JNK activation. Studies also determined whether SH-SY5Y cells over-expressing dominant negative forms of MLK family members would be protected from MPP+-induced JNK activation and cell death.

87

2. Materials and methods 2.1. Cell culture SH-SY5Y cells (J. Biedler, Memorial Sloan-Kettering Cancer Center, Rye, NY, USA) were seeded at a density of 4  104/cm2 in T150 flasks and propagated in Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco BRL, Gaithersburg, MD) containing 10% FBS and 2 mM L-glutamine (Gibco BRL). Growing SH-SY5Y cells were maintained in a humidified atmosphere at 37 jC in 10% CO2. Cells were routinely subcultured once a week and used for assays at passage 8– 37. For assays, SH-SY5Y cells were plated at a density of 1.2  105/cm2 onto poly-ornithine/mouse laminin (Sigma-Aldrich, St. Louis, MO/Becton Dickinson Biosciences, San Jose, CA) coated plates in Neurobasal medium (Gibco BRL) with B27 supplement (Gibco BRL) and 2 mM L-glutamine. Cells maintained in Neurobasal medium were kept in a humidified atmosphere at 37 jC and 5% CO2. The cells were allowed to attach for 1 h before the addition of 10 AM (final concentration) all trans-retinoic acid (RA, made in ethanol as a 10 mM stock solution; Sigma-Aldrich) that was used to induce neuronal differentiation. Chinese hamster ovary cells (CHO-K1; ATCC#CCL-61, American Type Culture Collection, Manassas, VA) were propagated as previously described [38]. 2.2. MPP+ treatment On 3 –5 days in vitro subsequent to plating/differentiation, SH-SY5Y cells were exchanged into fresh Neurobasal medium with B27 supplement and 2 mM L-glutamine without RA by serial dilution washing and treated for indicated times with MPP+ (made as an 11  stock in Neurobasal medium with supplements mentioned above; Sigma-Aldrich, previously from RBI). Typically, CEP-1347 (stored as 4 mM in DMSO in amber glass vials), was diluted to a 4  stock concentration in medium, then added to wells establishing final 1  concentrations and incubated with cells for 1 h prior to the addition of MPP+. 2.3. Lactate assay Lactate measurement in culture medium from differentiated SH-SY5Y cells was determined by a standard lactate assay (Sigma-Aldrich #826-10) following 4, 24 or 48 h of 3 mM MPP+ treatment. Data are represented as fold increase in lactate compared to untreated differentiated SH-SY5Y cells. 2.4. DNA Characterization For visualization of nuclear DNA, cell cultures were fixed (10 min) with 4% paraformaldehyde and subsequently incubated for 15 – 20 min with 1 Ag/ml bisBenzamide (Sigma-Aldrich) in PBS. Photomicrographs were taken

88

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

using a Nikon Eclipse TE300 microscope equipped with both Hoffman and epifluorescence optics. For DNA laddering studies, cellular DNA was isolated [29] and resolved on a 1.2% agarose gel containing 0.01% ethidium bromide. For RT-PCR analyses, RNA was isolated from differentiated SH-SY5Y cells and PC12 cells with RNAzol B according to the manufacturer’s directions (TelTest, Friendswood, TX). 2.5. Real-time RT-PCR Cells were lysed and RNA was isolated with a Qiagen RNeasy Mini Kit (#74104, Qiagen, Valencia, CA) according to the manufacturer’s instructions. Samples were run on a 1.2% agarose gel to check for RNA integrity. RNA was quantified spectrophotometrically (OD 260). cDNA was generated using oligo dT in Ambion’s Retroscript kit (Ambion, Austin, TX). 1.5 Ag of total RNA was used per 20 Al reaction. Gene specific primers were identified by using the Primer Express software (Applied Biosystems, Foster City, CA). Real-time PCR was performed in an ABI PrismR 7000 Sequence Detection System (Applied Biosystems) using SYBRR Green PCR Master Mix (Applied Biosystems). A 50-Al PCR reaction was run using manufacturer’s recommended cycling times. Real-time detection of gene expression is displayed as a threshold cycle or CT value. The CT value is the cycle at which the amplification of the gene enters into the exponential phase. Quantification of gene expression is possible with the assignment of a CT value. Hence, relative mRNA expression for each sample can be calculated by normalization to GAPDH CT values. The relative quantitation value is expressed as D (delta) CT. 2.6. Cell lysis and immunoblot analysis For measurement of activated, or phosphorylated JNK (pJNK), differentiated SH-SY5Y cells (1.5  106) were lysed in 125-Al ice cold FRAK buffer (1% Triton X-100, 50 mM NaCl, 30 AM sodium pyrophosphate, 50 mM sodium fluoride, 1 mM sodium vanadate, 10 mM Tris – HCl, pH 7.6) containing protease inhibitors (1 mM phenylmethylsulfonylfluoride (PMSF) and 20 Ag/ml aprotinin). Lysates were passed through a 28-1/2 gauge needle and the Triton insoluble fraction was then removed from the lysate by centrifugation at 4 jC (20 min, Effendorf microcentrifuge, 14,000 rpm). Supernatant was collected and an aliquot taken for protein determination (BCA assay, Pierce, Rockford, IL). Lysates were prepared for electrophoresis by adding Laemmeli sample buffer and heating for 10 min at 95 jC. Samples of equivalent total protein (20 Ag/lane) were run on 4– 20% Tris – Glycine gels (Invitrogen, Carlsbad, CA) and transferred (60 V/cm2) to 0.2 Am nitrocellulose (BioRad Laboratories, Hercules, CA). Membranes were blocked in 3% BSA/ Tris buffered saline (TBS) with 0.1% Triton X-100 followed by overnight incubation at 4 jC with a polyclonal pJNK antibody (New England Biolabs, Beverly, MA) diluted 1:1000 in block buffer. Following three 15-min washes in

0.5% Triton X-100/TBS membranes were incubated with a horseradish peroxidase conjugated secondary antibody (1:20,000, goat anti-rabbit; Southern Biotechnologies, Birmingham, AL), washed and incubated for 1 min with enhanced chemiluminescent (ECL) substrate (Amersham, Buckinghamshire, UK) for imaging with BioMax Film (Kodak, Rochester, NY). Total JNK levels across samples were determined following stripping and reprobing membranes with a monoclonal mouse antibody directed against non-phosphorylated JNK proteins (JNK1/2; 1:1000; BD PharMingen, Franklin Lakes, NJ). In experiments where adenoviral infection was included in the protocol, immunoblots were stripped a second and/or third time and reprobed with antibodies directed against MLK3 (Santa Cruz Biotechnology, Santa Cruz, CA; rabbit polyclonal sc-536; 1:1000) and/or h-galactosidase (h-gal; Invitrogen, Carlsbad, CA; rabbit polyclonal; 1:5000). Immunocytochemistry was performed on AdCMVdnMLK3-infected cells to detect dnMLK3 expression efficiency (anti-MLK3, Santa Cruz Biotechnology; 1:300). h-gal histochemistry was performed on AdCMVLacZ-infected cells to determine the expression efficiency of LacZ (Invitrogen h-gal staining kit #K1465-01). 2.7. Cell viability assessment Cell viability was determined based on measurement of lactate dehydrogenase (LDH) release into the culture medium. LDH assays were conducted in accordance with the manufacturer’s instructions (Cytotoxicity Detection Kit (LDH), Roche Diagnostics, Indianapolis, IN). Total LDH release was obtained by completely lysing cells in designated wells and data are represented as percent total LDH release. 2.8. Construction of recombinant replication-deficient adenovirus A HindIII to EcoRV fragment from vector pcDNA3EE containing dnMLK3 cDNA (kinase dead/dominant negative) [38] was inserted into respective sites of the vector pAdCMV [18], downstream from the cytomegalovirus (CMV) promoter and upstream of the cellular heavy chain enhancer (4F2) and the bovine growth hormone polyadenylation site. pAdCMV contains 0 –1 and 9 – 16 map units of the adenovirus 5 genome. pAdCMV containing mutant (dn) MLK3 was linearized with NheI and cotransfected, using the calcium phosphate precipitation method, with XbaI- and ClaIdigested adenovirus 5 (sub360) DNA into HEK293 cells, a trans-complementing cell line for E1 function. Viruses were purified by CsCl isopycnic centrifugation, dialyzed against HEPES-buffered saline (10 mM HEPES, 140 mM NaCl, 2 mM MgCl2, pH 7.5) containing 10% glycerol, and stored at 70 jC in small aliquots. AdCMVlacZ virus was a generous gift from Dr. Jerome Schaack (University of Colorado, Denver, CO [53]). AdCMVdnMLK2 and AdCMVdnDLK were prepared using the AdEasy Vector System (Quantum

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

Biotechnologies, also known as Qbiogene, Carlsbad, CA). Briefly, MLK2(K125A) or DLK(K162A) cDNA was cloned into the pShuttle-CMV transfer vector at the 5VAcc65I and 3VXbaI polylinker cloning sites and the positive clones identified were sequence-verified. These pShuttle-CMVMLK2(K125A) or DLK(K162A) constructs were linearized at the unique PmeI site, dephosphorylated and gel purified. These constructs were co-transformed with supercoiled adenovirus genome using highly competent BJ5183 E. coli cells via electroporation. The transformed colonies were grown on kanamycin plates, the smallest colonies picked from these plates and amplified in LB/kanamycin medium. Conventional alkaline lysis miniprep DNA protocols, utilizing phenol chloroform extractions and ethanol precipitations, were performed. Following restriction digest and PCR confirmation, this DNA was transformed into competent E. coli DH5a cells. Positive clones were digested with PacI to expose their ITRs and transfected into 293A cells using Qiagen SuperFect reagent (Qiagen, Valencia, CA). Viruses were purified and stored as described above.

89

infected using the same protocol that was performed on day 1. On day 4 cells were lysed and prepared for phosphoMKK4 ELISA [38] or immunoblot analysis. 2.10. Data computation Data was computed using Microsoft Excel and statistics were evaluated using either Student’s or Dunnett’s t-tests. Significance (*) was based on p < 0.05. The graphics package used was Prism GraphPad.

3. Results 3.1. CEP-1347 inhibition of MPP+-induced loss in neuronally differentiated SH-SY5Y cell survival, cellular morphology and apoptosis To establish cytotoxic effects of MPP+ in differentiated SH-SY5Y cells, increases in the release of cellular lactate dehydrogenase (LDH) were measured in the medium of

2.9. Adenoviral infection Differentiated SH-SY5Y cells were infected by removing the culture medium and incubating cells at 37 jC for 2.5 h with high-titer virus diluted in a small volume of Neurobasal medium containing B27 supplement and L-glutamine providing a multiplicity of infection (MOI) of 1000 based on optical particle units (OPU; [40]). This MOI was predetermined to infect f 90– 100% of the SH-SY5Y cells. The wells were gently rocked occasionally throughout the infection period. Following infection warmed Neurobasal medium containing B27 supplement and 2 mM L-glutamine was added to the wells bringing the media volume up to feeder levels (3 –4 ml/6-well dish; 200 Al/96-well). For proof of dominant negative (dn) status of the viral construct (AdCMVdnMLK3), CHO cells were treated in the following manner. CHO cells were infected twice due to their continual division. On day 0 cells were plated at 2  105 cells/well in six-well culture dishes. On day 1 cells were either uninfected, but subjected to the same conditions, or infected with either control adenoviral construct, AdCMVLacZ (2000 MOI), or AdCMVdnMLK3 (2000 MOI). On day 2 cells were transfected using Lipofectamine Plus (Gibco #10964-013), as previously described [38]. All cells were transfected with a total of 2 Ag of cDNA composed of the following: 4% MLK3, 20% dnMLK3, and 50% dnMKK4. The remaining percentage of cDNA was composed of vector. The ratio of MLK3 to dnMKK4 was predetermined to be in the linear range with respect to MLK3 expression and phosphorylation of the dnMKK4 substrate. A dominant negative MKK4 substrate was used to prevent downstream activation that would lead to cell death. The dominant negative nature of these clones has been previously described [38]. Four hours after the transfection, complete growth medium was added. On day 3, all cells were re-

Fig. 1. MPP+ concentration-dependent increases in release of LDH from RA differentiated SH-SY5Y cell cultures measured 48 h post exposure (A). CEP-1347-dependent decreases in LDH release are significantly different from 3 mM MPP+ at concentrations of 10 – 300 nM, indicating cell survival promotion (B). Data is representative of three independent experiments (*= significantly different from basal; **= significantly different from MPP+ treated; p < 0.05 Student’s t-test).

90

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

cultures treated with increasing concentrations of MPP+ up to a maximum of 10 mM. At 3 mM MPP+ significant increases in LDH were observed at 48 h post treatment (Fig. 1A). This was the lowest concentration of MPP+ that produced a significant 30 – 50% increase in LDH release (relative to total LDH) compared to untreated cultures. Although larger increases in LDH release were observed at a higher concentration of MPP+ (10 mM), the minimum concentration, 3 mM, which produced a significant increase in cytotoxicity was used to measure neuroprotective effects of CEP-1347. CEP-1347 provided significant protection from 3 mM MPP+-induced LDH release at concentrations of 10, 30, 100 and 300 nM (Fig. 1B). Across multiple experiments 30 – 100 nM CEP-1347 achieved maximal efficacy against MPP+induced cell death ranging from f 40 – 60% rescue. After 48 h of treatment with 3 mM MPP+ SH-SY5Y cellular morphology changes, going from adherent process bearing neuronal-looking cells (Fig. 2A) to rounded, clustered cells with decreased substrate adhesion and process extension (Fig. 2C). Moreover, two features indicative of apoptosis

were observed in SH-SY5Y cells following 48 h of treatment with 3 mM MPP+: (1) Condensed chromatin (Hoechst, Fig. 2B, D, and F) and (2) DNA laddering (Fig. 2D, insert). These observations confirm reports that SH-SY5Y cells undergo apoptosis in response to MPP+ treatment [30,33,46,55]. Of significance, the morphologic changes observed in SHSY5Y cells following exposure to 3 mM MPP+ were mostly prevented by pretreatment of cells with CEP-1347 (Fig. 2E). Moreover, CEP-1347 (30 nM) prevented nuclear chromatin condensation in a majority of cells when it was added 1 h prior to 3 mM MPP+ treatment (Fig. 2F), indicative of a partial inhibition of MPP+-induced apoptosis. 3.2. CEP-1347 acts downstream of MPP+ effects on mitochondrial demise in differentiated SH-SY5Y cells MPP+ is a mitochondrial toxin that inhibits respiration at complex I of the electron transport chain [43]. The resulting increase in NADH due to this inhibition leads to pyruvate being metabolized into lactate. To determine whether CEP-

Fig. 2. MPP+-induced changes in cellular morphology and apoptosis in differentiated SH-SY5Y cells. Untreated control cultures showing normal morphology with Hoffman optics (A) and normal nuclear bis-benzimide (Hoechst) staining (B). Cultures treated with 3 mM MPP+ show degenerating cells at 48 h (C) with condensed nuclear chromatin. (D) Arrows represent apoptotic cells. Treatment with 30 nM CEP-1347 1 h prior to 3 mM MPP+ exposure prevents morphologic (E) and chromatin changes (F). Insert panel D: DNA laddering following 3 mM MPP+ exposure: lanes (1) MW markers; (2) 0 h; (3) 24 h; (4) 48 h. Scale bar = 10 Am.

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

91

1347 altered the toxic effects of MPP+ by interfering with the initial events of mitochondrial complex I inhibition, lactate levels were measured in SH-SY5Y cell culture medium following MPP+ treatment in the presence or absence of CEP-1347. As early as 4 h after 3 mM MPP+ addition, lactate production increased f 2-fold over basal and this lactate induction was maintained to 48 h after 3 mM MPP+ addition (Fig. 3). This increase in lactate production was not prevented by addition of CEP-1347 to the cultures (Fig. 3). These data indicate that CEP-1347 does not interfere with MPP+ toxicity by blocking MPP+ inhibition of complex I and verifies that CEP-1347 is acting downstream of this initial insult. Therefore, the neuroprotective effects of CEP-1347 were not due to decreases in MPP+ interactions with the mitochondria. Fig. 3. Lack of CEP-1347 effects on cellular lactate production induced by 3 mM MPP+ in differentiated SH-SY5Y cells. Lactate levels in culture medium were measured at 4, 24 and 48 h post exposure to 3 mM MPP+ in quadruplicate. Groups as follows: (MPP) 3 mM MPP+; (MPP+/1347) 30 nM CEP-1347 (neuroprotective concentration) was tested with 3 mM MPP+. Graph represents one of two experiments with the same result.

Fig. 4. JNK activation following 3 mM MPP+ and inhibition by CEP1347. Immunoblot analysis of differentiated SH-SY5Y cell lysate for phosphorylated (activated) JNK and total JNK1/2 protein following 30 min of 3 mM MPP+ and as indicated below a 30-min pretreatment with a concentration response of CEP-1347 (A). The p46 pJNK band is quantitated in (B) and represented as a ratio of p46 optical density relative to total JNK1/2 protein density as a gel loading control. Numbered duplicate wells in A are as follows: (1) and (2) basal untreated cultures; (3) and (4) 3 mM MPP+; (5) and (6) MPP+ with 0.01 AM CEP1347; (7) and (8) MPP+ with 0.1 AM CEP-1347; (9) and (10) MPP+ with 0.3 AM CEP-1347; (11) and (12) MPP+ with 1 AM CEP-1347; (13) and (14) MPP+ with 10 AM CEP-1347. * indicates significantly different from basal ( p < 0.05); ** indicates significantly different from MPP+ ( p < 0.05).

3.3. CEP-1347 inhibits SAPK/JNK pathway activation following MPP+ treatment in neuronally differentiated SHSY5Y cells Evidence for activation of the SAPK/JNK pathway was evaluated by immunoblot analysis for pJNK. In SH-SY5Y cells the total JNK and pJNK antibodies detected two protein bands of 46 and 54 kDa (Fig. 4). These molecular weight size bands correspond to the reported sizes of cloned and expressed JNK [22]. Initial time course experiments showed JNK activation between 30 min and 4 h following MPP+ exposure to SH-SY5Y cells, with a maximal activation of 2.4-fold above untreated cultures at the 30-min time point. Typically in MPP+-treated SH-SY5Y cells the p46 kDA band

Fig. 5. Inhibition of plasmid overexpressed wild-type MLK3-stimulated pdominant negative (dn) MKK4 by plasmid dominant negative (dn) MLK3 and adenoviral dnMLK3 in CHO cells. Data represent duplicate samples from one experiment. *= significantly different from MLK3 alone. **= significantly different from MLK3-stimulated p-dnMKK4 ( p < 0.05, Student’s t-test).

92

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

showed a more robust and reliable activation than did the p54 kDA band. CEP-1347 inhibited the pJNK signal seen 30 min following treatment of cells with 3 mM MPP+ in a concentration-dependent manner from 0.1 to 10 AM with complete inhibition achieved at 0.1 AM and above (Fig. 4). Notably, the MPP+-induced pJNK signal fell below basal levels following treatment with CEP-1347. 3.4. Dominant negative MLK3 inhibits MPP+-induced toxicity and pJNK in neuronally differentiated SH-SY5Y cells To verify the dominant negative action of overexpressed AdCMVdnMLK3, a CHO cell system was infected and transfected with the following constructs. CHO cells infected

with AdCMVdnMLK3 were inhibited from wild-type plasmid transfected MLK3 phosphorylation of transfected dnMKK4, MKK4 being a downstream target of MLK3 (Fig. 5). A dominant negative MKK4 substrate was used to measure overexpressed MLK3 kinase activity so as to prevent death of the cells through downstream activation mechanisms. Due to the continual division of CHO cells, they were subjected to infection on two separate days bracketing the plasmid transfection day. The MOI was increased to 2000 per day to provide 90 –100% transfection efficiency demonstrated in parallel sister cultures. Control viral infection with AdCMVLacZ had no effect on the ability of MLK3 to phosphorylate dnMKK4. Following confirmation of the dominant negative status of the AdCMVdnMLK3 in CHO cells, overexpression of dnMLK3 in neuronally

Fig. 6. AdCMVdnMLK3 inhibition of MPP+-induced cell death in differentiated SH-SY5Y cells. MPP+ (3 mM) increased release of LDH from SH-SY5Y cells into culture medium measured 48 h post exposure. AdCMVdnMLK3 (1000 MOI, 48 h expression)-dependent decreases in LDH release are significantly different from 3 mM MPP+ (A). *= significantly different from basal. **= significantly different from 3 mM MPP+-treated. Significance based on p < 0.05, Student’s t-test. (B) Upper immunoblot shows overexpression of MLK3. Lanes: (1) uninfected SH-SY5Y cells, (2) AdCMVLacZ-infected SH-SY5Y cells (1000 MOI, 48 h), (3) AdCMVdnMLK3-infected SH-SY5Y cells (1000 MOI, 48 h). Lower immunoblot, same blot reprobed for phospho-JNK. (C) Separate immunoblot with same samples in A probed for hgal. Lanes: (1) uninfected SH-SY5Y cells, (2) AdCMVLacZ-infected (1000 MOI, 48 h). (D) Uninfected differentiated SH-SY5Y cells immunostained for MLK3 showing very low levels of endogenous MLK3 detected. (E) AdCMVdnMLK3 (1000 MOI, 48 h) infected differentiated SH-SY5Y cells immunostained with MLK3 antibody showing 90 – 100% infection efficiency and expression. Scale bar = 10 Am.

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

differentiated SH-SY5Y cells (Fig. 6, compare A lane 1— endogenous MLK3 with lane 3—overexpressed MLK3 and C endogenous immunocytochemically detected MLK3 vs. D overexpressed MLK3) partially suppressed MPP+-induced cell death, comparable to the protective effect of CEP-1347 on MPP+ toxicity (Fig. 1). MPP+-stimulated pJNK levels were also partially prevented by expression of AdCMVdnMLK3 in neuronally differentiated SH-SY5Y cells (Fig. 7). This dnMLK3 overexpression level did not decrease basal levels of pJNK (Fig. 6, compare A: lanes 1 and 2 vs. 3). The control construct, AdCMVLacZ, used at an equivalent MOI (1000) produced similar overexpression when blots were probed with the h-gal antibody instead of anti-MLK3 (Fig. 6B) and in histochemical evaluation of the cells (not shown). 3.5. Real-time RT-PCR expression analysis of SH-SY5Y cells The existence of endogenous MLK3 expression in SHSY5Y cells was implicit from immunoblot analysis of cell lysate with anti-MLK3 (Fig. 6). There are, however, no commercially available antibodies for human MLK1, 2 or ZPK(DLK). Therefore, real-time RT-PCR was performed on SH-SY5Y cell lysate with gene specific primers identified by the Primer Express software provided with the ABI PrismR 7000 Sequence Detection System. Gene specific

Fig. 7. AdCMVdnMLK3 inhibition of MPP+-induced pJNK response in differentiated SH-SY5Y cells. Immunoblot analysis of SH-SY5Y cell lysate for phosphorylated JNK following 30 min of 3 mM MPP+ (A, B). Numbered wells are as follows: (A1) basal untreated cultures infected with AdCMVLacZ (1000 MOI, 24 h); (A2) 3 mM MPP+-treated culture infected with AdCMVLacZ; (B1) basal untreated cultures infected with AdCMVdnMLK3 (1000 MOI, 24 h); (B2) 3 mM MPP+-treated culture infected with AdCMVdnMLK3. The p46 band is quantitated from duplicate distinct samples normalized to total JNK1/2 protein density and graphically represented as fold over control p46 activated JNK. * indicates significantly different from AdCMVLacZ ( p < 0.05).

93

Fig. 8. AdCMVdnMLK2 infection of differentiated SH-SY5Y cells. AdCMVdnMLK2 infection did not inhibit 3 mM MPP+-induced cell death in SH-SY5Y cells. MPP+ (3 mM) increased release of LDH from SH-SY5Y cell culture medium measured 48 h post exposure in cells infected with either AdCMVLacZ (1000 MOI, 48 h) or AdCMVdnMLK2 (250 MOI, 48 h). Immunoblot shows overexpression of MLK2 with an HA antibody probe. Lanes: (1 – 5) Adenoviral infection of SH-SY5Y cells for 24 h; (6 – 9) adenoviral infection for 48 h. Lane (1) AdCMVLacZ 1000 MOI; (2) and (6) AdCMVdnMLK2 250 MOI; (3) and (7) AdCMVdnMLK2 1000 MOI; (4) and (8) AdCMVdnMLK2 2000 MOI; (5) and (9) AdCMVdnMLK2 4000 MOI. *= significantly different from basal. Significance based on p < 0.05, Student’s t-test.

products for MLK1, 2, 3 and ZPK(DLK) were detected in differentiated SH-SY5Y cells. Two to three separate cell preparations with three independent measures were generated and normalized to GAPDH expression to determine the relative expression of MLK RNA in differentiated SHSY5Y cells. MLK1 and MLK3 were equivalently expressed (5.91 F 0.32 and 5.86 F 0.21 DCT, respectively) and MLK2

Fig. 9. AdCMVdnDLK infection of differentiated SH-SY5Y cells. AdCMVdnDLK infection did not inhibit 3 mM MPP+-induced cell death in SH-SY5Y cells. MPP+ (3 mM) increased release of LDH from SH-SY5Y cell culture medium measured 48 h post exposure in cells infected with either AdCMVLacZ (250 MOI, 48 h) or AdCMVdnDLK (500 MOI, 48 h). Immunoblot shows overexpression of DLK with an HA antibody probe. Lanes: (1 – 5) Adenoviral infection of SH-SY5Y cells for 24 h; (6 – 9) adenoviral infection for 48 h. Lane (1) AdCMVLacZ 1000 MOI; (2) and (6) AdCMVdnDLK 250 MOI; (3) and (7) AdCMVdnDLK 500 MOI; (4) and (8) AdCMVdnDLK 1000 MOI; (5) and (9) AdCMVdnDLK 2000 MOI. *= significantly different from basal. Significance based on p < 0.05, Student’s t-test.

94

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

and ZPK(DLK) were equivalently expressed 1 DCT level higher than MLK1 and MLK3 (6.83 F 1.17 and 6.89 F0.31, respectively). This represents a twofold lower expression of MLK2 and ZPK(DLK) in differentiated SH-SY5Y cells than MLK1 and MLK3, as each critical threshold value represents a doubling of RNA present upon replication in the PCR reaction. Statistically, differences in MLK2 vs. MLK3 expression did not reach significance ( p = 0.07), whereas significant differences were established for MLK1 vs. MLK2; MLK1 vs. ZPK(DLK) and MLK3 vs. ZPK(DLK). 3.6. Dominant negative MLK2 and dnDLK do not inhibit MPP+-induced toxicity in differentiated SH-SY5Y cells AdCMVdnMLK2 and AdCMVdnDLK represent dominant negative adenoviral constructs of MLK family members that are related to MLK3. The kinase domains of MLK3 and MLK2 have 75% sequence homology and MLK3 and DLK share 42% sequence homology. The leucine zipper regions share similar homology with MLK3 and MLK2 having 68% homology while MLK3 and DLK share only 32% homology. It is theoretically possible that dnMLK2 or dnDLK would cross react by heterodimerizing with endogenous MLK3 in SH-SY5Y cells and provide some protection against MPP+. This was not the case over a wide range of overexpressed protein (125 –1000 MOI). In the higher MOI conditions (500 – 1000 MOI), dnMLK2 and dnDLK were toxic, especially when combined with MPP+ (data not shown). At MOI levels that were not toxic to basal untreated cells, AdCMVdnMLK2 (Fig. 8) and AdCMVdnDLK (Fig. 9) did not protect differentiated SHSY5Y cells against MPP+-induced death. An adenoviral construct containing dnMLK1 was not tested in these studies.

4. Discussion CEP-1347 has been identified as a direct inhibitor of the mixed lineage kinase (MLK) family and has previously been found to attenuate MPTP-mediated DA neuronal death and activation of the JNK signaling pathway in vivo [38,51,52]. The present studies expand on those findings by demonstrating that CEP-1347 and a dominant-negative MLK3 adenovirus construct inhibit MPP+-induced death and JNK signaling in SH-SY5Y cells. These studies further implicate the MLK/JNK signaling pathway in MPP+-induced neuronal death in vitro and suggest that this pathway may be active in degenerating DA neurons in PD. Further, these studies demonstrate the potential value of CEP-1347 as a neuroprotective compound in PD. MPP+-treated SH-SY5Y cells are a useful in vitro model for studying neurodegenerative events that may occur in PD [7,30,33,46,55]. In the present study, CEP-1347 partially prevented MPP+-induced cell death in retinoic acid differ-

entiated SH-SY5Y cells at low nanomolar concentrations. CEP-1347 demonstrates neuroprotective properties in a variety of primary neurons in culture including dorsal root ganglion, sympathetic and motor neurons, and PC12 cells after trophic factor withdrawal, DNA damage, or oxidative stress [6,36,37]. CEP-1347 also maintains survival of motor neurons in several in vivo models of programmed cell death, such as developmental cell death of postnatal rat motor neurons or of chick lumbar motor neurons in ovo and adult rat hypoglossal neurons subjected to axotomy [19]. Morphologically, differentiated SH-SY5Y cells treated with 3 mM MPP+ lose their processes, cluster, and eventually lift off the plate. This morphological change was inhibited by pretreatment with CEP-1347. Additionally, apoptotic indicators were demonstrated to be present following MPP+ addition to RA differentiated SH-SY5Y cultures. CEP-1347 visually decreased the number of cells displaying condensed chromatin, indicating that CEP-1347 may be preventing programmed cell death initiated by MPP+ treatment. Inhibition of apoptotic cell death may be important because markers of apoptosis have been observed in many neurodegenerative diseases including PD [28]. However, others have shown that evidence of apoptosis was missing in samples of patients with PD [15]. While a snapshot of postmortem tissue is difficult to reconcile with an ongoing neurodegenerative disease process, coupled with the unknown clearance of apoptotic cells, it is difficult to ascribe PD to a strictly apoptotic process. In other studies CEP-1347 suppressed CHO cell-transfected MLK3-driven apoptotic death at concentrations that inhibited MLK3 kinase activity [38]. Transient transfection of naive and neuronally differentiated PC12 cells with MLK family members showed apoptotic responses including membrane blebbing, pyknotic nuclei and positive Hoechst staining in addition to cell death [68], thereby relating MLK family activation with potential apoptotic cell death. To begin to address the mechanism of CEP-1347 inhibition of the MPP+ insult it was important to evaluate whether the actual mitochondrial insult still occurred in the presence of CEP-1347. When MPP+ interferes with mitochondrial function at the level of complex I, lactate levels increase as a result of an increase in glycolysis and increases in NADH [45]. To determine whether CEP-1347 acted downstream of mitochondrial demise, lactate levels were measured following MPP+ treatment in the presence or absence of CEP1347. Lactate levels were increased twofold over basal levels at 4, 24 and 48 h post MPP+. This increase was unaffected by pretreatment with a neuroprotective concentration of CEP-1347. These data indicated that CEP-1347 effects on cellular death pathways were downstream of this initial toxic event. CEP-1347 is a known inhibitor of the SAPK/JNK signaling pathway and in vitro studies have demonstrated inhibition at the level of the MLK family [38]. Increased activity of JNK has been demonstrated following MPP+ treatment in undifferentiated SH-SY5Y cells [7,67]. In these

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

studies, undifferentiated cells were used in addition to slightly higher concentrations of MPP+ (5 mM). In addition, adenoviral gene transfer of the JNK binding domain of JNK-interacting protein-1 (a scaffold protein and inhibitor of JNK) inhibited MPP+-induced activation of JNK, c-Jun and caspase and inhibited cell death in undifferentiated SHSY5Y cells [67]. The present study confirms these findings by demonstrating that (1) MPP+ elevated pJNK in differentiated SH-SY5Y cells, (2) that this could be partially inhibited by adenoviral gene transfer of dominant negative MLK3 and (3) completely blocked by an inhibitor of the MLK family, CEP-1347/KT7515. It has been demonstrated that all three JNK isoforms are represented in SH-SY5Y cells [39]. What has not been determined is whether the expression of these isoforms changes upon differentiation or after MPP+ treatment. It is interesting that JNK activation was completely inhibited by pretreatment with CEP-1347, often with inhibition below basal conditions. CEP-1347 inhibition of JNK activation in differentiated SH-SY5Y cells following MPP+ treatment was concentration dependent with significant inhibition beginning at 10 nM and maximal inhibition at 300 –1000 nM. CEP-1347 does not directly inhibit JNK in CHO cells as measured by a c-Jun luciferase reporter construct [38] driven by c-Jun phosphorylation, which is one of the immediate downstream targets of JNK. This report [38] showed that only MLK3-induced JNK activation could be inhibited by CEP-1347 and not activation produced by MEKK1. More recently, there has been described a lack of JNK activation in SH-SY5Y cells following MPP+ [20]. These studies, however, were accomplished again in undifferentiated SH-SY5Y cells with much lower concentrations of MPP+ (5 AM), which the authors claim was insufficient to decrease mitochondrial function or activate oxidative stress. The MLK family has alternate downstream targets other than JNK such as the transcription factor NF-nB and p38 [10,24,42,61]. Others have shown dominant negative versions of MLK3 prevent JNK activation induced by Rac and Cdc42 but not JNK activation induced by MEKK1 [60]. Taken together, it seems that acute mitochondrial damage initiates multiple intracellular signaling pathways leading to cell death. Complete inhibition by CEP-1347 of the pJNK induced by MPP+ mitochondrial damage, while only preventing up to 70% of the cell death induced by MPP+, suggests that another cell death pathway (other than through JNK activation) is induced by MPP+ in these cells. CEP1347 can completely inhibit basal and MPP+-stimulated JNK activation while at the same time only partially prevents MPP+-induced cell death and condensed chromatin. This study demonstrates that the JNK pathway disruption is critical but may not be sufficient to completely protect from MPP+ toxicity. Others have shown that when JNK is activated following UV light the resulting apoptosis is JNK dependent [63]. They showed that the absence of JNK1/2, through JNK1/2 null murine embryonic fibroblasts prevented UV-induced apoptosis and DNA fragmentation

95

indicating that JNK is required for the normal apoptotic response of fibroblasts to UV light. JNK null fibroblasts were also resistant to the apoptotic effects of a genotoxin (methyl methanesulfonate) and anisomycin, while the Fasdeath signaling pathway remained intact. Utilization of an adenoviral dominant negative MLK3, which infects >90% of differentiated SH-SY5Y cells and allows considerable overexpression of this protein has shed some light on the role of MLK3 in MPP+-induced cell death. The dimerization of adenoviral overexpressed dominant negative MLK3 with endogenous MLK3 [35] is expected to prevent downstream phosphorylation of targets such as MKK4 and/or MKK7. Leung and Lassam [35] have shown that MLK3 homodimerization is crucial for downstream activation of the SAPK/JNK pathway. It is clear that MLK3 partially contributes to the toxic effects of MPP+ through activation of JNK as demonstrated by our dominant negative MLK3 data. Similar to CEP-1347, overexpression of dnMLK3 prevented some, but not all, of the MPP+induced cell death. In contrast to CEP-1347, dnMLK3 did not completely inhibit elevated levels of pJNK. CEP-1347 is an inhibitor of the MLK family and would therefore be expected to have effects beyond a single member such as MLK3. Through a real-time RT PCR analysis of differentiated SH-SY5Y cell lysate, expression of MLK1, MLK2, MLK3 and ZPK(DLK) was measured. Expression of MLK1 and MLK3 had the higher relative expression seen among these MLKs. It is theoretically possible for dnMLK2 or dnDLK to heterodimerize with endogenous MLK3 to prevent downstream activation of MKK4 or MKK7. Previous studies using the same kinase dead (dominant negative) mutations of MLK2 or DLK in plasmid transfection studies with MLK3-activated JNK in PC12 cells [68] showed cross reactivity. Others have shown cross reactivity with dominant negative DLK prevention of MLK3-induced JNK activity in 293T cells [58]. Even though these dominant negative adenoviral constructs overexpressed a considerable amount of the dominant negative proteins in differentiated SHSY5Y cells, they were not protective following MPP+ treatment; in fact, they became toxic themselves at high MOI concentrations. This toxicity was not seen with the control construct, AdCMVLacZ or the positive test construct, AdCMVdnMLK3, demonstrating that it was not toxicity of a viral origin. CEP-1347 may be acting upon other as yet undiscovered MLK family members or alternate kinases. This study demonstrates that the indolocarbazole CEP1347, likely through its inhibition of the MLK family, completely inhibited the SAPK/JNK pathway and partially prevented the neurotoxic effects of MPP+ in differentiated SH-SY5Y cells. While CEP-1347 shows tendencies towards an inverted U-shaped concentration response curve when measuring cell survival, this is not demonstrated when measuring JNK inhibition. At very high concentrations of CEP-1347, other pharmacological rather than physiological events could be occurring that are not addressed in this work.

96

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97

Adenoviral overexpression of dominant negative MLK3 also partially protected differentiated SH-SY5Y cells from cell death and pJNK activation indicating that MLK3 has a role in MPP+-induced cell death. These results taken with previous studies indicate that CEP-1347 may be protective in neurodegenerative diseases such as Parkinson’s Disease that may involve the SAPK/JNK pathway leading to apoptosis.

References [1] Y. Agid, Parkinson’s disease pathophysiology, Lancet 337 (1991) 1 – 3. [3] J.L. Biedler, S. Roffler-Tarlov, M. Schachner, L.S. Freedman, Multiple neurotransmitter synthesis by human neuroblastoma cell lines and clones, Cancer Res. 38 (1978) 3751 – 3757. [4] P.J. Blanchet, S. Konitsiotis, K. Hyland, L.A. Arnold, K.D. Pettigrew, T.N. Chase, Chronic exposure to MPTP as a primate model of progressive parkinsonism: a pilot study with a free radical scavenger, Exp. Neurol. 153 (1998) 214 – 222. [5] L.J. Bloem, T.R. Pickard, S. Aton, M. Donoghue, R.C. Beavis, M.D. Knierman, X. Wang, Tissue distribution and functional expression of a cDNA encoding a novel mixed lineage kinase, J. Mol. Cell Cardiol. 33 (2001) 1739 – 1750. [6] G.D. Borasio, S. Horstmann, J.M. Anneser, N.T. Neff, M.A. Glicksman, CEP-1347/KT7515, a JNK pathway inhibitor, supports the in vitro survival of chick embryonic neurons, NeuroReport 9 (1998) 1435 – 1439. [7] D.S. Cassarino, E.M. Halvorsen, R.H. Swerdlow, N.N. Abramova, W.D. Parker Jr., T.W. Sturgill, J.P. Bennett Jr., Interaction among mitochondria, mitogen-activated protein kinases and nuclear factorkB in cellular models of Parkinson’s disease, J. Neurochem. 74 (2000) 1384 – 1392. [8] S.J. Crocker, W.R. Lamba, P.D. Smith, S.M. Callaghan, R.S. Slack, H. Anisman, D.S. Park, c-Jun mediates axotomy-induced dopamine neuron death in vivo, PNAS 98 (2001) 13385 – 13390. [9] A. Cuenda, D.S. Dorow, Differential activation of stress-activated protein kinase kinases SKK4/MKK7 and SKK1/MKK4 by the mixed-lineage kinase-2 and mitogen-activated protein kinase kinase (MKK) kinase-1, Biochem. J. 333 (1998) 11 – 15. [10] R.J. Davis, Signal transduction by the JNK group of MAP kinases, Cell 103 (2000) 239 – 252. [11] B. Dipasquale, A.M. Marini, R.J. Youle, Apoptosis and DNA degradation induced by 1-methyl-4-phenylpyridinium in neurons, BBRC 181 (1991) 1442 – 1448. [12] C.P. Fall, J.P. Bennett Jr., Characterization and time-course of MPP+induced apoptosis in human SH-SY5Y neuroblastoma cells, J. Neurosci. Res. 55 (1999) 620 – 628. [13] G. Fan, S.E. Merritt, M. Kortenjann, P.E. Shaw, L.B. Holzman, Dual leucine zipper-bearing kinase (DLK) activates p46SAPK and p38mapk but not ERK2, J. Biol. Chem. 271 (1996) 24788 – 24793. [14] S.M. Farooqui, Induction of adenylate cyclase sensitive dopamine D2-receptors in retinoic acid induced differentiated human neuroblastoma SHSY-5Y cells, Life Sci. 55 (1994) 1887 – 1893. [15] I. Ferrer, R. Blanco, M. Carmona, B. Puig, M. Barrachina, C. Gomez, S. Ambrosio, Active, phosphorylation-dependent mitogen-activated protein kinase (MAPK/ERK), stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK), and p38 kinase expression in Parkinson’s disease and Dementia with Lewy bodies, J. Neural. Transm. 108 (2001) 1383 – 1396. [17] M. Gerlach, W. Gsell, J. Kornhuber, K. Jellinger, V. Krieger, F. Pantucek, R. Vock, P. Riederer, A post mortem study on neurochemical markers of dopaminergic, GABA-ergic and glutamatergic neurons in basal ganglia-thalamocortical circuits in Parkinson syndrome, Brain Res. 741 (1996) 142 – 152.

[18] G.D. Ghadge, R.P. Roos, U.J. Kang, R. Wollmann, P.S. Fishman, A.M. Kalynych, E. Barr, J.M. Leiden, CNS gene delivery by retrograde transport of recombinant replication-defective adenoviruses, Gene Ther. 2 (1995) 132 – 137. [19] M.A. Glicksman, A.Y. Chiu, C.A. Dionne, M. Harty, M. Kaneko, C. Murakata, R.W. Oppenheim, D. Prevette, D.R. Sengelaub, J.L. Vaught, N.T. Neff, CEP-1347/KT7515 prevents motor neuronal programmed cell death and injury-induced dedifferentiation in vivo, J. Neurobiol. 35 (1998) 361 – 370. [20] C. Gomez-Santos, I. Ferrer, J. Reiriz, F. Vinals, M. Barrachina, S. Ambrosio, MPP+ increases alpha-synuclein expression and ERK/ MAP-kinase phosphorylation in human neuroblastoma SH-SY5Y cells, Brain Res. 935 (2002) 32 – 39. [21] I. Gotoh, M. Adachi, E. Nishida, Identification and characterization of a novel MAP kinase kinase kinase, MLTK, J. Biol. Chem. 276 (2001) 4276 – 4286. [22] S. Gupta, T. Barrett, A.J. Whitmarsh, J. Cavanagh, H.K. Sluss, B. Derijard, R.J. Davis, Selective interaction of JNK protein kinase isoforms with transcription factors, EMBO J. 15 (1996) 2760 – 2770. [23] A. Hartley, J.M. Stone, C. Heron, J.M. Cooper, A.H. Schapira, Complex I inhibitors induce dose-dependent apoptosis in PC12 cells; relevance to Parkinson’s disease, J. Neurochem. 63 (1994) 1987 – 1990. [24] S.P. Hehner, T.G. Hofmann, A. Ushmorov, O. Dienz, I.W.-L. Leung, N. Lassam, C. Scheidereit, W. Droge, M.L. Schmitz, Mixed-lineage kinase 3 delivers CD3/CD28-derived signals into the IkB kinase complex, Mol. Cell. Biol. 20 (2000) 2556 – 2568. [25] R.E. Heikkila, A. Hess, R.C. Duvoisin, Dopaminergic neurotoxicity of 1-methyl-4 phenyl-1,2,3,6-tetrahydropyridine in mice, Science 224 (1984) 1451 – 1453. [26] R.E. Heikkila, L. Manzino, F.S. Cabbat, R.C. Duvoisin, Protection against the dopaminergic neurotoxicity of 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine by monoamine oxidase inhibitors, Nature 311 (1984) 467 – 469. [27] S. Hirai, M. Izawa, S. Osada, G. Spyrou, S. Ohno, Activation of the JNK pathway by distantly related protein kinases, MEKK and MUK, Oncogene 12 (1996) 641 – 650. [28] E.C. Hirsch, S. Hunot, B. Faucheux, Y. Agid, Y. Mizuno, H. Mochizuki, W.G. Tatton, N. Tatton, W.C. Olanow, Dopaminergic neurons degenerate by apoptosis in Parkinson’s disease, Mov. Disord. 14 (1999) 383 – 385. [29] D. Hockenbery, G. Nunez, C. Milliman, R.D. Schreiber, S.J. Korsmeyer, Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death, Nature 348 (1990) 334 – 336. [30] Y. Itano, Y. Nomura, 1-Methyl-4-phenyl-pyridinium ion (MPP+) causes DNA fragmentation and increases the Bcl-2 expression in human neuroblastoma, SH-SY5Y cells, through different mechanisms, Brain Res. 704 (1995) 240 – 245. [32] D.R. Kaplan, K. Matsumoto, E. Lucarelli, C.J. Thiele, Induction of TrkB by retinoic acid mediates biologic responsiveness to BDNF and differentiation of human neuroblastoma cells, Neuron 11 (1993) 321 – 331. [33] Y. Kitamura, T. Kosaka, J.-I. Kakimura, Y. Matsuoka, Y. Kohno, Y. Nomura, T. Taniguchi, Protective effects of the antiparkinsonian drugs talipexole and pramipexole against 1-methyl-4-phenylpyridinium-induced apoptotic death in human neuroblastoma SH-SY5Y cells, Mol. Pharmacol. 54 (1998) 1046 – 1054. [34] J.W. Langston, The etiology of Parkinson’s disease with emphasis on the MPTP story, Neurology 47 (1996) S153 – S160. [35] I.W. Leung, N. Lassam, Dimerization via tandem leucine zippers is essential for the activation of the nitrogen-activated protein kinase kinase kinase, MLK3, J. Biol. Chem. 273 (49) (1998) 32408 – 32415. [36] A.C. Maroney, M.A. Glicksman, A.N. Basma, K.M. Walton, E. Knight Jr., C.A. Murphy, B.A. Bartlett, T. Angeles, Y. Matsuda, N.T. Neff, C.A. Dionne, Motoneuron apoptosis is blocked by CEP1347 (KT 7515), a novel inhibitor of the JNK signaling pathway, J. Neurosci. 18 (1998) 104 – 111.

J.R. Mathiasen et al. / Brain Research 1003 (2004) 86–97 [37] A.C. Maroney, J.P. Finn, D. Bozyczko-Coyne, T.M. O’Kane, N.T. Neff, A.M. Tolkovsky, D.S. Park, C.Y. Yan, C.M. Troy, L.A. Greene, CEP-1347 (KT7515), an inhibitor of JNK activation, rescues sympathetic neurons and neuronally differentiated PC12 cells from death evoked by three distinct insults, J. Neurochem. 73 (1999) 1901 – 1912. [38] A.C. Maroney, J.P. Finn, T.J. Connors, J.T. Durkin, T. Angeles, G. Gessner, Z. Xu, S.L. Meyer, M.J. Savage, L.A. Greene, R.W. Scott, J.L. Vaught, CEP-1347 (KT7515), a semisynthetic inhibitor of the mixed lineage kinase family, J. Biol. Chem. 276 (2001) 25302 – 25308. [39] K. Mielke, A. Damm, D.D. Yang, T. Herdegen, Selective expression of JNK isoforms and stress-specific JNK activity in different neural cell lines, Mol. Brain Res. 75 (2000) 128 – 137. [40] N. Mittereder, K.L. March, B.C. Trapnell, Evaluation of the concentration and bioactivity of adenovirus vectors for gene therapy, J. Virol. 70 (1996) 7498 – 7509. [41] H. Mochizuki, N. Nakamura, K. Nishi, Y. Mizuno, Apoptosis is induced by 1-methyl-4-phenylpyridinium ion (MPP+) in ventral mesencephalic-striatal co-culture in rat, Neurosci. Lett. 170 (1994) 191 – 194. [42] M. Mota, M. Reeder, J. Chernoff, C.E. Bazenet, Evidence for a role of mixed lineage kinases in neuronal apoptosis, J. Neurosci. 21 (2001) 4949 – 4957. [43] W.J. Nicklas, I. Vyas, R.E. Heikkla, Inhibition of NADH-linked oxidation in brain mitochondria by 1-methyl-4-phenylpyridine, a metabolite of the neurotoxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine, Life Sci. 36 (1985) 2503 – 2508. [44] D. Offen, P.M. Beart, N.S. Cheung, C.J. Pascoe, A. Hocham, S. Gorodin, E. Melamed, R. Bernard, O. Bernard, Transgenic mice expressing human Bcl-2 in their neurons are resistant to 6-hydroxydopamine and 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine neurotoxicity, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 5789 – 5794. [45] S. Ofori, R.E. Heikkila, W.J. Nicklas, Attenuation by dopamine uptake blockers of the inhibitory effects of 1-methyl-4-phenyl-1,2,3,6tetrahydropyridine and some of its analogs on NADH-linked metabolism in mouse neostriatal slices, J. Pharmacol. Exper. Ther. 251 (1989) 258 – 266. [46] C.-W. Park, H.-S. Lee, Y.-S. Kim, Mechanism of MPP+-induced cytotoxicity in human neuroblastoma SH-SY5Y, J. Toxicol. Sci. 23 (Suppl. II) (1998) 184 – 188. [49] H. Sakuma, A. Ikeda, S. Oka, Y. Kozutsumi, J.P. Zanetta, T. Kawasaki, Molecular cloning and functional expression of a cDNA encoding a new member of mixed lineage protein kinase from human brain, J. Biol. Chem. 272 (1997) 28622 – 28629. [50] M.S. Saporito, R.E. Heikkila, S.K. Youngster, W.J. Nicklas, H.M. Geller, Dopaminergic neurotoxicity of 1-methyl-4-phenylpyridinium analogs in cultured mesencephalon: relationship to dopamine uptake affinity and inhibition of mitochondrial respiration, J. Pharmacol. Exp. Ther. 260 (1992) 1400 – 1409. [51] M.S. Saporito, E.M. Brown, M.S. Miller, S. Carswell, CEP-1347/KT7515, An inhibitor of c-jun N-terminal kinase activation, attenuates the 1-methyl-4-phenyl tetrahydropyridine-mediated loss of nigrostriatal dopaminergic neurons in vivo, J. Pharmacol. Exp. Ther. 288 (1999) 421 – 427. [52] M.S. Saporito, B.A. Thomas, R.W. Scott, MPTP activates c-jun NH2terminal kinase (JNK) and its upstream regulatory kinase MKK4 in nigrostriatal neurons in vivo, J. Neurochem. 75 (2000) 1200 – 1208. [53] J. Schaack, S. Langer, X. Guo, Efficient selection of recombinant

[54] [55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

[63]

[64]

[65]

[67]

[68]

[69]

97

adenovirus by vectors that express beta-galactosidase, J. Virol. 69 (1995) 3920 – 3923. A.H.V. Schapira, Evidence for mitochondrial dysfunction in Parkinson’s disease—a critical appraisal, Mov. Disord. 9 (1994) 125 – 138. J.P. Sheehan, P.E. Palmer, G.A. Helm, J.B. Tuttle, MPP+ induced apoptotic cell death in SH-SY5Y neuroblastoma cells: an electron microscope study, J. Neurosci. Res. 48 (1997) 226 – 237. R.H. Swerdlow, J.K. Parks, S.W. Miller, J.B. Tuttle, P.A. Trimmer, J.P. Sheehan, J.P. Bennett Jr., R.E. Davis, W.D. Parker Jr., Origin and functional consequences of the complex I defect in Parkinson’s disease, Ann. Neurol. 40 (1996) 663 – 671. T. Takahashi, Y. Deng, W. Maruyama, P. Dostert, M. Kawai, M. Naoi, Uptake of a neurotoxin-candidate, (R)-1,2-dimethyl-6,7-dihydroxy1,2,3,4-tetrahydroisoquinoline into human dopaminergic neuroblastoma SH-SY5Y cells by dopamine transport system, J. Neural Transm. Gen. Sect. 98 (1994) 107 – 118. S. Tanaka, H. Hanafusa, Guanine-nucleotide exchange protein C3G activates JNK1 by a ras-independent mechanism. JNK1 activation inhibited by kinase negative forms of MLK3 and DLK mixed lineage kinases, J. Biol. Chem. 273 (3) (1998) 1281 – 1284. N.A. Tatton, S.J. Kish, In situ detection of apoptotic nuclei in the substantia nigra compacta of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine-treated mice using terminal deoxynucleotidyl transferase labelling and acridine orange staining, Neuroscience 77 (1991) 1037 – 1048. H. Teramoto, O.A. Coso, H. Miyata, T. Igishi, T. Miki, J.S. Gutkind, Signaling from the small GTP-binding proteins Rac1 and Cdc42 to the c-Jun N-terminal kinase/stress-activated protein kinase pathway, J. Biol. Chem. 271 (1996) 27225 – 27228. L.A. Tibbles, Y.L. Ing, F. Kiefer, J. Chan, N. Iscove, J.R. Woodgett, N.J. Lassam, MLK-3 activates the SAPK/JNK and p38/RK pathways via SEK1 and MKK3/6, EMBO J. 15 (1996) 7026 – 7035. K.F. Tipton, T.P. Singer, Advances in our understanding of the mechanisms of the neurotoxicity of MPTP and related compounds, J. Neurochem. 61 (1993) 1191 – 1206. C. Tournier, P. Hess, D.D. Yang, J. Xu, T.K. Turner, A. Nimnual, D. Bar-Sagi, S.N. Jones, R.A. Flavell, R.J. Davis, Requirement of JNK for stress-induced activation of the cytochrome c-mediated death pathway, Science 288 (2000) 870 – 874. P.A. Trimmer, T.S. Smith, A.B. Jung, J.P. Bennett Jr., Dopamine neurons from transgenic mice with knockout of the p53 gene resist MPTP neurotoxicity, Neurodegeneration 5 (1996) 233 – 239. I. Vyas, R.E. Heikkila, W.J. Nicklas, Studies on the neurotoxicity of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine: inhibition of NADlinked substrate oxidation by its metabolite, 1-methyl-4-phenylpyridinium, J. Neurochem. 46 (1986) 1501 – 1507. X.G. Xia, T. Harding, M. Weller, A. Bieneman, J.B. Uney, J.B. Schulz, Gene transfer of the JNK interacting protein-1 protects dopaminergic neurons in the MPTP model of Parkinson’s disease, PNAS 98 (2001) 10433 – 10438. Z. Xu, A.C. Maroney, P. Dobrzanski, N.V. Kukekov, L.A. Greene, The MLK family mediates c-jun N-terminal kinase activation in neuronal apoptosis, Mol. Cell. Biol. 21 (2001) 4713 – 4724. L. Yang, R.T. Matthews, J.B. Schulz, T. Klockgether, A.W. Liao, J.-C. Martinou, J.B. Penney, B.T. Hyman, M.F. Beal, 1-methyl-4-phenyl1,2,3,6-tetrahydropyridne neurotoxicity is attenuated in mice overexpressing BCL-2, J. Neurosci. 18 (1998) 8145 – 8152.