Hiroshi Yamashita1

Kazuyuki Nakamura2, 3

Hirofumi Arai2

Hiroko Furumoto2

Masanori Fujimoto2

Shiro Kashiwagi4

Mitsunori Morimatsu1

1Department of Neurologyand Clinical Neuroscience

2Department of Biochemistryand Biomolecular Recognition

3Central Laboratory for BiomedicalResearch and Education,Yamaguchi University Schoolof Medicine

4Kurokawa Hospital,Yamaguchi, Japan

Electrophoretic studies on the phosphorylationof stathmin and mitogen-activated protein kinasesin neuronal cell death induced by oxidizedvery-low-density lipoprotein with apolipoprotein E

In the central nervous system, stressful conditions can easily cause the oxidation oflipoprotein particles, followed by the oxidative modification of apolipoproteins suchas apolipoprotein E (apoE) and the production of free radicals and aldehydes. Wehave confirmed that oxidized very-low-density lipoprotein (VLDL) inhibits the prolifera-tion, viability and differentiation of neuronal PC12 cells leading to cell death. The cellsinternalized intact apoE, but did not internalize oxidized apoE. The phosphorylation ofstathmin and various mitogen-activated protein (MAP) kinases including extracellularsignal-regulated protein kinase (ERK), p38, and c-Jun N-terminal kinase (JNK) wasexamined in PC12 cells exposed to native and oxidized VLDL, H2O2 (which generatesfree radicals), and 4-hydroxy-2-nonenal (HNE) (an aldehyde). Oxidized VLDL and H2O2

reduced stathmin phosphorylation while HNE increased it, suggesting that oxidizedVLDL and H2O2 stimulated similar signal transduction pathways. Based on the results,free radicals, but not aldehydes may play a major role in the neuronal cell deathinduced by lipoprotein oxidation. Furthermore, the phosphorylation status of MAPkinases indicated that the activation of the JNK cascade might be required for neuronalcell death.

Keywords: Immunoblotting / Neuronal cell death / Oxidative stress / Stathmin / Very-low-densitylipoprotein EL 4865

1 Introduction

Apolipoprotein E (apoE) is a component of very-low-den-sity-lipoprotein (VLDL), intermediate density lipoprotein(IDL), high density lipoprotein (HDL), and chylomicrons.ApoE, a 299 amino acid protein, plays a role in develop-ment and regeneration of the brain through regulatinglipid transport. There are three different alleles of humanapoE (�2, �3, and �4) encoding for three apoE phenotypes(E2, E3, and E4), respectively. The �4 allele has beenimplicated as an important risk factor in late-onset familialand sporadic Alzheimer’s disease (AD) [1–3], cerebralischemia [4] and coronary heart disease [5]. In addition,apoE immunoreactivity has been detected in senile

plaques and neurofibrillary tangles in the brain of AD pa-tients [6]. Several hypotheses have been proposed con-cerning the mechanism by which E4 contributes to AD [7–15]. Miyata et al. [13] reported that apoE protected rat neu-ronal cell lines from oxidative stress with E2�E3�E4. Thethree apoE phenotypes also possessed antioxidant activ-ity with the same relative potency (E2 � E3 � E4). Theauthors therefore suggested that decreased antioxidantactivity of E4 might contribute to the pathogenesis of AD.

On the other hand, Draczynska-Lusiak et al. [16] reportedthat oxidized lipoproteins induced apoptosis in PC12cells, following an activation of NF-�B binding activity[17]. It has been demonstrated that VLDL oxidation cata-lyzed by transition metal ions, such as ferrous (Fe2�) andcupric (Cu2�) ions, caused lipid peroxidation, followed byproduction of free radicals and aldehydes, as well asoxidative modifications of apoE. These components ofoxidized lipoproteins have been suggested to underliethe pathogenesis of neurodegenerative diseases such asAD and Parkinson’s disease [18–20]. However, it remainsto be determined which component is mainly responsiblefor neuronal cell death by oxidation of lipoproteins.

In this study, we evaluated the effect of oxidized VLDLincluding apoE3/E3 on neurons using PC12 cells. In addi-tion, we attempted to investigate the mechanism of the

Correspondence: Prof. Kazuyuki Nakamura, Department ofBiochemistry and Biomolecular Recognition, Yamaguchi Univer-sity School of Medicine, Minamikogushi 1-1-1, Ube, Yamaguchi755-8505, JapanE-mail: nakamura@po.cc.yamaguchi-u.ac.jpFax: �81-836-22-2212

Abbreviations: AD, Alzheimer’s disease; apoE, apolipoprotein E;ERK, extracellular signal-regulated protein kinase; HNE, 4-hydroxy-2-nonenal; JNK, c-Jun N-terminal kinase; MAP, mito-gen-activated protein; MTT, 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyl tetrazolium bromide; VLDL, very-low-density lipopro-tein

998 Electrophoresis 2002, 23, 998–1004

ª WILEY-VCH Verlag GmbH, 69451 Weinheim, 2002 0173-0835/02/07-08–04–998 $17.50+.50/0

cell death induced by oxidized VLDL with electrophoreticstudies on the phosphorylation of stathmin and mitogen-activated protein (MAP) kinases, including extracellularsignal-regulated protein kinase (ERK), p38, and c-JunN-terminal kinase (JNK) as indicators of signal transduc-tion.

2 Materials and methods

2.1 Chemicals

Anti-human and anti-rat apoE antibodies were purchasedfrom Chemicon International (Temecula, CA, USA) andSanta Cruz Biotechnology (Santa Cruz, CA, USA), respec-tively. 4-Hydroxy-2-noneal (HNE) was supplied by WakoPure Chemical (Osaka, Japan). Anti-HNE antibody waskindly provided by Dr. K. Uchida, Nagoya University,Nagoya, Japan. The phosphospecific antibodies againststathmin were kind gifts from Dr. A. Sobel, INSERM U440,Institut du Fer a Moulin, Paris, France. The monoclonalantibodies against phosphorylated ERK, phosphorylatedp38, and phosphorylated JNK were purchased fromSanta Cruz Biotechnology. The polyclonal antibodiesagainst ERK1, which react with ERK1 and to a lesserextent with ERK2, polyclonal antibodies against p38, andpolyclonal antibodies against JNK1, which react withJNK1, JNK2, and JNK3 were purchased from SantaCruz Biotechnology. Secondary antibodies conjugatedto horseradish peroxidase were purchased from ICNPharmaceutical (Aurora, OH, USA) and Santa Cruz Bio-technology. All other chemicals were of reagent grade.

2.2 Preparation of VLDL

Fresh blood from a 12-h fasted, healthy volunteer wascollected into vacutainer tubes containing 2Na-EDTA(1.5 mg/mL) and centrifuged at 3000 rpm for 10 min.The plasma was separated and stored at 4�C. TheVLDL was isolated from the plasma by ultracentrifugation(43 000 rpm) at 16�C for 16 h in a fraction of d � 1.006.The VLDL was dialyzed against a 137 mM NaCl and2.68 mM KCl solution. The protein content of the VLDLwas measured by Lowry’s method [21] using BSA as astandard. The phenotype of apoE was identified as E3/3by polyacrylamide gel isoelectric focusing followed byimmunoblot analysis using anti-human apoE antibodies[22].

2.3 Ferrous ion-induced peroxidation of VLDL

The isolated VLDL was incubated with FeSO4 (200 �M) for12 h at 37�C in aerobic conditions at pH 4.5. The mixturewas dialyzed against a 137 mM NaCl and 2.68 mM KCl

solution to remove FeSO4. The oxidative modification ofapoE in VLDL was confirmed by immunoblotting withanti-human apoE antibodies.

2.4 Cell culture and counting

PC12 cells, a rat pheochromocytoma cell line, were main-tained on a vacuum gas plasma-coating dish at 37�C in anatmosphere of 5% CO2/95% air in RPMI-1640 mediumsupplemented with 2 mM glutamine, 10% horse serum,5% fetal calf serum and 100 units/mL penicillin/strepto-mycin. At experimentation, the cells were grown to 40%confluence. The medium was then replaced with serum-free RPMI-1640 medium for 1 h and the cells were cul-tured in the incubation medium alone (control), or withVLDL (100 �g/mL), H2O2 (250 �M), or HNE (25 �M). Cellswere scraped from the dish after a 48 h treatment withmedium alone or with VLDL (100 �g/mL). After the cellsuspension was diluted 2-fold with trypan blue, unstainedcells (live cells) were counted using a hemocytometerunder a light microscope.

2.5 MTT assay

PC12 cells were grown to 40% confluence on vacuumgas plasma-coating 96-well plates. The media was re-placed with culture media containing various concen-trations of VLDL. After a further 48 h incubation, theVLDL-containing media were replaced with culture mediaand 3-[4,5-dimethyl-thiazol-2-yl]-2,5-diphenyl tetrazoliumbromide (MTT) dye was added. The plates were incu-bated at 37�c for 4 h. Blue formazan was solubilized inDMSO. The absorption was determined at 570 nm usinga microplate reader.

2.6 Preparation of cell lysates

After a 30 min treatment with the stressors, the cells wereharvested and centrifuged at 170�g for 5 min. The pelletwas washed three times with 20 volumes of cold 10 mM

PBS, pH 7.4. The cells were lysed in six volumes of com-bined solution (140 mM NaCl, 1 mM EGTA, 40 mM p-nitro-phenylphosphate, 1% NP-40, 0.1 �M okadaic acid, 1 mM

Na3VO4, and 1 mM dithiothreitol) shaking vigorously onice. The cell lysate was contrifuged at 15 000�g for30 min and the supernatant was stored at –80�C untiluse. The protein concentration of each sample wasdetermined by Lowry’s method [21].

2.7 Sample preparation for HNE detectionin VLDL

Samples for HNE detection shown in Fig. 2B were pre-pared as follows. The supernatant (sample for lane 2)and aggregates were separated from oxidized VLDL by

Electrophoresis 2002, 23, 998–1004 Analysis of neuronal cell death induced by oxidized VLDL 999

Gen

eral

Figure 1. Effect of native and oxidized VLDL on cellproliferation, viability, and differentiation. (A) Number ofcells after a 48-h exposure to VLDL (100 �g/mL). The cellswere counted as described in Section 2.4. The resultswere expressed as percent of control. * indicates a sta-tistically significant difference between treatment withnative and oxidized VLDL (p � 0.05). (B) Cell viability asdetermined by the MTTassay. The cells were treated withdifferent concentrations (0, 50, 100, 200 �g/mL) of VLDLfor 48 h. The results were expressed as percent of control.(C) Cell morphology, observed under a light microscopeafter treatment with VLDL (100 �g/mL) for 48 h. Thearrows in panel 3 indicate distinctive cells with poor neu-rodendrite outgrowth. Panel 1, control; panel 2, treatedwith native VLDL; panel 3, treated with oxidized VLDL.

centrifugation at 15 000�g for 15 min. After the aggre-gates were dissolved in six volumes of 2% SDS solution,further centrifugation at 15 000�g for 15 min was per-formed and the supernatant was used for lane 3. NativeVLDL, which did not include the aggregates, was used forlane 1.

2.8 SDS-PAGE and immunoblotting

SDS-polyacrylamide gel electrophoresis (SDS-PAGE) wascarried out using separating gels of 12.5% or 15.0% and2.6% according to the method of Laemmli [23]. The pro-teins in the gel were transferred to a polyvinylidenedifluoride membrane (Millipore, Bedford, MA, USA) [24].After blocking with TBS containing 5% skim milk, themembrane was washed with TBS containing 0.05%Tween 20 and reacted with the second antibody conju-gated to horseradish peroxidase for 2 h. The membranewas washed again and immunoreactive species werevisualized by enhanced chemiluminescence using a de-tection kit (Amersham Pharmacia Biotech Amersham,UK).

3 Results

3.1 Effect of native and oxidized VLDLon PC12 cells

PC12 cells were treated with medium alone, native VLDLor oxidized VLDL for 48 h and the number of cells wascounted (Fig. 1A). Proliferation was reduced in cells thathad been incubated with oxidized VLDL. Additionally,trypan blue positive cells (dead cells) were substantiallyincreased. As shown in Fig. 1B, cell viability was exam-ined by an MTT assay following 48 h culture in differentconcentrations of VLDL. Oxidized VLDL decreased cellviability in a dose-dependent manner. We also examinedthe morphology of these cells under a light microscope(Fig. 1C). In cells treated with oxidized VLDL, a significantdegree of cell death was observed and the neurodendritegrowth was poor. These findings demonstrate that oxi-dized VLDL is toxic to neuronal cells and leads to cellinjury and death.

3.2 Immunoblotting analysis of native andoxidized VLDL

ApoE in native and oxidized VLDL was analyzed by immu-noblotting (Fig. 2A). In native VLDL, intact apoE wasdetected as an approximately 34 kDa protein band (lane1). On the other hand, oxidative modification of apoEmediated by Fe2� was seen as a smeared band with only

1000 H. Yamashita et al. Electrophoresis 2002, 23, 998–1004

Figure 2. Immunoblotting pattern of human apoE andHNE in native and oxidized VLDL. (A) Immunoblottingpattern of human apoE. Samples containing 1.5 �g pro-tein were applied to each well. After SDS-PAGE (12.5%T,2.67%C), immunoblotting with an anti-human apoE anti-body was carried out. Lane 1: native VLDL, lane 2: oxi-dized VLDL. (B) Immunoblotting pattern of HNE. Sampleswere prepared as described under Materials and meth-ods. After SDS-PAGE (12.5%T, 2.67%C), immunoblottingwith HNE antibody was performed. Lane 1: native VLDL,lane 2: supernatant of oxidized VLDL, lane 3: aggregatesof oxidized VLDL.

an obscure 34 kDa band (lane 2). HNE formation by lipidperoxidation was checked with immunoblot analysis(Fig. 2B). An HNE-positive band was detected in theaggregates for oxidized VLDL (lane 3), but was not foundin the native VLDL or in the supernatant of oxidized VLDL(lane 1, 2). An approximately 66 kDa band was confirmedas albumin with immunoblotting using an anti-humanalbumin antibody (data not shown).

3.3 Uptake of human apoE by PC12 cells

We tested whether or not PC12 cells had taken up humanapoE in VLDL-containing medium (Fig. 3A). PC12 celllysates were subjected to electrophoresis and immuno-blotted. ApoE was detected in PC12 cells treated withnative VLDL using a primary antibody against humanapoE (lane 2), but it was not detected in the cells treatedwith oxidized VLDL (lane 3). There was, however, a possi-bility that the detected apoE had been produced by thecells. Therefore, immunoblot analysis with antibody torat apoE was also performed and no bands were found(Fig. 3B). Based on this result, apoE detected in Fig. 3Awas thought to be derived from the VLDL mixed with themedium.

Figure 3. Uptake of human apoE by PC12 cells. Aftertreatment with VLDL (100 �g/mL), the cell lysate wasanalyzed by immunoblotting using antibody against(A) human and (B) rat apoE. Equal amounts of sample(50 �g/well) were subjected to SDS-PAGE (15.0%T,2.67%C). Lane 1: control, lane 2: treated with nativeVLDL, lane 3: treated with oxidized VLDL.

Figure 4. Immunoblot analysis of stathmin Ser 38 phos-phorylation. Samples containing 50 �g protein were load-ed in each well and subjected to SDS-PAGE (15.0%T,2.67%C), Lane 1: control, lane 2: treated with nativeVLDL (100 �g/mL), lane 3: treated with oxidized VLDL(100 �g/mL), lane 4: control, lane 5: treated with HNE(25 �M), lane 6: treated with H2O2 (250 �M).

3.4 Immunoblot analysis of phosphorylatedstathmin

The level of stathmin phosphorylation was determined byimmunoblot analysis using antibodies against phos-phorylated stathmin. We tested the phosphorylation onSer 25 and Ser 38. Ser 25 phosphorylation was notdetected with each treatment (data not shown). As shownin Fig. 4, oxidized VLDL and H2O2 obviously reducedSer 38 phosphorylation (lane 3, 6), whereas HNE in-creased it (lane 5).

3.5 Immunoblot analysis of phosphorylatedMAP kinases

To estimate the activation of MAP kinases, phosphoryla-tion of MAP kinases was determined by immunoblot ana-lysis using monoclonal antibodies against phosphory-

Electrophoresis 2002, 23, 998–1004 Analysis of neuronal cell death induced by oxidized VLDL 1001

Figure 5. Immunoblot analysis of phosphorylated MAPkinases. The immunoblotting patterns of phosphorylatedMAP kinases were obtained using the monoclonal anti-body to (A) phosphorylated ERK, (C) phosphorylatedp38, and (E) phosphorylated JNK. The expression levelsof ERK, p38, and JNK were examined by immunoblottingusing the polyclonal antibodies to (B) ERK1, (D) p38, and(F) JNK1, respectively. Equal amounts of protein (80 �g/well) were used for SDS-PAGE (15.0%T, 2.67%C). Lane 1:control, lane 2: treated with native VLDL (100 �g/mL), lane3: treated with oxidized VLDL (100 �g/mL), lane 4: treatedwith HNE (25 �M), lane 5: control, lane 6: treated with H2O2

(250 �M).

lated ERK, phosphorylated p38, and phosphorylatedJNK. Phosphorylated ERK1/2, p38, and JNK1/2 weredetected. As shown in Fig. 5E, the phosphorylation ofJNK, especially that of the JNK1 isoform, was increasedby every treatment. In contrast, no treatments elevatedthe phosphorylation of ERK and p38 (Figs. 5A, C). Theexpression levels of ERK1/2, p38, and JNK1/2 were notsignificantly changed by each stress as shown in Fig. 5B,D, and F, respectively.

4 Discussion

Recent studies have suggested that oxidative stress maybe a principal factor underlying the pathogenesis of manydiseases, including neurodegenerative diseases such asAD and Parkinson’s disease [18–20, 25]. The brain is parti-cularly vulnerable to oxidative stress, because it has a highoxygen consumption rate as well as abundant lipids andlipoproteins, which are highly sensitive to oxidation. Drac-zynska-Lusiak et al. [16] reported that oxidative lipopro-teins induced neuronal cell death but the mechanism isnot understood. In this paper, we have shown that oxidizedVLDL descreased the proliferation and viability of a neu-ronal cell line and suppressed the neurodendrite growth(Figs. 1A–C).

The relationship between AD and apoE4 has been dis-cussed from various standpoints [7–15]. Miyata et al. [13]reported decreased antioxidant activity of E4, suggestingthat it could contribute to AD. Recently, we have shownthat apoE in VLDL was oxidatively modified and VLDLcomposed of E3/4 was more easily peroxidized than thatof E3/3 by a transition ion [26–28]. Pathologically, apoEimmunoreactivity was shown to be present in senileplaques, neurofibrillary tangles, and cerebral vessels inAD brains [6]. To determine whether or not oxidativelymodified apoE directly participated in neuronal damagein our study, we tested apoE uptake in PC12 cells, whichpossess apoE receptors belonging to the LDL receptorsuper family [29]. The result showed that intact apoE wasinternalized in the cells, whereas oxidatively modifiedapoE was not (Fig. 3A). This finding suggests that theneurotoxicity of oxidized VLDL is not due to uptake ofthese particles by the cells.

Lipoprotein oxidation by transition metal ions causes lipidperoxidation, leading to products such as free radicalsand aldehydes which are toxic to neuronal cells [18–20].We speculated that these oxidative products might beinvolved in the neuronal cell death induced by oxidizedVLDL. Peroxyl radical (ROO�) and alkoxyl radical (RO�)are formed as free radicals during lipid peroxidation. Freeradicals can cause membrane lipid peroxidation, directdamage to protein by oxidation or by protein-proteincross-linking, increases in intracellular free Ca2�, DNAdamage, inactivation of enzyme, increases in intracellularfree iron/copper and activation of signal transductionleading to cell death [20, 30]. In this experiment, we usedH2O2, which promotes the Fenton reaction to yield hy-droxyl radicals in the presence of Fe2� [31]. On the otherhand, HNE and malondialdehyde are produced as alde-hydes. It has been reported that HNE is cytotoxic to neu-roglial cultures by cross-linking cytoskeletal proteins [32]and that it induces neuronal apoptosis in PC12 cells [33].In the hippocampus and temporal cortex of patients withAD, HNE pyrrole adduct immunoreactivity is increased inneurofibrillary tangles and in neurons lacking neurofibril-lary tangles [34–37]. In our study, formation of HNE-pro-tein was exhibited in oxidized VLDL (Fig. 2B). In addition,the immunoblotting pattern was very similiar to that ofthe oxidatively modified apoE, indicating that HNE cova-lently cross-linked apoE with lipid peroxidation as Mon-tine et al. [38] have reported.

To investigate the mechanism of neuronal cell death in-duced by oxidized VLDL, we treated the cells with H2O2 orHNE and evaluated the phosphorylation patterns of stath-min. Stathmin is a ubiquitous cytoplasmic substrate forextracellular stimuli and has been proposed as a generalrelay, integrating diverse intracellular signaling pathways[39]. Stathmin has four serine phosphorylation sites which

1002 H. Yamashita et al. Electrophoresis 2002, 23, 998–1004

are targets for protein kinases regulating cell morphologyand function [40]. It has been suggested that Ser 16, 25,38, and 63 are substrates for Ca2�/calmodulin-dependentkinase, protein kinases C and protein kinase A [41–43],MAP kinase [40], cyclin-dependent kinases (cdks) [40,44], and protein kinase A [40, 43], respectively. Stathminis phosphorylated by extracellular stimuli such as heatshock [45]. However, there have been no reports on thephosphorylation of stathmin induced by oxidative stress.We examined the phosphorylation of stathmin on Ser 25and Ser 38. No Ser 25 phosphorylations were found in celllysates treated with oxidative stressors. Interestingly, oxi-dized VLDL and H2O2 markedly diminished the phosphor-ylation on Ser 38, whereas HNE increased it. These find-ings suggest that cdk’s, which regulate the cell cycle maybe a key substrate for oxidative stress. Hensley et al. [46]and Butterfield [47] reported that amyloid �-peptide (A�),which is aggregated on senile plaques in AD brains, couldgenerate free radicals in aqueous solution. It is widelybelieved that excess deposition of A� is a chief factorunderlying the pathogenesis of AD and these authorssuggest that radicalization might underlie the neurotoxi-city of A�. Our results indicate that radical generationmediated by lipid peroxidation or by other means mayplay an important role in inducing neuronal disorders andcontributing to the pathogenesis of AD.

To assess the effect of oxidized VLDL on MAP kinases,we examined the phosphorylation of ERK, p38, and JNK.These MAP kinases are activated by various externalsignals and are involved in early signal transduction path-ways. It has been reported that in PC12 cells, H2O2 acti-vates JNK1 [48] or ERK2 [49], while HNE induces selec-tive activation of JNK [50]. In the present sutdy, H2O2 andHNE selectively activated JNK, especially the JNK1 iso-form (Fig. 5). Moreover, the selective JNK1 activation wasevoked not only by oxidized VLDL but also by native VLDL.Recent studies have demonstrated that JNK is an essen-tial component of a signal transduction pathway leading toapoptosis [51] or cell differentiation [52, 53] in PC12 cells.Taken together with these data, our results suggest thatthe activation of JNK may be necessary but not sufficientfor neuronal cell death induced by oxidized VLDL.

Received November 9, 2001

5 References

[1] Corder, E. H., Saunders, A. M., Strittmatter, W. J., Schmechel,D. E., Gaskell, P. C., Small, G. W., Roses, A. D., Haines, J. L.,Pericak-Vance, M. A., Science 1993, 261, 921–923.

[2] Saunders, A. M., Strittmatter, W. J., Schmechel, D., St.George-Hyslop, P. H., Pericak-Vance, M. A., Joo, S. H., Rosi,B. L., Gusella, J. F., Crapper-MacLachlan, D. R., Alberts, M.J., Hulette, C., Crain, B., Goldgaber, D., Roses, A. D., Neurol-ogy 1993, 43, 1467–1472.

[3] Strittmatter, W. J., Saunders, A. M., Schmechel, D., Pericak-Vance, M., Enghild, J., Salvesen, G. S., Roses, A. D., Proc.Natl. Acad. Sci. USA 1993, 90, 1977–1981.

[4] Sheng, H., Laskowitz, D. T., Bennett, E., Schmechel, D. E.,Bart, R. D., Saunders, A. M., Pearlstein, R. D., Roses, A. D.,Warner, D. S., J. Cereb. Blood Flow Metab. 1998, 18, 361–366.

[5] Xhignesse, M., Lussier-Cacan, S., Sing, C. F., Kessling, A. M.,Davignon, J., Arterioscl. Thromb. 1991, 11, 1100–1110.

[6] Namba, Y., Tomonaga, M., Kawasaki, H., Otomo, E., Ikeda,K., Brain Res. 1991, 541, 163–166.

[7] Wisniewski, T., Castaño, E. M., Golabek, A., Vogel, T., Fran-gione, B., Am. J. Pathol. 1994, 145, 1030–1035.

[8] Ma, J., Yee, A., Brewer, H. B., Jr., Das, S., Potter, H., Nature1994, 372, 92–94.

[9] Strittmatter, W. J., Weisgraber, K. H., Goedert, M., Saunders,A. M., Huang, D., Corder, E. H., Dong, L.-M., Jakes, R.,Alberts, M. J., Gilbert, J. R., Han, S.-H., Hulette, C., Einstein,G., Schmechel, D. E., Pericak-Vance, M. A., Roses, A. D.,Exp. Neurol. 1994, 125, 163–171.

[10] Nathan, B. P., Bellosta, S., Sanan, D. A., Weisgraber, K. H.,Mahley, R. W., Pitas, R. E., Science 1994, 264, 850–852.

[11] Nathan, B. P., Chang, K.-C., Bellosta, S., Brisch, E., Ge, N.,Mahley, R. W., Pitas, R. E., J. Biol. Chem. 1995, 270, 19791–19799.

[12] Marques, M. A., Tolar, M., Harmony, J. A., Crutcher, K. A.,Neuroreport 1996, 7, 2529–2532.

[13] Miyata, M., Smith, J. D., Nature Genet. 1996, 14, 55–61.[14] Jordán, J., Galindo, M. F., Miller, R. J., Reardon, C. A., Getz,

G. S., LaDu, M. J., J. Neurosci. 1998, 18, 195–204.[15] Michikawa, M., Yanagisawa, K., J. Neurosci. Res. 1998, 54,

58–67.[16] Draczynska-Lusiak, B., Doung, A., Sun, A. Y., Mol. Chem.

Neuropathol. 1998, 33, 139–148.[17] Draczynska-Lusiak, B., Chen, Y. M., Sun, A. Y., Neuroreport

1998, 9, 527–532.[18] Evans, P. H., Br. Med. Bull. 1993, 49, 577–587.[19] Markesbery, W. R., Free Rad. Biol. 1997, 23, 134–147.[20] Halliwell, B., Gutteridge, J. M. C., Free Radicals in Biology

and Medicine, Oxford University Press, New York 1999.[21] Lowry, O. H., Rosebrough, N. J., Farr, A. L., Randall, R. J.,

J. Biol. Chem. 1951, 193, 265–275.[22] Kataoka, S., Paidi, M., Howard, B. V., Clin. Chem. 1994, 40,

11–13.[23] Laemmli, U. K., Nature 1970, 227, 680–685.[24] Nomura, S., Kashiwagi, S., Ito, H., Mimura, Y., Nakamura,

K., Electrophoresis 1993, 14, 1318–1321.[25] Lyras, L., Cairns, N. J., Jenner, P., Halliwell, B., J. Neuro-

chem. 1997, 68, 2061–2069.[26] Hara, S., Tanaka, T., Yamada, M., Nagasaka, Y., Nakamura,

K., in: Ishimura, Y., Shimada, H., Suematsu, M. (Eds), Oxy-gen Homeostasis and its Dynamics, Springer-Verlag, Tokyo1998, pp. 597–602.

[27] Arai, H., Kashiwagi, S., Nagasaka, Y., Uchida, K., Hoshii, Y.,Nakamura, K., Arch. Biochem. Biophys. 1999, 367, 1–8.

[28] Kashiwagi, S., Nakamura, K., Arai, H., Yamashita, H., Ito, H.,Electrophoresis 1999, 20, 1418–1424.

[29] Kim, D. H., Iijima, H., Goto, K., Sakai, J., Ishii, H., Kim, H. J.,Suzuki, H., Kondo, H., Saeki, S., Yamamoto, T., J. Biol.Chem. 1996, 271, 8373–8380.

[30] Stadtman, E. R., Berlett, B. S., Chem. Res. Toxicol. 1997, 10,486–494.

[31] Halliwell, B., Gutteridge, J. M. C., Methods Enzymol. 1984,105, 47–56.

Electrophoresis 2002, 23, 998–1004 Analysis of neuronal cell death induced by oxidized VLDL 1003

[32] Montine, T. J., Amarnath, V., Martin, M. E., Strittmatter, W. J.,Graham, D. G., Am. J. Pathol. 1996, 148, 89–93.

[33] Kruman, I., Bruce-Keller, A. J., Bredesen, D., Waeg, G.,Mattson, M. P., J. Neurosci. 1997, 17, 5089–5100.

[34] Montine, K. S., Kim, P. J., Olson, S. J., Markesbery, W. R.,Montine, T. J., J. Neuropathol. Exp. Neurol. 1997, 56, 866–871.

[35] Montine, K. S., Olson, S. J., Amarnath, V., Whetsell, W. O.,Jr., Graham, D. G., Montine, T. J., Am. J. Pathol. 1997, 150,437–443.

[36] Sayre, L. M., Zelasko, D. A., Harris, P. L. R., Perry, G., Salo-mon, R. G., Smith, M. A., J. Neurochem. 1997, 68, 2092–2097.

[37] Montine, K. S., Reich, E., Neely, M. D., Sidell, K. R., Olson,S. J., Markesbery, W. R., Montine, T. J., J. Neuropathol. Exp.Neurol. 1998, 57, 415–425.

[38] Montine, T. J., Huang, D. Y., Valentine, W. M., Amarnath, V.,Saunders, A., Weisgraber, K. H., Graham, D. G., Strittmatter,W. J., J. Neuropathol. Exp. Neurol. 1996, 55, 202–210.

[39] Sobel, A., Boutterin, M. C., Beretta, L., Chneiweiss, H.,Doye, V., Peyro-Saint-Paul, H., J. Biol. Chem. 1989, 264,3765–3772.

[40] Beretta, L., Dobránsky, T., Sobel, A., J. Biol. Chem. 1993,268, 20076–20084.

[41] Marklund, U., Brattsand, G., Osterman, Ö., Ohlsson, P. I.,Gullberg, M., J. Biol. Chem. 1993, 268, 25671–245680.

[42] Brattsand, G., Marklund, U., Nylander, K., Roos, G., Gull-berg, M., Eur. J. Biochem. 1994, 220, 359–368.

[43] Gradin, H. M., Larsson, N., Marklund, U., Gullberg, M.,J. Cell Biol. 1998, 140, 131–141.

[44] Marklund, U., Brattsand, G., Shingler, V., Gullberg, M.,J. Biol. Chem. 1993, 268, 15039–15049.

[45] Fujimoto, M., Nagasaka, Y., Tanaka, T., Nakamura, K.,Electrophoresis 1998, 19, 2515–2520.

[46] Hensley, K., Carney, J. M., Mattson, M. P., Aksenova, M.,Harris, M., Wu, J. F., Floyd, R. A., Butterfield, D. A., Proc.Natl. Acad. Sci. USA 1994, 91, 3270–3274.

[47] Butterfield, D. A., Chem. Res. Toxicol. 1997, 10, 495–506.[48] Mielke, K., Damm, A., Yang, D. D., Herdegen, T., Mol. Brain

Res. 2000, 75, 128–137.[49] Guyton, K. Z., Liu, Y., Gorospe, M., Xu, Q., Holbrook, N. J.,

J. Biol. Chem. 1996, 271, 4138–4142.[50] Soh, Y., Jeong, K. S., Lee, I. J., Bae, M. A., Kim, Y. C., Song,

B. J., Mol. Pharmacol. 2000, 58, 535–541.[51] Xia, Z., Dickens, M., Raingeaud, J., Davis, R. J., Greenberg,

M. E., Science 1995, 270, 1326–1331.[52] Heasley, L. E., Storey, B., Fanger, G. R., Butterfield, L.,

Zamarripa, J., Blumberg, D., Maue, R. A., Mol. Cell. Biol.1996, 16, 648–656.

[53] Leppä, S., Saffrich, R., Ansorge, W., Bohmann, D., EMBO J.1998, 17, 4404–4413.

1004 H. Yamashita et al. Electrophoresis 2002, 23, 998–1004