Transfection of the plasma-type platelet-activating factor acetylhydrolase gene attenuates glutamate-induced apoptosis in cultured rat cortical neurons

Brain Research 885 (2000) 128–132 www.elsevier.com / locate / bres

Short communication

Transfection of the plasma-type platelet-activating factor acetylhydrolase gene attenuates glutamate-induced apoptosis in cultured rat cortical neurons Yutaka Hirashima a , *, Hikaru Ueno b , Ken Karasawa c , Kazuaki Yokoyama c , Morio Setaka c , Akira Takaku a b

a Department of Neurosurgery, Toyama Medical and Pharmaceutical University, Toyama-shi, Toyama 930 -0194, Japan Department of Cardiology, Graduate School of Medical Sciences, Kyushu University, Fukuoka-shi, Fukuoka 812 -8582, Japan c Faculty of Pharmaceutical Sciences, Teikyo University, Sagamiko-cho, Kanagawa 199 -0106, Japan

Accepted 15 August 2000

Abstract Using an adenoviral vector, we induced overexpression of the plasma type of platelet-activating factor acetylhydrolase in cultured rat neurons. Neurons overexpressing this enzyme showed a decrease in glutamate-induced injury, mainly, apparent as decreased apoptosis. Reduction of lipid peroxidation by this enzyme and protection of mitochondrial function were demonstrated, and these may be the basis of the resistance to glutamate-induced neuronal injury that we observed.  2000 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Neuronal death Keywords: Neurons; Apoptosis; Platelet-activating factor (PAF) acetylhydrolase; Glutamate; Mitochondrial damage

Platelet-activating factor (PAF) acetylhydrolase (PAFAH) removes the acetyl group at the sn-2 position of PAF. Forms of mammalian PAFAH include an intracellular (tissue) type and an extracellular (plasma) type. In turn, tissue cytosol includes at least two types of tissue PAFAH, isoforms Ib and II [7]. Isoform Ib is a tertiary G protein complex-like heterotrimeric enzyme, while isoform II is a 40-kDa monomer. Plasma PAFAH is a 45-kDa monomeric enzyme in humans and a 63-kDa monomeric enzyme in guinea pigs [5,12,24]. Isoform II has an amino acid sequence with 41% identity to that of human plasma PAFAH [8,9,24]. Although human and guinea pig plasma PAFAH show some differences in physical properties such as association of lipoproteins or molecular mass, they share many similar biochemical properties [11]. Isoform II preferentially hydrolyzes oxidized phospholipids as well as PAF [8], and therefore may function as an antioxidant [15]. *Corresponding author. Tel.: 181-76-434-2281; fax: 181-76-4345034. E-mail address: [email protected] (Y. Hirashima).

Plasma PAFAH also has the ability to catabolize oxidized phospholipids [11,21,22]. Although plasma PAFAH is secreted in vitro from various cells such as macrophages [4] and hepatocytes [23], the exact cellular source(s) of this enzyme in vivo remains unclear. Recently, Asano et al. reported that most of the PAFAH activity in human plasma originates from hematopoietic lineage cells [2] Neurotoxicity initiated by overstimulation of glutamate receptors and subsequent depolarization of mitochondria has been suggested to contribute to oxidative stress in neuronal injury [6,26]. Previous investigation has indicated that two types of neuronal death, apoptosis and necrosis, can result from glutamate receptor-mediated excitotoxicity [1]. Mild insults lead to transient mitochondial depolarization, reversible energy failure, and apoptosis, while intense insults produce irreversible mitochondrial depolarization, permanent energy collapse, and ionic imbalances resulting in cellular swelling and necrosis [1]. Overexpression of isoform II in Chinese hamster ovary K1 cells has been reported to suppress tert-butylhydroperoxide (t-BuOOH)induced apoptotic cell death, most likely by its antioxidant

0006-8993 / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 00 )02852-3

Y. Hirashima et al. / Brain Research 885 (2000) 128 – 132

effects [15]. In the present study, we determined whether glutamate-induced neuronal death was reduced in neuronal cultures by overexpression of the plasma PAFAH gene transfected with an adenovirus vector in place of tissuetype PADAH gene and estimated whether oxidative stress due to impairment of mitochondrial function contributed to neuronal apoptosis. We purchased 5-bromo-4-chloro-3-indolyl-b-D-galactopyranoside (X-gal) from Wako (Osaka, Japan). Thiobarbituric acid (TBA) reagent as well as 1,1,3,3-tetraethoxypropane, 3-[4,5-dimethylthiazole-2-yl]-2,5-diphenyl tetrazolium bromide (MTT reagent), and Hoechst 33258 was obtained from Sigma (St. Louis, MO). A protein assay kit was obtained from Bio-Rad (Hercules, CA). Replication-defective E1- and E3-adenoviral vectors without an inserted gene (Ad1W) or alternatively containing either guinea pig plasma PAFAH gene (AdPAFAH)[12] or b-galactosidase gene (AdLacZ)[14] under control of cytomegalovirus enhancer and a chicken b-actin promoter together constituting a CA promoter [17] were prepared as described previously [25]. Cerebral cortical neuronal cells were prepared from 17-day-old rat embryos according to a previously reported procedure [18,19]. The dissociated cells were plated in polyethylenimine-coated 96-well and 24-well culture plates and 2-well glass slides. Incubation with recombinant adenovirus (10–50 multiplicity of infection (m.o.i.)) was carried out for 6 h using 4-day-old cultures. Cells next were exposed to 1 mM glutamate for 60 min on day 7 in culture, and the medium was replaced. Cultures then were maintained for 24 h before neuronal death was assessed. PAFAH activity was assayed as described by Stafforini et al. [20]. X-gal staining was performed to demonstrate gene expression for b-galactosidase, as previously described in detail [25]. Briefly, cells on glass slides were prefixed with 2% formaldehyde and 0.2% glutaraldehyde in phosphatebuffered saline (PBS), incubated in 5-bromo-4-chloro-3indolyl-b-D-galactopyranoside, and postfixed with 2% formaldehyde and 0.2% glutaraldehyde in PBS, with a thorough rinse with PBS before each step. The TBA test protocol of Lee and Csallany [13] was used with the modification that phosphoric acid was replaced with 10% trichloroacetic acid (TCA) in 1% HCl [3]. After addition of the TCA solution samples, 0.5 ml of TBA reagent (1% w / v TBA in 50 mM NaOH) was added, and the mixture was heated for 15 min at 1008C. After cooling, the chromophore was extracted into 1.5 ml butan1-ol, and absorbance was measured at 532 nm. Since the TBA test measures malondialdehyde (MDA), we prepared a standard plot for MDA in the TBA test system, which was linear up to 20 nmol MDA per sample [3]. The plot was used to convert absorbance values to nmol of MDA. MDA content was normalized to protein concentration. Mitochondrial function was assessed by the MTT assay. Using 96-well microplates, a modified MTT assay was

129

performed essentially as described by Musser et al. [16]. MTT reagent was dissolved at 5 mg / ml in PBS, filtersterilized using 0.22 mm disposable filters (Millex-GV; Millipore, Bedford, MA), and stored at 48C. MTT stock solution (25 ml) was added to each well and incubated at 378C for 4 h to allow mitochondria to form a colored, insoluble formazan product. After the supernatant was removed by aspiration, 100 ml of dimethylsulphoxide was added to each well to solubilize the formazan crystals. The plate was agitated to fully dissolve the crystals, and absorbance at 550 nm was measured by a model 340 ATTC microplate reader (SLT, Saltzburg, Austria). Cells on glass slides were stained with Hoechst 33258 (10 mM) for 5 min and analyzed morphologically under a nonconfocal fluorescence microscope (Olympus BX50, Tokyo, Japan). A photomicrograph was taken immediately with color reversal film (Fujichrome Provia 400, Fuji, Tokyo, Japan). Apoptotic cells identifiable by nuclear fragmentation were counted from the color print. Data are presented as the mean6S.D. For statistical comparisons, one-way analyses of variance followed by Fisher’s protected least-squares difference test were used for multiple comparisons. P-values ,0.05 were considered to indicate significant differences. We first examined the expression of LacZ by X-Gal staining in neurons exposed to AdLacZ at various m.o.i. for 72 h. Neurons showed deep blue coloration with intensity dependent on vector quantity from 10 to 50 m.o.i. (data not shown). The percentage of neurons with positive X-gal staining at 50 m.o.i. was 70.8610.9% (n54). Next, we measured PAFAH activity in neurons infected with AdPAFAH. Like LacZ, the activity of PAFAH increased with increasing m.o.i. (Fig. 1). PAFAH activity in culture

Fig. 1. Platelet-activating factor acetylhydrolase (PAFAH) activity in neurons transfected with AdPAFAH (recombinant adenovirus vector containing the guinea pig plasma PAFAH gene). Values are expressed as the mean6S.D. (n54). **, P,0.01 and ***, P,0.001 compared with controls. †††, P,0.001 compared with AdPAFAH at 10 m.o.i.; ¶¶, P,0.01 compared with AdPAFAH at 50 m.o.i. (Significance testing: one-way analysis of variance followed by Fisher’s protected least-squares difference test.).

130

Y. Hirashima et al. / Brain Research 885 (2000) 128 – 132

medium showed no difference between media from neurons without infection (control), neurons infected with Ad1W, and neurons infected with AdPAFAH. Activity in medium was negligible compared with activity in cell suspensions (data not shown). Plasma type of PAFAH should be produced and secreted from some type of cells [2,4,23] and some enzyme activity was expected in the culture medium infected with AdPAFAH. The discrepancy may be explained by the pooling system of the enzyme in neurons. The result suggested that transfection at m.o.i. 50 for 72 h achieves sufficient cellular expression of plasma PAFAH. Oxidation of lipids, (lipid peroxidation) was assayed with TBA. In experiments without glutamate, neurons infected with Ad1W were associated with more reaction product (MDA) than neurons without infection (control) or those infected with AdPAFAH. Infection with adenovirus then, may induce oxidation of lipids (Fig. 2) [11,15]. Although glutamate treatment resulted in increased MDA in neurons without infection (control) or infection with Ad1W, MDA content was not significantly increased by glutamate loading in neurons infected with AdPAFAH (Fig. 2). PAFAH may attenuate increased membrane lipid peroxidation resulting from both viral infection and glutamate treatment. Extent of neuronal injury was assayed in terms of mitochondrial function in the MTT assay [16]. Neurons infected with AdPAFAH and those without infection (control) showed significantly higher amounts of MTT product than neurons infected with Ad1W (Fig. 3). These results suggest that without glutamate loading neuronal mitochondrial function is impaired by viral infection, while the antioxidant action of plasma PAFAH may protect

Fig. 2. Malondialdehyde (MDA) content in neurons with and without exposure to 1 mM glutamate (Glu).Values are expressed as the mean6S.D. (n54). AdPAFAH represents neurons transfected with recombinant adenovirus vector containing the guinea pig plasma plateletactivating factor acetylhydrolase gene; Ad1W, neurons infected with adenoviral vectors without an inserted gene; and control, neurons without infection. ¶, P,0.05 compared with Ad1W/ Glu(2). *, P,0.05 compared with control / Glu(2). †, P,0.05 and ††, P,0.01 compared with AdPAFAH / Glu(1). (Significance testing:one-way analysis of variance followed by Fisher’s protected least-squares difference test.).

Fig. 3 . Tetraepoxypropane dimethylthiazolediphenyl tetrazolium (MMT) product in neurons with and without exposure to 1 mM glutamate (Glu).Values are expressed as the mean6S.D. (n54). AdPAFAH represents neurons transfected with recombinant adenovirus vector containing the guinea pig plasma platelet-activating factor acetylhydrolase gene; Ad1W, neurons infected with adenoviral vectors without an inserted gene; and control, neurons without infection.; transfected with recombinant adenovirus vector containing the guinea pig plasma PAFAH gene, Ad1W; infected with only adenoviral vector, and control; without infection. **, P,0.01 compared with Ad1W/ Glu(2). §§§, P,0.001 compared with Ad1W/ Glu(2). ¶¶¶, P,0.001 compared with control / Glu(2). ††, P, 0.01 and †††, P,0.001 compared with control / Glu(1) and Ad1W/ Glu(1). (Significance testing:one-way analysis of variance followed by Fisher’s protected least-squares difference test.). OD550, optical density at 550 nm.

neuronal mitochondrial function from this insult in addition to attenuating oxidation of membrane phospholipids [11,15].

Fig. 4 . Apoptosis of neurons with and without exposure to 1 mM glutamate (Glu). Values are expressed as the mean6S.D. (n54). AdPAFAH represents neurons transfected with recombinant adenovirus vector containing the guinea pig plasma platelet-activating factor acetylhydrolase gene; Ad1W, neurons infected with adenoviral vectors without an inserted gene; and control, neurons without infection. ††, P,0.01 compared with Ad1W/ Glu(2). §§§, P,0.001 compared with Ad1W/ Glu(2). ¶¶¶, P,0.001 compared with control / Glu(2). ***, P, 0.001 compared with AdPAFAH / Glu(1).(Significance testing:one-way analysis of variance followed by Fisher’s protected least-squares difference test.).

Y. Hirashima et al. / Brain Research 885 (2000) 128 – 132

131

Fig. 5. Hoechst 33258 staining of cultured cortical neurons with and without exposure to 1 mM glutamate (Glu). Apoptotic neurons with fragmented nuclei were widespread in cultures infected with adenoviral vectors without an inserted gene (Ad1W/ Glu(1)) or with no infection (control / Glu(1)), while neurons infected with AdPAFAH(AdPAF / Glu(1)) did not show marked apoptotic features. A, control / Glu(2); B, Ad1W/ Glu(2); C, AdPAFAH / Glu(2); D, control / Glu(1); E, Ad1W/ Glu(1) and F; AdPAFAH / Glu(1).

Damage to neurons from 1 mM glutamate exposure was estimated by the MTT assay 24 h after exposure to glutamate for 60 min. Significantly more MTT product was present 24 h after exposure to 1 mM glutamate for 60 min in neurons infected with AdPAFAH than in those infected with Ad1W or those without infection (Fig. 3). Tissue-type, 40 kDa monomer PAFAH (isoform II) acts as an antioxidant phospholipase that protects cells from oxidative stress [15]. Plasma-type PAFAH has an amino acid sequence similar to isoform II [8,9,24] and also has the ability to catabolize oxidized phospholipids [11,21,22]. Overstimulation of glutamate receptors induces cell death by reactive oxygen intermediates that result from impairment of mitochondrial function [6,26]. The present results suggest that like tissue-type isoform II, plasma-type PAFAH can protect neurons from apoptosis by reducing oxidative stress. Mild glutamate excitotoxic insults lead to transient mitochondrial depolarization and apoptosis, while intense insults produce irreversible mitochondrial depolarization and necrosis [1]. According to the MTT assay, impairment

of mitochondrial function was relatively mild after 1 mM glutamate exposure in the present study, in accordance with our previous observation that neuronal death in culture was mainly apoptosis 24 h after exposure to 1 mM glutamate [10]. With Hoechst 33258 staining of neurons 24 h after exposure of 1 mM glutamate for 60 min, apoptotic neurons with fragmented nuclei were widespread in cultures infected with Ad1W or with no infection (control), while neurons infected with AdPAFAH did not show marked apoptotic features (Figs. 4 and 5). Thus, like the tissue type, 40-kDa PAFAH monomer, plasma-type PAFAH has an antioxidant action that protects cells from death due to oxidative stress.

References [1] M. Ankarcrona, J.M. Dypbukt, E. Bonfoco, B. Zhivotovsky, S. Orrenius, S.A. Lipton, P. Nicotera, Glutamate-induced neuronal death: a succession of necrosis or apoptosis depending on mitochondrial function, Neuron 15 (1995) 961–973.

132

Y. Hirashima et al. / Brain Research 885 (2000) 128 – 132

[2] K. Asano, S. Okamoto, K. Fukunaga, T. Shiomi, T. Mori, M. Iwata, Y. Ikeda, K. Yamaguchi, Cellular source(s) of platelet-activatingfactor acetylhydrolase activity in plasma, Biochem. Biophys. Res. Commun. 261 (1999) 511–514. [3] R. Brett, M.G. Rumsby, Evidence of free radical damage in the central nervous system of guinea pigs at the prolonged acute and early relapse stages of chronic relapsing allergic encephalomyelitis, Neurochem. Int. 23 (1993) 35–44. [4] M.R. Elstad, D.M. Stafforini, T.M. Mcintyre, S.M. Prescott, G.A. Zimmerman, Platelet-activating factor acetylhydrolase increases during macrophage differentiation, J. Biol. Chem. 264 (1989) 8467– 8470. [5] Z.S. Derewenda, Y.S. Ho, PAF-acetylhydrolases, Biochim. Biophys. Acta 1441 (1999) 229–236. [6] L.L. Dugan, S.L. Sensi, L.M. Canzoniero, S.D. Handran, S.M. Rothman, T.-S. Lin, M.P. Coldberg, D.W. Choi, Mitochondrial production of reactive oxygen species in cortical neurons following exposure to N-methyl-D-aspartate, J. Neurosci. 15 (1995) 6377– 6388. [7] M. Hattori, H. Arai, K. Inoue, Purification and characterization of bovine brain platelet-activating factor acetylhydrolase, J. Biol. Chem. 268 (1993) 18748–18753. [8] K. Hattori, M. Hattori, H. Adachi, M. Tsujimoto, H. Arai, K. Inoue, Purification and characterization of platelet-activating factor acetylhydrolase II from bovine liver cytosol, J. Biol. Chem. 270 (1995) 22308–22313. [9] K. Hattori, H. Adachi, A. Matsuzawa, K. Yamamoto, M. Tsujimoto, J. Aoki, M. Hattori, H. Arai, K. Inoue, cDNA cloning and expression of intracellular platelet-activating factor (PAF) acetylhydrolase II. Its homology with plasma PAF acetylhydrolase, J. Biol. Chem. 271 (1996) 33032–33038. [10] Y. Hirashima, M. Kurimoto, K. Nogami, S. Endo, M. Saitoh, O. Ohtani, T. Nagata, A. Muraguchi, A. Takaku, Correlation of glutamate-induced apoptosis with caspase activities in cultured rat cerebral cortical neurons, Brain Res. 849 (1999) 109–118. [11] K. Karasawa, M. Yato, M. Setaka, S. Nojima, Purification and characterization of platelet-activating factor acetylhydrolase from peritoneal fluid obtained from guinea pigs after endotoxin shock, J. Biochem. (Tokyo) 116 (1994) 374–379. [12] K. Karasawa, O. Kuge, K. Kawasaki, M. Nishijima, Y. Nakano, M. Tomita, K. Yokoyama, M. Setaka, S. Nojima, Cloning, expression and characterization of plasma platelet-activating factoracetylhydrolase from guinea pig, J. Biochem. (Tokyo) 120 (1996) 838–844. [13] H.-S. Lee, A.S. Csallany, Measurement of free and bound malondialdehyde in vitamin E-deficient and -supplemented rat liver tissues, Lipid 22 (1987) 104–107. [14] J.-J. Li, H. Ueno, Y. Pan, H. Tomita, H. Yamamoto, Y. Kanegae, I.

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

Saito, A. Takeshita, Percutaneous transluminal gene transfer into canine myocardium in vivo by replication-defective adenovirus, Cardiovasc. Res. 30 (1995) 97–105. A. Matsuzawa, K. Hattori, J. Aoki, H. Arai, K. Inoue, Protection against oxidative stress-induced cell death by intracellular plateletactivating factor-acetylhydrolase II, J. Biol. Chem. 272 (1997) 32315–32320. D.A. Musser, A.R. Oseroff, The use of tetrazolium salts to determine sites of damage to the mitochondrial electron transport chain in intact cells following in vitro photodynamic therapy with photofrin II, Photochem. Photobiol. 59 (1994) 621–626. H. Niwa, K. Yamamura, J. Miyazaki, Efficient selection for highexpression transfectants with a novel eukaryotic vector, Gene 108 (1991) 193–200. K. Nogami, Y. Hirashima, S. Endo, A. Takaku, Involvement of platelet-activating factor (PAF) in glutamate neurotoxicity in rat neuronal cultures, Brain Res. 754 (1997) 72–78. T. Ohmori, Y. Hirashima, M. Kurimoto, S. Endo, A. Takaku, In vitro hypoxia of cortical and hippocampal CA1 neurons: glutamate, nitric oxide, and platelet activating factor participate in the mechanism of selective neuronal death in CA1 neurons, Brain Res. 743 (1996) 109–115. D.M. Stafforini, S.M. Prescott, T.M. McIntyre, Human plasma platelet-activating factor acetylhydrolase: Purification and characterization, J. Biol. Chem. 262 (1987) 4223–4230. D.M. Stafforini, E.N. Rollins, S.M. Prescott, T.M. McIntyre, The platelet-activating factor acetylhydrolase from human erythrocytes Purification and properties, Biol. Chem. 268 (1993) 3857–3865. K.E. Stremler, D.M. Stafforini, S.M. Prescott, T.M. McIntyre, Human plasma platelet-activating factor acetylhydrolase. oxidatively fragmented phospholipids as substrates, J. Biol. Chem. 266 (1991) 11095–11103. E.B. Tarbet, D.M. Stafforini, M.R. Elstad, G.A. Zimmerman, T.M. McIntyre, S.M. Prescott, Liver cells secrete the plasma form of platelet-activating factor acetylhydrolase, J. Biol. Chem. 266 (1991) 16667–16673. L.W. Tjoelker, C. Wilder, C. Eberhardt, D.M. Stafforini, G. Dietsch, B. Schimpf, S. Hooper, H.L. Trong, S. Cousens, G.A. Zimmerman, Y. Yamada, T.M. McIntyre, S.M. Prescott, P.W. Gray, Anti-inflammatory properties of a platelet-activating factor acetylhydrolase, Nature 374 (1995) 549–553. H. Ueno, J.-J. Li, H. Tomita, H. Yamamoto, Y. Pan, Y. Kanegae, I. Saito, A. Takeshita, Quantitative analysis of repeat adenovirusmediated gene transfer into injured canine femoral arteries, Arterioscler. Thromb. Vasc. Biol. 15 (1995) 2246–2253. R.J. White, I.J. Reynolds, Mitochondrial depolarization in glutamatestimulated neurons: an early signal specific to excitotoxic exposure, J. Neurosci. 16 (1996) 5688–5697.