.loirrnal o/ Nrurochemislry Raven Press, Ltd.. New York Q 1993 International Society for Neurochemistrq

Rapid Communication

Aurintricarboxylic Acid Protects Hippocampal Neurons from NMDA- and Ischemia-Induced Toxicity In Vivo

Jill M. Roberts-Lewis, Val R. Marcy, Yonghua Zhao, Jeffi-y L. Vaught, Robert Siman, and Michael E. Lewis

Cephulon, Inc., West Chester, Pennsylvania, U.S.A.

Abstract: The polymeric dye aurintricarboxylic acid (ATA) has been shown to protect various cell types from apoptotic cell death, reportedly through inhibition of a calcium-dependent endo- nuclease activity. Recent studies have indicated that there may be some commonalities among apoptosis, programmed cell death, and certain other forms of neuronal death. To begin to explore the possibility of common biochemical mechanisms un- derlying ischemia- or excitotoxin-induced neuronal death and apoptosis in vivo, gerbils or rats subjected to transient global isch- emia or NMDA microinjection, respectively, received a simulta- neous intracerebral infusion of ATA or vehicle. As a biochemical marker of neuronal death, spectrin proteolysis, which is mediated by activation of calpain I , was measured in hippocampus after 24 h. ATA treatment resulted in a profound reduction of both NMDA- and ischemia-induced spectrin proteolysis, consistent with the possibility of some common mechanism in apoptosis and other forms of neuronal death in vivo. Key Words: Aurintricarboxylic acid-NMDA-Ischemia-Apoptosis-Spectrin-Calpain I. J. Neurochem. 61,378-381 (1993).

Excessive stimulation of excitatory amino acid receptors can result in a precipitous rise in intracellular calcium levels and subsequent (delayed) neuronal death in vulnerable re- gions of the CNS (Siesjo, 1988; Choi and Rothman, 1990). Excitatory amino acid-induced perturbations in calcium ho- meostasis have clearly been implicated in neuronal death related to ischemia and may be involved in other neurode- generative diseases as well (Seubert and Lynch, 1990; Si- man, 1990; Farooqui and Horrocks, 199 1 ). This “excito- toxin”-induced degeneration may be mediated in part by the activation of some or many calcium-dependent en- zymes, such as proteases, protein kinases, phospholipases, and endonucleases (Bazan, 1989; Manev et al., 1990; Seu- bert and Lynch, 1990; Farooqui and Horrocks, 199 1).

Programmed cell death and apoptosis are believed to be fundamental embryological mechanisms defining both body shape and the immune system (Lockshin and Zakeri, 199 1; Tomei, 199 1). Typically, programmed cell death is an active process, requiring gene expression and protein synthe- sis to be fulfilled (Lockshin and Zakeri, 199 1). Apoptosis is characterized by internucleosomal DNA cleavage and mor- phological alterations that are distinct from those involved in necrotic cell death (Kerr and Harmon, 199 1). In the ner- vous system, programmed cell death and apoptosis regulate the size and distribution of a neuronal population during development (Oppenheim, 1989). Depending on the state of differentiation of a cell, apoptosis might also occur

in a transcription- and translation-independent manner (Mesner et al., 1992). Thus, it is possible that apoptotic-like cell death may persist, or be triggered, in the mature nervous system. For example, infusion of glutamate into the hippo- campus of adult rats has been reported to elicit DNA frag- mentation (Kure et al., 199 1). Moreover, the protein synthe- sis and endonuclease inhibitor aurintricarboxylic acid (ATA) prevents cell death induced by nerve growth factor deprivation in pheochromocytoma (PC12) cells grown in serum-free medium, as well as in cultured sympathetic neu- rons (Batistatou and Greene, 1991; Mesner et al., 1992).

Both apoptosis and ischemia- or excitotoxin-induced neu- ronal death appear to involve calcium-dependent mecha- nisms and activation of early immediate genes (Nowak, 1985; Buttyan et al., 1988; Siesjo, 1988; Jorgensen et al., 1989; McConkey et al., 1989). Moreover, both forms ofcell death have been shown in some studies to be blocked with protein synthesis inhibitors (Martin et al., 1988; Shigeno et al., 1990; but also see Leppin et al., 1992). In the present study, we sought to explore further the possibility that isch- emia- or excitotoxin-induced neuronal death in the mature nervous system might share mechanisms or specific intra- cellular targets in common with apoptosis.

MATERIALS AND METHODS

NMDA-induced hippocampal damage Twenty female Sprague-Dawley rats (Hilltop Lab Ani-

mals, Scottdale, PA, U.S.A.), weighing 200-250 g, were anesthetized with pentobarbital sodium (Nembutal; 50 mg/ kg i.p.) and given 7 p g of ATA simultaneously with 3 p g of NMDA in a total of 1 pl of saline into the right dorsal hippo- campus at stereotaxic coordinates -3.3 mm posterior from bregma, lateral 2.3 mm from the midline, and ventral 4.3 mm from the top of the skull. A control infusion of NMDA without ATA was made into the left hippocampus, and sham animals received an infusion of saline instead of NMDA or ATA at the same stereotaxic coordinates. Body

Resubmitted manuscript received April 8, 1993; accepted April 12, 1993.

Address correspondence and reprint requests to Dr. J . M. Rob- erts-Lewis at Cephalon, Inc., 145 Brandywine Parkway, West Ches- ter, PA 19382, U.S.A.

Abbreviations used: ATA, aurintricarboxylic acid; BDP, break- down product; i.c.v., intracerebroventricular.

3 78

ATA PROTECTS HIPPOCAMPAL NEURONS IN VIVO 3 79

temperature was monitored with a rectal probe and main- tained within normal physiological limits by the use of a heat lamp during surgery and an incubator during the post- operative recovery period. The rats were anesthetized and killed either at 24 h for biochemical analysis (n = 10) or at 1 week for histological analysis (n = 10). This procedure re- sults in a significant, maximal increase in hippocampal (CA 1) spectrin degradation by 24 h and extensive degenera- tion of pyramidal cells by 4 days following NMDA infusion. All animal use procedures were approved by the Institu- tional Animal Care and Use Committee and conducted in strict compliance with the NIH Guide for the Care and Use of Laboratory Animals.

Ischemia-induced hippocampal damage Twelve male Mongolian gerbils (Tumblebrook Farms,

West Brookfield, MA, U.S.A.), weighing 50-80 g, were anes- thetized with isoflurane (3%, carried with oxygen, 2 L/min) and given 4 pg of ATA in 1 pl of saline, or saline alone, into the right lateral ventricle of the brain [intracerebroventricu- lar (i.c.v.)] at -0.5 mm posterior, 0.5 mm lateral to bregma, and 2.5 mm ventral from the top of the skull. The intraven- tricular location of the injection bolus was validated with dye injections. Ten minutes following i.c.v. drug or vehicle infusion, while still under anesthesia, each animal was sub- jected to transient global forebrain ischemia. Suture thread was tied around the common carotid arteries to achieve total occlusion of blood flow for a period of 5 min. At the end of the ischemic period, the suture threads were re- moved, and the incision was closed with wound clips after it was ascertained that blood flow was completely restored to both carotid arteries. Body temperature was monitored with a rectal probe and maintained within normal physiological limits by the use of a heat lamp throughout the surgery. The gerbils were anesthetized and killed 24 h later and evaluated for spectrin degradation. This procedure results in a signifi- cant increase in hippocampal (CAI) spectrin degradation by 24 h and virtually total deterioration of the CA 1 subfield of the hippocampus by 4 days following the ischemic insult.

Biochemical analysis Spectrin proteolysis has previously been established as a

relevant biochemical marker ofexcitotoxin- or ischemia-in- duced neuronal degeneration (Siman et al., 1989; Lee et al., 199 1). Breakdown of spectrin by the calcium-activated neu- tral cysteine protease calpain I produces four molecular mass fragments of 140-150 kDa that react with antibodies to the intact a-subunit of spectrin (-240 kDa) (Siman and Noszek, 1988). Spectrin proteolysis was evaluated in homog- enates of the hippocampus using an immunoblot analysis (Siman et al., 1989). In brief, the dorsal hippocampus was dissected from each brain and homogenized in 20 mMTris- HCI (pH 7.4) containing 1 mM EDTA, 1 mM EGTA, and 0.1 mM phenylmethlysulfonyl fluoride. Proteins from ali- quots of each homogenate were separated by sodium do- decyl sulfate-polyacrylamide gel electrophoresis, and poly- clonal antibodies against a-spectrin were used to quantify NMDA- or ischemia-induced spectrin breakdown products (BDPs) present in each hippocampal homogenate by digi- tizing the BDP bands with computerized image analysis.

Histological analysis Rat brains were rapidly removed and frozen on dry ice.

Sixteen coronal sections from each brain were cut on a cryo- stat, thaw-mounted, stained with thionin, and examined mi-

A. 1 2 3 4 1

L R L R L R L R W

L = NMDA R = NMDA + ATA

25U

-7-

NMDA NMDA + ATA

FIG. 1. ATA attenuates NMDA-induced spectrin proteolysis in the hippocampus. A Western blot depicts NMDA-induced spec- trin BDPs in hippocampal homogenates. Each pair of lanes repre- sents the left and right hippocampus from one animal. All animals except shams received intrahippocampal infusions of NMDA into both left and right sides, together with either vehicle (left hippo- campus) or ATA (right hippocampus). Shams (shown in the last lane) received an intrahippocampal infusion of saline, but no NMDA or ATA. B: Densitometric analysis of the western blot in A. Data are mean f SEM (bars) values, in percent increase in level of spectrin BDPs over sham control values for vehicle- (NMDA) or ATA-treated (NMDA + ATA) hippocampi. Sham was defined as 100%. ATA treatment resulted in a significant reduction in amount of hippocampal spectrin BDPs compared with vehicle- treated controls ( p < 0.001 ; 2-tailed paired t test).

croscopically. Damage to the hippocampus was quantified with the imaging system by measuring the area of intact cells in the CA1 subfield of the hippocampus relative to the total area subtended by CAI for each section. The average ratio of intact/total CAI was then calculated for each animal.

Materials Polyclonal antisera against the &-subunit of rat brain

spectrin were prepared as previously described (Siman and Noszek, 1988). All electrophoresis and western blotting re- agents were obtained from Bio-Rad. ATA was obtained from Aldrich, and NMDA from Sigma.

RESULTS AND DISCUSSION Intracerebral treatment of rats with ATA virtually abol-

ished both NMDA- and ischemia-induced hippocampal spectrin proteolysis. The neuroprotective effect of ATA is shown in a western blot (Fig. 1 A) of NMDA-induced spec- trin BDPs. A densitometric analysis of this blot (Fig. 1B) indicated that although NMDA alone produced a 128% in- crease in hippocampal spectrin BDPs above sham control (non-NMDA-treated) values, intrahippocampal coinfusion of ATA with NMDA blocked the NMDA-induced increase in spectrin proteolysis by 73% ( p < 0.001). The ameliorating effect of ATA treatment on NMDA-induced neuronal de-

J. Neurochem.. Vol. 61. No. 1, 1993

380 J. M. ROBERTS-LEWIS ET AL

B

NMDA NMDA + ATA

FIG. 2. ATA attenuates NMDA-induced degeneration of hippo- campal pyramidal cells. A: Coronal section from a rat brain 1 week after an intrahippocampal infusion of NMDA alone (left side) or NMDA with ATA (right side). Arrows define the extent of pyra- midal cell degeneration on the nontreated (left) side. B: Morpho- metric analysis of the hippocampal degeneration shown in A. The area of intact pyramidal cells in the CAI subfield was determined as a percentage of the total CAI sector for each section. Histo- grams depict overall means f SEM (bars) values for each side (left or right). ATA treatment virtually abolished the pyramidal cell degeneration induced by NMDA ( p i 0.001 ; 2-tailed paired t test).

generation was also demonstrated using a histological analy- sis of the hippocampus at 1 week following NMDA and ATA treatments (Fig. 2). In a Nissl-stained coronal section through the hippocampus of a representative animal, exten- sive necrosis of the CA 1 pyramidal cell layer can be seen on the vehicle-injected side (NMDA only), whereas the ATA- treated side shows significant sparing of pyramidal cells in the same region ofthe hippocampus (Fig. 2A). The extent of sparing of pyramidal cells in the CA 1 subfield of the hippo- campus was nearly 100% overall (Fig. 2B) compared with 50% in vehicle-treated control animals ( p < 0.001).

A similar effect of ATA was also seen in the hippocampus 24 h following transient global ischemia. The difference in magnitude of ischemia-induced spectrin BDPs between ATA- and vehicle-treated animals is shown using a western blot analysis (Fig. 3A). The amount of ischemia-induced BDPs was reduced by 93% in hippocampal tissue from ani- mals that received an i.c.v. infusion of ATA immediately before a 5-min ischemic insult, compared with vehicle- treated controls (Fig. 3B; p < 0.00 1).

The mechanism by which ATA reduces the effects of these injuries, both of which appear to involve NMDA re- ceptor-mediated elevation of intracellular calcium concen- trations (Siesjo, 1988; Choi and Rothman, 1990), could be due to one or a combination of many of the known biologi- cal activities of this compound. For example, ATA, which

exists as a complex, heterogeneous mixture of polymers, inhibits protein-nucleic acid complex formation in cell-free systems through competition between the nucleic acid and polymeric ATA for binding to the active site (Gonzalez et al., 1980). Thus, these interactions may result in the inhibi- tion of transcription and translation, as well as reducing the calcium-dependent endonuclease activity that appears to be a key factor in initiating apoptotic cell death (Wyllie et al., 1980). Protein synthesis inhibitors have been reported to prevent ischemia-induced neuronal degeneration (Shigeno et al., 1990; but also see Leppin et al., 1992); therefore, blockade of translation by ATA could be related to its neuro- protective activity. Alternatively, it has been reported that excitotoxins can stimulate hippocampal DNA fragmenta- tion in vivo (Kure et al., 1991; but also see Ignatowicz et al., 1991), so inhibition of an endonuclease activity could also be an important mechanism of action for ATA in these models of neuronal death. However, there are also potential extracellular mechanisms of ATA that have not yet been explored, such as possible activity as an antioxidant, ion chelator, calcium channel antagonist, or glutamate receptor antagonist. We have found that ATA was also effective in preventing hippocampal damage following kainic acid infu- sion (authors' unpublished data), thus demonstrating that the compound is effective against activation of at least two glutamate receptor subtypes. Samples and Dubinsky (1993) have demonstrated that treatment of hippocampal neurons with ATA in vitro abolished glutamate excitotoxicity, con- sistent with the findings reported here.

The higher-molecular-weight polymeric form of ATA may account for some or all ofthese pharmacological activi- ties. For example, higher-molecular-weight fractions of

B.

Il" i T

Vehicle ATA

croup

FIG. 3. ATA attenuates ischemia-induced spectrin proteolysis in the hippocampus. A Western blot depicts ischemia-induced spectrin BDPs in hippocampal homogenates. Each lane repre- sents the hippocampus from one animal, treated with either vehi- cle or ATA immediately before ischemia. 8: Densitometric analy- sis of the western blot shown in A. Data are mean f SEM (bars) values, in percent increase in level of spectrin BDPs over sham control values for hippocampal tissue from vehicle (n = 6) or ATA- treated (n = 6) animals. Sham was defined as 100%. ATA treat- ment resulted in a significant decrease in the magnitude of isch- emia-induced spectrin proteolysis compared with vehicle-treated controls ( p < 0.001 ; 2-tailed; unpaired t test).

J. Neurochem., Val. 61, No. 1. 1993

ATA PROTECTS HIPPOCAMPAL NEURONS IN VIVO 381

ATA have been correlated with increased potency in pre- venting protein-nucleic acid interactions (Gonzalez et al., 1979), and the anti-HIV activity of different-molecular- weight species of ATA also increases with the molecular weight (Cushman et al., 199 I). Consistent with this observa- tion, we have found that a high-molecular-weight dialysis fraction of ATA (M, 2 3,500) resulted in equal or better neuroprotection against NMDA-induced toxicity in vivo compared with the full polymeric mixture (authors’ unpub- lished data). It remains to be determined which molecular weight species or pharmacological properties of ATA might be specifically involved in its neuroprotective activity.

Arguably, ischemia-induced neuronal degeneration does not show the classic morphological and ultrastructural changes that are uniquely characteristic of programmed cell death or apoptosis (see Deshpande et al., 1992). However, it is intriguing that apoptosis and ischemia- or excitotoxin-in- duced neuronal death apparently share in common a n “ATA-sensitive” mechanism, suggesting that similar bio- chemical mechanisms may be operational in both forms of cellular degeneration.

Acknowledgment: We would like to thank Dr. Mo- hammed Iqbal for preparation of the high-molecular-weight dialysis fraction of ATA, Dr. James C. Kauer for his expert opinion concerning the possible biological targets of ATA, and Dr. Janet Dubinsky for many insightful discussions and a critical review of the manuscript.

REFERENCES Batistatou A. and Greene L. A. (199 1) Aurintricarboxylic acid res-

cues PC12 cells and sympathetic neurons from cell death caused by nerve growth factor deprivation: correlation with supression of endonuclease activity. J. Cell Biol. 11546 1-47 1.

Bazan N. G. (1989) Arachidonic acid in the modulation ofexcitable membrane function and at the onset ofbrain damage. Ann. N Y Acad. Sci. 559, 1-16.

Buttyan R., Zakeri Z., Lockshin R. A., and Wolgemuth D. (1988) Cascade induction of c-fos, c-myc, and heat shock 70 K tran- scripts during regression of the rat ventral prostrate gland. Mol. Endocrinol. 2,650-657.

Choi D. W. and Rothman S. M. ( 1990) The role ofglutamate neuro- toxicity in hypoxic-ischemic neuronal death. Annu. Rev. Neu- rosci. 13, 171-182.

Cushman M., Wang P., Chang S. H., Wild C., De Clercq E., Schols D., Goldman M. E., and Bowen J. A. (199 1) Preparation and anti-HIV activities of aurintricarboxylic acid fractions and ana- logues: direct correlation of antiviral potency with molecular weight. J. Med. Chem. 34, 329-337.

Deshpande J., Bergstedt K., Linden T., Kalimo H., and Wieloch T. ( I 992) Ultrastructural changes in the hippocampal CAI region following transient cerebral ischemia: evidence against pro- grammed cell death. Exp. Brain Res. 88, 9 1 - 105.

Farooqui A. A. and Horrocks L. A. (1991) Excitatory amino acid receptors, neural membrane phospholipid metabolism and neurological disorders. Brain Res. Rev. 16, 171-191.

Gonzalez R. G., Blackburn B. J., and Schleich T. (1979) Fraction- ation and structural elucidation of the active components of aurintricarboxylic acid, a potent inhibitor of protein nucleic acid interactions. Biochim. Biophys. Acta 562, 534-545.

Gonzalez R. G., Haxo R. S., and Schleich T. (1980) Mechanism of action of the polymeric aunntricarboxylic acid, a potent inhibi- tor of protein-nucleic acid interactions. Biochemistry 19,

Ignatowicz E., Vessani A.-M., Rizzi M., and DIncalci M. (1991) Nerve cell death induced in vivo by kainic acid and quinolinic acid does not involve apoptosis. Neuroreport 2, 65 1-654.

4299-4303.

Jorgensen M. B., Deckert J., Wright D. C., and Gehlert D. R. ( I 989) Delayed c-fos proto-oncogene expression in the rat hippocam- pus induced by transient global cerebral ischemia: an in situ hybridization study. Brain Res. 484, 393-398.

Kerr J. F. R. and Harmon B. V. (199 1) Definition and incidence of apoptosis: an historical perspective, in Apoptosis. The Molecu- lar Basis ofcell Death (Tomei L. D. and Cope F. O., eds), pp. 5-29. Cold Spring Harbor Laboratory Press, Cold Spring Har- bor, New York.

Kure S., Tominaga T., Yoshimoto T., Tada K., and Narisawa K. ( 199 1 ) Glutamate triggers internucleosomal DNA cleavage in neuronal cells. Biochem. Biophys. Res. Commun. 179,39-45.

Lee K. S., Frank S., Vanderklish P., Arai A,, and Lynch G. (1991) Inhibition of spectrin proteolysis protects hippocampal neu- rons from ischemia. Proc. Natl. Acad. Sci. USA 88, 7233- 7237.

Leppin C., Finiels-Marlier F., Crawley J. N., Montpied P., and Paul S. M. (1 992) Failure of a protein synthesis inhibitor to modify glutamate receptor-mediated neurotoxicity in vivo. Brain Res.

Lockshin R. A. and Zakeri 2. ( I 99 I ) Programmed cell death and apoptosis, in Apoptosis. The Molecular Basis of Cell Death (Tomei L. D. and Cope F. O., eds), pp. 47-60. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

Manev H., Costa E., Wroblewski J. T., and Guidotti A. (1990) Abusive stimulation of excitatory amino acid receptors: a strat- egy to limit neurotoxicity. FASEB J. 4, 2789-2797.

Martin D. P., Schmidt R. E., Di Stephano P. S., Lowry 0. H., Carter J. G., and Johnson E. M. (1988) Inhibitors of protein synthesis and RNA synthesis prevent neuronal death caused by nerve growth factor deprivation. J. Cell Biol. 106,829-844.

McConkey D. J., Nicotera P., Hartzell P., Bellomo G., Wylie A. H., and Orrenius S. (1989) Glucocorticoids activate a suicide pro- cess in thymocytes through an elevation of cytosolic Ca2+ con- centration. Arch. Biochem. Biophys. 269, 365-370.

Mesner P. W., Winters T. R., and Green S. H. ( 1 992) Nerve growth factor withdrawal-induced cell death in neuronal PC12 cells resembles that in sympathetic neurons. J. Cell Biol. 119, 1669- 1680.

Nowak T. S. Jr. (1985) Synthesis of a stress protein following tran- sient ischemia in the gerbil. J. Neurochem. 45, 1635-1641.

Oppenheim R. W. (1989) The neurotrophic theory and naturally occumng motoneuron death. Trends Neurosci. 12,252-255.

Samples S. D. and Dubinsky J. M. (1993) Aurintricarboxylic acid protects hippocampal neurons from glutamate toxicity in vi- tro. J. Neurochem. 61, 382-385.

Seubert P. and Lynch G. ( I 990) Plasticity to pathology: brain cal- pains as modifiers of synaptic stucture, in Intracellular Cal- cium-Dependent Proteolysis (Mellgren R. L. and Murachi T., eds), pp. 251-263. CRC Press, Boca Raton, Florida.

Shigeno T., Yamasaki Y., Kato G., Kusaka K., Mima T., Takakura K., Graham D. I., and Furukawa S. (1990) Reduction of de- layed neuronal death by inhibition of protein synthesis. Neuro- sci. Lett. 120, 1 17-1 19.

Siesjo B. K. (1988) Historical overview. Calcium, ischemia, and death of brain cells. Ann. NY Acad. Sci. 522, 638-66 1.

Siman R. (1990) Role of calpain I in excitatory amino acid-induced degenerative structural changes, in Neurotoxicity of Excitatory Amino Acids (Guidotti A., ed), pp. 145-16 1. Raven Press, New York.

Siman R. and Noszek J. C. (1988) Excitatory amino acids activate calpain I and induce structural protein breakdown in vivo. Neuron 1,279-287.

Siman R., Noszek J. C., and Kegerise C. (1989) Calpain I activation is specifically related to excitatory amino acid induction of hippocampal damage. J. Neurosci. 9, 1579- 1590.

Tomei L. D. ( 199 1 ) Apoptosis: a program for death or survival?, in Apoptosis. The Molecular Basis of Cell Death (Tomei L. D. and Cope F. O., eds), pp. 279-3 16. Cold Spring Harbor Labora- tory Press, Cold Spring Harbor, New York.

Wyllie A. H., Kerr J. F. R., and Cume A. R. (1980) Cell death the significance of apoptosis. Int. Rev. Cytof. 68, 25 1-306.

581, 168-170.

J. Neurochem., Vol. 61, No. I , 1993