- Email: [email protected]
An adenoviral vector expressing the glucose transporter protects cultured striatal neurons from 3-nitropropionic acid
Brain Research 859 Ž2000. 21–25 www.elsevier.comrlocaterbres
Research report
An adenoviral vector expressing the glucose transporter protects cultured striatal neurons from 3-nitropropionic acid Sheri L. Fink, Dora Y. Ho, John McLaughlin, Robert M. Sapolsky
)
Department of Biological Sciences, Stanford UniÕersity, Stanford, CA 95406, USA Accepted 23 November 1999
Abstract Considerable interest has focused on the possibility of using gene transfer techniques to introduce protective genes into neurons around the time of necrotic insults. We have previously used herpes simplex virus amplicon vectors to overexpress the rat brain glucose transporter, Glut-1 ŽGT., and have shown it to protect against a variety of necrotic insults both in vitro and in vivo, as well as to buffer neurons from the steps thought to mediate necrotic injury. It is critical to show the specificity of the effects of any such transgene overexpression, in order to show that protection arises from the transgene delivered, rather than from the vector delivery system itself. As such, we tested the protective potential of GT overexpression driven, in this case, by an adenoviral vector, against a novel insult, namely exposure of primary striatal cultures to the metabolic poison, 3-nitropropionic acid Ž3NP.. We observed that GT overexpression buffered neurons from neurotoxicity induced by 3NP. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Adenovirus; Neuron death; 3-Nitropropionic acid; Striatum; Gene therapy; Glucose transport; Neuronal energetics
1. Introduction There is now considerable understanding of the mechanisms mediating necrotic neuronal injury, in particular the cascade of excessive synaptic glutamate and free cytosolic calcium w30x. With this understanding comes the potential for therapeutic intervention. Particularly exciting is the possibility of gene transfer techniques, typically involving the use of viral vectors w14,21,23,38x. Reports of protection against models of necrotic insults using viral vectors involve targeting the calcium excess Žwith overexpression of the calcium binding protein, calbindin D28K. w24,36,37x, the protein malfolding Žwith hsp72. w13,45x, the ROS production Žwith superoxide dismutase. w24x, the apoptotic elements Žwith apoptosis inhibitors such as Bcl-2 or NAIP. w22,27,28,32,44x, or the inflammatory elements w8,17x. In addition to these approaches, we have studied the neuroprotective potential of overexpression of the rat brain glucose transporter, Glut-1 ŽGT.. The rationale for the construction of such a vector was the evidence that energy availability influences the efficacy with which an endangered neuron can contain glutamate and calcium, and thus )
Corresponding author. Fax: q1-650-725-5356; e-mail: [email protected]
the extent of toxicity caused by necrotic insults w4,41x. We have observed that overexpression of GT enhances glucose transport in both monolayer cultures and in vivo w19x. Such overexpression at the time of an insult helps maintain metabolism, and decreases glutamate release, cytosolic calcium concentrations, and oxygen radical accumulation w16,25x. Moreover, GT overexpression decreases neuron loss following hypoglycemia, glutamate administration, or antimetabolite exposure in cultured hippocampal, spinal cord, and septal neurons as well as following kainic-acidinduced seizure, focal ischemia, and exposure to antimetabolites in vivo w9,19,20,25,26x. These studies with glucose transporter overexpression have all used a herpes amplicon as a vector for delivery. An important point in the nascent discipline of neuronal gene therapy is to show the specificity of the effects of any such transgene overexpression, i.e., that it is the gene delivered which is efficacious, rather than the specifics of the particular delivery system. As such, in this paper, we explore the versatility of the neuroprotective potential of GT overexpression with two modifications: Ža. overexpression is driven, in this case, by an adenoviral vector, rather than a herpes amplicon vector; Žb. we test the protective effects of this vector against a unique insult, namely damage to striatal neurons induced by 3-nitropropionic
0006-8993r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 2 4 0 1 - 4
22
S.L. Fink et al.r Brain Research 859 (2000) 21–25
acid Ž3NP., a mitochondrial poison that irreversibly inhibits succinate dehydrogenase w2x.
2. Materials and methods 2.1. Cell cultures Mixed neuronalrglial cultures were prepared from the striatum of fetal Sprague–Dawley rats ŽHarlan, Indianapolis, IN. on day 18 of gestation. The experimental protocol concerning the use of rats was approved by the Division of Laboratory Animal Medicine at Stanford University. The method of preparing the cultures has been described w12x. Briefly, striatum was dissected from fetuses and dissociated with papain Ž10 unitsrml; Worthington Biochemical, Freehold, NJ.. The cells were then spun down, resuspended in MEM-PAK Ža modified minimal essential medium from the cell culture facility of the University of California, San Francisco, CA. supplemented with 10% horse serum ŽHyclone Laboratories, Logan, UT., and filtered through a cell strainer Ž70 mM mesh, Becton Dickinson, Lincoln Park, NJ.. The cells were then plated in wells of tissue culture plates at an approximate density of 1.6– 1.9 = 10 5 cellsrcm2 , and maintained in MEM-PAK supplemented with 10% Žvrv. horse serum. Cultures were used at 10–11 days of age, at which time approximately 50–70% of cells was non-neuronal w20,25x. A total of 293 cells ŽATCC CRL 1573. are grown in MEM ŽGibco BRL; cat. no. 61100-061. supplemented with 10% bovine newborn serum ŽGibco BRL.. All cells were maintained in a 5% CO 2r95% O 2 atmosphere at 378C in a humidified incubator. 2.2. Construction and preparation of adenoÕirus Õectors, AdRSVb gal and AdRSVGT The recombinant adenoviruses, AdRSVbgal and AdRSVGT, were constructed according to the method of Graham and Prevec w15x. First, two shuttle vectors, pDIE1RSV-GT and pDIE1RSV-bgal, were constructed ŽFig. 1.. To create pDIE1RSV-GT, a BamHI–BglII fragment from pON820 ŽViera and Mocarski, unpublished. containing the Rous sarcoma virus ŽRSV. promoter is inserted into the BamHI site of the shuttle vector, pDE1sp1A w6x. The resulting vector is named pDIE1RSV. Then, an XbaI fragment containing the glucose transporter gene and the polyadenylation signal from HCMV IE1 was excised from pa4GTY aY w25x and cloned into the XbaI site of pDIE1RSV. The resulting construct is pDIE1RSV-GT. To create pDIE1RSV-bgal, the BamHI–BglII fragment containing the RSV promoter was cloned into the BamHI site of pON1 w40x, 5X to the Escherichia coli lacZ coding sequence. A BamHI fragment from the resulting construct Žcontaining the RSV promoter, the lacZ gene and the
Fig. 1. Generation of AdRSVGT. The two parent plasmids used for the generation of AdRSVGT are pDIE1RSV-GT and pJM17. In pDIE1RSVGT, the glucose transporter gene gt Žwhite arrow. under the control of the RSV promoter Ždotted arrow. is inserted into the E1 region of Ad5 sequences Žsolid black bars.. The flanking Ad5 homologies on either side of the transgene represent 0–0.1 map unit Žmu. and 9.8–16.1 mu, respectively. pJM17 w34x contains the entire sequence of Ad5 interrupted by the plasmid, pBRX Žstriped bar., which encodes ampicillin and tetracycline resistance. The junction of the two ends of Ad5 genome is indicated as 100r0 mu. pJM17 is non-packagable since it has exceeded the packaging limit of adenovirus. Co-transfection of pDIE1RSV-GT and pJM17 into 293 cells resulted in homologous recombination between the two plasmids and the generation of packagable AdRSVGT. AdRSVbgal is similarly generated using plasmids, pDIE1RSV-bgal amd pJM17. In pDIE1REV-bgal, the gt gene is replaced by the lacZ gene.
SV40 polyadenylation sequences. was then inserted into the BamHI site of pDE1sp1A to create pDIE1RSV-bgal. To generate the recombinant viruses, AdRSVGT and AdRSVbgal, the two shuttle vectors, pDIE1RSV-GT and pDIE1RSV-bgal ŽFig. 1., were transfected into 293 cells along with pJM17, a bacterial plasmid comprised of the circular form of Ad5 genome in an unpackagable form w34x. Upon recombination of the shuttle vectors with pJM17, infectious virus are generated. Viral plaques resulting from the co-transfection were plaque-purified three times and amplified according to the protocol of Graham and Prevec w15x. Preparation of recombinant adenovirus stocks was based on a protocol obtained from Dr. Frank Graham’s laboratory with modifications. Briefly, infected 293 cells Žgrown on 150 mm dish. were harvested when 100% cytopathic effect was reached. Cells from 30–40 dishes were pelleted and resuspended in 5 ml of 0.1 M Tris ŽpH 8.1.. Sodium deoxycholate, MgCl 2 and DNAse I were then added to a final concentration of 0.5%, 20 mM and 50 mgrml, respectively. Then 1.8 ml of saturated CsCl solution Žin 10 mM Tris pH 8.1r1 mM EDTA. was added to each 3.1 ml of cell lysate. The virus was banded twice by centrifugation in a Ti70.1 rotor ŽBeckman. for 16–20 h at 48C and 35,000 rpm. The final viral band was collected and desalted with a DG-10 Econo-pack column ŽBioRad, Hercules, CA.. The virus was eluted in hepes-buffered saline Ž20 mM hepesr150 mM NaCl, pH 7.8. with 3% sucrose and titrated on 293 cells by standard plaque assay.
S.L. Fink et al.r Brain Research 859 (2000) 21–25
2.3. Toxicity experiments Striatal cultures grown in 48-well tissue culture plates were infected on day 7 in culture with AdRSVGT or AdRSVbgal Ž6.6 = 10 4 PFUrwell.; under these conditions, 20% of neurons and 5% of glia are infected ŽMcLaughlin, in preparation.. At 3 days post-infection, the cultures were challenged with 0, 0.5 or 1.0 mM 3NP in DMEM containing 0.5 or 1.5 mM glucose. The cultures were maintained in this medium for approximately 22 h, fixed, and immunostained with a mouse monoclonal antibody against MAP-2 Ž1:1000 dilution; Sigma. and the Vectastain ABC kit for peroxidase ŽVector Laboratories, Burlingame, CA.. Counts were made across the diameter of each well; only neurons positive for the MAP2 antigen and with intact processes were counted. The neuron counts for each condition were expressed as the percentage of mean control Žsame vector, same glucose concentration, no 3NP. values and averaged unless otherwise noted. Comparisons are by Mann–Whitney t-tests. Data are expressed as mean " S.E.M. values.
3. Results Two adenovirus recombinants, AdRSVGT and AdRSTbgal, were constructed to express GT or bgal, respectively, under the control of the RSV promoter. The E1 region of the viral genome was replaced by the transgene of interest Ždetailed in Fig. 1. and consequently, both recombinants are E1 deletion mutants. In pilot studies, GT
Fig. 2. Effects of AdRSVGT infection on survival of striatal neurons following 3NP andror hypoglycemic exposure. Striatal cultures were infected with AdRSVGT or AdRSVbgal. At 3 days post-infection, the cells were incubated with DMEM containing various concentrations of glucose, with or without 0.5 or 1.0 mM 3NP, for approximately 22 h. The cells were then fixed and immunostained against MAP-2. The numbers of intact neurons were counted and expressed as the percent mean values of results in their respective control groups, which were infected with the same vector, maintained under same glucose concentrations, but without 3NP. Under those control conditions, neuron number in wells treated with AdRSVGT or AdRSVbgal did not differ. U ps 0.007; UU ps 0.0002 by Mann–Whitney t-test.
23
or bgal expression from the recombinants was assessed in primary neuronal cultures using immunofluorescence techniques. Furthermore, bgal expression from AdRSVbgal was assessed at several timepoints by reacting cells with X-gal and was found to be increased from days 1 to 6 post-infection Ždata not shown.. We then tested whether overexpression of a glucose transporter would prove protective against 3NP. In agreement with prior reports under these culture conditions w12x, low glucose conditions Ž0.5 mM glucose, 0 mM 3NP. were not neurotoxic, and the toxicity of 3NP was greatly potentiated under low glucose conditions ŽFig. 2.. We observed that adRSVGT decreased the toxicity of 3NP exposure.
4. Discussion Because of the post-mitotic nature of neurons, those lost to neurological insults cannot be replaced. Thus, the ability to intervene at the time of an insult with a gene therapeutic approach meant to save neurons has considerable clinical promise. The particular therapeutic strategy of overexpressing the glucose transporter is a logical one. Glucose is the primary energy source utilized by the brain w18,39x. Because glucose is not stored in neurons, the capability of a neuron to take up glucose during an insult could crucially determine its fate. Glucose transport into target cells in the brain becomes rate-limiting during some necrotic insults w3,7x, and neurons will upregulate glucose transporter expression post-ischemia as part of the adaptive profile of activating stress proteins w31,33,42x. We observe that overexpression of the glucose transporter gene, Glut-1, driven by an adenoviral vector, can protect cultured striatal neurons from 3NP. The toxin is a mitochondrial poison that irreversibly inhibits succinate dehydrogenase, disrupting oxidative metabolism w2x, and producing necrotic neuron loss both in animals w5x and in cultured cells w12,43x, with a characteristic profile of pyknosis. As a dramatic measure of the energetic basis of the toxicity of 3NP, glucose availability will shift its LD50 by more than an order of magnitude in primary neuronal cultures w12x. Additional glucose uptake would be expected to protect the function of many energy-requiring cellular processes after necrotic insults. In our prior studies overexpressing GT with a herpes amplicon vector, we observed a more prolonged maintenance of metabolism during an insult in hippocampal cultures Žas assessed with silicon microphysiometry w25x., as well as decreases in glutamate release, cytosolic calcium mobilization, and oxygen radical accumulation w16x. Collectively, these effects are likely to result in a decreased likelihood of neuron death. Herpes amplicon vectors have a strong preference for infecting neurons over glia w13,19x, whereas adenoviral vectors infect a significant subset of glia w1,10,29x. While only a small
24
S.L. Fink et al.r Brain Research 859 (2000) 21–25
percentage of glia was infected under the conditions in this study, part of the neuroprotection observed may still have arisen from enhancing salutary and energy-dependent actions of glia, such as the high-affinity removal of glutamate from the synapse. The magnitude of neuroprotection is likely to have been considerable. Fig. 2 indicates that under conditions where 3NP-induced toxicity in control vector-treated wells was virtually complete, treatment with AdRSVGT resulted in survival of approximately 20% of neurons. As noted, under these conditions of infection, approximately 20% of neurons is infected with AdRSVGT, suggesting that a very high percentage of infected neurons is likely to have survived Ža less parsimonious explanation would be that infection of a particular neuron with AdRSVGT would not spare that neuron from 3NP toxicity, but would result in survival of an uninfected neighbor.. Our observation of neuroprotection with this vector extends prior work in which herpes-based GT vectors protected against hypoglycemia, glutamate, 3-acetylpyridine, stroke, and seizure, even when introduced after an insult w9,19,20,25,26x; moreover, hippocampal neurons, when saved by GT overexpression, are able to sustain normal spatial maze learning in rats Ž345, as well as typical synaptic plasticity w11x. The efficacy of both herpes and adenovirus vectors in neuroprotection is further evidence that expression of the GT gene itself, rather than a particular method of overexpression, leads to neuroprotection. This is important, both for the heuristic value of further appreciating the energetic component to necrotic neuronal injury, as well as for designing interventions that may ultimately be of clinical use.
Acknowledgements We thank Frank Graham ŽMcMaster University, Hamilton, ON, Canada. for providing pDE1sp1A and pJM17, Edward Mocarski ŽStanford University, Stanford, CA., for providing pON1 and pON820 and Sheila Brooke, John McLaughlin and Timothy Meier for their excellent technical assistance. Funding was provided by RO1 NS 32848 to R.M.S.
References w1x S. Akli, C. Caillaud, E. Vigne, L.D. Stratford-Perricaudet, L. Poenaru, M. Perricaudet, A. Kahn, M.R. Peschanski, Transfer of a foreign gene into the brain using adenovirus vectors, Nat. Genet. 3 Ž1993. 224–228. w2x T.A. Alston, L. Mela, H.J. Bright, 3-Nitropropionate, the toxic substance of Indigofera, is a suicide inactivator of succinate dehydrogenase, Proc. Natl. Acad. Sci. U.S.A. 74 Ž1977. 3767–3771. w3x R.N. Auer, B.K. Siesjo, Biological differences between ischemia, hypoglycemia, and epilepsy, Ann. Neurol. 24 Ž1988. 699–707.
w4x M.F. Beal, Does impairment of energy metabolism result in excitotoxic neuronal death in neurodegenerative illnesses?, Ann. Neurol. 31 Ž1992. 119–130. w5x M.F. Beal, Neurochemical and histologic characterization of striatal excitotoxic lesions produced by the mitochondrial toxin, 3-nitropropionic acid, J. Neurosci. 13 Ž1993. 4181–4192. w6x A.J. Bett, W. Haddara, L. Prevec, F.L. Graham, An efficient and flexible system for construction of adenovirus vectors with insertions or deletions in early regions 1 and 3, Proc. Natl. Acad. Sci. U.S.A. 91 Ž1994. 8802–8806. w7x A. Betz, F. Iannott, J. Hoff, Ischemia reduces blood-to-brain glucose transport in the gerbil, J. Cereb. Blood Flow Metab. 3 Ž1983. 200–206. w8x A.G. Betz, B. Yang, J. Davidson, Attenuation of stroke size in rats using an adenoviral vector to induce overexpression of interleukin-1 receptor antagonist in brain, J. Cereb. Blood Flow Metab. 15 Ž1995. 547. w9x R. Dash, M. Lawrence, D. Ho, R. Sapolsky, A herpes simplex virus vector overexpressing the glucose transporter gene protects the rat dentate gyrus from an antimetabolite toxin, Exp. Neurol. 137 Ž1996. 43–48. w10x B.L. Davidson, E.D. Allen, K.F. Kozarsky, J.M. Wison, B.J. Roessler, A model system for in vivo gene transfer into the central nervous system using an adenoviral vector, Nat. Genet. 3 Ž1993. 219–223. w11x T. Dumas, T. Meier, R. Sapolsky, Differential sparing of synaptic plasticity via gene therapy with overexpression of a glucose transporter versus of Bcl-2, submitted. w12x S.L. Fink, D.Y. Ho, R.M. Sapolsky, Energy and glutamate dependency of 3-nitropropionic acid neurotoxicity in culture, Exp. Neurol. 138 Ž1996. 298–304. w13x S.L. Fink, L. Chang, D. Ho, R. Sapolsky, Defective herpes simplex virus vectors expressing the rat brain stress-inducible heat shock protein 72 protect cultured neurons from severe heat shock, J. Neurochem. 68 Ž1997. 961. w14x A.I. Geller, X.O. Breakefield, A defective HSV-1 vector expresses Escherichia coli b-galactosidase in cultured peripheral neurons, Science 241 Ž1988. 1667–1668. w15x F.L. Graham, L. Prevec, Manipulation of adenovirus vectors, Methods in Molecular Biology, Gene Transfer and Expression Protocols, Vol. 7, The Human Press, Clifton, NJ, 1991. w16x A. Gupta, D. Ho, S. Fink, M. Roy, R. Sapolsky, An adenoviral vector expressing the glucose transporter protects cultured neurons from the cellular events mediating necrotic neuron death, submitted. w17x P. Hagan, J. Barks, M. Yabut, B. Davidson, B. Roessler, F. Silverstein, Adenovirus-mediated overexpression of IL-1 receptor antagonist reduces susceptibility to excitotoxic brain injury in perinatal rats, Neuroscience 75 Ž1996. 1033. w18x L. Hertz, L. Peng, Energy metabolism at the cellular level of the CNS, Can. J. Physiol. Pharmacol. 70 Ž1992. S145–S157. w19x D.Y. Ho, E.S. Mocarski, R.M. Sapolsky, Altering central nervous system physiology with a defective herpes simplex virus vector expressing the glucose transporter gene, Proc. Natl. Acad. Sci. U.S.A. 90 Ž1993. 3655–3659. w20x D.Y. Ho, T.C. Saydam, S.L. Fink, M.S. Lawrence, R.M. Sapolsky, Defective herpes simplex virus vectors expressing the rat brain glucose transporter protect cultured neurons from necrotic insults, J. Neurochem. 65 Ž1995. 842–850. w21x D.Y. Ho, R. Sapolsky, Gene therapy in the nervous system, Scientific American ŽJuly 1997. 116. w22x W. Jia, Y. Wang, D. Qiang, F. Tufaro, R. Remington, M. Cynader, Mol. Brain Res. 42 Ž1996. 350. w23x Kaplitt, A. Loewy ŽEds.., Viral Vectors, Academic Press, 1995. w24x M. Kindy, J. Yu, R. Miller, R. Roos, G. Ghadge, Adenoviral vectors in ischemic injury, in: J. Krieglstein ŽEd.., Pharmacology of Cerebral Ischemia, 1996. w25x M.S. Lawrence, D.Y. Ho, R. Dash, R.M. Sapolsky, Herpes simplex
S.L. Fink et al.r Brain Research 859 (2000) 21–25
w26x
w27x
w28x
w29x
w30x w31x w32x
w33x
w34x
virus vectors overexpressing the glucose transporter gene protect against seizure-induced neuron loss, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 7247–7251. M.S. Lawrence, G.H. Sun, D.M. Kunis, T.C. Saydam, R. Dash, D.Y. Ho, R.M. Sapolsky, G.K. Steinberg, Overexpression of the glucose transporter gene with a herpes simplex viral vector protects striatal neurons against stroke, J. Cereb. Blood Flow Metab. 16 Ž1996. 181–185. M.S. Lawrence, D. Ho, G. Sun, G. Steinberg, R. Sapolsky, Overexpression of Bcl-2 with herpes simplex virus vectors protects CNS neurons against neurologic insults in vitro and in vivo, J. Neurosci. 16 Ž1996. 486. M.S. Lawrence, G. Sun, D. Ho, L. McIntosh, D. Kunis, J. McLaughlin, R. Sapolsky, G. Steinberg, Herpes simplex viral vectors expressing Bcl-2 are neuroprotective when delivered following a stroke, J. Cereb. Blood Flow Metab. 17 Ž1997. 740. G. Le Gal La Salle, J.J. Robert, S. Berrard, V. Ridoux, L.D. Stratford-Perricaudet, M. Perricaudet, J. Mallet, An adenovirus vector for gene transfer into neurons and glia in the brain, Science 259 Ž1993. 988–990. J.-M. Lee, G. Zipfel, D. Choi, The changing landscape of ischaemic brain injury mechanisms, Nature 399 Ž1999. A7–A14, Suppl. W.-H. Lee, C.A. Bondy, Ischemic injury induces brain glucose transporter gene expression, Endocrinology 133 Ž1993. 2540–2544. M. Linnik, P. Zahos, M. Geschwind, H. Federoff, Expression of bcl-2 from a defective herpes simplex virus-1 vector limits neuronal death in focal cerebral ischemia, Stroke 26 Ž1995. 1670. A.L. McCall, A.M. Van Bueren, V. Nipper, M. Moholt-Siebert, H. Downes, N. Lessov, Forebrain ischemia increases Glut1 protein in brain microvessels and parenchyma, J. Cereb. Blood Flow Metab. 16 Ž1996. 69–76. J. McGrory, D. Bautista, F.L. Graham, A simple technique for the rescue of early region I mutations into infectious human adenovirus type 5, Virology 163 Ž1988. 614–617.
25
w36x T. Meier, D. Ho, T. Park, R. Sapolsky, Gene transfer of calbindin D28K cDNA via herpes simplex virus amplicon vector decreases calcium ion mobilization and enhances neuronal survival following glutamatergic challenge but not following cyanide, J. Neurochem. 70 Ž1998. 1013. w37x T.J. Meier, D.Y. Ho, R.M. Sapolsky, Increased expression of calbindin D28k via herpes simplex virus amplicon vector decreases calcium ion mobilization and enhances neuronal survival after hypoglycemic challenge, J. Neurochem. 69 Ž1997. 1039–1047. w38x L. Naldini, U. Blomer, P. Gallay, D. Ory, R. Mulligan, F. Gage, I. Vema, D. Trono, In vivo gene delivery and stable transduction of non-dividing cells by a lentiviral vector, Science 272 Ž1996. 263. w39x L. Sokoloff, The brain as a chemical machine, Prog. Brain Res. 94 Ž1992. 19–33. w40x R.R. Spaete, N. Frenkel, The herpes simplex virus amplicon: a new eucaryotic defective virus cloning-amplifying vector, Cell 30 Ž1982. 295–304. w41x L. Turski, W. Turski, Toward an understanding of the role of glutamate in neurodegenerative disorders: energy metabolism and neuropathology, Experientia 49 Ž1993. 1064. w42x S.J. Vannucci, L.B. Seaman, R.C. Vannucci, Effects of hypoxia– ischemia on GLUT1 and GLUT3 glucose transporters in immature rat brain, J. Cereb. Blood Flow Metab. 16 Ž1996. 77–81. w43x M. Weller, S.M. Paul, 3-Nitropropionic acid is an indirect excitotoxin to cultured cerebellar granule neurons, Eur. J. Pharmacol. Environ. Toxicol. Pharmacol. Sect. 248 Ž1993. 223–228. w44x D. Xu, S. Crocker et al., Elevation of neuronal expression of NAIP reduces ischemic damage in the rat hippocampus, Nat. Med. 3 Ž1997. 997. w45x M. Yenari, S. Fink, G. Sun, L. Chang, M. Patel, D. Kunis, D. Onley, D. Ho, R. Sapolsky, G. Steinberg, Gene therapy with hsp72 is neuroprotective in rat models of stroke and epilepsy, Ann. Neurol. 44 Ž1998. 584.
Research report
An adenoviral vector expressing the glucose transporter protects cultured striatal neurons from 3-nitropropionic acid Sheri L. Fink, Dora Y. Ho, John McLaughlin, Robert M. Sapolsky
)
Department of Biological Sciences, Stanford UniÕersity, Stanford, CA 95406, USA Accepted 23 November 1999
Abstract Considerable interest has focused on the possibility of using gene transfer techniques to introduce protective genes into neurons around the time of necrotic insults. We have previously used herpes simplex virus amplicon vectors to overexpress the rat brain glucose transporter, Glut-1 ŽGT., and have shown it to protect against a variety of necrotic insults both in vitro and in vivo, as well as to buffer neurons from the steps thought to mediate necrotic injury. It is critical to show the specificity of the effects of any such transgene overexpression, in order to show that protection arises from the transgene delivered, rather than from the vector delivery system itself. As such, we tested the protective potential of GT overexpression driven, in this case, by an adenoviral vector, against a novel insult, namely exposure of primary striatal cultures to the metabolic poison, 3-nitropropionic acid Ž3NP.. We observed that GT overexpression buffered neurons from neurotoxicity induced by 3NP. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Adenovirus; Neuron death; 3-Nitropropionic acid; Striatum; Gene therapy; Glucose transport; Neuronal energetics
1. Introduction There is now considerable understanding of the mechanisms mediating necrotic neuronal injury, in particular the cascade of excessive synaptic glutamate and free cytosolic calcium w30x. With this understanding comes the potential for therapeutic intervention. Particularly exciting is the possibility of gene transfer techniques, typically involving the use of viral vectors w14,21,23,38x. Reports of protection against models of necrotic insults using viral vectors involve targeting the calcium excess Žwith overexpression of the calcium binding protein, calbindin D28K. w24,36,37x, the protein malfolding Žwith hsp72. w13,45x, the ROS production Žwith superoxide dismutase. w24x, the apoptotic elements Žwith apoptosis inhibitors such as Bcl-2 or NAIP. w22,27,28,32,44x, or the inflammatory elements w8,17x. In addition to these approaches, we have studied the neuroprotective potential of overexpression of the rat brain glucose transporter, Glut-1 ŽGT.. The rationale for the construction of such a vector was the evidence that energy availability influences the efficacy with which an endangered neuron can contain glutamate and calcium, and thus )
Corresponding author. Fax: q1-650-725-5356; e-mail: [email protected]
the extent of toxicity caused by necrotic insults w4,41x. We have observed that overexpression of GT enhances glucose transport in both monolayer cultures and in vivo w19x. Such overexpression at the time of an insult helps maintain metabolism, and decreases glutamate release, cytosolic calcium concentrations, and oxygen radical accumulation w16,25x. Moreover, GT overexpression decreases neuron loss following hypoglycemia, glutamate administration, or antimetabolite exposure in cultured hippocampal, spinal cord, and septal neurons as well as following kainic-acidinduced seizure, focal ischemia, and exposure to antimetabolites in vivo w9,19,20,25,26x. These studies with glucose transporter overexpression have all used a herpes amplicon as a vector for delivery. An important point in the nascent discipline of neuronal gene therapy is to show the specificity of the effects of any such transgene overexpression, i.e., that it is the gene delivered which is efficacious, rather than the specifics of the particular delivery system. As such, in this paper, we explore the versatility of the neuroprotective potential of GT overexpression with two modifications: Ža. overexpression is driven, in this case, by an adenoviral vector, rather than a herpes amplicon vector; Žb. we test the protective effects of this vector against a unique insult, namely damage to striatal neurons induced by 3-nitropropionic
0006-8993r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 2 4 0 1 - 4
22
S.L. Fink et al.r Brain Research 859 (2000) 21–25
acid Ž3NP., a mitochondrial poison that irreversibly inhibits succinate dehydrogenase w2x.
2. Materials and methods 2.1. Cell cultures Mixed neuronalrglial cultures were prepared from the striatum of fetal Sprague–Dawley rats ŽHarlan, Indianapolis, IN. on day 18 of gestation. The experimental protocol concerning the use of rats was approved by the Division of Laboratory Animal Medicine at Stanford University. The method of preparing the cultures has been described w12x. Briefly, striatum was dissected from fetuses and dissociated with papain Ž10 unitsrml; Worthington Biochemical, Freehold, NJ.. The cells were then spun down, resuspended in MEM-PAK Ža modified minimal essential medium from the cell culture facility of the University of California, San Francisco, CA. supplemented with 10% horse serum ŽHyclone Laboratories, Logan, UT., and filtered through a cell strainer Ž70 mM mesh, Becton Dickinson, Lincoln Park, NJ.. The cells were then plated in wells of tissue culture plates at an approximate density of 1.6– 1.9 = 10 5 cellsrcm2 , and maintained in MEM-PAK supplemented with 10% Žvrv. horse serum. Cultures were used at 10–11 days of age, at which time approximately 50–70% of cells was non-neuronal w20,25x. A total of 293 cells ŽATCC CRL 1573. are grown in MEM ŽGibco BRL; cat. no. 61100-061. supplemented with 10% bovine newborn serum ŽGibco BRL.. All cells were maintained in a 5% CO 2r95% O 2 atmosphere at 378C in a humidified incubator. 2.2. Construction and preparation of adenoÕirus Õectors, AdRSVb gal and AdRSVGT The recombinant adenoviruses, AdRSVbgal and AdRSVGT, were constructed according to the method of Graham and Prevec w15x. First, two shuttle vectors, pDIE1RSV-GT and pDIE1RSV-bgal, were constructed ŽFig. 1.. To create pDIE1RSV-GT, a BamHI–BglII fragment from pON820 ŽViera and Mocarski, unpublished. containing the Rous sarcoma virus ŽRSV. promoter is inserted into the BamHI site of the shuttle vector, pDE1sp1A w6x. The resulting vector is named pDIE1RSV. Then, an XbaI fragment containing the glucose transporter gene and the polyadenylation signal from HCMV IE1 was excised from pa4GTY aY w25x and cloned into the XbaI site of pDIE1RSV. The resulting construct is pDIE1RSV-GT. To create pDIE1RSV-bgal, the BamHI–BglII fragment containing the RSV promoter was cloned into the BamHI site of pON1 w40x, 5X to the Escherichia coli lacZ coding sequence. A BamHI fragment from the resulting construct Žcontaining the RSV promoter, the lacZ gene and the
Fig. 1. Generation of AdRSVGT. The two parent plasmids used for the generation of AdRSVGT are pDIE1RSV-GT and pJM17. In pDIE1RSVGT, the glucose transporter gene gt Žwhite arrow. under the control of the RSV promoter Ždotted arrow. is inserted into the E1 region of Ad5 sequences Žsolid black bars.. The flanking Ad5 homologies on either side of the transgene represent 0–0.1 map unit Žmu. and 9.8–16.1 mu, respectively. pJM17 w34x contains the entire sequence of Ad5 interrupted by the plasmid, pBRX Žstriped bar., which encodes ampicillin and tetracycline resistance. The junction of the two ends of Ad5 genome is indicated as 100r0 mu. pJM17 is non-packagable since it has exceeded the packaging limit of adenovirus. Co-transfection of pDIE1RSV-GT and pJM17 into 293 cells resulted in homologous recombination between the two plasmids and the generation of packagable AdRSVGT. AdRSVbgal is similarly generated using plasmids, pDIE1RSV-bgal amd pJM17. In pDIE1REV-bgal, the gt gene is replaced by the lacZ gene.
SV40 polyadenylation sequences. was then inserted into the BamHI site of pDE1sp1A to create pDIE1RSV-bgal. To generate the recombinant viruses, AdRSVGT and AdRSVbgal, the two shuttle vectors, pDIE1RSV-GT and pDIE1RSV-bgal ŽFig. 1., were transfected into 293 cells along with pJM17, a bacterial plasmid comprised of the circular form of Ad5 genome in an unpackagable form w34x. Upon recombination of the shuttle vectors with pJM17, infectious virus are generated. Viral plaques resulting from the co-transfection were plaque-purified three times and amplified according to the protocol of Graham and Prevec w15x. Preparation of recombinant adenovirus stocks was based on a protocol obtained from Dr. Frank Graham’s laboratory with modifications. Briefly, infected 293 cells Žgrown on 150 mm dish. were harvested when 100% cytopathic effect was reached. Cells from 30–40 dishes were pelleted and resuspended in 5 ml of 0.1 M Tris ŽpH 8.1.. Sodium deoxycholate, MgCl 2 and DNAse I were then added to a final concentration of 0.5%, 20 mM and 50 mgrml, respectively. Then 1.8 ml of saturated CsCl solution Žin 10 mM Tris pH 8.1r1 mM EDTA. was added to each 3.1 ml of cell lysate. The virus was banded twice by centrifugation in a Ti70.1 rotor ŽBeckman. for 16–20 h at 48C and 35,000 rpm. The final viral band was collected and desalted with a DG-10 Econo-pack column ŽBioRad, Hercules, CA.. The virus was eluted in hepes-buffered saline Ž20 mM hepesr150 mM NaCl, pH 7.8. with 3% sucrose and titrated on 293 cells by standard plaque assay.
S.L. Fink et al.r Brain Research 859 (2000) 21–25
2.3. Toxicity experiments Striatal cultures grown in 48-well tissue culture plates were infected on day 7 in culture with AdRSVGT or AdRSVbgal Ž6.6 = 10 4 PFUrwell.; under these conditions, 20% of neurons and 5% of glia are infected ŽMcLaughlin, in preparation.. At 3 days post-infection, the cultures were challenged with 0, 0.5 or 1.0 mM 3NP in DMEM containing 0.5 or 1.5 mM glucose. The cultures were maintained in this medium for approximately 22 h, fixed, and immunostained with a mouse monoclonal antibody against MAP-2 Ž1:1000 dilution; Sigma. and the Vectastain ABC kit for peroxidase ŽVector Laboratories, Burlingame, CA.. Counts were made across the diameter of each well; only neurons positive for the MAP2 antigen and with intact processes were counted. The neuron counts for each condition were expressed as the percentage of mean control Žsame vector, same glucose concentration, no 3NP. values and averaged unless otherwise noted. Comparisons are by Mann–Whitney t-tests. Data are expressed as mean " S.E.M. values.
3. Results Two adenovirus recombinants, AdRSVGT and AdRSTbgal, were constructed to express GT or bgal, respectively, under the control of the RSV promoter. The E1 region of the viral genome was replaced by the transgene of interest Ždetailed in Fig. 1. and consequently, both recombinants are E1 deletion mutants. In pilot studies, GT
Fig. 2. Effects of AdRSVGT infection on survival of striatal neurons following 3NP andror hypoglycemic exposure. Striatal cultures were infected with AdRSVGT or AdRSVbgal. At 3 days post-infection, the cells were incubated with DMEM containing various concentrations of glucose, with or without 0.5 or 1.0 mM 3NP, for approximately 22 h. The cells were then fixed and immunostained against MAP-2. The numbers of intact neurons were counted and expressed as the percent mean values of results in their respective control groups, which were infected with the same vector, maintained under same glucose concentrations, but without 3NP. Under those control conditions, neuron number in wells treated with AdRSVGT or AdRSVbgal did not differ. U ps 0.007; UU ps 0.0002 by Mann–Whitney t-test.
23
or bgal expression from the recombinants was assessed in primary neuronal cultures using immunofluorescence techniques. Furthermore, bgal expression from AdRSVbgal was assessed at several timepoints by reacting cells with X-gal and was found to be increased from days 1 to 6 post-infection Ždata not shown.. We then tested whether overexpression of a glucose transporter would prove protective against 3NP. In agreement with prior reports under these culture conditions w12x, low glucose conditions Ž0.5 mM glucose, 0 mM 3NP. were not neurotoxic, and the toxicity of 3NP was greatly potentiated under low glucose conditions ŽFig. 2.. We observed that adRSVGT decreased the toxicity of 3NP exposure.
4. Discussion Because of the post-mitotic nature of neurons, those lost to neurological insults cannot be replaced. Thus, the ability to intervene at the time of an insult with a gene therapeutic approach meant to save neurons has considerable clinical promise. The particular therapeutic strategy of overexpressing the glucose transporter is a logical one. Glucose is the primary energy source utilized by the brain w18,39x. Because glucose is not stored in neurons, the capability of a neuron to take up glucose during an insult could crucially determine its fate. Glucose transport into target cells in the brain becomes rate-limiting during some necrotic insults w3,7x, and neurons will upregulate glucose transporter expression post-ischemia as part of the adaptive profile of activating stress proteins w31,33,42x. We observe that overexpression of the glucose transporter gene, Glut-1, driven by an adenoviral vector, can protect cultured striatal neurons from 3NP. The toxin is a mitochondrial poison that irreversibly inhibits succinate dehydrogenase, disrupting oxidative metabolism w2x, and producing necrotic neuron loss both in animals w5x and in cultured cells w12,43x, with a characteristic profile of pyknosis. As a dramatic measure of the energetic basis of the toxicity of 3NP, glucose availability will shift its LD50 by more than an order of magnitude in primary neuronal cultures w12x. Additional glucose uptake would be expected to protect the function of many energy-requiring cellular processes after necrotic insults. In our prior studies overexpressing GT with a herpes amplicon vector, we observed a more prolonged maintenance of metabolism during an insult in hippocampal cultures Žas assessed with silicon microphysiometry w25x., as well as decreases in glutamate release, cytosolic calcium mobilization, and oxygen radical accumulation w16x. Collectively, these effects are likely to result in a decreased likelihood of neuron death. Herpes amplicon vectors have a strong preference for infecting neurons over glia w13,19x, whereas adenoviral vectors infect a significant subset of glia w1,10,29x. While only a small
24
S.L. Fink et al.r Brain Research 859 (2000) 21–25
percentage of glia was infected under the conditions in this study, part of the neuroprotection observed may still have arisen from enhancing salutary and energy-dependent actions of glia, such as the high-affinity removal of glutamate from the synapse. The magnitude of neuroprotection is likely to have been considerable. Fig. 2 indicates that under conditions where 3NP-induced toxicity in control vector-treated wells was virtually complete, treatment with AdRSVGT resulted in survival of approximately 20% of neurons. As noted, under these conditions of infection, approximately 20% of neurons is infected with AdRSVGT, suggesting that a very high percentage of infected neurons is likely to have survived Ža less parsimonious explanation would be that infection of a particular neuron with AdRSVGT would not spare that neuron from 3NP toxicity, but would result in survival of an uninfected neighbor.. Our observation of neuroprotection with this vector extends prior work in which herpes-based GT vectors protected against hypoglycemia, glutamate, 3-acetylpyridine, stroke, and seizure, even when introduced after an insult w9,19,20,25,26x; moreover, hippocampal neurons, when saved by GT overexpression, are able to sustain normal spatial maze learning in rats Ž345, as well as typical synaptic plasticity w11x. The efficacy of both herpes and adenovirus vectors in neuroprotection is further evidence that expression of the GT gene itself, rather than a particular method of overexpression, leads to neuroprotection. This is important, both for the heuristic value of further appreciating the energetic component to necrotic neuronal injury, as well as for designing interventions that may ultimately be of clinical use.
Acknowledgements We thank Frank Graham ŽMcMaster University, Hamilton, ON, Canada. for providing pDE1sp1A and pJM17, Edward Mocarski ŽStanford University, Stanford, CA., for providing pON1 and pON820 and Sheila Brooke, John McLaughlin and Timothy Meier for their excellent technical assistance. Funding was provided by RO1 NS 32848 to R.M.S.
References w1x S. Akli, C. Caillaud, E. Vigne, L.D. Stratford-Perricaudet, L. Poenaru, M. Perricaudet, A. Kahn, M.R. Peschanski, Transfer of a foreign gene into the brain using adenovirus vectors, Nat. Genet. 3 Ž1993. 224–228. w2x T.A. Alston, L. Mela, H.J. Bright, 3-Nitropropionate, the toxic substance of Indigofera, is a suicide inactivator of succinate dehydrogenase, Proc. Natl. Acad. Sci. U.S.A. 74 Ž1977. 3767–3771. w3x R.N. Auer, B.K. Siesjo, Biological differences between ischemia, hypoglycemia, and epilepsy, Ann. Neurol. 24 Ž1988. 699–707.
w4x M.F. Beal, Does impairment of energy metabolism result in excitotoxic neuronal death in neurodegenerative illnesses?, Ann. Neurol. 31 Ž1992. 119–130. w5x M.F. Beal, Neurochemical and histologic characterization of striatal excitotoxic lesions produced by the mitochondrial toxin, 3-nitropropionic acid, J. Neurosci. 13 Ž1993. 4181–4192. w6x A.J. Bett, W. Haddara, L. Prevec, F.L. Graham, An efficient and flexible system for construction of adenovirus vectors with insertions or deletions in early regions 1 and 3, Proc. Natl. Acad. Sci. U.S.A. 91 Ž1994. 8802–8806. w7x A. Betz, F. Iannott, J. Hoff, Ischemia reduces blood-to-brain glucose transport in the gerbil, J. Cereb. Blood Flow Metab. 3 Ž1983. 200–206. w8x A.G. Betz, B. Yang, J. Davidson, Attenuation of stroke size in rats using an adenoviral vector to induce overexpression of interleukin-1 receptor antagonist in brain, J. Cereb. Blood Flow Metab. 15 Ž1995. 547. w9x R. Dash, M. Lawrence, D. Ho, R. Sapolsky, A herpes simplex virus vector overexpressing the glucose transporter gene protects the rat dentate gyrus from an antimetabolite toxin, Exp. Neurol. 137 Ž1996. 43–48. w10x B.L. Davidson, E.D. Allen, K.F. Kozarsky, J.M. Wison, B.J. Roessler, A model system for in vivo gene transfer into the central nervous system using an adenoviral vector, Nat. Genet. 3 Ž1993. 219–223. w11x T. Dumas, T. Meier, R. Sapolsky, Differential sparing of synaptic plasticity via gene therapy with overexpression of a glucose transporter versus of Bcl-2, submitted. w12x S.L. Fink, D.Y. Ho, R.M. Sapolsky, Energy and glutamate dependency of 3-nitropropionic acid neurotoxicity in culture, Exp. Neurol. 138 Ž1996. 298–304. w13x S.L. Fink, L. Chang, D. Ho, R. Sapolsky, Defective herpes simplex virus vectors expressing the rat brain stress-inducible heat shock protein 72 protect cultured neurons from severe heat shock, J. Neurochem. 68 Ž1997. 961. w14x A.I. Geller, X.O. Breakefield, A defective HSV-1 vector expresses Escherichia coli b-galactosidase in cultured peripheral neurons, Science 241 Ž1988. 1667–1668. w15x F.L. Graham, L. Prevec, Manipulation of adenovirus vectors, Methods in Molecular Biology, Gene Transfer and Expression Protocols, Vol. 7, The Human Press, Clifton, NJ, 1991. w16x A. Gupta, D. Ho, S. Fink, M. Roy, R. Sapolsky, An adenoviral vector expressing the glucose transporter protects cultured neurons from the cellular events mediating necrotic neuron death, submitted. w17x P. Hagan, J. Barks, M. Yabut, B. Davidson, B. Roessler, F. Silverstein, Adenovirus-mediated overexpression of IL-1 receptor antagonist reduces susceptibility to excitotoxic brain injury in perinatal rats, Neuroscience 75 Ž1996. 1033. w18x L. Hertz, L. Peng, Energy metabolism at the cellular level of the CNS, Can. J. Physiol. Pharmacol. 70 Ž1992. S145–S157. w19x D.Y. Ho, E.S. Mocarski, R.M. Sapolsky, Altering central nervous system physiology with a defective herpes simplex virus vector expressing the glucose transporter gene, Proc. Natl. Acad. Sci. U.S.A. 90 Ž1993. 3655–3659. w20x D.Y. Ho, T.C. Saydam, S.L. Fink, M.S. Lawrence, R.M. Sapolsky, Defective herpes simplex virus vectors expressing the rat brain glucose transporter protect cultured neurons from necrotic insults, J. Neurochem. 65 Ž1995. 842–850. w21x D.Y. Ho, R. Sapolsky, Gene therapy in the nervous system, Scientific American ŽJuly 1997. 116. w22x W. Jia, Y. Wang, D. Qiang, F. Tufaro, R. Remington, M. Cynader, Mol. Brain Res. 42 Ž1996. 350. w23x Kaplitt, A. Loewy ŽEds.., Viral Vectors, Academic Press, 1995. w24x M. Kindy, J. Yu, R. Miller, R. Roos, G. Ghadge, Adenoviral vectors in ischemic injury, in: J. Krieglstein ŽEd.., Pharmacology of Cerebral Ischemia, 1996. w25x M.S. Lawrence, D.Y. Ho, R. Dash, R.M. Sapolsky, Herpes simplex
S.L. Fink et al.r Brain Research 859 (2000) 21–25
w26x
w27x
w28x
w29x
w30x w31x w32x
w33x
w34x
virus vectors overexpressing the glucose transporter gene protect against seizure-induced neuron loss, Proc. Natl. Acad. Sci. U.S.A. 92 Ž1995. 7247–7251. M.S. Lawrence, G.H. Sun, D.M. Kunis, T.C. Saydam, R. Dash, D.Y. Ho, R.M. Sapolsky, G.K. Steinberg, Overexpression of the glucose transporter gene with a herpes simplex viral vector protects striatal neurons against stroke, J. Cereb. Blood Flow Metab. 16 Ž1996. 181–185. M.S. Lawrence, D. Ho, G. Sun, G. Steinberg, R. Sapolsky, Overexpression of Bcl-2 with herpes simplex virus vectors protects CNS neurons against neurologic insults in vitro and in vivo, J. Neurosci. 16 Ž1996. 486. M.S. Lawrence, G. Sun, D. Ho, L. McIntosh, D. Kunis, J. McLaughlin, R. Sapolsky, G. Steinberg, Herpes simplex viral vectors expressing Bcl-2 are neuroprotective when delivered following a stroke, J. Cereb. Blood Flow Metab. 17 Ž1997. 740. G. Le Gal La Salle, J.J. Robert, S. Berrard, V. Ridoux, L.D. Stratford-Perricaudet, M. Perricaudet, J. Mallet, An adenovirus vector for gene transfer into neurons and glia in the brain, Science 259 Ž1993. 988–990. J.-M. Lee, G. Zipfel, D. Choi, The changing landscape of ischaemic brain injury mechanisms, Nature 399 Ž1999. A7–A14, Suppl. W.-H. Lee, C.A. Bondy, Ischemic injury induces brain glucose transporter gene expression, Endocrinology 133 Ž1993. 2540–2544. M. Linnik, P. Zahos, M. Geschwind, H. Federoff, Expression of bcl-2 from a defective herpes simplex virus-1 vector limits neuronal death in focal cerebral ischemia, Stroke 26 Ž1995. 1670. A.L. McCall, A.M. Van Bueren, V. Nipper, M. Moholt-Siebert, H. Downes, N. Lessov, Forebrain ischemia increases Glut1 protein in brain microvessels and parenchyma, J. Cereb. Blood Flow Metab. 16 Ž1996. 69–76. J. McGrory, D. Bautista, F.L. Graham, A simple technique for the rescue of early region I mutations into infectious human adenovirus type 5, Virology 163 Ž1988. 614–617.
25
w36x T. Meier, D. Ho, T. Park, R. Sapolsky, Gene transfer of calbindin D28K cDNA via herpes simplex virus amplicon vector decreases calcium ion mobilization and enhances neuronal survival following glutamatergic challenge but not following cyanide, J. Neurochem. 70 Ž1998. 1013. w37x T.J. Meier, D.Y. Ho, R.M. Sapolsky, Increased expression of calbindin D28k via herpes simplex virus amplicon vector decreases calcium ion mobilization and enhances neuronal survival after hypoglycemic challenge, J. Neurochem. 69 Ž1997. 1039–1047. w38x L. Naldini, U. Blomer, P. Gallay, D. Ory, R. Mulligan, F. Gage, I. Vema, D. Trono, In vivo gene delivery and stable transduction of non-dividing cells by a lentiviral vector, Science 272 Ž1996. 263. w39x L. Sokoloff, The brain as a chemical machine, Prog. Brain Res. 94 Ž1992. 19–33. w40x R.R. Spaete, N. Frenkel, The herpes simplex virus amplicon: a new eucaryotic defective virus cloning-amplifying vector, Cell 30 Ž1982. 295–304. w41x L. Turski, W. Turski, Toward an understanding of the role of glutamate in neurodegenerative disorders: energy metabolism and neuropathology, Experientia 49 Ž1993. 1064. w42x S.J. Vannucci, L.B. Seaman, R.C. Vannucci, Effects of hypoxia– ischemia on GLUT1 and GLUT3 glucose transporters in immature rat brain, J. Cereb. Blood Flow Metab. 16 Ž1996. 77–81. w43x M. Weller, S.M. Paul, 3-Nitropropionic acid is an indirect excitotoxin to cultured cerebellar granule neurons, Eur. J. Pharmacol. Environ. Toxicol. Pharmacol. Sect. 248 Ž1993. 223–228. w44x D. Xu, S. Crocker et al., Elevation of neuronal expression of NAIP reduces ischemic damage in the rat hippocampus, Nat. Med. 3 Ž1997. 997. w45x M. Yenari, S. Fink, G. Sun, L. Chang, M. Patel, D. Kunis, D. Onley, D. Ho, R. Sapolsky, G. Steinberg, Gene therapy with hsp72 is neuroprotective in rat models of stroke and epilepsy, Ann. Neurol. 44 Ž1998. 584.