The U.S. Government has a paid-up license in this invention and the right in limited circumstances to require the patent owner to license others on reasonable terms as provided for by the terms of grants DA-00266, DA-271-90-7408, DA-00074, and NS-01528 awarded by the National Institutes of Health.

FTELD OF THE INVENTION

The invention relates to the use of inhibitor of nitric oxide cytotoxicity to prevent neurodegeneration induced by immunodeficiency viruses. Such inhibitors may block production of nitric oxide or formation of peroxynitric from nitric oxide

BACKGROUND OF THE -DWENTION

AIDS and other diseases caused by mammalian immunodeficiency viruses are often associated with multiple neurological abnormalities including deficits in cognitive and motor functions (Navia, B.A. (1986) Ann. Neurol. 19, 525-535;

Navia, B.A. (1986) Ann. Neurol. 19, 517-524; and Price, R.W. (1988) Science 239, 586-592)). While the human immunodeficiency virus type 1 (HTV-l) can enter the central nervous system, the pathogenesis of the HIV-l -associated cognitive motor complex, the AIDS dementia complex, has remained elusive (Lipton, S.A. (1992) Trends Neurosci. 15, 75-79)), as HIV-l infection has been detected in macrophages and microglia but not in neurons (Koenig, S., et al. (1986) Science 233, 1089-1093; Giulian, D., et al. (1990) Science 250, 1593-1596;

Watkins, B.A., et al. (1990) Science 249, 549-553; Gabuzda, D.H., et al. (1986) Ann. Neurol. 20, 289-295)). Despite HTV-l not directly infecting neurons, there is profound neuronal loss in the cortex and retina (Wiley, C.A. (1991) Ann. Neurol. 29, 651-657; and Pomerantz, RJ. (1987) N. Engl. J. Med. 317, 1643- 1647)).

Neuronal cell death might involve the HTV-l coat protein, gpl20, which is shed by the virus and which can elicit neurotoxicity in very low concentrations in primary hippocampal (Dreyer, E.B. (1990) Science 248, 364-367; and Brenneman, D.E. (1988) Nature (London) 335, 639-642)) and retinal ganglion (Dreyer, E.B. (1990) Science 248, 364-367; Lipton, S.A. (1991) Ann. Neurol. 30,

110-114; and Lipton, S.A. (1991) Neuron 7, 111-118)) cultures. Death of gρl20- treated retinal ganglion cells is preceded by a marked increase in intracellular calcium.

L-type calcium channel antagonists (Dreyer, E.B. (1990) Science 248, 364-367; and Lipton, S.A. (1991) Ann. Neurol. 30, 110-114)) bloc pl20-induced retinal ganglion cell death. N-methyl-D-aspartate (NMDA) receptor antagonists also attenuate gpl20-induced neurotoxicity (Lipton, S.A. (1991) Neuron 7, 111-118). There is a need in the art for additional therapeutic agents for attenuating neurotoxicity induced by immunodeficiency viruses.

SUMMARY OF THE INVENTION

It is an object of the invention to provide a method of slowing the progression of encephalopathy induced by an immunodeficiency virus.

This and other objects of the invention are provided by one or more of the embodiments described below. In one embodiment of the invention, a method for slowing the progression of encephalopathy induced by an immunodeficiency virus is provided. The method comprises: administering to a mammal infected with an immunodeficiency virus a drug which inhibits production of nitric oxide by neuronal cells. In another embodiment of the invention, a superoxide dismutase gene is administered to a mammal infected with an immunodeficiency virus.

Thus, the present invention provides the art with methods for treating neurological degeneration associated with infection by immunodeficiency viruses, such as AIDS dementia.

5 BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 shows that gpl20 neurotoxicity is dependent on extracellular glutamate. Figure 1(A) shows the effect of various glutamate concentrations in the presence of 100 pM gpl20. Figure 1(B) shows the effect of various gpl20 concentrations in the presence of 25 μM. glutamate. All data points represent

10 means ±. SEM of at least two individual experiments with a total of n ≥ 8.

Figure 2 proposes a mechanism for the neurotoxicity induced by gpl20, the HIV-l coat protein.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

15 It is a discovery of the present invention that the neurotoxicity of mammalian immunodeficiency viruses is mediated via nitric oxide synthase and superoxide anions. Thus, drugs and biologicals which inhibit the formation of NO, lower the intercellular concentration of NO, or lower the intercellular concentration of superoxide anions, decrease the amount of neuronal degeneration.

20 According to the present invention the neurological effects of mammalian immunodeficiency viruses are ameliorated. Such viruses are retroviruses, such as human immunodeficiency virus (HTV), simian immunodeficiency virus (STY) and the like. The neurological effects include neuronal necrosis, loss of motor skills, and loss of cognitive skills. These effects can be assessed by any means known

25 by those of skill in the art, including cytochemistry, immunocytochemistry, motor function tests, cognitive function tests, etc.

Production of nitric oxide by nitric oxide synthase (NOS) can be assayed r as is known in the art, for example, by assaying formation of citrulline from

L-arginine. See Bredt and Snyder (1990) Proc. Natl. Acad. Sci. USA £7:682-685.

'30 The production of nitric oxide can also be assessed by its effect on neuronal viability, motor skills, cognitive skills, and neuronal function.

Known inhibitors of NOS which may be used according to the invention include L-N ^β -nitτoarginine and its analogs which reversibly compete with L-arginine. Suitable analogs include L-i- noemylomithine and methyl ester of L-N"-m^oarginine. Inhibitors of NOS more generally are compounds which compete for the substrate binding site of NOS or other sites on the enzyme, and include both reversible and irreversible inhibitors. The present invention contemplates the use of any physiologically acceptable agent which inhibits NOS activity. The effectiveness of a compound, and its relative potency as an NOS inhibitor, can be tested and routinely determined by measuring inhibition of NOS activity by monitoring the conversion of arginine to citrulline by NOS in, for example, cerebellar homogenates. A reduction in citrulline formation indicates inhibitory activity of the compound. The percent reduction in citrulline formation, compared to the amount of citrulline formed in the absence of the compound being tested, indicates the potency of the compound as an NOS inhibitor. Other inhibitors of

NOS have also been developed. Inhibitors have been prepared from arginine, nitroarginine and guanidinoalkanoic acids, wherein the guanidine group, the amino group, the carboxy group and the backbone have been systematically altered. Various types of structural alterations affect the selectivity of inhibitors toward brain or macrophage enzyme.

Among the nitroarginine analogs tested, nitroarginine was the most potent inhibitor. In general, the introduction of the nitro group on the guanidine moiety appears to result in selective inhibition of the brain enzyme. Moreover, it was found that the nature of substituents on the α-amino group seemed to dictate the inhibition potency of the compounds. Whereas a free α-amino group offers good inhibitors with 5 to 200-fold selectivity toward brain, the substitution of the amino group with a bulky protective group, such as benzyloxy, results in total loss of activity. A small substituent, such as a formyl group, appears to be favorable.

In the guanidinoalkanoic acid series, it has been found that 6-guanidinohexanoic acid (6-GHA) and 5-guanidinopentanoic acid (5-GPA) are potent inhibitors and show good selectivity toward the macrophage enzyme but

were inactive against the brain NOS. The nitro analogs of 6-GHA and 5-GPA were found to be generally inactive, as were guanidinoalkanoic acid analogs made rigid by the cyclization of the guanidine group.

The dosage and length of treatment depends on the disease stage of the mammal being treated. The duration of treatment may be a day, a week or longer and may last over the entire lifetime of the patient. The inhibitors are administered in a therapeutically effective amount, a typical human dosage of nitroarginine ranging from about 0.01 mg/kg of body weight of nitroarginine to about 10 mg/kg of nitroarginine, in single or divided doses. The dosage will vary depending on the NOS inhibitor to be used and its relative potency. Dosage and length of treatment are readily determinable by the skilled practitioner based on the condition and stage of disease.

In therapeutic use, NOS inhibitors may be administered by any route whereby drugs are conventionally administered. Such routes of administration include intraperitoneally, intravenously, intramuscularly, subcutaneously, intrathecally and intraventricularly, as well as orally.

Typical preparations for administration include sterile aqueous or nonaqueous solutions, suspensions and emulsions. Examples of nonaqueous solvents are propylene glycol, polyethylene glycol, vegetable oils such as olive oil and injectable organic esters such as ethyl oleate. Aqueous carriers include water alcoholic aqueous and buffered media. Intravenous vehicles include fluid and nutrient replenishers, electrolyte replenishers, such as those based on Ringer's dextrose, and the like. Preservatives and other additives may also be present such as, for example, antimicrobials, anti-oxidants, chelating agents, inert gases, and the like. Oral preparations, such as in capsules, tablets, and other forms, include additives such as cellulose, silica gel and stearic acid.

To be effective therapeutically, an NOS inhibitor desirably should be able to penetrate the blood brain barrier when peripherally administrated. However, because of the charged guanidine group, most arginine derivatives would not be expected to penetrate into the brain. Arginine itself has only poor access to the brain following peripheral administration. However, arginine derivatives do exist

which are able to penetrate into the brain and inhibit enzyme activity in the brain. In fact, as little as 5 mg/kg of nitroarginine markedly inhibits enzyme activity in the brain. It may be that the nitro group facilitates penetration of the blood brain barrier and access to the brain following peripheral administration. NOS inhibitors which are unable to penetrate the blood brain barrier can be effectively administered by, for example, an intraventricular route of delivery.

Neuronal NOS is calmodulin-dependent. Thus, drugs which inhibit calmo¬ dulin will inhibit the production of NO by neuronal NOS. Examples of such drags include camidazolium, N-(6-aminohexyl)-5-chloro-l-naphthalenesulfonamide,and the phenothiazine class of drugs.

It is also known in the art that immunophilin-binding drugs inhibit NO formation by neuronal NOS. This is due to their ability to bind their respective immunophilin and subsequently inhibit the activity of the calcium-activated phosphatase calcineurin. Such drugs include FK-506 and cyclosporin which bind to FKBP and cyclophilin, respectively. The effect of these drugs is presumably due to the increased phosphorylation of NOS resulting from the decreased calcineurin activity. Phosphorylated NOS is less active than the dephosphorylated form.

FK-506, complexed to FKBP, binds to calcineurin, inhibiting its phosphatase activity. This prevents the dephosphorylation of NOS, which decreases NOS catalytic activity. With lowered NO production, adjacent neurons remain viable. Other immunophilin-binding drugs act by a similar mechanism.

Both FK-506 and cyclosporin A, two immunophilin-binding calcineurin inhibitors, prevent neurotoxicity in proportion to their relative potencies as calcineurin inhibitors. In addition to these compounds, other immunophilin- binding drugs have been developed. Such drugs include FK-520, FK-523, 15-0- DeMe-FK-520,(4R)-[(E)-L-butenyl]-4,N-dimethyl-L-threonine. (Liu, Biochemistry, 31:3896-3902 (1992)).

To be effective therapeutically, an immunophilin-binding drug desirably should be able to penetrate the blood-brain barrier when peripherally administrated. However, some immunophilin-binding drugs, like cyclosporin A,

do not readily penetrate into the brain. Immunophilin-binding drugs which are unable to penetrate the blood-brain barrier can be effectively administered by, for example, an intraventricular route of delivery.

According to another aspect of the invention, a superoxide dismutase gene is administered to elevate the level of activity of the SOD enzyme in the nervous system. See Chan et al (1990) Stroke 21:80-82. The enzyme lowers the concentration of superoxide anions which combine with NO to form peroxynitrite that degenerates into the toxic hydroxyl and nitric dioxide-free radicals.

The related disclosures of U.S. patent applications S.N. 08/082,848 and 07/881,900 are expressly incorporated herein.

EXAMPLES

The following examples are provided to exemplify various aspects of the invention and are not intended to limit the scope of the invention. Example 1

This example demonstrates that gpl20 neurotoxicity is dependent on extracellular glutamate.

In our initial experiments, gpl20-induced neurotoxicity in cortical cultures was variable with the potency of gpl20 apparentiy related to the interval following the exchange of the growth medium. When we exposed neuronal cultures to gpl20, 12 hours after exchanging the standard growth medium, we observed minimal cell death. Lipton et al. (Lipton, S.A., Sucher, N.J., Kaiser, P.K. & Dreyer, E.B. (1991) Neuron 7, 111-118) showed that gpl20 toxicity in retinal ganglia cell cultures depends on extracellular glutamate. Accordingly, we exposed cells to 100 pM gpl20 with various concentrations of glutamate. gpl20 toxicity is absolutely dependent on extracellular glutamate (Fig. 1A). No toxicity is evident in the absence of glutamate while 25 μM glutamate plus 100 pM gpl20 produces »75% of the maximal cell death obtained at 50-75 μM glutamate plus 100 pM gpl20. The maximal percentage of cells killed by gpl20 in the presence of glutamate is «60%. All subsequent experiments used 25 M glutamate, as 50 μM glutamate itself elicits some toxicity.

gpl20 is extremely potent in eliciting cell death with significant effects at

0.1 pM gpl20, half-maximal influences at 10 pM gpl20, and maximal toxicity at

0.1-1 nM gpl20 with «50% of cells dying (Fig. IB). Toxicity following 100 pM gpl20 is attenuated 75% by immunoprecipitating with antiserum to gpl20 but not by nonimmune mouse IgG.

Cell Cultures. Primary cell cultures were prepared from fetal rats on gestation day 14 (Dichter, M.A. (1978) Brain Res. 149, 279-293). The cortex was dissected under a microscope, incubated for 20 minutes in 0.027% trypsin/saline (5% phosphate-buffered saline/40 mM sucrose 30 mM glucose/10 mM Hepes, pH 7.4), and transferred to modified Eagle's medium (MEM)/10% horse serum/10% fetal bovine serum/2 mM glutamine. Cells were dissociated by trituration, counted, and plated in 15-mm multiwell (Nunc) plates coated with polyornithine at a density of 3-4 x 10 ³ cells per well. Four days after plating the cells were treated with 5-fluoro-2'-deoxyuridine (10 g/ml) for 3 days to inhibit proliferation of nonneuronal cells. Cells were maintained in MEM 5% horse serum/2 mM glutamine in humidified 8% CO ₂ /92% air at 37°C. The medium was changed twice a week with freshly prepared medium in which glutamine was added at the time of feeding. Mature neurons (>21 days in culture) were used in all experiments. (Tissue culture reagents were obtained from GIBCO/BRL.) Cytotoxicity. Cells were exposed to test solutions as described (Dawson,

V.L., Dawson, T.M., London, E.D., Bredt, D.S., & Snyder, S.H. (1991) Proc. Natl. Acad. Sci. USA 88, 6368-6371). The cells were washed three times with a Tris-buffered control salt solution (CSS) (120 mM NaCl/5.4 mM KC1/1.8 mM CaCl ₂ /25 mM Tris-HCl/15 mM glucose, pH 7.4), exposed to test solutions for 5 minutes and then washed with CSS followed by MEM containing 21 mM glucose.

The cells were then returned to the incubator. At 20-24 hours after exposure to test solutions, the cells were exposed to 0.4% trypan blue in CSS to stain the residue of nonviable cells. Two to four photoprints at xlO to x20 were made of each well. Viable and nonviable cells were counted with at least 500-1500 cells countered per well. At least two experiments were performed using four separate wells so that a minimum of 4,000-12,000 neurons were counted for each data

point. Ten percent of the photomicrographs were counted by an additional observer blinded to the arrangement of photomicrographs, study design, and treatment protocol. An interrater reliability of >90% was consistently observed for the cell counting. Materials. HTV-lSFj gpl20 was obtained from the National Institutes of

Health AIDS Reagent Program through Monique Dubois-Dalcq (National Institute of Neurological and Communicative Disorders and Stroke National Institutes of Health). Recombinant gpl20 and mouse anti-gpl20 antibody were obtained from American Biotechnologies. GP-120. Immunoprecipitation of gpl20 was performed by incubating

0.96-ng samples of gpl20 with 4.8 μg of a mouse monoclonal anti-gpl20 (IgG) antibody (American Biotechnologies, Cambridge, MA) overnight at 4°C. Control experiments were performed by incubating 0.96 ng of gpl20 with 4.8 μg of mouse IgG or subjecting 0.96 ng of gpl20 to identical conditions except for the addition of antibody. After overnight incubation, the solutions described above were incubated with 100 μl of goat anti-mouse IgG agarose beads (Sigma) overnight at 4°C. After centrifugation, the supernatants were used for cytotoxicity assays. Example 2

This example demonstrates that gpl20 toxicity involves glutamate receptors and calcium.

In retinal ganglion cell cultures, NMD A antagonists block gpl20 toxicity (Lipton, S.A., Sucher, N.J., Kaiser, P.K. & Dreyer, E.B. (1991) Neuron 7, 111-118). i cortical cultures, the NMDA antagonist MK801 (10 μM) reduces gρl20 (100 pM)-induced toxicity by 70% (Table 1). 6,7-Dinitroquinoxaline-2,3- dione (DNQX), which blocks non-NMDA glutamate receptors, reduces neurotoxicity to a similar extent as MK801 (purchased from Research Biochemicals, Natick, MA). The combination of MK801 and DNQX results in even greater protection, reducing neurotoxicity by >90%. gpl20 toxicity in retinal ganglion cells is diminished by the L-voltage- dependent calcium channel antagonists nifedipine and nimodipine as well as by calcium-free medium. (Dreyer, E.B., Kaiser, P.K., Offermann, J.T. & Lipton,

S.A. (1990) Science 248, 364-367) and (Lipton, S.A., Sucher, N.J., Kaiser, P.K. & Dreyer, E.B. (1991) Neuron 7, 111-118)). We observe a 50% reduction in gpl20 (100 pM) toxicity in cortical cultures with 100 μM nifedipine and complete protection when calcium is omitted from the CSS medium. Release of calcium from intracellular calcium pools may also contribute to neurotoxicity. In cultured cortical neurons, dantrolene inhibits glutamate neurotoxicity as well as increases intracellular calcium that are both dependent and independent of external calcium (Frandsen, A. & Schousboe, A. (1992) Proc. Natl. Acad. Sci. USA 89, 2590- 2594) and (Lei, S.Z., Zhang, D., Abele, A.E. & Lipton, S.A. (1992) Brain Res. 598, 196-202)). Dantrolene (30 μM) reduces gpl20 (100 pM) neurotoxicity nearly to control values (Table 1). Since inositol phospholipid turnover increases intracellular Ca ²⁺ , we monitored effects of gpl20 on inositol phospholipid turnover in cortical slices (Godfrey, P.P. (1989) Biochem. J. 258, 621-624) but observed no changes.

Table l

Inhibition of gpl20 neurotoxicity by glutamate receptor antagonists and calcium channel blockers

Cell death %

100 pM gpl20 + 25 μM glutamate 51.0 +_ 4.2

+ 10 μM MK801 16.2 ± 4.7*

+ 100 μM DNQX 21.7 ± 6.7*

+ 10 μM MK801 + 100 μM DNQX 5.5 ± 5.7* + 100 μM nifedipine 30.8 ± 6.2*

+Ca ⁺ free medium 8.1 +_ 5.3*

+30 μM dantrolene 6.0 +_ 3.9*

Data are means +_ SEM (h=8-20). Cell death was determined by 0.4% trypan blue exclusion by viable cells (see text). Significance was determined by Student's t test for independent means. *P< 0.001

Inositol Phospholipid Turnover. Inositol phospholipid turnover was assayed as described (Godfrey, P.P. (1989) Biochem. J. 258, 621-624. Rat brain cortical slices (400 x 400 μm) were prepared, allowed to recover (Mourey, R.J., Dawson, T.M., Barrow, R.K., Enna. A.E., & Snyder, S.H. (1992) Mol.

Pharmacol. 42, 619-626), and then incubated for 1 hour in Krebs/Hepes buffer containing 2-[ ³ H]cytidine at 0.4 μCi/ml (1 Ci = 37 GBq (DuPont/NEN)). Lithium chloride (final concentrations, 10 mM) was then added followed 10 minutes later by gpl20 (100 pM). The reaction was terminated after 1 hour of gpl20 exposure. [ ³ H]cytidine diphosphate diacylglycerol ([ ³ H]CDP-DAG) was extracted and radioactivity was counted. Simultaneous experiments were performed with carbachol (1 mM). Typical ratios of carbachol-stimulated to basal [ ³ H]CDP-DAG were 10: 1.

Example 3

This example demonstrates that gpl20 neurotoxicity involves NO. Cortical cultures were depleted of endogenous arginine by incubating them 20-24 hours before addition of gpl20 in MEM with arginine deleted and glutamine added to block arginine synthesis (Sessa, W.C., Hecker, M., Mitchell, J.A. &

Vane, J.R. (1990) Proc. Natl. Acad. Sci. USA 87, 8607-8611). In arginine-free medium, gpl20 toxicity is reduced by nearly 70% (Table 2). Nitroarginine (100 μM), a potent inhibitor of NO synthase (NOS), reduces cell death by 70%. Protection by nitroarginine is reversed by L-arginine (1 mM). NOS contains tightly bound flavin groups βredt, D.S., Ferris, CD. & Snyder, S.H. (1992) J.

Biol. Chem. 267, 10976-10981). Flavm-containing enzymes are potently inhibited by DPI (diphenyleneiodonium purchased from Kodak). DPI (500 nM) blocks gpl20 toxicity by 90%. NO that passes between cells can be captured by hemoglobin, as NO binds with high affinity to iron in heme. Hemoglobin (0.5 mM) reduces gpl20 toxicity «90%. (Reduced hemoglobin was prepared by the method of Martin et al (Martin, W., Villani, G.M., Jothianandan, D. & Furchgott, R.F. (1985) J. Pharmacol. Exp. Ther. 232, 708-716).)

NO toxicity may result from the combination of NO with superoxide to form peroxynitrite, which degenerates to hydroxyl and nitrogen dioxide (NQj) free radicals, which are highly reactive (Beckman, J.S., Beckman, T.W., Chen, J.,

Marshall, P.A. & Freeman, B.A. (1990) Proc Natl. Acad. Sci. USA 87, 1620- 1624; Radi, R., Beckman, J.S., Bush, K.M. 8c Freeman, B.A. (1991) J. Biol. Chem. 266, 4244-4250; and Radi, R., Beckman, J.S., Bush, K.M. 8c Freeman, B.A. (1991) Arch. Biochem. Biophys. 288, 481-487). To remove superoxide anions, we treated cultures with SOD (Table 2). SOD blocks toxicity to a slightly greater extent than nitroarginine. Coapplication of SOD and nitroarginine produces a further reduction in toxicity.

Microglia, astrocytes, and macrophages can produce NO (Nathan, C. (1992) FASEB J. 6, 3051-3064; Chao, C.C., Hu, S., Molitor, T.W., Shaskan, E.G. & Peterson, P.K. (1992) 7. Immunol. 149, 2736-2741; Simmons, M.L. 8c

Murphy, S. (1992) J. Neurochem. 59, 897-905; and Galen, E., Feinstein, D.L.

& Reis, D.J. (1992) Proc. Natl. Acad. Sci. USA 89, 10945-10949). To differentiate between NO formed by neurons versus nonneuronal cells, we previously took advantage of the differential sensitivity of NOS-containing neurons to various glutamate derivatives. Although markedly resistant to NMDA toxicity, NOS neurons are uniquely sensitive to the toxic affects of quisqualate (Dawson,

V.I., Dawson, T.M., Bartley, D.A., Uhl, G.R. & Snyder, S.H. (1993) J. Neurosci. U.-2651-2661; and Koh, J.-Y. 8c Choi, D.W. (1988) /. Neurosci. 8, 2153-2163). In cortical cultures, quisqualate (20 μM) kills >90% of NOS neurons but only 15-20% of the total neuronal population (Dawson, V.I., Dawson, T.M., Bartley, D.A., Uhl, G.R. & Snyder, S.H. (1993) J. Neurosci.,

7J.-2651-2661). Accordingly, we treated our cultures with 20 μM quisqualate for 5 minutes and 24 hours later exposed them to 100 pM gpl20 (Table 2). Quisqualate pretreatment reduces gpl20 toxicity by 65%, implying that NOS neurons are the primary source of NO in cortical cultures that mediates gpl20 toxicity.

Table 2

Modulation of gpl20 neurotoxicity by NOS inhibitors or SOD

Cell death %

100 pM gpl20 + 25 μM glutamate 51.0 +_ 4.2

+L-Arg-free MEM 16.8 ± 5.6*

+ 100 μM N-Arg 16.5 ± 6.7*

+ 100 μM N-Arg + 1 mM L-Arg 50.2 ± 6.5*

+500 nM DPI 4.7 ± 4.3* +500 μM Hb 12.6 ± 6.5*

+ 100 units of SOD 9.3 +_ 8.0*

+ 100 μM N-Arg + 100 units of SOD 5.3 ± 4.4*

+20 μM Quis pretreatmeήt 24.1 +_ 6.0*

Data are means +. SEM (n«8-20). Significance was determined by Student's t test for independent means. N-Arg, nitroarginine; Quis, quisqualate; SOD, superoxide dismutase. *P< 0.001

Example 4

This example demonstrates that gpl20 influences cGMP levels via NMDA (N-methyl-D-aspartate) receptors and NO.

NO enhances cGMP levels in cortical cultures and numerous other tissues by stimulating guanylyl cyclase activity (Dawson, T.M., Dawson, V.L. & Snyder, S.H. (1992) Ann. Neurol. 32, 297-311; Snyder, S.H. (1992) Science 257,

494-496; and Garthwaite, J. (1991) Trends Neurol. Sci. 14, 60-67). If gρl20 were to stimulate NOS activity, one would anticipate elevation of cGMP levels. In cortical cultures gpl20 increases cGMP levels 3- to 4-fold (Table 3). The increase is completely abolished by nitroarginine and nearly abolished by N-methylarginine,

another selective inhibitor of NOS. Hemoglobin also blocks the increase in cGMP levels. MK801 prevents gpl20 enhancement of cGMP levels. Whereas SOD

(superoxide dismutase) blocks gpl20-induced neurotoxicity, it augments gpl20 stimulation of cGMP formation so that coapplication of gpl20 and SOD elevates cGMP levels 9-fold (Table 3). SOD removes superoxide anion that combines with

NO so that more NO is available to stimulate guanylyl cyclase. In previous studies, SOD increased NMDA stimulation of cGMP levels only 30-50% in cortical cultures (Dawson, V.L., Dawson, T.M., London, E.D., Bredt, D.S. & Snyder, S.H. (1991) Proc. Natl. Acad. Sci. USA 88, 6368-6371). The greater stimulation of cGMP levels by gpl20 implies that gpl20 may increase superoxide formation more than NMDA.

Table 3 cGMP formation after stimulation by gpl20 in the absence or presence of glutamate receptor antagonist, NOS inhibitors, and SOD

cGMP % basal

100 pM gpl20 345 ± 146 + 100 μM N-Arg 16.9 +_ 1.5

+500 μM NMA 133 ± 27.2

+10 μM MK801 98.5 ± 6.3

+500 μM Hb 13.5 +_ 0.5

+ 100 units of SOD 900 + 262

Data are means +_ SEM (n=6-8). N-arg, nitroarginine, NMA, N-methylarginine.

cGMP Assay. The formation of cGMP in primary cortical cultures was determined by a RIA (Amersham). Neuronal cultures were washed three times with CSS. This was followed by application of the test solutions for 5 minutes in

the presence of 100 μM isobutylmethylxanthine to inhibit phosphodiesterases. The reaction was stopped with 15% trichloroacetic acid followed by ether extraction. cGMP levels were measured according to the manufacturer's instructions.

We have shown that NO plays a role in gpl20 toxicity. Depletion of arginine from the incubation medium blocks toxicity, as does the NOS inhibitor nitroarginine. DPI, which inhibits NOS by binding to its flavin groups, also prevents toxicity. Hemoglobin, which binds extracellular NO, blocks neurotoxicity, implying that NO involved in mediating neurotoxicity passes between cells. The blockade of toxicity by SOD presumably involves the removal of superoxide anions, which can combine with NO to form peroxynitrite that degenerates into the toxic hydroxyl and NO ₂ free radicals (Beckman, J.S., Beckman, T.W., Chen, J., Marshall, P. A. & Freeman, B.A. (1990) Proc. Natl. Acad. Sci. USA 87, 1620-1624; Radi, R., Beckman, J.S., Bush, K.M. & Freeman, B.A. (1991) J. Biol. Chem. 266, 4244-4250; and Radi, R., Beckman, J.S., Bush, K.M. & Freeman, B.A. (1991) Arch. Biochem. Biophys. 288, 481-487).

Mediation of gpl20 toxicity by NO implies that gpl20 stimulates NO synthesis. The enhancement of cGMP levels by gpl20 and its attenuation by NOS inhibitors further demonstrates the role of NO in gpl20 neurotoxicity.

Based on these findings, a model is proposed which accounts for gpl20- induced neurocytotoxicity. See Figure 2. According to the model, the HIV-l coat protein, gpl20, which is shed by the virus, elicits neurotoxicity by interacting with macrophages/microglia and astrocytes to release cytokines and/or arachidonic acid metabolites (Genis, P., Jett, M., Bernton, E.W., Boyle, T., Gelbard, H.A., Dzenko, K., Keane, R.W., Resnick, L., Mizrachi, Y., Volsky, D.J., Epstein, L.G. & Gendelman, H.E. (1992) J. Exp. Med. 176, 1703-1718; and Wahl, L.M.,

Corcoran, M.L., Pyle, S.W., Arthur, L.O., Harel-Bellan, A. & Farrar, W.L. (1989) Proc. Natl. Acad. Sci. USA 86, 621-625). These cytokines and/or arachidonic acid metabolites may act synergistically with glutamate to activate NMDA receptors, which increase intracellular calcium levels. NOS is subsequently activated.