of January 2, 2019.This information is current as

Neuropathogenesis of HIV-1 InfectionDementia Complex: Implications for the Increased Peroxynitrite Activity in AIDS

S. L. M. NottetGray, Jan Verhoef, Peter Portegies, Marc Tardieu and Hans Leonie A. Boven, Lucio Gomes, Christiane Hery, Françoise

http://www.jimmunol.org/content/162/7/43191999; 162:4319-4327; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/162/7/4319.full#ref-list-1

, 20 of which you can access for free at: cites 49 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 1999 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on January 2, 2019

http://ww

w.jim

munol.org/

Dow

nloaded from

Increased Peroxynitrite Activity in AIDS Dementia Complex:Implications for the Neuropathogenesis of HIV-1 Infection1

Leonie A. Boven,2* Lucio Gomes,* Christiane Hery,† Francoise Gray,‡ Jan Verhoef,*Peter Portegies,§ Marc Tardieu, † and Hans S. L. M. Nottet3*

Oxidative stress is suggested to be involved in several neurodegenerative diseases. One mechanism of oxidative damage is mediatedby peroxynitrite, a neurotoxic reaction product of superoxide anion and nitric oxide. Expression of two cytokines and two keyenzymes that are indicative of the presence of reactive oxygen intermediates and peroxynitrite was investigated in brain tissue ofAIDS patients with and without AIDS dementia complex and HIV-seronegative controls. RNA expression of IL-1b, IL-10, in-ducible nitric oxide synthase, and superoxide dismutase (SOD) was found to be significantly higher in demented compared withnondemented patients. Immunohistochemical analysis showed that SOD was expressed in CD68-positive microglial cells whileinducible nitric oxide synthase was detected in glial fibrillary acidic protein (GFAP)-positive astrocytes and in equal amounts inmicroglial cells. Approximately 70% of the HIV p24-Ag-positive macrophages did express SOD, suggesting a direct HIV-inducedintracellular event. HIV-1 infection of macrophages resulted in both increased superoxide anion production and elevated SODmRNA levels, compared with uninfected macrophages. Finally, we show that nitrotyrosine, the footprint of peroxynitrite, wasfound more intense and frequent in brain sections of demented patients compared with nondemented patients. These resultsindicate that, as a result of simultaneous production of superoxide anion and nitric oxide, peroxynitrite may contribute to theneuropathogenesis of HIV-1 infection. The Journal of Immunology,1999, 162: 4319–4327.

H uman immunodeficiency virus (HIV) infection of thecentral nervous system leads to severe neurologicalcomplications in about 25% of adults and half of chil-

dren with AIDS (1, 2). Viral replication in the brain, which, in-triguingly, occurs only in macrophages and microglia and not inneurons (3–5), results in dendritic pruning, simplification of syn-aptic contacts, and frank cell loss. This suggests an indirect mech-anism to be the cause of neuronal damage. Indeed, HIV-inducedintracellular events in macrophages are shown to result in the se-cretion of several neurotoxins (6–8).

One of the neurotoxins that is suggested to be involved in neu-ronal damage is nitric oxide (NO)4 (9). The proinflammatory cy-tokine IL-1b, which is released in brain tissue of demented AIDSpatients (10–12), has been shown to induce NO by up-regulatinginducible nitric oxide synthase (iNOS) (13–15). Indeed, coculturesof HIV-infected macrophages and astrocytes were shown to re-lease NO when compared with cocultures of uninfected macro-

phages and astrocytes (16), suggesting that interactions betweenHIV-infected macrophages/microglia and astrocytes are critical inthe induction of factors that lead to neuronal injury (17, 18). Fur-thermore, it has been shown that iNOS, as well as NO, was in-duced in primary cultures of rat mixed brain cells in response tostimulation with gp41 (19).

Recently, evidence has been presented that direct neurotoxiceffects of NO are modest but are tremendously enhanced by re-acting with superoxide anion to form peroxynitrite (20–24). Su-peroxide anion is reported to be produced by myeloid-monocyticcell lines upon HIV-1 infection (25), and, to keep the concentrationof this reactive free radical low, superoxide dismutase (SOD), asuperoxide anion scavenger, is produced (26). Cytosolic copper-zinc SOD (CuZnSOD) is responsible for degrading reactive su-peroxide anion by catalyzing the dismutation of superoxide anioninto molecular oxygen and hydrogen peroxide, therefore playingan important role in the defense mechanisms against oxidativestress (26). However, NO reacts with superoxide anion at a neardiffusion-limited rate, and, when present in large amounts, it there-fore outcompetes SOD completely (21). The reaction product per-oxynitrite is a potent oxidizer that is responsible for the nitration oftyrosine residues of structural proteins (21). Neurofilament, whichis a protein that provides structural stability to neurons, is one ofthe target proteins of peroxynitrite, and nitration will result in dis-rupted neurofilament assembly and thus neuronal damage (21, 26).Interestingly, this reaction is catalyzed by SOD, the enzyme thatnormally scavenges an excess of superoxide anion and therebyprevents the formation of peroxynitrite (21, 26, 27). This mightexplain the paradoxical neurotoxicity found in transgenic miceoverexpressing extracellular SOD activity (28, 29).

Since the combination of NO and superoxide anion results in thegeneration of highly neurotoxic peroxynitrite, we here investigatetheir role in AIDS dementia complex. Levels of iNOS and SOD,two of the key enzymes in oxidative stress that are indicative of thepresence of NO and superoxide anion, respectively, were studied

*Eijkman-Winkler Institute, Section of Neuroimmunology, Utrecht University, Utre-cht, The Netherlands;†Laboratoire Universitaire “Virus, neurone et immunite,” Uni-versite Paris-Sud, Paris, France;‡Laboratory of Neuropathology, Faculte´ de MedecineParis-Ouest, Garches, France;§Department of Neurology, Academic Medical Center,University of Amsterdam, Amsterdam, The Netherlands

Received for publication June 30, 1998. Accepted for publication January 11, 1999.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisementin accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by grants from the Netherlands Ministry of Public Health(S 96035/1234) to H.S.L.M.N. as part of the stimulation program on AIDS researchof the Dutch Program Committee for AIDS research, and the Royal NetherlandsAcademy of Sciences and Arts.2 Address correspondence and reprint requests to Dr. Leonie A. Boven, Eijkman-Winkler Institute, Section of Neuroimmunology, AZU, hp G04.614, Heidelberglaan100, 3584 CX Utrecht, The Netherlands. E-mail address: L.A.Boven@lab.azu.nl3 H.S.L.M.N. is a fellow of the Royal Netherlands Academy of Sciences and Arts.4 Abbreviations used in this paper: NO, nitric oxide; iNOS, inducible NO synthase;GFAP, glial fibrillary acidic protein; RFU, relative fluorescence unit; SOD, superoxidedismutase; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; NT, nitrotyrosine.

Copyright © 1999 by The American Association of Immunologists 0022-1767/99/$02.00

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

in brain tissue of demented and nondemented AIDS patients. Inaddition, we examined the localization of these two enzymes bydouble immunohistochemical staining on brain slices of dementedAIDS patients. To confirm our in vivo results, we investigatedwhether HIV-1 infection of macrophages in vitro also resulted inchanges in oxidative processes, by looking at superoxide anionproduction and SOD mRNA expression. Finally, immunohisto-chemical staining for nitrotyrosine, a footprint for peroxynitrite,was performed to investigate whether peroxynitrite was present inthe brains of demented and nondemented AIDS patients.

Materials and MethodsBrain tissue

Tissue specimens of the frontal cortex was obtained from autopsied brainof thirteen HIV-1-infected and five control cases (those who died of causesnot related to HIV-1 infection). All the HIV-1-infected individuals haddeveloped AIDS at the time of death and showed decreased levels of CD41

T cells (,300). Seven of them developed cognitive and motor impair-ments. None of the patients received any antiretroviral therapy. The clinicaldata on all individual patients are shown in Table I. AIDS or HIV-associ-ated dementia was a premortem clinical diagnosis made by an AIDS spe-cialty physician or neurologist. The severity of dementia was scored on theMemorial Sloan-Kettering scale. All demented patients scored at least 2 orhigher. In addition, from all patients, the neurological status was deter-mined retrospectively. Clinical premortem diagnoses were confirmed bypostmortem obduction and by staining and neuropathological examinationof frozen sections of brain tissue. The diagnosis HIV encephalitis wasmade when p24 positive multinucleated giant cells were observed. The lossof brain tissue (atrophy) was scored using the computed tomography (CT)scan. Furthermore, all patients showed a wide variety of opportunistic in-fections (Table I) and died because of various reasons. All patients werewell characterized at the time of death, and the cause of death was neverdirectly associated with diseases of the central nervous system. Viral levelswere measured by RT-PCR and showed a high degree of variation amongall AIDS patients (data not shown). There was no correlation between thepresence or absence of high viral levels and the stage of disease. Whenthere were no differences between brain tissue of the studied patients andnormal brain tissue, it was scored as “no significant difference.”

Frontal cortex specimens were stored at270°C until RNA isolation wasperformed. From adjacent regions, 6-mm sections were made, fixated incold acetone, and stored at270°C until immunostaining was performed.Sections from each individual were stained with hematoxylin and eosin(H&E) and analyzed by a neuropathologist. The neuropathological findingsare given in Table I.

HIV-1 infection of macrophages

PBMC were isolated from heparinized blood from HIV-1-, HIV-2-, andhepatitis B-seronegative donors and obtained on Ficoll-Hypaque densitygradients. Cells were washed twice, and monocytes were purified by coun-tercurrent centrifugal elutriation. Cells were.98% monocytes by criteriaof cell morphology on May-Grunwald-Giemsa-stained cytosmears and bynonspecific esterase staining usinga-naphtylacetate (Sigma, St. Louis,MO) as substrate. Monocytes were cultured in suspension at a concentra-tion of 2 3 106 cells/ml in Teflon flasks (Nalgene, Rochester, NY) inDMEM with 10% heat-inactivated human AB serum negative for anti-HIVAbs, 10 mg/ml gentamicin, and 10 mg/ml ciprofloxacin (Sigma). As pre-viously described, HIV-1 infection of nonadherent macrophages, espe-cially when using a low multiplicity of infection, appears much more re-producible than infection of macrophages that were first allowed to adhere(30). After 7 days, monocyte-derived macrophages (MDM) were recoveredfrom the Teflon flasks and infected with HIV-1Ba-L at a multiplicity ofinfection of 0.01. Two hours later, macrophages were removed from theTeflon flasks, washed twice to remove unbound virus, and used for studiesto determine levels of superoxide anion production by chemiluminescenceand SOD expression by RNA PCR in HIV-infected macrophages.

Chemiluminescence

Cells were recovered from Teflon flasks and washed twice with HBSScontaining 1% FCS (Life Technologies, Grand Island, NY). A total of 23105 cells was exposed to an equal volume of HBSS in the presence of 250nM bis-N-methylacridinum (Lucigenin, Sigma). Superoxide dismutase(SOD, Sigma) was added to HIV-infected macrophages to demonstratethat chemiluminescence was superoxide anion specific. Lucigenin reactswith superoxide anion (31), and this reaction is accompanied by photonemission. The number of photons emitted during stimulation of themacrophages was measured as light emission in a luminometer (PackardInstruments, Brussels, Belgium) and expressed as cpm, as describedpreviously (32).

RNA PCR detection of cytokines and enzymes

Brain tissue and 23 106 macrophages were homogenized and lysed, re-spectively, in 1 ml and 0.4 ml TRIzol (Life Technologies) according to themanufacturer’s guidelines. In experiments where the levels of SOD ex-pression of macrophages were determined, the lysed cells were stored inTrizol at270°C. When lysates of all time points were obtained, total RNAwas isolated. Total RNA was dissolved in diethylpyrocarbonate (DEPC)-treated water, and 1mg of RNA was used for the synthesis of complemen-tary DNA. The RNA was previously heated for 5 min at 70°C, chilled onice, and added to a mixture containing 13reverse transcriptase (RT) buffer(Promega, Madison, WI), 200 U of reverse transcriptase, 0.1 M DTT (LifeTechnologies), 2.5 mM dNTP (Boehringer Mannheim, Indianapolis, IN),80 U random hexamer oligonucleotides (Boehringer Mannheim), and 10 U

Table I. Clinical and neuropathological data

PatientAge (yr)/

SexPostmortemDelay (h) Clinical Status MSKa Neuropathologic Findings

1 26/M 2 AIDS 0 No significant changes2 36/M 24 AIDS 0 HIV encephalitis3 44/M 2 AIDS 0 Cerebral toxoplasmosis4 34/M 6 AIDS 0 Cerebral toxoplasmosis/cryptococcoses5 46/M 24 AIDS 0 Cerebral toxoplasmosis6 33/M ND AIDS 0 Cryptococcoses, toxoplasmoses, CMV7 36/M ND AIDS 3 Major atrophy8 55/M 12 AIDS 3 Atrophy9 39/M 4 AIDS 3 HIV encephalitis

10 48/M 2 AIDS 2 Atrophy11 62/M 12 AIDS 3 Major atrophy12 37/M ND AIDS 2 Minor atrophy13 40/M 30 AIDS 3 Major atrophy, HIV encephalitis14 71/F 18 Seronegative No significant changes15 71/M ND Seronegative Diffuse gliosis16 81/F 6 Seronegative No significant changes17 38/M 9 Seronegative No significant changes18 56/F 9 Seronegative No significant changes

a HIV-associated dementia was a premortem clinical diagnosis made by a neurologist. The severity of dementia was scored on the MemorialSloan-Kettering (MSK) scale. Patients 1–6 and 14–18 did not suffer from motor or cognitive impairments. ND, not determined.

4320 INCREASED PEROXYNITRITE ACTIVITY IN AIDS DEMENTIA COMPLEX

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

RNAsin (Promega). The complete mixture was now incubated for 60 minat 37°C and then heated for 5 min at 90°C. The final reaction volume wasdiluted 1:8 by adding distilled water. Amplification of the cDNA was ac-complished using one primer biotinylated on the 59terminal nucleotide tofacilitate later capture using streptavidin. The PCR primer pair was chosento span at least one intron. To the PCR reaction mixture the followingcomponents were added: 0.25 mM dNTP mix (Boehringer Mannheim),1 3 PCR buffer (50 mM KCl, 10 mM Tris-HCl, 1.5 mM MgCl2; Promega),0.2 mM sense and antisense primers (Table II), 5ml cDNA, and 1 UTaqpolymerase (Promega). Denaturation, annealing, and elongation tempera-tures for PCR were 94°C, 60°C, and 72°C for 1, 1, and 2 min each, usinga DNA thermal cycler (Perkin-Elmer, Norwalk, CT). Negative controlswere included in each assay to confirm that none of the reagents werecontaminated with cDNA or previous PCR products. PCR was also per-

formed on RNA samples to exclude genomic DNA contamination. To con-firm single band product, positive reactions were subjected to 40 cyclesamplification and electrophoresis, followed by ethidium bromide staining.Then, for semiquantification, every primer pair was tested at different cyclenumbers to determine the linear range. GAPDH, SOD, and IL-1b mRNAlevels were high, and 20 cycles was enough to measure the PCR productin its linear range, whereas iNOS cDNA had to be subjected to 30 cyclesand IL-10 even to 40 cycles to be in the linear range.

Aliquots of 5 ml of the biotinylated PCR product were semiquantita-tively analyzed using a fluorescent digoxigenin detection ELISA kit(Boehringer Mannheim) according to manufacturer’s protocol. In short, thebiotinylated strand of denatured PCR product is captured by immobilizedstreptavidin. Then, a digoxigenin-labeled probe is added, followed by analkaline phosphatase-labeled Ab against digoxigenin. After addition of the

FIGURE 1. Cytokine and enzyme mRNA levels in the frontal cortex of postmortem brain tissue of HIV-infected nondemented patients (H), HIV-infected demented patients (HD), and seronegative control patients (C) expressed as RFU. Significantly elevated gene expression for IL-1b (A, p , 0.025),IL-10 (B, p , 0.01), iNOS (C, p , 0.05), and SOD (D, p , 0.01) was found in demented patients compared with nondemented HIV-infected patients.p values were calculated using a Student’st test. Results are representative of at least three independent PCR experiments.

Table II. Sequences of the oligonucleotide primers and probes in RT-PCR

Target(product size) Primers Sequence

GAPDH (195 bp) Sense CCATGGAGAAGGCTGGGGAntisense CAAAGTTGTCATGGATGACCProbe CTGCACCACCAACTGCTTAGC

IL-1b (328 bp) Sense GCATCCAGCTACGAATCTCCGACCAntisense CACTTGTTGCTCCATATCCTGTCCCProbe GGACCAGACATCACCAAGCTTTTTTGCTG

IL-10 (328 bp) Sense AAGCTGAGAACCAAGACCCAGACATCAAGGCGAntisense AGCTATCCCAGAGCCCCAGATCCGATTTTGGProbe GCCTGAGGGTCTTCAGGTTCTCCCCCAGGG

SOD (236 bp) Sense AGGACTGACTGAAGGCCAntisense CCAATGATGCAATGGTCTCCProbe GATGAAGAGAGGCATGTTGG

iNOS (236 bp) Sense ACTTTGATCAGAAGCTGTCCCAntisense CAAAGGCTGTGAGTCCTGCACProbe CTGTGAGACGTTTGATGTCC

HIV-1 tat/rev (123 bp) Sense GGCTTAGGCATCTCCTATGGCAntisense TGTCGGGTCCCCTCGTTGCTGGProbe CTTTGATAGAGAAACTTGATGAGTCTG

4321The Journal of Immunology

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

substrate, fluorescence was measured in relative fluorescence units (RFU)in a fluorescence multiwell plate reader (Perseptive Biosystems, Framing-ham, MA) at excitation 450 nm/emission 550 nm. All data were normal-ized against GAPDH mRNA level, which was used as an internal standard.

Statistical analysis

Data were compared, and a two-tailed Student’st test was used to deter-mine p values.

Immunohistochemical analysis of brain tissue

Frozen sections of brain tissue were analyzed for SOD, iNOS, nitrotyrosine(NT), and HIV p24 Ag expression. Brain slices were first incubated for18 h at 4°C with the first Ab (anti-SOD human liver, Calbiochem, SanDiego, CA; anti-iNOS mac NOS, Transduction Laboratories, Lexington,KY; anti-HIV-1 p24, Dupont-NEN, Boston, MA; anti-NT polyclonal, Up-state, Biotechnology, Lake Placid, NY). Astrocytes and microglial cellswere stained with anti-glial fibrillary acidic protein (GFAP; AmershamLife Science, Rainham, England) and anti-human phagocyte macrophage/microglia CD68/Ki-M7 (BMA, Valbiotech, France), respectively. Thebinding was subsequently revealed after another incubation of 45 min at18°C with a corresponding alkaline phosphatase conjugated anti-IgG Aband fast red substrate (Boehringer Mannheim).

ResultsExpression of IL-1b, IL-10, iNOS, and SOD mRNA in braintissue of demented and nondemented HIV patients and controlpatients

A semiquantitative fluorescence assay was used to study the levelsof expression of IL-1b, IL-10, iNOS, and SOD in brain tissue ofthe frontal cortex of the individual patients described in Table I.The mRNA levels of all gene products, expressed in RFU, aredepicted in Fig. 1,A-D. To detect all PCR products in their linearrange, cDNA was subjected to 30, 40, 20, and 20 cycles for iNOS,IL-10, SOD, and IL-1b, respectively, indicating that high levels ofIL-1b and SOD were present in all patients. IL-1b, a proinflam-matory cytokine, was detected in the HIV-demented group at sig-nificantly higher levels than in the group of nondemented HIVpatients (p , 0.025) and in the control group (p , 0.025). Thisfinding confirms that, also in these demented patients, cerebralimmune activation seems to occur (10–12), which may eventuallyprove to be a crucial event in the neuropathogenesis of HIV-1infection (33–37).

Interestingly, IL-10 was also expressed at significantly higherlevels in the HIV-demented group compared with the groups ofcontrol (p , 0.01) and nondemented HIV patients (p , 0.01).

FIGURE 2. Detection of enzyme mRNAs in the frontal cortex of postmortem brain tissue of HIV-infected nondemented patients (h), HIV-infecteddemented patients (hd), and seronegative control patients (C) expressed as RFU. Although not significant, a trend between the levels of SOD (A) and iNOS(B) expression can be observed within the group of demented AIDS patients.

4322 INCREASED PEROXYNITRITE ACTIVITY IN AIDS DEMENTIA COMPLEX

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

SOD mRNA levels were also expressed significantly more in theHIV-demented patients compared with the control patients (p ,0.02) and the nondemented HIV-infected patients (p , 0.01). Fi-nally, iNOS expression was also detected at higher levels in thedemented HIV-infected patients when compared with nonde-mented HIV-infected patients (p , 0.05), but not to the controlgroup. When the individual levels of enzyme RNAs were com-pared, a trend between the levels of SOD and iNOS can be ob-served in the group of demented AIDS patients (Fig. 2). With theexception of case 12, it was found that the moderate to high levelsof SOD in patients 7, 8, 9, and 11 (Fig. 2A) correspond to moderateand high levels of iNOS (Fig. 2B). This trend is not observed in thenondemented patients or in the control group. Brain specimens ofmost nondemented patients even do not have detectable SOD lev-els. In the two cases (cases 3 and 6) where SOD expression wasdetected, only in one (case 3) iNOS was found. Thus, these datasuggest that the presence of both SOD and iNOS seems to behighly associated with the occurrence of clinical dementia.

Immunohistochemical localization of increased SOD and iNOSgene products in brain tissue of AIDS patients

The expression of SOD, iNOS, and HIV-1 p24 Ags was analyzedin frozen sections of frontal lobes of five of the studied patients(patients 1, 3, 7, 9, and 14). Expression of SOD and iNOS Ags wasdetected in patients 3, 7, and 9, whereas they were barely detect-able in patient 1 and control patient 14, a result that paralleled thatof the RNA detection method. Since staining patterns between thedifferent patients did not differ substantially, only single stains ofpatient 9 for SOD and iNOS are shown in Fig. 3,A and C, re-spectively. In general, SOD labeling was more intense than that ofiNOS, which was also observed by the RNA detection method.SOD was found in cells of both perivascular and parenchymalareas. To better define the SOD-positive cells, double immunohis-tochemical staining was performed on cases 3 and 9. SOD immu-noreactivity patterns did not differ for both patients, except forminor differences in staining intensity. Therefore, only the result ofthe double staining of brain tissue of patient 9 is shown in Fig. 3B.SOD expression was mostly localized in CD68-positive microglialcells in the parenchyma (Fig. 3B) and in the perivascular areaswhere the frequency of SOD-positive cells was less (data notshown). SOD staining was infrequent and faint in GFAP-positiveastrocytes (data not shown). To better define the iNOS-positivecells, double labeling experiments were performed on the samecases. By double labeling, we found that both CD68-positive mi-croglial cells (Fig. 3D) and GFAP-positive astrocytes (data notshown) expressed iNOS Ags with roughly the same frequency andintensity in patient 9, whereas iNOS reactivity was detectable butfaint in patient 3 (data not shown). Subsequently, double immu-nohistochemical staining was performed on case 9 with SOD andHIV-1 p24-specific mAbs. More than 60% of the p24 Ag-positivecells also contained SOD Ag (Fig. 3E). Since HIV-1 productivelyreplicates only in brain macrophages (3–5), these findings suggestthat SOD expression and possibly superoxide anion productionoccurred mostly in HIV-1-infected brain macrophages.

HIV-1 infection of macrophages results in increased SODexpression

Since elevated SOD levels were detected in brain macrophages intissue obtained from demented AIDS patients, the ability of HIV-1to affect SOD expression in macrophages was investigated in vitro.Therefore SOD expression in HIV-infected macrophages wascompared with replicate uninfected cultures by RT-PCR analysis.As shown in Fig. 4A, the expression of SOD mRNA in uninfectedmacrophage cultures decreased in time, whereas the HIV-infected

macrophages continued to express SOD. Importantly, the timecourse pattern of HIV mRNA levels was similar to that of SODmRNA levels (Fig. 4B). To confirm that HIV RNA expressionindeed led to HIV-1 production, the release of p24 Ag in the cul-ture supernatants was detected by ELISA (Fig. 4C).

FIGURE 3. Immunohistochemical staining of brain tissue sections ob-tained from an AIDS patient with AIDS dementia complex.A, Parenchy-mal microglial cells stained for SOD Ag (brown) in the frontal lobe ofpatient 9 (see Table I).B, Parenchymal microglial cells double-stained forCD68/KiM-7 Ag (red) and SOD Ag (brown) in the frontal lobe of patient9. C, Parenchymal microglial cells stained for iNOS Ag (brown) in thefrontal lobe of patient 9.D, Parenchymal microglial cells double-stainedfor CD68/KiM-7 Ag (brown) and iNOS Ag (red dots) in the frontal lobe ofpatient 9.E, Parenchymal cells expressing both p24 Ag (red) and SOD(brown) in the frontal lobe of patient 9 identified by arrowheads.

4323The Journal of Immunology

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

HIV-1 infection of macrophages results in enhanced superoxideanion production

To investigate whether the relative increased levels of SOD ex-pression in HIV-infected macrophages could be a consequence ofincreases in superoxide anion production, the production of super-oxide anion by HIV-1-infected macrophages was compared withthat of uninfected macrophages using chemiluminescence. Imme-diately after HIV-1 infection, the ratio of the amount of superoxideanion production between HIV-infected and uninfected macro-phages measured after 30 min was 1.2. (Fig. 5A). Four days afterHIV-1 infection, the amount of superoxide anion production byHIV-infected macrophages increased when compared with that ofthe control macrophages (Fig. 5B). The 30-min ratio of the amountof superoxide anion production between the virus-infected andcontrol cells was 1.8. Eight days after viral inoculation of the mac-rophages, the amount of superoxide anion production by HIV-infected macrophages was substantially increased, when compared

with that of the uninfected cells (Fig. 5C). The ratio of the amountof superoxide anion production between HIV-infected and unin-fected macrophages measured after 30 min was 2.9. To demon-strate that the chemiluminescence signal was indeed superoxideanion specific, we added SOD, resulting in a dose-dependent de-crease of the signal (Fig. 6). The chemiluminescence signal wascompletely abolished when 100mg/ml SOD was added (data notshown), indicating that, in our experiments, lucigenin indeed reactsspecifically with superoxide anion.

Brain sections of patients suffering from AIDS dementia complexshow intense parenchymal and perivascular NT staining

Since the reaction between NO and superoxide anion results in theformation of peroxynitrite and both these molecules appear to bepresent in the brains of demented AIDS patients, elevated levels ofperoxynitrite are to be expected. Therefore, brain sections of pa-tients 4, 9, and 11 were stained for NT, which is observed whenlarge amounts of peroxynitrite were produced. Whereas the non-demented AIDS patient 4 did not show any substantial reactivitywith the NT polyclonal, the demented AIDS patient 11 showedstrong staining for NT parenchymal as well as perivascular (Fig.7). In addition, patient 9 stained heavily for NT (data not shown)although less than patient 11. Interestingly, patient 9 showed lower

FIGURE 4. Kinetic analysis of SOD and HIV-1 mRNA levels and viralp24 production in macrophages at different time points after HIV-1 infec-tion. A, SOD mRNA levels in HIV-infected macrophages (open symbols)and in mock-infected macrophages (filled symbols).B, Expression ofHIV-1 tat/rev RNA in HIV-infected macrophages.C, Virus replicationmeasured by HIV-1 p24 Ag ELISA of culture supernatants.

FIGURE 5. Superoxide anion production by HIV-1-infected (opensymbols) and mock-infected (filled symbols) macrophages measured bychemiluminescence immediately (A), four days (B), and eight days (C)after viral inoculation of the cells.

4324 INCREASED PEROXYNITRITE ACTIVITY IN AIDS DEMENTIA COMPLEX

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

expression of iNOS and SOD than patient 11 (Fig. 2), suggestinga possible relation between the degree of iNOS/SOD expressionand the presence of nitrosylated proteins. As a control for nonspe-cific staining, the primary Ab was omitted. In addition, as a controlfor NT staining, the primary Ab was preincubated with NT. Bothcontrol experiments resulted in inhibition of staining (data notshown).

DiscussionIt is generally assumed that AIDS dementia complex is a diseasein which immune activation of glial cells plays an important role(10–12, 33–37). This study confirms earlier observations thatIL-1b mRNA levels are increased in brain tissue of dementedAIDS patients. Surprisingly, we also detected elevated expressionof IL-10 in these brain samples. IL-10 is a potent inhibitor ofcytokine secretion by macrophages/microglia and was not to beexpected in the group of patients with elevated IL-1b levels. How-ever, this further supports the concept of the existence of immuneactivation in the brains of demented AIDS patients, despite thepresence of endogenous antiinflammatory mediators.

In this study, it is demonstrated that, although the ability ofmacrophages to produce superoxide anion in vitro decreases intime, elevated amounts of superoxide anion are produced upon invitro HIV-1 infection of macrophages compared with control mac-rophages. Since macrophages and microglia function as a long-term reservoirs for HIV-1 (38), these cells apparently possess amechanism to protect themselves against the toxic effects of su-peroxide anion. Indeed, we show here that elevated levels of su-peroxide anion coincide with elevated levels of cytosolic copper-zinc SOD, an important intracellular scavenger of superoxideanion, in HIV-infected macrophages compared with control mac-rophages. Since these in vitro data demonstrate that HIV infectionof macrophages result in both increased superoxide anion produc-tion and in increased SOD expression, changes in SOD mRNAexpression in vivo may also be indicative of changes in superoxideanion production. In vivo, SOD mRNA levels were found to beelevated in demented AIDS patients, and immunohistochemicalanalysis of brain tissue revealed that SOD is localized mostly inHIV-infected brain macrophages. These data suggest that super-oxide anion production by HIV-infected macrophages may also beincreased in vivo.

SOD is known to be involved in several other neurodegenerativediseases, like amyotrophic lateral sclerosis, Down’s syndrome, and

Alzheimer’s disease (39–43). Although SOD can scavenge super-oxide anion, this reaction will not take place in the presence of NO.When produced in large amounts, iNOS is the only molecule thatcan effectively out-compete SOD for superoxide anion by gener-ating NO, a highly diffusible molecule that is able to react withsuperoxide anion to form peroxynitrite (20–24). Recently, iNOShas been shown to be involved in the pathogenesis of AIDS de-mentia complex (16, 19) and to be directly or indirectly responsi-ble for neuronal damage (9, 44, 45). In addition to the elevatedmRNA levels of SOD, we here also show that iNOS mRNA levelsare significantly elevated in brain tissue of demented AIDS pa-tients compared with nondemented AIDS patients. This suggeststhat, besides superoxide anion, levels of NO may be elevated aswell in brains of demented AIDS patients.

In general, cellular interactions between astrocytes and immune-activated macrophages/microglia are believed to be responsible forthe production of neurotoxic as well as neurotrophic factors (46–48). Although we show that in vivo microglia are able to expressiNOS, the ability of human macrophages to produce NO remainshighly controversial. Despite the presence of iNOS in human mac-rophages, the production of NO by these cells in vitro is presum-ably very low (49) or even absent (13, 15). Indeed, we were alsonot able to demonstrate any NO production or expression of iNOSby HIV-infected macrophages in vitro (data not shown). However,recently it was demonstrated that macrophages isolated from ac-tive multiple sclerosis lesions showed immunoreactivity for iNOSand were able to produce NO (50). In addition, it has been reportedthat IFN-a can induce iNOS and NO in human monocytes (51).This implicates that there might be a trigger involved in vivo thatis not present in vitro. NOS has been detected in primary humanastrocytes, and, for these cells, IL-1b is the key proinflammatory

FIGURE 6. Superoxide anion generated by HIV-infected macrophagesis scavenged by superoxide dismutase (SOD) in a dose-dependent way.Macrophages were exposed to the indicated concentrations of SOD, andchemiluminescence was measured after 30 min of incubation. Superoxideanion generation by macrophages was completely undetectable when 100mg/ml SOD was added (data not shown).

FIGURE 7. Immunohistochemical staining for NT on brain tissue sec-tions obtained from an AIDS patients with AIDS dementia complex (A)and a nondemented AIDS patient (B).

4325The Journal of Immunology

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

cytokine involved in the induction of this molecule (13, 14). In-terestingly, elevated levels of macrophage-derived IL-1b havebeen detected in brain tissue of demented AIDS patients (11, 12),suggesting that, in AIDS dementia complex, immune-activatedmacrophages are able to evoke the release of NO from astrocytes(Fig. 8). Together with the macrophage-derived superoxide anion,this astrocyte-derived NO may result in the formation of the highlyneurotoxic peroxynitrite. Importantly, we show that NT, the foot-print of peroxynitrite, is detected more frequently and more in-tensely in brain sections of demented AIDS patients comparedwith nondemented AIDS patients, indicating that peroxynitrite wasindeed present in the brains of these patients. In conclusion, neu-ronal damage and death may be the result of interactions betweenboth immune-activated microglia and astrocytes and the subse-quent production of combined toxic reactive oxygen intermediateslike peroxynitrite (Fig. 8). Thus, although HIV-1 replicates in mac-rophages, astrocytes might also participate in the neuropathogen-esis of HIV-1 infection.

AcknowledgmentsWe thank Dr. Corline de Groot for advice and assistance with the nitro-tyrosine immunohistochemistry, Bert Tigges and Machiel de Vos for out-standing technical support, and Dr. Marc E. Jones for critical reading of themanuscript.

References1. Price, R. W., B. Brew, J. Sidtis, M. Rosenblum, A. C. Scheck, and P. Cleary.

1988. The brain in AIDS: central nervous system HIV-1 infection and AIDSdementia complex.Science 239:586.

2. Navia, B. A., B. D. Jordan, and R. W. Price. 1986. The AIDS dementia complex.I. Clinical features.Ann. Neurol. 19:517.

3. Koenig, S., H. E. Gendelman, J. M. Orenstein, M. C. Dal Canto,G. H. Pezeshkpour, M. Yungbluth, F. Janotta, A. Aksamit, M. A. Martin, andA. S. Fauci. 1986. Detection of AIDS virus in macrophages in brain tissue fromAIDS patients with encephalopathy.Science 233:1089.

4. McGeer, P. L., S. Itagaki, B. E. Boyes, and E. G. McGeer. 1988. Reactive mi-croglia are positive for HLA-DR in the substantia nigra of Parkinson’s and Alz-heimer’s disease brains.Neurology 38:1285.

5. Wahl, S. M., J. B. Allen, N. McCartney Francis, M. C. Morganti Kossmann,T. Kossmann, L. Ellingsworth, U. E. Mai, S. E. Mergenhagen, andJ. M. Orenstein. 1991. Macrophage- and astrocyte-derived transforming growthfactorb as a mediator of central nervous system dysfunction in acquired immunedeficiency syndrome.J. Exp. Med. 173:981.

6. Giulian, D., K. Vaca, and C. A. Noonan. 1990. Secretion of neurotoxins bymononuclear phagocytes infected with HIV-1.Science 250:1593.

7. Nottet, H. S. L. M., M. Jett, C. R. Flanagan, Q. H. Zhai, Y. Persidsky, A. Rizzino,E. W. Bernton, P. Genis, T. Baldwin, J. Schwartz, C. J. LaBenz, andH. E. Gendelman. 1995. A regulatory role for astrocytes in HIV-1 encephalitis:an overexpression of eicosanoids, platelet-activating factor, and tumor necrosisfactor-a by activated HIV-1-infected monocytes is attenuated by primary humanastrocytes.J. Immunol. 154:3567.

8. Giulian, D., J. Yu, X. Li, D. Tom, J. Li, E. Wendt, S. N. Lin, R. Schwarcz, andC. Noonan. 1996. Study of receptor-mediated neurotoxins released by HIV-1-infected mononuclear phagocytes found in human brain.J. Neurosci. 16:3139.

9. Dawson, V. L., T. M. Dawson, E. D. London, D. S. Bredt, and S. H. Snyder.1991. Nitric oxide mediates glutamate neurotoxicity in primary cortical cultures.Proc. Natl. Acad. Sci. USA 88:6368.

10. Tyor, W. R., J. D. Glass, J. W. Griffin, P. S. Becker, J. C. McArthur, L. Bezman,and D. E. Griffin. 1992. Cytokine expression in the brain during the acquiredimmunodeficiency syndrome.Ann. Neurol. 31:349.

11. Genis, P., M. Jett, E. W. Bernton, T. Boyle, H. A. Gelbard, K. Dzenko,R. W. Keane, L. Resnick, Y. Mizrachi, D. J. Volsky, et al. 1992. Cytokines andarachidonic metabolites produced during human immunodeficiency virus (HIV)-infected macrophage-astroglia interactions: implications for the neuropathogen-esis of HIV disease.J. Exp. Med. 176:1703.

12. Epstein, L. G., and H. E. Gendelman. 1993. Human immunodeficiency virus type1 infection of the nervous system: pathogenetic mechanisms.Ann. Neurol. 33:429.

13. Liu, J., M. L. Zhao, C. F. Brosnan, and S. C. Lee. 1996. Expression of type IInitric oxide synthase in primary human astrocytes and microglia: role of IL-1band IL-1 receptor antagonist.J. Immunol. 157:3569.

14. Lee, S. C., D. W. Dickson, W. Liu, and C. F. Brosnan. 1993. Induction of nitricoxide synthase activity in human astrocytes by interleukin-1b and interferon-g.J. Neuroimmunol. 46:19.

15. Janabi, N., S. Chabrier, and M. Tardieu. 1996. Endogenous nitric oxide activatesprostaglandin F2a production in human microglial cells but not in astrocytes: astudy of interactions between eicosanoids, nitric oxide, and superoxide anion(O2

2) regulatory pathways.J. Immunol. 157:2129.16. Bukrinsky, M. I., H.S.L.M. Nottet, H. Schmidtmayerova, L. Dubrovsky,

C. R. Flanagan, M. E. Mullins, S. A. Lipton, and H. E. Gendelman. 1995. Reg-ulation of nitric oxide synthase activity in human immunodeficiency virus type 1(HIV-1)-infected monocytes: implications for HIV-associated neurological dis-ease.J. Exp. Med. 181:735.

17. Nottet, H. S. L. M., D. R. Bar, H. Hassel, J. Verhoef, and L. A. Boven. 1997.Cellular aspects of HIV-1 infection of macrophages leading to neuronal injury inin-vitro models for HIV-1 encephalitis.J. Leukocyte Biol. 62:1.

18. Fine, S. M., R. A. Angel, S. W. Perry, L. G. Epstein, J. D. Rothstein, S. Dewhurst,and H. A. Gelbard. 1996. Tumor necrosis factora inhibits glutamate uptake byprimary human astrocytes: implications for pathogenesis of HIV-1 dementia.J. Biol. Chem. 271:15303.

19. Adamson, D. C., B. Wildemann, M. Sasaki, J. D. Glass, J. C. McArthur,V. I. Christov, T. M. Dawson, and V. L. Dawson. 1996. Immunologic NO syn-thase: elevation in severe AIDS dementia and induction by HIV-1 gp41.Science274:1917.

20. Lipton, S. A., Y. B. Choi, Z. H. Pan, S. Z. Lei, H. S. Chen, N. J. Sucher,J. Loscalzo, D. J. Singel, and J. S. Stamler. 1993. A redox-based mechanism forthe neuroprotective and neurodestructive effects of nitric oxide and related ni-troso-compounds.Nature 364:626.

21. Beckman, J. S., and W. H. Koppenol. 1996. Nitric oxide, superoxide, and per-oxynitrite: the good, the bad, and ugly.Am. J. Physiol. 271:C1424.

22. Bonfoco, E., D. Krainc, M. Ankarcrona, P. Nicotera, and S. A. Lipton. 1995.Apoptosis and necrosis: two distinct events induced, respectively, by mild andintense insults withN-methyl-D-aspartate or nitric oxide/superoxide in corticalcell cultures.Proc. Natl. Acad. Sci. USA 92:7162.

23. Chao, C. C., S. Hu, and P. K. Peterson. 1995. Glia, cytokines, and neurotoxicity.Crit. Rev. Neurobiol. 9:189.

24. Fukuto, J. M., and L. J. Ignarro. 1997. In vivo aspects of nitric oxide (NO)chemistry: Does peroxynitrite (2OONO) play a major role in cytotoxicity?Ac-counts Chem. Res. 30:149.

25. Kimura, T., M. Kameoka, and K. Ikuta. 1993. Amplification of superoxide aniongeneration in phagocytic cells by HIV-1 infection.FEBS Lett. 326:232.

26. Coyle, J. T., and P. Puttfarcken. 1993. Oxidative stress, glutamate, and neuro-degenerative disorders.Science 262:689.

27. Ischiropoulos, H., L. Zhu, J. Chen, M. Tsai, J. C. Martin, C. D. Smith, andJ. S. Beckman. 1992. Peroxynitrite-mediated tyrosine nitration catalyzed by su-peroxide dismutase.Arch. Biochem. Biophys. 298:431.

FIGURE 8. Cellular interactions between HIV-infected brain macro-phages and astrocytes. 1) HIV-infected and immune-activated brain mac-rophages express high levels of IL-1b (10–12). 2) An increase in IL-1bwill lead to an induction of iNOS in astrocytes resulting in elevated nitricoxide (NO) production (13–15). 3) NO diffuses readily out of the cellwhere it reacts instantly with 4) superoxide anion, produced at elevatedlevels at the membrane of HIV-infected macrophages, to form 5) the highlyneurotoxic peroxynitrite (20–24). 6) The presence of iNOS in brain mac-rophages has been shown in this study. 7) However, this may not contributeto substantial formation of NO, since the ability of human macrophages toproduce any significant amounts of NO is still controversial (49, 50). Asshown here, HIV-infected macrophages produce high amounts of 4) su-peroxide anion and 8) express high levels of SOD. However, since super-oxide anion reacts with NO at a reaction rate that is six times faster thanthat of SOD (21), SOD will not be fully capable of preventing peroxynitritefrom being formed by 9) scavenging superoxide anion.

4326 INCREASED PEROXYNITRITE ACTIVITY IN AIDS DEMENTIA COMPLEX

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from

28. Oury, T. D., Y. S. Ho, C. A. Piantadosi, and J. D. Crapo. 1992. Extracellularsuperoxide dismutase, nitric oxide, and central nervous system O2 toxicity. Proc.Natl. Acad. Sci. USA 89:9715.

29. Tu, P. H., M. E. Gurney, J. P. Julien, V. M. Y. Lee, and J. Q. Trojanowski. 1997.Oxidative stress, mutantSOD1, and neurofilament pathology in transgenic mousemodels of human motor neuron disease.Lab. Invest. 76:441.

30. Nottet, H. S.L.M., I. I. Moelans, N. M. de Vos, L. de Graaf, M. R. Visser, andJ. Verhoef. 1997.N-acetyl-L-cysteine-induced up-regulation of HIV-1 gene ex-pression in monocyte-derived macrophages correlates with increased NF-kBDNA binding activity.J. Leukocyte Biol. 61:33.

31. Gyllenhammar, H. 1987. Lucigenin chemiluminescence in the assessment of neu-rotrophil superoxide production.J. Immunol. 97:209.

32. Nottet, H. S. L. M., L. de Graaf, N. Machiel de Vos, L. J. Bakker, J. A. van Strijp,M. R. Visser, and J. Verhoef. 1993. Down-regulation of human immunodefi-ciency virus type (HIV-1) production after stimulation of monocyte-derived mac-rophages infected with HIV-1.J. Infect. Dis. 167:810.

33. Vitkovic, L., A. da Cunha, and W. R. Tyor. 1994. Cytokine expression andpathogenesis in AIDS brain. InHIV, AIDS and the Brain.R. W. Price andS. W. Perry, eds. Raven Press, New York, p. 203.

34. Dickson, D. W., L. A. Mattiace, K. Kure, K. Hutchins, W. D. Lyman, andC. F. Brosnan. 1991. Microglia in human disease, with an emphasis on acquiredimmune deficiency syndrome.Lab. Invest. 64:135.

35. Nuovo, G. J., and M. L. Alfieri. 1996. AIDS dementia is associated with massive,activated HIV-1 infection and concomitant expression of several cytokines.Mol.Med. 2:358.

36. Nottet, H. S. L. M., and H. E. Gendelman. 1995. Unraveling the neuroimmunemechanisms for the HIV-1-associated cognitive/motor complex.Immunol. Today16:441.

37. Eddleston, M., and L. Mucke. 1993. Molecular profile of reactive astrocytes:implications for their role in neurologic disease.Neuroscience 54:15.

38. Gendelman, H. E., J. M. Orenstein, L. M. Baca, B. Weiser, H. Burger,D. C. Kalter, and M. S. Meltzer. 1989. The macrophage in the persistence andpathogenesis of HIV infection.AIDS 3:475.

39. Beckman, J. S., M. Carson, C. D. Smith, and W. H. Koppenol. 1993. ALS, SODand peroxynitrite.Nature 364:584-WE.

40. De La Torre, R., A. Casado, E. Lopez Fernandez, D. Carrascosa, V. Ramirez, andJ. Saez. 1996. Overexpression of copper-zinc superoxide dismutase in trisomy 21.Experientia 52:871.

41. Margaglione, M., R. Garofano, F. Cirillo, A. Ruocco, E. Grandone,G. Vecchione, G. Milan, G. Di Minno, A. De Blasi, and A. Postiglione. 1995.Cu/Zn superoxide dismutase in patients with non-familial Alzheimer’s disease.Aging Milano. 7:49.

42. Urakami, K., K. Sato, A. Okada, T. Mura, T. Shimomura, T. Takenaka,Y. Wakutani, T. Oshima, Y. Adachi, K. Takahashi, et al. 1995. Cu, Zn superoxidedismutase in patients with dementia of the Alzheimer type.Acta Neurol. Scand.91:165.

43. Good, P. F., P. Werner, A. Hsu, C. W. Olanow, and D. P. Perl. 1996. Evidenceof neuronal oxidative damage in Alzheimer’s disease.Am. J. Pathol. 149:21.

44. Boje, K. M., and P. K. Arora. 1992. Microglial-produced nitric oxide and reactivenitrogen oxides mediate neuronal cell death.Brain Res. 587:250.

45. Giulian, D., K. Vaca, and M. Corpuz. 1993. Brain glia release factors with op-posing actions upon neuronal survival.J. Neurosci. 13:29.

46. Norenberg, M. D. 1994. Astrocyte responses to CNS injury.J. Neuropathol. Exp.Neurol. 53:213.

47. Moretto, G., R. Y. Xu, D. G. Walker, and S. U. Kim. 1994. Co-expression ofmRNA for neurotrophic factors in human neurons and glial cells in culture.J. Neuropathol. Exp. Neurol. 53:78.

48. Mallat, M., and B. Chamak. 1994. Brain macrophages: neurotoxic or neurotro-phic effector cells?J. Leukocyte Biol. 56:416.

49. Weinberg, J. B., M. A. Misukonis, P. J. Shami, S. N. Mason, D. L. Sauls,W. A. Dittman, E. R. Wood, G. K. Smith, B. McDonald, K. E. Bachus, et al.1995. Human mononuclear phagocyte inducible nitric oxide synthase (iNOS):analysis of iNOS mRNA, iNOS protein, biopterin, and nitric oxide production byblood monocytes and peritoneal macrophages.Blood 86:1184.

50. de Groot, C. J., S. R. Ruuls, J. W. Theeuwes, C. D. Dijkstra, and P. Van der Valk.1997. Immunocytochemical characterization of the expression of inducible andconstitutive isoforms of nitric oxide synthase in demyelinating multiple sclerosislesions.J. Neuropathol. Exp. Neurol. 56:10.

51. Sharara, A.I., D. J. Perkins, M. A. Misukonis, S. U. Chan, J. A. Dominitz, andJ. B. Weinberg. 1997. Interferon (IFN)-a activation of human blood mononuclearcells in vitro and in vivo for nitric oxide synthase (NOS) type 2 mRNA andprotein expression: possible relationship of induced NOS2 to the anti-hepatitis Ceffects of IFN-a in vivo. J. Exp. Med. 186:1495.

4327The Journal of Immunology

by guest on January 2, 2019http://w

ww

.jimm

unol.org/D

ownloaded from