Kynurenine pathway metabolism in human astrocytes: a paradox for neuronal protection

There is good evidence that the kynurenine pathway (KP) and one of its products, quinolinic acid (QUIN), play a role in the pathogenesis of neurological diseases, in particular AIDS dementia complex. Although QUIN has been shown to be produced in neurotoxic concentrations by macrophages and microglia, the role of astrocytes in QUIN production is controversial. Using cytokine-stimulated cultures of human astrocytes, we assayed key enzymes and products of the KP. We found that human astrocytes lack kynurenine hydroxylase so that large amounts of kynurenine and the QUIN antagonist kynurenic acid were produced. However, the amounts of QUIN that were synthesized were subsequently completely degraded. We then showed that kynurenine in concentrations comparable with those produced by astrocytes led to significant production of QUIN by macrophages. These results suggest that astrocytes alone are neuroprotective by minimizing QUIN production and maximizing synthesis of kynurenic acid. However, it is likely that, in the presence of macrophages and/or microglia, astrocytes become indirectly neurotoxic by the production of large concentrations of kynurenine that can be secondarily metabolized by neighbouring or infiltrating monocytic cells to form the neurotoxin QUIN.

Characterisation of kynurenine pathway metabolism in human astrocytes and implications in neuropathogenesis

The role of astrocytes in the production of the neurotoxin quinolinic acid (QUIN) and other products of the kynurenine pathway (KP) is controversial. Using cytokine-stimulated human astrocytes, we assayed key enzymes and products of the KP. We found that astrocytes lack kynurenine-hydroxylase so that large amounts of kynurenine (KYN) and kynurenic acid (KYNA) were produced, while minor amounts of QUIN were synthesised that were completely degraded. We then showed that kynurenine added to macrophages led to significant production of QUIN. These results suggest that astrocytes alone are neuroprotective by minimising QUIN production and maximising synthesis of KYNA. However, it is likely that, in the presence of macrophages and/or microglia, astrocytes are neurotoxic by producing large concentrations of KYN that can be metabolised by neighbouring monocytic cells to QUIN.

Quinolinic acid is produced by macrophages stimulated by platelet activating factor, Nef and Tat

Activated macrophages produce quinolinic acid (QUIN), a neurotoxin, in several inflammatory brain diseases including AIDS dementia complex. We hypothesized that IL1-beta, IL6, transforming growth factor (TGF-beta2 and platelet activating factor could increase macrophage QUIN production. And that the HIV-1 proteins Nef, Tat and gp41 may also increase synthesis of QUIN by macrophages. At 72 h there were significant increases in QUIN production in the cells stimulated with PAF (914 +/- 50 nM) and Nef (2781 +/- 162 nM), with somewhat less production by Tat stimulation (645 +/- 240 nM). The increases in QUIN production approximated in vitro concentrations of QUIN shown to be neurotoxic and correlated closely with indoleamine 2,3-dioxygenase induction. IL1-beta, IL6, TGF-beta2 and gp41 stimulation produced no significant increase in QUIN production. These results suggest that some of the neurotoxicity of PAF, nef and tat may be mediated by QUIN.

Kynurenine pathway metabolism in human astrocytes

The involvement of astrocytes in Kynurenine pathway (KP) metabolism is still poorly understood. In the present study, we investigated the ability of human fetal astrocytes in vitro to produce quinolinic and picolinic acids using mass spectrometry. In parallel, we estimated the level of expression of five major KP enzymes using RT-PCR. The results demonstrated that astrocytes express most KP enzymes, except for kynurenine-hydroxylase. This in vitro study provides novel informations regarding the ability of human fetal astrocytes to degrade L-tryptophan along the KP.

Human Herpesvirus 8 and Protection from AIDS Dementia Complex

Infection with human herpesvirus 8 (HHV-8) has been associated with the development of three distinct conditions: Kaposi's sarcoma, body cavitybased lymphoma and Castleman's disease. HHV-8 produces chemokinelike proteins including viral macrophage inflammatory protein II, which has been shown to block human immunodeficiency virus 1 (HIV-1) infection of CD4-positive cells expressing CCR-3. As CCR-3 is a receptor for HIV-1 into microglial cells, it has been hypothesized that HHV-8 infection may inhibit HIV-1 infection of the brain, thereby decreasing the incidence of AIDS dementia complex. We reviewed published studies of the incidence of AIDS dementia complex in individuals with and without Kaposi's sarcoma. The data are consistent in showing a negative association between Kaposi's sarcoma and AIDS dementia complex and, although sparse, support the hypothesis that productive HHV-8 infection decreases HIV-1 infection of the brain sufficiently to decrease the incidence of AIDS dementia complex. This negative association should be examined in further cohorts of HIV-1-infected subjects, to exclude alternative explanations.

AIDS dementia complex

Human immunodeficiency virus (HIV) infection is often complicated by the development of AIDS dementia complex (ADC). This article examines the typical and atypical presentations of ADC, along with aspects of the prevalence and natural history of the disorder. Salient aspects of the neuropathology, neurovirology, neuroimmunology, and pathogenesis are also considered. The intricacies of management of ADC, especially in the era of highly active antiretroviral therapy, are fully evaluated. Finally, this information is synthesized into an approach to the diagnosis of ADC in a particular patient.

Changes to AIDS dementia complex in the era of highly active antiretroviral therapy

OBJECTIVES

To determine the protective efficacy of highly active antiretroviral therapy (HAART) against AIDS dementia complex (ADC) relative to other initial AIDS-defining illnesses (ADIs), Australian AIDS notification data over recent years were examined.

METHODS

All initial ADIs in Australia over the period 1992-1997 were included. Three initial ADI groups were established: ADC; other predominantly central nervous system (CNS) ADIs (toxoplasmosis and cryptococcosis); and non-CNS ADIs. For each ADI grouping, the proportion of total ADls, and median CD4 cell count in the pre-HAART era (1992-1995) were compared with the HAART era (1996 and 1997).

RESULTS

Initial ADls peaked in Australia in 1994 (n = 1049), with a gradual decline to 1996 (n = 722), and a marked decline in 1997 (n = 367). ADC constituted 4.4% of initial ADIs over the period 1992-1995, but increased after the introduction of HAART to 6.0% in 1996 and 6.5% in 1997 (P = 0.02). In contrast, the proportion of other CNS ADIs (1992-1995, 8.1%; 1996, 6.0%; 1997, 8.2%; P = 0.41) was stable over the period 1992-1997. The median CD4 cell count at ADC diagnosis increased from 70/mm3 in 1992-1995 to 120/mm3 in 1996 and 170/mm3 in 1997 (P = 0.04). Although the median CD4 cell count also increased significantly over this period for both other CNS ADIs (40-60/mm3; P = 0.02), and non-CNS ADIs (60-70/mm3; P = 0.02), the increase was small.

CONCLUSION

A proportional increase in ADC compared with other ADIs and a marked increase in the median CD4 cell count at ADC diagnosis have occurred since the introduction of HAART in Australia. These changes suggest that HAART has a lesser impact on ADC than on other ADIs, with the poor CNS penetration of many antiretroviral agents a possible explanation.

Anomalies in Nef expression within the central nervous system of HIV-1 positive individuals/AIDS patients with or without AIDS dementia complex

In determining levels of expression of HIV-1 Nef protein within the central nervous system (CNS) we assessed antibody responses to the protein both peripherally and in CNS. Antibodies to Nef were not detected within the CNS despite detection of antibodies to both gp41 and Nef in peripheral blood and representative virus isolates derived from CNS and peripheral blood (PB) samples containing full length nef sequence and virus-infected cells expressing Nef protein. We conclude from this that expression of Nef within the CNS is such that little or no antibody production occurs and that these differences indicate that Nef protein may not be directly contributing to the AIDS dementia complex. Expression of Nef protein in PHA-activated peripheral blood mononuclear cells from CNS derived isolates was different to that of coincidental PB derived isolates in that partial surface expression was observed for the latter. The results suggest that antigenic presentation of Nef within the CNS is anomalous and that Nef protein expression, at least for the limited number of in vitro derived isolates tested, has a different localization pattern.

Chronic exposure of human neurons to quinolinic acid results in neuronal changes consistent with AIDS dementia complex

OBJECTIVE

Concentrations of quinolinic acid, an N-methyl-D-aspartate agonist, are often elevated for long periods of time in the cerebrospinal fluid (CSF) and brain tissue of patients with AIDS dementia complex (ADC). This study was designed to test the hypothesis that chronic exposure of human neurons to quinolinic acid levels equivalent to those in the CSF of ADC patients is neurotoxic.

DESIGN AND METHODS

Human fetal brain 14-16 weeks post-menses was cultured in medium with no detectable levels of quinolinic acid. After 4 weeks, 350 or 1200 nmol/l quinolinic acid was added to the feeding medium for a further 5 weeks. Neurotoxicity was evaluated using immunohistochemistry, transmission and scanning electron microscopy, and image analysis.

RESULTS

A total of 1200 nmol/l quinolinic acid caused altered cell associations, a decrease in cell density and decreased microtubule-associated protein (MAP)-2 immunoreactivity compared with cultures exposed to 350 nmol/l quinolinic acid or controls. Image analysis of neurons in randomly selected fields revealed significantly swollen cells (P < 0.0001) compared with those treated with 350 nmol/l quinolinic acid or controls. Dendritic varicosities and discontinuous microtubular arrays were present in neurons exposed to both quinolinic acid concentrations, but not in control cultures.

CONCLUSIONS

This study is the first to assess quinolinic acid levels in the experimental medium, and demonstrates that chronic exposure of human neurons to concentrations of quinolinic acid equivalent to those in the CSF of patients with ADC leads to alterations in dendritic ultrastructure and MAP-2 immunoreactivity, which is consistent with ADC pathology.

Cerebral perfusion scanning in treating AIDS dementia: a pilot study

Acquired immunodeficiency syndrome (AIDS) dementia complex (ADC) is a common effect of the AIDS virus. We studied the regional cerebral blood flow of patients with early ADC and its response to atevirdine mesylate.

METHODS

Ten men with early ADC, who had failed or were intolerant to zidovudine or didanosine therapy, were treated with atevirdine mesylate for 12 wk. Cerebral perfusion SPECT using 99mTc-HMPAO was performed at Week 0 and Week 12. SPECT images were analyzed qualitatively and semiquantitatively.

RESULTS

The cerebral perfusion abnormalities in early ADC were usually mild and characteristically involved the cortices and periventricular regions bilaterally and symmetrically. Four patients were able to complete the protocol. Three of these patients responded to atevirdine clinically, two of whom showed improvement in their Week 12 SPECT images. The other responder had an essentially unchanged image. The patient who did not respond to atevirdine showed a definite deterioration in cerebral perfusion.

CONCLUSION

Cerebral perfusion SPECT is useful in detecting and assessing therapeutic responses in ADC. The preliminary results of atevirdine in treating ADC are promising and need further investigation.

Kynurenine pathway inhibition reduces neurotoxicity of HIV-1-infected macrophages

The AIDS dementia complex (ADC) is a consequence of excessive immune activation driven at least in part by systemic HIV infection and probably brain infection. Quinolinic acid (QUIN) is a neurotoxic tryptophan metabolite produced by macrophages in response to stimulation with cytokines or infection with HIV-1. Consequently it has been implicated in ADC pathogenesis. However, macrophages infected with HIV-1 synthesize numerous neurotoxic substances. Therefore we conducted experiments using human fetal brain tissue to determine the relative importance of QUIN as a neurotoxin in ADC. Human macrophages were infected with HIV-1 in vitro using a viral isolate from a demented patient. 6-Chloro-D-tryptophan, an inhibitor of QUIN biosynthesis, was added to half the macrophage cultures to block formation of QUIN. Supernatants containing QUIN (SQpos) or in which QUIN biosynthesis had been inhibited (SQneg) were then added to human fetal brain aggregate cultures. Toxicity was evaluated using lactate dehydrogenase efflux, trypan blue exclusion, immunohistochemistry, image analysis, and electron microscopy. Each technique showed a reduction of toxicity in SQneg-treated cultures. These studies confirm the significance of QUIN as a neurotoxin in ADC and suggest that neuroprotective strategies may have a place in the treatment of this disease.

Direct ex vivo flow cytometric analysis of human microglial cell CD4 expression: examination of central nervous system biopsy specimens from HIV-seropositive patients and patients with other neurological disease

OBJECTIVE

To define a clear ex vivo flow cytometric phenotype for adult human microglia that would distinguish it from all other macrophage lineage cells in the central nervous system (CNS) or blood, and to utilize this phenotype to examine the activation state and CD4 expression of microglia freshly derived from CNS tissue of HIV-positive patients and those with other neurological diseases.

DESIGN

Fresh human CNS tissue from both HIV-uninfected and HIV-infected individuals was obtained by biopsy or resection, and cells isolated immediately, labelled for flow cytometry and analysed.

METHODS

A Percoll density gradient isolation technique and phenotypic characteristics used for rodent microglia were applied and modified.

RESULTS

Resident microglia could clearly be defined by the flow cytometric phenotype CD45low CD4- CD11b+ CD11chigh major histocompatibility complex (MHC) class II+ CD26- CD14-. Assuming normally low-level MHC class II expression in the healthy CNS, it was likely that MHC class II positivity reflected underlying pathology necessitating biopsy or resection and appeared to be a 'leaky' activation marker. Microglia activation was observed in specimens from only six (35%) out of 17 HIV-uninfected but all four (100%) HIV-infected patients, defined strictly as any level of upregulation of CD4 expression, to produce the phenotype CD45low/medium CD4low CD11b+ CD1.1Chigh MHC class II+/+2 CD26- CD14-. Where examined by immunohistology, CD68 was also upregulated in these cases.

CONCLUSIONS

When activated in situ, microglia express low levels of CD4 and this is always seen in tissue from HIV-infected patients. Using the flow cytometric phenotype established here, microglia from HIV-infected tissue can now be isolated in pure form and studied directly ex vivo.

Quinolinic acid production by macrophages stimulated with IFN-gamma, TNF-alpha, and IFN-alpha

Quinolinic acid (QUIN) has been associated with several inflammatory neurologic disorders, including AIDS dementia complex (ADC). Recent studies suggest that activation of macrophages with either HIV-1 or interferon-gamma (IFN-gamma) can lead to QUIN production. However, the importance of other cytokines, especially those related to the macrophage and that are especially important in ADC pathogenesis, remains unclear. We, therefore, sought to determine the role of tumor necrosis factor-alpha (TNF-alpha) and IFN-alpha in the production of QUIN. Primary human macrophages were stimulated with two different concentrations of these cytokines alone, in combination with each other, and with IFN-gamma. QUIN concentrations in the supernatants were then measured by mass spectrometry at 24, 48, and 72 hs. Results at 72 h showed significant increases in QUIN production in the cells stimulated with IFN-gamma (10297 +/- 170 nmol/L) and also in those stimulated with IFN-alpha (3600 +/- 113 nmol/L), whereas TNF-alpha-stimulated macrophages produced low levels of QUIN (1108 +/- 23 nmol/L). Macrophages stimulated with the cytokine combinations TNF-alpha and IFN-gamma, IFN-alpha, and IFN-gamma, and TNF-alpha and IFN-alpha also resulted in increases in QUIN production (11471 +/- 77.6 nmol/L, 16656 +/- 184 nmol/L, and 3369 +/- 120.5 nmol/L, respectively). The increases in QUIN production in all of the cytokine treatments approached or exceeded in vivo concentrations of QUIN that have been shown to be neurotoxic. These data further support a role for QUIN in cytokine-mediated neuronal death in inflammatory disorders of the brain, especially ADC.

Levels of human immunodeficiency virus type 1 RNA in cerebrospinal fluid correlate with AIDS dementia stage

The relationship between the presence and severity of AIDS dementia complex (ADC) and the levels of human immunodeficiency virus type 1 (HIV-1) RNA in cerebrospinal fluid (CSF) were assessed. Nineteen patients with ADC (stages 1-3), 6 without ADC (group 1), and 10 (group 2) without ADC but with cryptococcal meningitis or progressive multifocal leukoencephalopathy were studied. There was a significant relationship between increasing CSF virus burden and ADC severity (P = .0006) but not with plasma burden and ADC severity. In group 2, CSF HIV-1 RNA levels in patients with cryptococcal meningitis were elevated. These results show that CSF HIV-1 RNA concentrations correlate well with ADC severity but may also be increased by central nervous system infections, such as cryptococcal meningitis.

Association between HIV distal symmetric polyneuropathy and Mycobacterium avium complex infection

OBJECTIVES

Pronounced infiltration of activated macrophages occurs in the peripheral nerves of patients with HIV distal symmetric polyneuropathy (DSPN). Mycobacterium avium complex (MAC) is a common facultative intracellular parasite of the macrophage in advanced HIV disease and may induce macrophage activation. Whether MAC disease is associated with DSPN was examined prospectively.

METHODS

One hundred and fifty consecutive patients with HIV infection were assessed for the probability of DSPN. Blood cultures for MAC were performed, independently of neurological assessment, as part of the investigation of unexplained fever, anaemia, weight loss, or, less commonly, diarrhoea.

RESULTS

There were 20 patients with possible, 14 with probable, and 22 with definite HIV DSPN. Blood cultures for MAC were performed on 80 patients, of whom 39 were positive and 41 negative. The test for trend, when corrected for CD4 count, disclosed a significant association (P = 0.01). There was no statistically significant association between DSPN and cytomegalovirus (CMV) disease.

CONCLUSION

Coinfection of the macrophage by MAC may further activate the HIV infected macrophage thereby accelerating the elaboration of neural toxins or MAC infection of the macrophage itself may lead to the production of neural toxins.

Myositis associated with a newly described microsporidian, Trachipleistophora hominis, in a patient with AIDS

Microsporidia are zoonotic protozoa which were rare human pathogens prior to 1985, when Enterocytozoon bieneusi was described in human immunodeficiency virus-infected patients with chronic diarrhea. Another species, Encephalitozoon (Septata) intestinalis, is associated with diarrhea and chronic sinusitis, and approximately 25 cases have been reported in the literature. However, other microsporidial infections in human immunodeficiency virus-infected patients remain extremely rare. We report the first case of a Pleistophora sp.-like microsporidian infection presenting as a progressive severe myosotis associated with fever and weight loss. The organism was demonstrated by light microscopy and electron microscopy in corneal scrapings, skeletal muscle, and nasal discharge. Electron microscopy showed an electron-dense surface coat with "sunflare"-like projections surrounding all stages of development of meronts (two to four nuclei, dividing by binary fission), sporonts, and sporoblasts. Division of sporonts, in which sporonts separate from the thick outer coat, creating a sporophorous vesicle, is by binary fission, differentiating this organism from Pleistophora sp. The spore measures 4.0 by 2.5 microns and has a rugose exospore. A new genus and species, Trachipleistophora hominis, has been established for this parasite. The patient was treated with albendazole, sulfadiazine, and pyrimethamine, and the clinical symptoms resolved.

Pilot study of the efficacy of atevirdine in the treatment of AIDS dementia complex

OBJECTIVES

To determine the efficacy of the non-nucleoside reverse transcriptase inhibitor atevirdine in the treatment of AIDS dementia complex (ADC).

DESIGN

Open label pilot study.

METHODS

Ten patients with ADC (stage 1 or 2) who were intolerant to zidovudine or dideoxyinosine, or in whom the antiretrovirals had failed to prevent further decline in CD4 cell levels, were entered into the study. Atevirdine, 1800 mg daily in three divided doses, was given over a 12-week period. Patients were assessed neurologically and neuropsychologically every 4 weeks. Cerebrospinal fluid analysis was performed at entry and at weeks 4 and 12. Technetium-99 exametazine single photon emission computed tomographic cerebral perfusion scans and magnetic resonance imaging brain scans were performed at weeks 0 and 12.

RESULTS

Five patients completed the 12 week protocol. Four of these five responded to atevirdine, as judged by quantified neurological and neuropsychological assessments. Atevirdine was well tolerated apart from the development of rash, anxiety, intermittent diarrhoea and fatigue.

CONCLUSIONS

These preliminary results suggest that atevirdine is efficacious in the treatment of ADC. Larger blinded studies of this class of drug in the treatment of ADC are required.

Human immunodeficiency virus type 1-related transient neurological deficits

PURPOSE

To characterize human immunodeficiency virus type 1 (HIV-1)-related transient neurological deficit (TND).

DESIGN

A 3-year prospective study based at a tertiary referral center.

METHODS

Patients with TND in the absence of an opportunistic infection, neoplasm, neurosyphilis, or seizure were recruited and further investigated. The frequency of hospital admission for these patients was compared with those who were HIV-1 seronegative and with those who were admitted for HIV-1-related thromboembolic events.

RESULTS

Twenty-seven patients were identified: mean age of 39 +/- 9 years; CD4+ cell count of 130 +/- 80/microL. Seven patients had no history of an AIDS-defining illness. Hemiparesis and hemisensory disturbance occurred in 23 patients (85%); 15 had preexisting ADC, 7 stage 1 and 8 stage 2; 3 developed ADC after 18 months. Thirteen patients had multiple attacks and 5 had more than 20; 2 patients developed a cerebral infarct. No cause for the TND was identified in 8 patients (30%). Anticardiolipin antibodies were found in 70% and low protein S levels in 53% which was significantly more often than in a neurologically normal group with similarly advanced HIV-1 disease. The frequency of admission was 0.8% whereas the frequency for similar TND in HIV-1-seronegative patients was 0.4% and the frequency for HIV-1-related thromboembolic events was 0.9%.

CONCLUSIONS

Transient neurological deficits occur in advanced HIV-1 disease and are often associated with ADC, elevated concentrations of anticardiolipin antibodies, and low protein S levels. Future studies will need to define the precise role of these associations in the pathogenesis of HIV-1-related TND.

Predictive markers of AIDS dementia complex: CD4 cell count and cerebrospinal fluid concentrations of beta 2-microglobulin and neopterin

A predictive marker for AIDS dementia complex (ADC) in a cohort of neurologically asymptomatic human immunodeficiency virus type 1 (HIV-1)-infected patients with < 200 CD4 cells/microL was sought. Patients were assessed neurologically and neuropsychologically at entry. Blood and cerebrospinal fluid (CSF) were taken for assay of beta 2-microglobulin (beta 2M), and neopterin and T cell subsets were assessed from blood. Patients were evaluated every 4 months. Of 37 patients recruited, 35 had sufficient follow-up data. Seventeen patients progressed to ADC stage > or = 1. In univariate analyses, concentrations of CSF beta 2M and neopterin and CD4 cell count were each significantly associated with ADC development. In a multivariate analysis, concentrations of CSF beta 2M remained significant, with levels > 5 mg/L carrying approximately 17 times the risk of ADC. CSF beta 2M and neopterin levels and CD4 cell count are useful in identifying patients at risk of ADC and as such can be used to target high-risk patients so therapy can be optimized.

The relationship between AIDS dementia complex and the presence of macrophage tropic and non-syncytium inducing isolates of human immunodeficiency virus type 1 in the cerebrospinal fluid

We sought to determine the clinical significance of macrophage tropic and non-syncytium inducing isolates of human immunodeficiency virus type 1 (HIV-1) in the cerebrospinal fluid (CSF) of patients with and without AIDS dementia complex (ADC). HIV-1 was isolated from the CSF of 31 patients with and without ADC. The isolates were then characterised as to the degree of macrophage tropism by quantitation of p24 production and the presence of syncytium inducing (SI) or non-syncytium inducing (NSI) isolates by MT2 assay and SupT1 coculture. The degree of macrophage tropism varied according to the donor macrophage that was used except in strongly macrophage tropic isolates. Moderate and severe ADC (stage > or = 2) was associated with the presence of highly macrophage tropic isolates in the CSF (P = 0.01). The sensitivity and specificity values of a highly macrophage tropic isolate in the CSF for ADC stage > or = 2 were 82% and 66% respectively while the predictive value was 64%. Three of four asymptomatic patients with such highly macrophage tropic isolates in the CSF subsequently developed ADC after an average of 4 months. Twenty-eight isolates from the CSF and 23 from the blood were NSI regardless of the presence or absence of ADC. The predictive value of an SI isolate in the blood reflecting an SI isolate in the CSF was 37.5% while the predictive value of an NSI isolate in the blood reflecting an NSI in the CSF was 100%. These data suggest that host factors are essential in determining the degree of macrophage tropism in HIV-1 and that such tropism is important for the presence and possibly subsequent development of ADC. The CSF usually has NSI isolates regardless of the presence of ADC and irrespective of the presence of such isolates in the blood thereby suggesting that the CSF is behaving virologically as a separate compartment to the blood.

Neurocytotoxity of quinolinic acid in human brain cultures

Quinolinic acid (QUIN) is a tryptophan metabolite which has been found to be an excitotoxin in rats, although its toxicity in humans is unknown. QUIN has been implicated in the pathogenesis of AIDS dementia complex. This study examined the effect of QUIN on human fetal brain tissue in vitro. After at least 14 days in vitro, QUIN was added to the cultures in the feeding medium, and lactate dehydrogenase (LDH) efflux at 20 h and neuronal morphology were used as a measure of neuronal injury. LDH levels in media from cultures exposed to QUIN concentrations of 5 and 10 mM were consistently elevated compared to controls. Overall, LDH levels in media from cultures exposed to lower QUIN concentrations did not differ significantly from controls. These data are comparable to animal in vitro studies, and support the hypothesis that QUIN is toxic to human central nervous system neurons and further strengthen its potential role in the pathogenesis of AIDS dementia complex.

Quinolinic acid production is related to macrophage tropic isolates of HIV-1

We sought to determine whether the neurotoxin quinolinic acid (QUIN) was produced by macrophages or lymphocytes infected with isolates of HIV-1 with varying degrees of macrophage tropism derived from patients with varying stages of AIDS dementia complex (ADC). Highly macrophage tropic isolates and minimally macrophage tropic isolates were used to inoculate macrophages and QUIN production was measured. Similarly, QUIN production from macrophages was monitored using a purified cell free highly macrophage tropic isolate and laboratory isolates SF33 and SF2. Each of these experiments was also performed with lymphocytes. We found that macrophages infected with macrophage tropic isolates of HIV-1 led to QUIN production while lymphocytes did not produce QUIN. The ability of the HIV-1 infected macrophages to produce QUIN was related to the viral inoculum and the degree of macrophage tropism of the isolate. The severity of ADC in the patient from whom a particular isolate was derived was not per se a determining factor for QUIN production. Purified cell free ADC isolates also led to QUIN production by macrophages thereby suggesting that HIV-1 infection alone is capable of inducing QUIN production.

AIDS dementia complex and HIV-1 brain infection: clinical-virological correlations

To evaluate the presence and distribution of central nervous system infection by human immunodeficiency virus type 1 (HIV-1), we used immunohistochemical methods to map the HIV-1 p24 core protein in the brains of 55 autopsied patients with acquired immunodeficiency syndrome (AIDS). In a subset of 40 of these patients who had undergone antemortem neurological evaluation of the AIDS dementia complex (ADC), we analyzed the relation between the severities of the viral infection and clinical dysfunction. Viral antigen was detected in macrophages and cells with morphological and immunohistochemical characteristics of microglia as well as multinucleated cells. The distribution of antigen-positive cells preferentially involved certain deep brain structures, especially the globus pallidus, other basal ganglia nuclei, and the central white matter. Overall, the presence and frequency of infected cells were highly correlated with the histological findings of multinucleated-cell encephalitis and in general with the clinical ADC stage. However, infection was often more limited than might be "anticipated" from the severity of patients' clinical dysfunction: Only 61% of patients with at least ADC stage 1 had detectable antigen and of these only approximately 30% of the brain sections were antigen positive. These results suggest a pathogenetic model of ADC where virus- or cell-coded toxins amplify the effect of limited brain infection.

Cerebrospinal fluid nitrite/nitrate levels in neurologic diseases

Nitric oxide has been proposed to mediate cytotoxic effects in inflammatory diseases. To investigate the possibility that overproduction of nitric oxide might play a role in the neuropathology of inflammatory and noninflammatory neurological diseases, we compared levels of the markers of nitric oxide, nitrite plus nitrate, in the CSF of controls with those in patients with various neurologic diseases, including Huntington's and Alzheimer's disease, amyotrophic lateral sclerosis, and HIV infection. We found that there were no significant increases in the CSF levels of these nitric oxide metabolites, even in patients infected with HIV or in monkeys infected with poliovirus, both of which have significantly elevated levels of the neurotoxin quinolinic acid and the marker of macrophage activation, neopterin. However, CSF quinolinic acid, neopterin, and nitrite/nitrate levels were significantly increased in a small group of patients with bacterial and viral meningitis.

Cerebrospinal fluid HIV-1 p24 antigen and culture: sensitivity and specificity for AIDS-dementia complex

We prospectively evaluated 94 patients with AIDS-dementia complex (ADC) and a smaller group of 27 patients with other HIV-1 related neurological conditions to determine the usefulness of cerebrospinal fluid (CSF) p24 antigen and HIV-1 culture in the diagnosis of ADC. The presence of ADC correlated with detectable CSF p24, but not with a positive culture. However, only 54% of the patients with severe or end stage (stages 3 and 4) ADC had detectable CSF p24 and only 25% had a positive culture. Among those with detectable CSF p24, there was no discernible relationship between the severity of ADC and the amount of CSF p24. The diagnostic sensitivity of CSF p24 in ADC was 21% whereas the specificity was 98%. CSF HIV-1 culture had a sensitivity of 30% and a specificity of 80%. To address the possibility of binding of p24 in immune complexes, thereby escaping detection, an acid hydrolysis procedure was performed on the CSF and serum samples. This did not, however, make an appreciable difference in the detection rate of p24. To delineate whether the finding of cell free virus in the CSF was associated with ADC, CSF culture for HIV-1 was performed on both cell depleted and cell associated fractions. It was uncommon for CSF to be culture positive in only the cell free fraction and there was no relationship to the presence or severity of ADC.

The clinical spectrum and pathogenesis of HIV encephalopathy, myelopathy, and peripheral neuropathy

Human immunodeficiency virus infection commonly affects the nervous system, with the most important features being a dementing illness, a myelopathy, and a neuropathy. In the past year, important advances have been made in the epidemiology and treatment of these conditions. Moreover, a better understanding of the pathogenetic mechanisms has emerged, with particular emphasis being placed upon gp120, tumor necrosis factor, and quinolinic acid and other partly characterized neurotoxins. This review highlights these advances in the context of the clinical syndromes.

Human immunodeficiency virus-related headache

We retrospectively assessed the frequency and clinical characteristics of headache occurring in human immunodeficiency virus (HIV)-infected patients in whom we had excluded all other causes, and determined the frequency of a similar headache in an HIV-negative group. Over a 1-year period, the HIV-related headache occurred in 2.8% of total admissions to the HIV service as opposed to 0.2% of admissions to the neurology service. The affected patients had advanced HIV infection (CD4+ cell count = 58.9 +/- 80.3; normal, > 500 x 10(-6)/l). The headache had features similar to those ascribed to HIV aseptic meningitis and was not related to the presence of AIDS dementia complex or its subsequent development over a 6-month follow-up. We propose that this is a distinct clinical entity related to HIV infection that is similar, if not identical, to HIV aseptic meningitis except for the lack of a CSF pleocytosis, probably reflecting the lymphocyte depletion that is characteristic of the more advanced state of HIV infection.

HIV-related neurological disease

Early neurological complications present an opportunity to make an early diagnosis of HIV infection. Patients may present with subtle symptoms and signs of neurological disease, but many HIV-related complications are readily treated if they are correctly diagnosed. As there are also neurological side effects of anti-HIV drugs, the management of patients with neurological complications and HIV disease requires accurate diagnosis.

Increased neopterin levels in brains of patients with human immunodeficiency virus type 1 infection

Postmortem levels of native neopterin (D-erythro-neopterin) were measured in cerebral cortical samples from 44 human immunodeficiency virus type 1-infected and eight uninfected, nonneurological control patients. Cerebral cortical gray and white matter neopterin levels for the controls ranged from 0.5 to 7.2 pmol/mg of protein in contrast to neopterin levels in brains of the virus-infected patients, which frequently were more than threefold and occasionally more than 30-fold higher than mean control levels. Cortical neopterin levels did not correlate with severity of the acquired immunodeficiency syndrome dementia complex, but subcortical levels correlated with the presence of active human immunodeficiency virus type 1 infection, as reflected by pathological evidence of multinucleated giant cell encephalitis. Evidence of opportunistic cytomegalovirus infections in approximately 25% of the human immunodeficiency virus type 1-infected patients was associated with enhanced levels of neopterin in frontal cortex.

Inter-relationships between quinolinic acid, neuroactive kynurenines, neopterin and beta 2-microglobulin in cerebrospinal fluid and serum of HIV-1-infected patients

Quinolinic acid (QUIN) is an neurotoxic N-methyl-D-aspartate receptor agonist and an L-tryptophan metabolite of the kynurenine pathway. Increased concentrations of QUIN occur in both cerebrospinal fluid (CSF) and blood of patients infected with human immunodeficiency virus (HIV)-1, particularly those with neurologic disturbances. In the present study of HIV-1 infected patients in Walter Reed stages 4, 5 and 6, reductions in L-tryptophan accompanied proportional increases in L-kynurenine and QUIN in both serum and CSF. Further, close inter-correlations exist between QUIN kynurenic acid and L-kynurenine with both beta 2-microglobulin and neopterin in CSF and serum. These correlations support the hypotheses that the kynurenine pathway is activated in association with inflammation and induction of indoleamine-2,3-dioxygenase. There were no relationships between CSF QUIN, L-kynurenine or kynurenic acid with the ratio of serum:CSF albumin concentrations, which indicates that the increases in CSF QUIN, L-kynurenine or kynurenic acid were not dependent on a breakdown of the blood-brain barrier. Kynurenic acid is also a kynurenine pathway metabolite that can attenuate the excitotoxic effects of QUIN when present in higher molar concentrations. While CSF kynurenic acid levels were increased in HIV-1-infected patients, the magnitude of the increases were smaller than those of QUIN and the molar concentrations of kynurenic acid were consistently lower than QUIN by at least one order of magnitude. We conclude that immune activation increases the levels of neuroactive kynurenines within the central nervous system of HIV-1-infected patients secondary to activation of indoleamine-2,3-dioxygenase.