Abstract

The relationship between monocyte immune responses and cognitive impairment during progressive human immunodeficiency virus type 1 (HIV-1) infection was investigated in 28 subjects receiving highly active antiretroviral therapy. The mean±SEM CD4+ T lymphocyte count and virus load for all patients were 237±41 cells/mm3 and 77,091±195,372 HIV-1 RNA copies/mL, respectively. Levels of soluble tumor necrosis factor–α type II receptor (sTNF-RII) and soluble CD14 (sCD14) were measured in plasma by ELISA and were correlated with results from neuropsychological, magnetic resonance imaging, and magnetic resonance spectroscopy tests. Plasma sCD14 and sTNF-RII levels were elevated in subjects with cognitive impairment and in those with brain atrophy. Furthermore, both factors were correlated with spectroscopic choline:creatine ratios. These findings support the idea that peripheral immune responses are linked to cognitive dysfunction during advanced HIV-1 disease

Highly active antiretroviral therapy (HAART) has decreased but not eliminated the incidence of human immunodeficiency virus type 1 (HIV-1)–associated dementia (HAD) [1]. This disorder remains a serious complication in infected individuals [2]. Despite advances made in elucidating the neuropathogenesis of HIV-1, few tests are available that can diagnose or monitor the progression of HAD

Previous studies demonstrated that HIV load, CD4+ T lymphocyte counts, and β2-microglobulin and neopterin levels correlate with disease progression [3] and prognosis. Similarly, cerebrospinal fluid levels of platelet-activating factor [4] and quinolinic acid increase with the development of HAD [5] but are not diagnostic. Advances in neuroimaging have proven to be useful in diagnosing and monitoring patients with cognitive, behavioral, and motor dysfunction [6–10]. Several reports have demonstrated a correlation between brain atrophy, cognitive dysfunction [11, 12], and disease progression [13]. Magnetic resonance imaging (MRI) can show structural deficits during late-stage disease [12, 14]; however, proton magnetic resonance spectroscopy (1H MRS) can demonstrate changes in brain metabolites when subtle neurological deficits are manifest [7, 8]. For example, the ratio of N-acetyl aspartate (NAA) to creatine (CR) reflects neuronal density and is a reproducible marker of HAD [6, 14–16]. Moreover, ratios both of choline (CHO) to CR and of myoinositol (MI) to CR, reflections of glial density, can monitor neurological decline, especially during the earliest stages of cognitive impairment [7, 8]

Mononuclear phagocytes (MPs)—microglia and perivascular and brain macrophages—are the predominant productively infected cells in the brain. Moreover, HIV-1 infection and subsequent MP activation results in the neurotoxic secretory activities thought to be responsible for the cognitive and motor deficits of HAD [17–22]. The exact mechanism by which these cells become activated remains unclear. The emergence of CD14+CD16+ and CD14+CD69+ monocyte subsets in the blood of HAD patients suggests that immune activation may occur in the periphery before the occurrence of cognitive impairment [23]. Moreover, this observation implies that peripheral immune responses may participate in HAD pathogenesis. In this regard, we believe that products of activated monocytes, including chemokines and pro-inflammatory cytokines, affect the cell migratory and neurotoxic activities responsible for HIV-1 neuropathogenesis [24]. The detection of these soluble factors in blood may correlate with the development of HAD

Serum tumor necrosis factor (TNF)–α [25] and soluble TNF-α receptors (sTNF-Rs) [26] have been shown to correlate with CD4+ lymphocyte numbers and HIV disease progression [3, 27]. Soluble receptors can neutralize TNF-α, preventing it from activating cell surface receptors [28, 29]. In this way, high levels of plasma sTNF-R may prevent a potentially harmful TNF-α response [28]. In contrast, sTNF-R could enhance TNF-α stability by providing a reservoir for long-term immune activity [30]. During HIV infection, large amounts of TNF-α are secreted by activated MPs [31]. CD14 is found principally on human monocytes and exists in both membrane-bound and soluble forms. Increased release of soluble CD14 (sCD14) has been observed from monocytes stimulated by lipopolysaccharide [32]. Increased levels of sCD14 correlate with HIV disease progression [33, 34]

To investigate whether soluble factors secreted in response to monocyte activation are predictive for HAD, we studied 28 HIV-1–infected patients in late-stage disease who were receiving HAART. Correlations between sCD14 and sTNF-Rs were made in patients with advanced HIV infection with and without cognitive dysfunction

Subjects and Methods

SubjectsTwenty-eight patients with advanced HIV disease were recruited from the HIV Clinic at the University of Nebraska Medical Center. The mean±SEM CD4+ T lymphocyte count for all patients was 237±41 cells/mm3, and the mean virus load was 77,091±195,372 HIV-1 RNA copies/mL (determined by use of a Roche Diagnostic Systems assay). Laboratory tests included lymphocyte subset analysis and measurement of HIV load in plasma. In vivo brain metabolites were examined by 1H MRS. At the time of evaluation, all patients were free of acute systemic or opportunistic infections. Six HIV-1–seronegative healthy control subjects served to standardize the laboratory tests

Neuropsychological testsA battery of neuropsychological tests sensitive to HAD [35] was administered. The Rey Auditory Verbal Learning Test was used to measure verbal learning (total words recalled on trials 1–5) and memory (delayed recall trial) [36]. Complex attention was assessed by using part B of the Trail Making Test [37] and the color-word interference trial from the Stroop Test [38]. Psychomotor speed was assessed with the Symbol Digit Modalities Test [39], and Grooved Pegboard Test [40] performances were used to measure fine motor skills. Each subject’s results were converted to age-corrected z scores, using appropriate normative information from the sources cited above, except for part B of the Trail Making Test and the Grooved Pegboard Test. For those two tests, normative data from Heaton and colleagues [41] were relied on to compute z scores. Subjects with performances 1.5 SD below the age-adjusted normative mean for any test were considered to be impaired on that test. Subjects who demonstrated impairment in any domain (i.e., verbal learning, verbal memory, complex attention, psychomotor speed, or fine motor skills) were further classified as cognitively impaired. Subjects with all test scores <1.5 SD below the age-adjusted normative mean were classified as cognitively intact

MRI and1H MRSScout brain MRI and 1H MRS were performed on all patients. T2-weighted MRIs were evaluated by a board-certified neuroradiologist who did not know the subject’s clinical status or neuropsychological test results. Determinations of atrophy were made by measuring the width of the superior frontal sulcus. All measurements ⩾3 mm were considered to be atrophic [42]. To adjust for age-related atrophy, those subjects >1.5 SD above the mean age of the patient cohort (n=3) were not considered in this analysis. A preliminary T1-weighted axial MRI of the brain was used for selection of a voxel of tissue (6–8 cc). Single-voxel proton spectra were acquired from the midline of the posterior parietal lobes (gray matter) by use of PROBE-p software (on the 1.5 Tesla Signal system; General Electric). A fully automated PROBE-p PRESS sequence was used. Metabolite measurements were given as relative peak intensities of NAA, MI, and CHO with respect to CR and phosphocreatine

Measurements of plasma levels of sCD14, TNF-α, sTNF type I receptor (sTNF-RI), and sTNF type II receptor (sTNF-RII)Whole blood was collected by peripheral draw and then was placed into tubes containing either sodium polyanetholesulfonate or a clotting agent. Plasma or serum was separated from cells by centrifugation and then was collected and frozen. Levels of sCD14, TNF-α, sTNF-RI, and sTNF-RII were measured by ELISA (R&D Systems), following the manufacturer’s instructions. Samples were measured in triplicate during 2 separate experiments

Statistical analysisThe 1-tailed Mann-Whitney U test was used to compare 2 groups. Significance was considered to be P<.05. Correlation analysis was performed using the Spearman&amp;rank correlation coefficient. All data are presented as the mean±SEM, unless otherwise stated

Results

Subject characterizationClinical and demographic characteristics for the subjects are shown in table 1. Seventeen subjects had brain atrophy, as determined by MRI, and 8 had normal scan results. The mean age, virus load, and monocyte and CD4+ T cell counts were similar in each study group (table 1). The mean duration of HAART use for all patients was 20.4±2.0 months. Twenty subjects were considered to be cognitively impaired (table 2), and 8 showed no signs of cognitive dysfunction

Table 1

Clinical characteristics of all study subjects

Table 1

Clinical characteristics of all study subjects

Table 2

Summary of neuropsychological test results for 28 human immunodeficiency virus type 1–infected subjects receiving highly active antiretroviral therapy

Table 2

Summary of neuropsychological test results for 28 human immunodeficiency virus type 1–infected subjects receiving highly active antiretroviral therapy

Plasma levels of TNF-α and TNF-RTNF-α is one of the first cytokines to be secreted by MPs after activation. It has been shown to enhance HIV-1 replication and to kill infected cells [43, 44]. Of importance, increased levels of TNF-α mRNA have been found in the brains of patients with HAD, compared with infected individuals without neurological disease [45, 46]

HIV-1–infected subjects demonstrated significantly elevated plasma levels of TNF-α, compared with levels for seronegative control subjects; however, no significant differences were observed between infected patients with and without cognitive impairment (figure 1A). Overall, the absolute levels of detected TNF-α were low. Subjects with cognitive impairment (n=20) had mean TNF-α levels of 7.53±0.9 pg/mL, versus 7.82±1.0 pg/mL for those with normal cognitive function (n=8) and 1.70±0.2 pg/mL for HIV-seronegative healthy control subjects (n=6)

Figure 1

Plasma levels of tumor necrosis factor (TNF)–α, soluble TNF-α type I receptor (sTNF-RI), sTNF-α type II receptor (sTNF-RII), and soluble CD14 (sCD14) in human immunodeficiency virus type 1 (HIV-1)–infected subjects with abnormal neuropsychological test results. TNF-α (A) sTNF-RI (B) sTNF-RII (C) and sCD14 (D) were measured in subject plasma by ELISA. Subjects were grouped on the basis of cognitive impairment, as determined by neuropsychological testing. Subjects with test scores >1.5 SD below the mean (n=20) were considered to be impaired. Cytokine and receptor levels were compared among the cognitively impaired (+) and unimpaired (−) seropositive groups and the seronegative control subjects (n=6). Data were analyzed by use of the 1-tailed Mann-Whitney U test. Error bars SEM. **P<.01, seronegative control subjects vs. HIV-infected patients

Figure 1

Plasma levels of tumor necrosis factor (TNF)–α, soluble TNF-α type I receptor (sTNF-RI), sTNF-α type II receptor (sTNF-RII), and soluble CD14 (sCD14) in human immunodeficiency virus type 1 (HIV-1)–infected subjects with abnormal neuropsychological test results. TNF-α (A) sTNF-RI (B) sTNF-RII (C) and sCD14 (D) were measured in subject plasma by ELISA. Subjects were grouped on the basis of cognitive impairment, as determined by neuropsychological testing. Subjects with test scores >1.5 SD below the mean (n=20) were considered to be impaired. Cytokine and receptor levels were compared among the cognitively impaired (+) and unimpaired (−) seropositive groups and the seronegative control subjects (n=6). Data were analyzed by use of the 1-tailed Mann-Whitney U test. Error bars SEM. **P<.01, seronegative control subjects vs. HIV-infected patients

Similar levels of TNF-α for patient groups may be the result of physiologic compensatory responses. As TNF-α levels increase, soluble TNF receptors present in patient plasma may serve to neutralize TNF-α or bind it to the surface of other cells. If either were true, this would prevent the detection of TNF-α in plasma. This possibility was examined by measuring sTNF-RI and sTNF-RII. Indeed, both receptors were significantly elevated in plasma from infected patients, compared with levels in HIV-1–seronegative control subjects (figure 1B and 1C). However, no significant differences were found between subject groups. Cognitively impaired subjects (n=20) had sTNF-RI and sTNF-RII levels of 2.0±0.3 and 4.2±0.4 ng/mL, respectively, whereas cognitively unimpaired HIV-1–seropositive individuals (n=8) had levels of 1.7±0.1 and 3.6±0.5 ng/mL, respectively. HIV-1–seronegative control subjects had mean levels of 1.1±0.2 and 1.4±0.2 ng/mL

Plasma levels of sCD14CD14 is released from monocytes in a soluble form after viral infection and immune activation [33, 47]. Levels of sCD14 in patient serum increase with disease progression [34] and are elevated after gp120 stimulation [33]. In our study cohort, sCD14 levels were significantly increased in the 20 cognitively impaired subjects, compared with levels in the 8 HIV-1–infected subjects without cognitive dysfunction and in the seronegative control subjects (2.8±0.1 vs. 2.1±0.2 and 1.2±0.1 μg/mL, respectively; P<.01; figure 1D)

Links between sTNF-RII and sCD14 levels and brain atrophyLinks between MP activation and neuronal injury make it possible that soluble factors in the periphery may be associated with brain atrophy. Figure 2A shows a significant increase in sTNF-RII levels in subjects with brain atrophy (n=17; 4.6±0.5 ng/mL), compared with levels in subjects with normal MRI scan results (n=8; 3.1±0.3 ng/mL; P<.05). Similar findings were demonstrated for sCD14 (figure 2C). Subjects with brain atrophy (n=17) had mean sCD14 levels of 2.8±0.02 μg/mL, whereas those with normal MRI results (n=8) had levels of 2.3±0.03 μg/mL (P<.05). No significant correlations were observed for TNF-α or sTNF-RI (data not shown)

Figure 2

Plasma levels of soluble tumor necrosis factor–α type II receptor (sTNF-RII) and soluble CD14 (sCD14) among human immunodeficiency virus type 1–infected patients with (+) and without (−) brain atrophy. Levels of sTNF-RII (A) and sCD14 (B) were measured in plasma by ELISA. Subjects were grouped on the basis of the presence (n=17) or absence (n=8) of brain atrophy on magnetic resonance imaging, as determined by a certified neuroradiologist blinded to the subject’s condition. Data were analyzed by use of the 1-tailed Mann-Whitney U test. Error bars SEM. *P<.05

Figure 2

Plasma levels of soluble tumor necrosis factor–α type II receptor (sTNF-RII) and soluble CD14 (sCD14) among human immunodeficiency virus type 1–infected patients with (+) and without (−) brain atrophy. Levels of sTNF-RII (A) and sCD14 (B) were measured in plasma by ELISA. Subjects were grouped on the basis of the presence (n=17) or absence (n=8) of brain atrophy on magnetic resonance imaging, as determined by a certified neuroradiologist blinded to the subject’s condition. Data were analyzed by use of the 1-tailed Mann-Whitney U test. Error bars SEM. *P<.05

Links between brain metabolites and soluble immune factorsSeveral studies have reported that MRS can provide diagnostic information not available from conventional MRI scans [14, 48, 49]. Although we were unable to detect statistically significant differences between metabolite ratios in the cognitively impaired and unimpaired subject groups, there were trends of elevated CHO:CR ratios (figure 3A), but few changes in MI:CR (figure 3B) or NAA:CR ratios (data not shown). CHO:CR was correlated with both sCD14 (R=.4; P<.05) and TNF-RII (R=.5; P<.01; figure 4A and 4D), but not with MI:CR. sTNF-RII showed positive trends that failed to reach significance (R=.3; P=.06)

Figure 3

Ratios of myoinositol (MI) and choline (CHO) to creatine (CR) in human immunodeficiency virus type 1–infected patients with (+) or without (−) cognitive impairment. The relative peak intensities of CHO (A) and MI (B) with respect to CR, in the posterior parietal region of the brain, were determined for all subjects. Subjects were grouped on the basis of the presence (n=20) or absence (n=8) of cognitive impairment. Boxes encompass the interquartile ranges; error bars indicate the ranges of values. Horizontal lines median values for each group. P>.05

Figure 3

Ratios of myoinositol (MI) and choline (CHO) to creatine (CR) in human immunodeficiency virus type 1–infected patients with (+) or without (−) cognitive impairment. The relative peak intensities of CHO (A) and MI (B) with respect to CR, in the posterior parietal region of the brain, were determined for all subjects. Subjects were grouped on the basis of the presence (n=20) or absence (n=8) of cognitive impairment. Boxes encompass the interquartile ranges; error bars indicate the ranges of values. Horizontal lines median values for each group. P>.05

Figure 4

Brain metabolites, compared with plasma levels of soluble tumor necrosis factor–α type II receptor (sTNF-RII) and soluble CD14 (sCD14). The relative peak intensities of choline (CHO) and myoinositol (MI) with respect to creatine (CR), in the posterior parietal region of the brain, were determined for all subjects. Plasma levels of sTNF-RII and sCD14 were determined. Correlation analyses were performed, using the Spearman&amp;rank correlation coefficient, among CHO:CR (A) and MI:CR (C) and sCD14 (A and C). Similar correlation analyses were performed for CHO:CR (B) and MI:CR (D) and sTNF-RII (B and D)

Figure 4

Brain metabolites, compared with plasma levels of soluble tumor necrosis factor–α type II receptor (sTNF-RII) and soluble CD14 (sCD14). The relative peak intensities of choline (CHO) and myoinositol (MI) with respect to creatine (CR), in the posterior parietal region of the brain, were determined for all subjects. Plasma levels of sTNF-RII and sCD14 were determined. Correlation analyses were performed, using the Spearman&amp;rank correlation coefficient, among CHO:CR (A) and MI:CR (C) and sCD14 (A and C). Similar correlation analyses were performed for CHO:CR (B) and MI:CR (D) and sTNF-RII (B and D)

Relationship between HAART and soluble immune factorsAll subjects had received HAART for an average of 20±2 months. No correlations were observed between the duration of HAART use and neuropsychological test results, spectroscopic metabolites, virus load, or sCD14 level. There was, however, a negative correlation between the duration of HAART use and the level of sTNF-RII (R=-.3; P<.05)

We considered control of the virus to be successful if patients had <500 HIV-1 RNA copies/mL. The 13 subjects with virus loads of <500 copies/mL had significantly higher levels of sCD14 than did the 15 subjects with virus loads of ⩾500 copies/mL (2.85 ± 0.017 vs. 2.40 ± 0.018 μg/mL; P<.05). Of interest, 11 of the 13 patients for whom virus was successfully controlled were cognitively impaired

Discussion

In this study, we used cross-validating approaches to determine cognitive dysfunction in HIV-1–infected subjects with, or at risk for, HAD. Neuropsychological tests, MRI, and MRS assessed mental function, brain atrophy, and neuronal and glial metabolites. The results showed that HIV-1–infected subjects with cognitive impairment had elevated plasma levels of sCD14 and sTNF-RII, compared with levels among seronegative control subjects or HIV-1–infected individuals without mental impairment. Moreover, sCD14 and sTNF-RII levels correlated with CHO:CR ratios and brain atrophy, as determined by 1H MRS and MRI examinations

Use of HAART has decreased, but not eliminated, the incidence and severity of neurological complications seen in HIV-1–infected patients [1]. Although people with HIV-1 are living longer and healthier lives, resistance to antiretroviral therapy, as a consequence of virus strain mutation and/or impaired ability of drugs to penetrate the blood-brain barrier, strongly suggests that HAD will continue to be a significant complication of advanced HIV disease [50]. Despite the availability of highly specialized viral, cognitive, and immune tests, HAD remains difficult to diagnose. This difficulty emphasizes the need for surrogate markers to monitor disease progression and therapeutic responses

Of importance, although use of HAART can decrease levels of plasma cytokines, monocyte subsets [51], sTNF receptors [52], and brain metabolites [53], these likely plateau over time. The subjects in this cohort had been using HAART for an average of 20 months. There was no significant difference in duration of HAART use for any of the study groups used in this analysis. In addition, no correlation was observed between months of HAART use and neuropsychological test results, spectroscopic metabolites, or virus load. There was, however, a negative correlation between months of HAART use and sTNF-RII levels. This finding is in accordance with other reports [52]

When the subjects were grouped on the basis of successful response to HAART, those subjects with virus loads of <500 HIV-1 RNA copies/mL had significantly elevated levels of sCD14 but not of TNF-α or sTNF-RII. In addition, 11 of the 13 subjects in this category were cognitively impaired. Although this observation supports the idea that correlations observed between levels of sCD14 and cognitive dysfunction are independent of HAART, more studies are required to confirm this preliminary observation. Certainly, the influence of HAART on soluble factor production and cognitive dysfunction in HIV-1–infected patients remains an important question under intensive investigation [1, 24, 53–55]

A number of reports strongly suggest that HIV enters the brain early after viral infection [56, 57] and only years later, through persistent infection, establishes a significant reservoir within brain macrophages and microglia [58, 59]. Recent attention has focused on the interplay between immune-activated monocytes and neurological impairments. Reports suggest that peripheral monocytes are primed by virus and subsequent immune activation, during late-stage HIV infection, for transendothelial migration of infected cells across an impaired blood-brain barrier [60–62]. Once within the brain, the cells may be further activated by pro-inflammatory cytokines, chemokines, or opportunistic infection to cause neuronal injury [63]

Up-regulation of monocytes expressing CD16 and CD69 in subjects with HAD [23] supports the idea that a particular subset of monocytes, found in the periphery, emerges during disease and may play a role in the onset or progression of dementia. sCD14 is shed from activated monocytes [32], and sTNF-RII is released from cell surfaces in the presence of high levels of TNF-α [28], a cytokine secreted by activated monocytes. sCD14 and sTNF-RII have already been linked to HIV-1 disease progression [3, 25–27, 33]. Our findings of increased plasma sCD14 and sTNF-RII in subjects with cognitive impairment further support an interplay between neurological dysfunction and peripheral immune responses

We did not observe significantly elevated CHO:CR or MI:CR ratios in subjects with neurocognitive impairment, as determined by neuropsychological testing. We did, however, see trends of elevated CHO:CR ratios. This finding is in agreement with those from previous reports that demonstrate that CHO:CR is more consistently elevated in patients with HAD [14, 64]

Significant correlations between sTNF-RII and sCD14 levels and results of the MRS test, which is used to assess early neurological disease [7, 8], suggest that these markers may be useful in evaluating mildly impaired patients, such as those in this study. The strong links between sCD14 and sTNF-RII levels and the CHO:CR ratio are of particular interest, because this ratio reflects both glial activation and MP migration [7]. Because the influx of MPs into the central nervous system might occur before cognitive dysfunction, it is likely that the soluble factors become elevated early in the course of neurological disease. Moreover, since monocyte numbers were similar between subject groups, our findings suggest that macrophage activation is a driving force for disease

The specific cause-and-effect relationship between peripheral markers and neurological dysfunction remains incompletely defined. One idea is that sTNF-RII is a protective and compensatory response. sTNF-RI and -RII are cleaved from cell surfaces in response to increased TNF-α levels [65]. Thus, it is possible that these receptors represent a protective mechanism whereby TNF-α is neutralized [28]. Consequently, soluble TNF-α receptors may act as indirect markers of macrophage activation. Like sTNF-RII, CD14 is shed at the time of activation, likely as a means of modulating CD14 expression [47]. Viral infection and resulting immune activation may act as a stimulus for sCD14 release. Indeed, levels of cell surface CD14 and sCD14 increase during disease progression [33, 34]

Further studies are necessary to ascertain whether soluble markers change with neurological disease progression. However, our findings suggest that determination of levels of inflammatory factors in the blood, together with MRI, MRS, and neuropsychological testing, may prove to be useful in the treatment and monitoring of HAD. Most importantly, this report highlights the importance of disordered monocyte immune function in HIV-1 neuropathogenesis

Acknowledgments

We kindly thank the patients for participating in this study; Susan Morgello and Michael Boska for constructive comments and ideas; Lori Todd, Alicia Lopez, Clancy Williams, and My Hanh Che for technical support; and Julie Ditter and Robin Taylor for outstanding administrative and secretarial support

References

1
Sacktor
N
Lyles
RH
Skolasky
R
, et al.  . 
HIV-associated neurologic disease incidence changes: multicenter AIDS Cohort Study, 1990–1998
Neurology
 , 
2001
, vol. 
56
 (pg. 
257
-
60
)
2
McArthur
JC
Sacktor
N
Selnes
O
Human immunodeficiency virus–associated dementia
Semin Neurol
 , 
1999
, vol. 
19
 (pg. 
129
-
50
)
3
Stein
DS
Lyles
RH
Graham
NMH
, et al.  . 
Predicting clinical progression or death in subjects with early stage HIV infection: a comparative analysis of quantification of HIV RNA, soluble tumor necrosis factor type II receptors, neopterin and B2-microglobulin
J Infect Dis
 , 
1997
, vol. 
176
 (pg. 
1161
-
7
)
4
Gelbard
H
Nottet
H
Dzenko
K
, et al.  . 
Platelet-activating factor: a candidate human immunodeficiency virus type-1 infection neurotoxin
J Virol
 , 
1994
, vol. 
68
 (pg. 
4628
-
35
)
5
Heyes
MP
Brew
BB
Martin
A
, et al.  . 
Quinolinic acid in cerebrospinal fluid and serum in HIV-1 infection: relationship to clinical and neurological status
Ann Neurol
 , 
1991
, vol. 
29
 (pg. 
202
-
9
)
6
Swindells
S
Ong
CS
, et al.  . 
Prospective utility of cerebral proton magnetic resonance spectroscopy in monitoring HIV infection and its associated neurological impairment
AIDS Res Hum Retroviruses
 , 
1994
, vol. 
10
 (pg. 
977
-
82
)
7
Jarvik
JG
Lenkinski
RE
Grossman
RI
Gomori
JM
Schnall
MD
Proton MR spectroscopy of HIV-infected patients: characterization of abnormalities with imaging and clinical correlation
Radiology
 , 
1993
, vol. 
186
 (pg. 
739
-
44
)
8
Tracey
I
Carr
CA
Guimaraes
AR
Worth
JL
Navia
BA
Brain choline containing compounds are elevated in HIV-positive patients before the onset of AIDS dementia complex
Neurology
 , 
1996
, vol. 
46
 (pg. 
783
-
8
)
9
Navia
BA
Lee
RPL
Chang
L
, et al.  . 
Metabolite changes in the basal ganglia provide a signature for the AIDS dementia complex: a multi-center proton magnetic resonance study [abstract]
Neurology
 , 
1999
, vol. 
52
 pg. 
A254
 
10
Yiannoutsos
C
Miller
E
Chang
L
, et al.  . 
Proton MRS and neuropsychological assessments in AIDS dementia complex: a measure of subcortical function [abstract]
Neurology
 , 
1999
, vol. 
52
 (pg. 
A253
-
4
)
11
McArthur
JC
Hoover
DR
Bacellar
H
, et al.  . 
Dementia in AIDS patients: incidence and risk factors
Neurology
 , 
1993
, vol. 
43
 (pg. 
2245
-
52
Multicenter AIDS Cohort Study
12
Duncan
R
Quencer
RM
Pall
L
Asymptomatic and neurologically symptomatic HIV-seropositive subjects: results of long-term MR imaging and clinical follow-up
Radiology
 , 
1993
, vol. 
188
 (pg. 
727
-
33
)
13
Handelsman
L
Song
IS
Losonczy
M
, et al.  . 
Magnetic resonance abnormalities in HIV infection: a study in the drug-user risk group
Psychiatry Res
 , 
1993
, vol. 
47
 (pg. 
175
-
86
)
14
Chong
WK
Sweeney
B
Wilkinson
ID
, et al.  . 
Proton spectroscopy of the brain in HIV infection: correlation with clinical, immunologic, and MR imaging findings
Radiology
 , 
1993
, vol. 
188
 (pg. 
119
-
24
)
15
Jarvik
JG
Lenkinski
RE
Saykin
AJ
Jaans
A
Frank
I
Proton spectroscopy in asymptomatic HIV-infected adults: initial results in a prospective cohort study
J Acquir Immune Defic Syndr Hum Retrovirol
 , 
1996
, vol. 
13
 (pg. 
247
-
53
)
16
Meyerhoff
DJ
Bloomer
C
Cardenas
V
Norman
D
Weiner
MW
Elevated subcortical choline metabolites in cognitively and clinically asymptomatic HIV+ patients
Neurology
 , 
1999
, vol. 
52
 (pg. 
995
-
1003
)
17
Tornatore
C
Nath
A
Amemiya
K
Major
EO
Persistent human immunodeficiency virus type 1 infection in human fetal glial cells reactivated by T-cell factor(s) or by the cytokines tumor necrosis factor alpha and interleukin-1 beta
J Virol
 , 
1991
, vol. 
65
 (pg. 
6094
-
100
)
18
Moses
AV
Bloom
FE
Pauza
CD
Nelson
JA
HIV infection of human brain capillary endothelial cells occurs via a CD4/galactosylceramide-independent mechanism
Proc Natl Acad Sci USA
 , 
1993
, vol. 
90
 (pg. 
10474
-
8
)
19
Ranki
A
Nyberg
M
Ovod
V
, et al.  . 
Abundant expression of HIV Nef and Rev proteins in brain astrocytes in vivo is associated with dementia
AIDS
 , 
1995
, vol. 
9
 (pg. 
1001
-
8
)
20
Nath
A
Hartloper
V
Furer
M
Fowke
KR
Infection of human fetal astrocytes with HIV-1: viral tropism and the role of cell to cell contact in viral transmission
J Neuropathol Exp Neurol
 , 
1995
, vol. 
54
 (pg. 
320
-
30
)
21
Ma
M
Geiger
JD
Nath
A
Characterization of a novel binding site for the human immunodeficiency virus type 1 envelope protein gp120 on human fetal astrocytes
J Virol
 , 
1994
, vol. 
68
 (pg. 
6824
-
8
)
22
Zheng
J
Thylin
MR
Persidsky
Y
, et al.  . 
HIV-1 infected immune competent mononuclear phagocytes influence the pathways to neuronal demise
Neurotoxicol Res
  
(in press)
23
Pulliam
L
Gascon
R
Stubblebine
M
Mcguire
D
McGrath
MS
Unique monocyte subset in patients with AIDS dementia
Lancet
 , 
1997
, vol. 
349
 (pg. 
692
-
5
)
24
Gendelman
HE
Zheng
J
Coulter
CL
, et al.  . 
Suppression of inflammatory neurotoxins by highly active antiretroviral therapy in human immunodeficiency virus–associated dementia
J Infect Dis
 , 
1998
, vol. 
178
 (pg. 
1000
-
7
)
25
Aukrust
P
Liabakk
NB
Muller
F
Lien
E
Espevik
T
Serum levels of tumor necrosis factor-alpha (TNF alpha) and soluble TNF receptors in human immunodeficiency virus type 1 infection—correlations to clinical, immunologic, and virologic parameters
J Infect Dis
 , 
1994
, vol. 
169
 (pg. 
420
-
4
(erratum: J Infect Dis 1994; 169:1186–7)
26
Godfried
MH
van der Poll
T
Weverling
GJ
, et al.  . 
Soluble receptors for tumor necrosis factor as predictors of progression to AIDS in asymptomatic human immunodeficiency virus type 1 infection
J Infect Dis
 , 
1994
, vol. 
169
 (pg. 
739
-
45
)
27
Fahey
JL
Taylor
JM
Manna
B
, et al.  . 
Prognostic significance of plasma markers of immune activation, HIV viral load and CD4 T-cell measurements
AIDS
 , 
1998
, vol. 
12
 (pg. 
1581
-
90
)
28
Van Zee
KJ
Kohno
T
Fischer
E
Rock
CS
Moldawer
LL
Tumor necrosis factor soluble receptors circulate during experimental and clinical inflammation and can protect against excessive tumor necrosis factor alpha in vitro and in vivo
Proc Natl Acad Sci USA
 , 
1992
, vol. 
89
 (pg. 
4845
-
9
)
29
Coyne
CP
Willetto
C
Fenwick
BW
A mechanism of TNFR type II (75 kDa) “shedding” in macrophages
Shock
 , 
1999
, vol. 
11
 (pg. 
19
-
28
)
30
Aderka
D
Engelmann
H
Maor
Y
Brakebusch
C
Wallach
D
Stabilization of the bioactivity of tumor necrosis factor and its soluble receptors
J Exp Med
 , 
1992
, vol. 
175
 (pg. 
323
-
9
)
31
Herbein
G
Keshav
S
Collin
M
Montaner
LJ
Gordon
S
HIV-1 induces tumour necrosis factor and IL-1 gene expression in primary human macrophages independent of productive infection
Clin Exp Immunol
 , 
1994
, vol. 
95
 (pg. 
442
-
9
)
32
Landmann
R
Knopf
HP
Link
S
Sansano
S
Schumann
R
Human monocyte CD14 is upregulated by lipopolysaccharide
Infect Immun
 , 
1996
, vol. 
64
 (pg. 
1762
-
9
)
33
Lien
E
Aukrust
P
Sundan
A
Muller
F
Froland
SS
Elevated levels of serum-soluble CD14 in human immunodeficiency virus type 1 (HIV-1) infection: correlation to disease progression and clinical events
Blood
 , 
1998
, vol. 
92
 (pg. 
2084
-
92
)
34
Nockher
WA
Bergmann
L
Scherberich
JE
Increased soluble CD14 serum levels and altered CD14 expression of peripheral blood monocytes in HIV-infected patients
Clin Exp Immunol
 , 
1994
, vol. 
98
 (pg. 
369
-
74
)
35
McArthur
J
Grant
I
Gendelman
HE
Epstein
L
HIV neurocognitive disorders
The neurology of AIDS
 , 
1998
New York
Chapman & Hall
(pg. 
499
-
523
)
36
Lezak
M
Neuropsychological assessment
1983
2d ed
New York
Oxford University Press
37
Partington
JE
Leiter
RG
Partington’s Pathway Test
Psychol Serv Center Bull
 , 
1949
, vol. 
1
 (pg. 
9
-
20
)
38
Mitrushina
MN
Boone
KB
D’Elia LF
Elia
LF
Stroop Test
Handbook of normative data for neuropsychological assessment
 , 
1999
New York
Oxford University Press
(pg. 
74
-
100
)
39
Smith
A
Symbol Digit Modalities Test
Los Angeles: Western Psychological Services
 , 
1991
40
Mitrushina
MN
Boone
KB
D’Elia LF
Elia
LF
Grooved Pegboard Test
Handbook of normative data for neuropsychological assessment
 , 
1999
New York
Oxford University Press
(pg. 
428
-
47
)
41
Heaton
RK
Grant
I
Matthews
CG
Comprehensive norms for an expanded Halstead-Reitan Battery: demographic corrections, research findings and clinical applications
Odessa, FL: Psychological Assessment Resources
 , 
1991
42
Taveras
JM
Wood
EH
Diagnostic neuroradiology
Baltimore: Williams & Wilkins
 , 
1964
43
Matsuyama
T
Kobayashi
N
Yamamoto
N
Cytokines and HIV infection: is AIDS a tumor necrosis factor disease
AIDS
 , 
1991
, vol. 
5
 (pg. 
1405
-
17
[editorial]
44
Folks
TM
Justement
J
Kinter
A
Dinarello
CA
Fauci
AS
Cytokine-induced expression of HIV-1 in a chronically infected promonocyte cell line
Science
 , 
1987
, vol. 
238
 (pg. 
800
-
2
)
45
Tyor
WR
Glass
JD
Griffin
JW
, et al.  . 
Cytokine expression in the brain during acquired immune deficiency syndrome
Ann Neurol
 , 
1992
, vol. 
31
 (pg. 
349
-
60
)
46
Wesselingh
SL
Power
C
Glass
JD
, et al.  . 
Intracerebral cytokine messenger RNA expression in acquired immunodeficiency syndrome dementia
Ann Neurol
 , 
1993
, vol. 
33
 (pg. 
576
-
82
)
47
Bazil
V
Strominger
JL
Shedding as a mechanism of down-modulation of CD14 on stimulated human monocytes
J Immunol
 , 
1991
, vol. 
147
 (pg. 
1567
-
74
)
48
McComb
RD
Gendelman
HE
Utility of cerebral proton magnetic resonance spectroscopy in differential diagnosis of HIV-related dementia
J Neurovirol
 , 
1995
, vol. 
1
 (pg. 
268
-
74
)
49
Salvan
AM
Vion-Dury
J
Confort-Gouny
S
Nicoli
F
Lamoureux
S
Brain proton magnetic resonance spectroscopy in HIV-related encephalopathy: identification of evolving metabolic patterns in relation to dementia and therapy
AIDS Res Hum Retroviruses
 , 
1997
, vol. 
13
 (pg. 
1055
-
66
)
50
Carpenter
CC
Cooper
DA
Fischl
MA
, et al.  . 
Antiretroviral therapy in adults: updated recommendations of the International AIDS Society—USA Panel
JAMA
 , 
2000
, vol. 
283
 (pg. 
381
-
90
)
51
Amirayan-Chevillard
N
Tissot-Dupont
H
Capo
C
, et al.  . 
Impact of highly active anti-retroviral therapy (HAART) on cytokine production and monocyte subsets in HIV-infected patients
Clin Exp Immunol
 , 
2000
, vol. 
120
 (pg. 
107
-
12
)
52
Bonnet
F
Saves
M
Moriat
P
, et al.  . 
Evolution of plasma levels of soluble interleukin-2 (sIL2-R) and tumor necrosis factor type II (sTNFII-R) receptors is correlated with immunologic or virologic responses of patients treated by protease inhibitor–containing regimen [abstract 63]
Program and abstracts of the 8th Conference on Retroviruses and Opportunistic Infections
 , 
2001
Alexandria, VA
Foundation for Retrovirology and Human Health
pg. 
62
 
53
Chang
L
Ernst
T
Leonido-Yee
M
, et al.  . 
Highly active antiretroviral therapy reverses brain metabolite abnormalities in mild HIV dementia
Neurology
 , 
1999
, vol. 
53
 (pg. 
782
-
9
)
54
Clifford
DB
Human immunodeficiency virus–associated dementia
Arch Neurol
 , 
2000
, vol. 
57
 (pg. 
321
-
4
)
55
Dore
GJ
Correll
PK
Li
Y
Kaldor
JM
Cooper
DA
Changes to AIDS dementia complex in the era of highly active antiretroviral therapy
AIDS
 , 
1999
, vol. 
13
 (pg. 
1249
-
53
)
56
Davis
LE
Hjelle
BL
Miller
VE
, et al.  . 
Early viral brain invasion in iatrogenic human immunodeficiency virus infection
Neurology
 , 
1992
, vol. 
42
 (pg. 
1736
-
9
)
57
Navia
BA
Price
RW
The acquired immunodeficiency syndrome dementia complex as the presenting or sole manifestation of human immunodeficiency virus infection
Arch Neurol
 , 
1987
, vol. 
44
 (pg. 
65
-
9
)
58
Gabuzda
DH
Ho
DD
Monte
MSDL
Rota
TR
Sobel
RA
Immunohistochemical identification of HTLV-III antigen in brains of patients with AIDS
Ann Neurol
 , 
1986
, vol. 
20
 (pg. 
289
-
95
)
59
Koenig
S
Gendelman
HE
Orenstein
JM
, et al.  . 
Detection of AIDS virus in macrophages in brain tissue from AIDS patients with encephalopathy
Science
 , 
1986
, vol. 
233
 (pg. 
1089
-
93
)
60
Liu
Y
Tang
XP
McArthur
JC
Scott
J
Gartner
S
Analysis of human immunodeficiency virus type 1 gp160 sequences from a patient with HIV dementia: evidence for monocyte trafficking into brain
J Neurovirol
 , 
2000
, vol. 
6
 
Suppl 1
(pg. 
S70
-
81
)
61
Nottet
HS
Persidsky
Y
Sasseville
VG
, et al.  . 
Mechanisms for the transendothelial migration of HIV-1–infected monocytes into brain
J Immunol
 , 
1996
, vol. 
156
 (pg. 
1284
-
95
)
62
Gartner
S
HIV infection and dementia
Science
 , 
2000
, vol. 
287
 (pg. 
602
-
4
)
63
Zheng
J
Gendelman
HE
The HIV-1 associated dementia complex: a metabolic encephalopathy fueled by viral replication in mononuclear phagocytes
Curr Opin Neurol
 , 
1997
, vol. 
10
 (pg. 
319
-
25
)
64
Lopez-Villegas
D
Lenkinski
RE
Frank
I
Biochemical changes in the frontal lobe of HIV-infected individuals detected by magnetic resonance spectroscopy
Proc Natl Acad Sci USA
 , 
1997
, vol. 
94
 (pg. 
9854
-
9
)
65
Howard
OM
Clouse
KA
Smith
C
Goodwin
RG
Farrar
WL
Soluble tumor necrosis factor receptor: inhibition of human immunodeficiency virus activation
Proc Natl Acad Sci USA
 , 
1993
, vol. 
90
 (pg. 
2335
-
9
)
Presented in part: 8th Conference on Retroviruses and Opportunistic Infections, Chicago, 4–8 February 2001 (abstract 7)
Informed consent was obtained from the patients, and human experimental guidelines of the US Department of Health and Human Services and the University of Nebraska Medical Center (UNMC) institutional review board were followed
Financial support: National Institutes of Health (NS-31492, MH-57556, NS-34239, and NS-36126 to H.E.G.; RR-15635 to J.Z. and H.E.G.); UNMC Research Support