Department of Medical Microbiology and Immunology, University of Alberta, Edmonton, Alberta, Canada
Correspondence: Peter Dickie, 1-40 HMRC, University of Alberta, Edmonton, AB T6G 2S2, Canada. E-mail: peter.dickie{at}ualberta.ca
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gag/pol) proviral DNA were tested for the systemic
production of nitric oxide (NO). Serum levels of NO metabolites were
reduced about 50% in HIV transgenic mice compared with nontransgenic
sibling mice. This difference persisted when NO production was induced
with peritoneal injections of bacterial endotoxin (LPS). Peritoneal
inflammatory macrophages, but not resident peritoneal macrophages,
derived from HIV-1 transgenic mice and activated in vitro with LPS and
IFN-
(or tumor necrosis factor 
and IFN-) also produced about
50% less NO than did macrophages harvested from nontransgenic
littermates. Isogenic, transgenic mice bearing mutated nef
or vpr genes had normal serum levels of NO metabolites and
their macrophages produced normal levels of NO when stimulated. An
explanation for the reduced NO response of HIV[Vpr+Nef+] macrophages
was not apparent from measured levels of iNOS expression, viral gene
expression, or arginase activity in activated macrophages. Inhibition
of nitric oxide synthase (NOS) isoforms with L-NAME or
aminoguanidine blocked time-dependent increases in HIV gene expression
in activated macrophages cultured ex vivo. Inhibition with
L-NAME occurred despite high levels of NO generated by
iNOS, and exogenously supplied NO induced HIV gene expression only
weakly, suggesting that cNOS had the greater influence on proviral gene
induction. This system is presented as a model of HIV-1 proviral gene
expression and dysfunction in macrophages.
Key Words: transcription • arginase • TNF- • NOS • Nef • Vpr
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There are studies showing that the production of NO is elevated in
human immunodeficiency virus (HIV)-1-infected patients
[9
, 10
], a phenomenon replicated in acute
simian immunodeficiency virus (SIV)-infection models
[11
]. A similar induction of NO was observed in acute
HIV infections of human cells in vitro [12
,
13
]. Elevated levels of NO could have a variety of
deleterious, immunologic effects contributing to AIDS, including the
deactivation of lymphocytes, the induction of apoptosis, and the
suppression of T helper cell type 1 (Th1) responses
[14
15
16
]. HIV Env protein alone can activate NO
production in cultured monocytes, astrocytoma, and glial cells
[17
18
19
], and this may be critical to the development of
AIDS dementia complex (ADC). In support of this hypothesis, recent
studies have pointed to the coexistence of viral protein, iNOS, and
macrophage/microglia in brain centers [20
,
21
]. One neuropathologic effect of HIV gp120 coat protein
may be the stimulation of neuronal N-methyl-D-aspartate (NMDA)
receptors that is mediated in part by NO [22
,
23
]. The HIV protein Tat up-regulates nuclear factor
(NF)-
B expression, a transactivator of the iNOS gene, and exogenous
Tat has been shown to up-regulate NO production in microglial cells
[24
]. However, it remains uncertain whether NO
production is elevated in HIV-infected cells [25
] or
whether infected cells induce NO production in bystander cells as
demonstrated in macrophage-astrocyte co-cultures [26
].
In other scenarios, an underproduction of NO has been associated with HIV infection. For example, the infection of monocytes/macrophages from normal individuals with Rhodococcus equi induced NO production, whereas the infection of cells derived from HIV-positive patients failed to induce NO [27 ]. In AIDS patients with toxoplasmic encephalitis, it was discovered that central nervous system (CNS)-related NO production was not elevated and possibly nonoptimal to control the recurrence of Toxoplasma gondii infection [28 ]. A similar observation was made in HIV-positive patients with pulmonary tuberculosis. In this case, patients had normal serum levels of NO metabolites, whereas HIV-negative patients with pulmonary tuberculosis displayed elevated levels of NO metabolites [29 ]. Reduced NO production was also apparent in a cohort of HIV-positive patients as determined from the concentration of NO metabolites in exhaled air [30 ]. Collectively, these observations suggest that HIV infection could result in inadequate NO responses that contribute to secondary infections.
NO has an established, anti-HIV effect, demonstrated in acute
infections of human peripheral blood mononuclear cells (PBMCs)
[31
] and astrocytoma cells [32
]. Numerous
mechanisms for this exist, including the inhibition of HIV protease and
reverse transcriptase [33
, 34
] and the
inactivation of NF-B [35
] by blocking the
inactivation of I-B, an inhibitor of NF-B, or of DNA-binding by
NF-B [36
]. It is interesting that there are at least
two studies suggesting that NO may activate virus in chronically
infected cultures. NO donor compounds stimulated virus replication in
latently infected U1 macrophage cells [31
], and the
cross-linking of CD23 (Fc
RII), known to stimulate NO production,
induced p24 production in a variant of U1 macrophages
[37
].
HIV-1 transgenic mice model a form of chronically, nonproductively infected cells [38 ] in which viral gene expression can be induced with microbial infection [39 ]. In animals carrying a noninfectious, proviral DNA (based on the clone pEVd1443 [40 ], deleted in gag and pol), inducible viral gene expression was supported by peritoneal macrophages [38 ]. Thus, uncomplicated by the influences of an active HIV infection and secondary infections, this model has enabled a study of the intrinsic effect of chronic, low-level HIV gene expression on NO metabolism in animals and in macrophages derived from them. In addition, this system has been used to examine the role of NO in the induction of HIV transcription. We discovered that NO production in response to cytokine and lipopolysaccharide (LPS) activation is significantly reduced in HIV transgenic mice, and this phenomenon is dependent on the HIV nef and vpr genes. We also provide experimental evidence for the NO-dependent induction of HIV gene expression in this system.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gag/Pol] or T26, bore a
proviral DNA deleted in gag and pol, but the
genes for Env and the six accessory gene products remained intact
[38
, 40
]. Variants of this line bearing
mutations in nef (X5) or vpr (V13) were included
[38
]. We identified transgenic offspring in lines T26
and V13 on the basis of congenital cataracts that characterized
HIV-positive mice expressing Nef [41
]. Transgenic X5
offspring were identified by polymerase chain reaction amplification of
the nef gene in tail DNA preparations [38
].
The three lines were maintained in the hemizygous state with respect to
the HIV transgene. They were housed in conventional holding facilities
and treated in accordance with Canadian Council on Animal Care
regulations. All animals used experimentally were between 7 and 12
weeks old.
Peritoneal macrophages were harvested from untreated animals (resident
macrophages) or mice injected intraperitoneally (i.p.) with 1.5 ml 3%
thioglycollate medium 4 days before harvesting (inflammatory
macrophages). Peritoneal cavities were washed with 6 ml sterile,
ice-cold, phosphate-buffered saline (PBS), and the cells were collected
by centrifugation. Peritoneal exudate cells were washed twice with cold
PBS and counted before resuspending in RPMI medium supplemented with
10% fetal calf serum, 1 mM glutamate, and ß-mercaptoethanol
(4x10-5 M). Cells were plated in tissue
culture-treated, 96-well plates at 2 x 105 cells per
well, in 24-well dishes at 2 x 106 cells per well, or
in 6-well dishes at 5 x 106 cells/well, depending on
the experiment. Cells were incubated at 37°C, 5% CO2,
for 3 h before washing 3x with warm PBS to remove nonadherent
cells. Macrophage monolayers were stimulated with Escherichia
coli LPS (serotype 055:B5, Sigma Chemical Co., St. Louis, MO) at 1
µg/ml, tumor necrosis factor- (TNF-; Life Technologies,
Bethesda, MD) at 25 µg/ml, and interferon- (IFN-; Life
Technologies) at 10 U/ml, or were left untreated.
Modulation of NO in mice and macrophage cultures
As a source of exogenous NO, the NO-donor,
S-nitro-N-acetylpenicillamine (SNAP; Sigma), was added to the culture
medium overlaying macrophage monolayers. SNAP was added at a
concentration of 0.4 mM (from an 80-mM stock solution, in dimethyl
sulfoxide) and replenished every 6 h. Control cultures received
carrier alone. Two competitive inhibitors of NOS were used to block the
endogenous generation of NO: L-NAME (NG-nitro-arginine-methyl ester;
Sigma) and aminoguanidine (AG; Sigma). For in vivo experiments, LPS
(200 µg/mouse) was administered i.p. in a volume of 0.1 ml. Control
animals received carrier (saline) alone.
MTT ([3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium
bromide] assay
As a measure of cell viability and for the normalization
of experimental data, the MTT assay of mitochondrial activity was
applied. MTT (Sigma) was dissolved in PBS to 5 mg/ml and filtered. A
10-µl vol was added to macrophage monolayers in 96-well plates, and
the plates were incubated at 37°C for 4 h. After fluid was
withdrawn, the blue precipitate was dissolved in 100 µl acid-propanol
(0.1 N HCl). Absorbance was read at 570 nm in a microplate reader.
Measurement of NO production and arginase activity
Screening for nitrate in mouse sera provided estimates of
systemic NO production [42
]. Briefly, mouse serum (100
µl of a 1/20 dilution in 0.2 M HEPES, pH 7.4) was incubated with 50
µl of 0.2 M HEPES and 50 µl of 5 x 109
Pseudomonas oleovorans diluted in phenol red-free DMEM (as a
source of nitrate reductase) for 90 min at 37°C. After clarifying the
sample by centrifugation, nitrite was measured by combining 100 µl
sample and 100 µl Griess reagent [0.5% sulfanilamide and 0.05%
N-(1-naphthyl)ethylenediamine dihydrochloride in 2.5%
H3PO4] [43
] in 96-well dishes
and by reading the absorbance at 570 nm in an ELISA plate reader. To
measure NO produced by macrophage monolayers, 100 µl culture medium
was combined with 100 µl Griess reagent in micro-well plates. Nitrite
concentrations were determined by comparisons with a standard curve of
NaNO in the range of 1–100 µM.
To measure arginase activity in macrophage monolayers
[44
], cells were lysed with 100 µl 0.1% Triton X-100
added to wells of a 96-well microtiter plate seeded with 2 x
105 exudate cells. After 30 min on ice, 100 µl 25 mM
Tris-HCl (pH 7.4) was added, 100 µl of this mixture was combined with
10 µl of 10 mM MnCl2, and the enzyme was activated by
heating to 56°C for 10 min. The lysate was then mixed with 100 µl
of 0.5 M arginine (pH 9.7) and incubated at 37°C for 60 min. The
reaction was stopped with 900 µl H2SO4
(96%)/H3PO4 (85%)/H2O (1/3/7).
Then, 40 µl -isonitrosopropriophenone (in 100% ethanol) was
added, and the mixture was heated to 95°C for 30 min. Absorbance was
measured at 570 nm. One unit of enzyme catalyzed the formation of 1
µmol urea per minute.
Statistical significance of data was calculated using a one-tailed test for a t distribution (Student’s t-test).
Northern and Western analyses
Total RNA was extracted using a guanidinium isothiocyanate
procedure [45
] from macrophage monolayers plated at an
initial density of 2 x 106 peritoneal exudate
cells/well in a 24-well plate. RNA (total well contents) was
fractionated in 1% agarose (2.2 M formaldehyde) and transferred to
nylon membranes (Nytran®, Schleicher and Schuell, Inc., Keene,
NH). After UV cross-linking, HIV RNA was detected by
hybridization with a 32P-labeled Nef cDNA and exposure to
X-ray film. Quantitation was accomplished using NIH Image 1.61 software
analysis of digitalized images. Blots were stripped and re-probed with
a radio-labeled, human ß-actin probe (Clontech, Palo Alto, CA).
For immunoblot analyses, 5 x 106 plated cells were washed twice with PBS and lysed on ice for 30 min with 0.1% Triton X-100, 10 mM Tris•HCl (pH 7.4). Cell debris was removed by centrifugation in a benchtop microfuge at 4°C, and the clarified samples were measured for protein content using Bradford reagent (Bio-Rad, Hercules, CA). For the detection of iNOS, 1 µg protein was electrophoresed through 10% acrylamide, transferred to cellulose membranes, and subjected to immunoblotting using a monoclonal antibody for iNOS (Transduction Laboratories, Lexington, KY). Membranes were blocked with 5% nonfat dry milk (in PBS) and exposed for 1 h to anti-iNOS (1/1000 dilution). Washed membranes were subsequently treated with horseradish peroxidase-conjugated donkey anti-mouse immunoglobulin (Ig)G (1/2000). Immunoreactive peptides were visualized following reaction with a chemiluminescent substrate (Pierce, Rockford, IL) and exposure of blots to Hyperfilm (Amersham, Arlington Heights, IL). Pre-stained protein molecular-weight ladder was obtained from Life Technologies.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (26K): [in a new window] |
Figure 1. HIV gene transcription in transgenic macrophages. Peritoneal exudate
cells (resident or inflammatory) were plated at an initial density of
2 x 106 cells/well, and adherent cells were harvested
12 h after treatment as indicated. Total RNA was extracted,
fractionated by gel electrophoresis, probed with a nef cDNA
probe specific for HIV transcripts, and measured by densitometry.
Measurements of HIV transcripts were normalized to actin transcripts,
and levels in Vpr+ Nef- inflammatory macrophages were arbitrarily set
at 1. As different actin expression levels characterized resident and
inflammatory macrophages, relative RNA expression cannot be compared
between the two populations as plotted (see text). Shown is the
cumulative data of three experiments with macrophages pooled from at
least three transgenic mice per group. Error bars represent one
standard deviation. Three HIV transgenic lines defined by their
HIVd1443 proviral transgene are included: T26, Vpr+Nef+; X5, Vpr+Nef-;
and V13, Vpr-Nef+.
|
was determined
by assaying culture medium for nitrite, a NO metabolite, 24 h
after stimulation (Fig. 2
). With inflammatory macrophages specifically, NO production was
reduced in HIV transgenic (T26) mice when stimulated with LPS or with
LPS plus IFN-. Normalization of NO production in T26 macrophages to
that by control (nontransgenic FVB/N) macrophages in six independent
experiments confirmed the significance of the reduced capacity of T26
macrophages to generate NO (P<0.01). Subnormal NO
production was dependent on the nef and vpr genes
because mutations in either gene restored NO production to normal, if
not greater-than-normal, levels in the macrophages of proviral HIV
mouse lines X5 and V13. This defect was specific for inflammatory
macrophages because NO production was normal in HIV-positive, resident
macrophage populations. The reduced capacity of wild-type HIV
macrophages to produce NO in response to activation had in vivo
relevance. We observed that serum nitrate levels in untreated animals
were reduced significantly (about 50%; P<0.025) in
wild-type HIV mice relative to nontransgenic siblings and mutant HIV
transgenic lines (Fig. 3
). When NO production was induced in vivo with i.p. injections of
bacterial LPS, NO metabolite levels increased rapidly in all mice (by
8 h; Fig. 3
). However, lower levels of NO production were observed
again in treated, wild-type HIV animals (P<0.025) but not
in the mutant lines.
 View larger version (29K): [in a new window] |
Figure 2. NO production in HIV transgenic macrophages. Adherent peritoneal
exudate cells (inflammatory or resident) were treated with E.
coli LPS (1 µg/ml) and/or IFN- (10 U/ml), and culture fluids
were assayed in triplicate for the presence of nitrite as a measure of
NO production after 24 h. Results shown are the cumulative data of
six experiments with macrophages pooled from at least three mice per
group. Transgenic macrophages are compared with similarly treated cells
from nontransgenic, sibling mice (FVB/N).
|
 View larger version (34K): [in a new window] |
Figure 3. NO production in HIV transgenic mice. Serum was collected from groups
of at least six mice (equal numbers of males and females) representing
nontransgenic, sibling controls (FVB/N) and the three HIV transgenic
lines. Serum nitrates/nitrites were measured in triplicate 8 h
after the i.p. injection of E. coli LPS (200 µg/mouse).
Untreated animals received saline alone. Average measurements of the
six mice are shown.
|
[1
], and levels of iNOS protein induced by LPS/IFN-
stimulation were measured by Western analysis (Fig. 4
). In three experiments, using elicited peritoneal macrophages
pooled from three to five mice, iNOS expression was undetected in
untreated cultures. Upon stimulation, iNOS protein was visible in all
macrophage cultures, but the level of iNOS in wild-type (T26) murine
macrophages was comparable with nontransgenic cultures (Fig. 4)
. The
induction of iNOS in the mutant HIV lines was also near normal. Thus,
the reduced capacity of HIV macrophages to produce NO was apparently
not a result of the failure of these cells to respond to activation by
inducing iNOS expression.
 View larger version (40K): [in a new window] |
Figure 4. iNOS induction in macrophages from HIV transgenic macrophages.
Inflammatory peritoneal-exudate cells were harvested 16 h
following plating with or without stimulation with LPS and IFN- (1
µg/ml and 10 U/ml, respectively). Cell lysates were assayed for
protein content, and 1 µg protein was fractionated in 10% acrylamide
and subjected to Western analysis. The primary antibody was an iNOS mAb
that detected purified iNOS monomer at 130 kDa (Std). HIV macrophages
are compared with nontransgenic, sibling mice (FVB/N). The
autoradiogram represents one of three experiments using macrophages
pooled from at least three mice. Cumulative data from the three
experiments are presented in the box.
|
and IFN-. This confirmation suggested that wild-type HIV
macrophages responded differently to stimulation, not necessarily in
the production of iNOS but in the regulation of iNOS activity. One
manner of regulating iNOS activity is through substrate availability,
controlled in part by the activity of arginase, an enzyme that converts
arginine to ornithine and urea [46
, 47
].
Supplementing elicited macrophage cultures with arginine (1 mM) failed
to correct the deficit in NO production (unpublished results).
Alternatively, total arginase activity in lysates of macrophage
monolayers was measured. Constitutive levels of arginase activity were
comparable amongst control and HIV-positive cells. The low-level
induction of arginase activity that typically follows endotoxin
activation [48
] was observed in control FVB/N
macrophages (Fig. 5)
. However, arginase activity was uniformly reduced,
not induced, in activated HIV cultures. The reduction was statistically
significant in all induced cultures (P<0.01 for all with
the exception of P<0.025 for X5 cultures treated with
IFN-/LPS). Thus, the unique deficiency of wild-type HIV macrophages
to produce normal levels of NO could not be attributed to elevations in
arginase activity.
 View larger version (30K): [in a new window] |
Figure 5. NO production relative to arginase activity in plated macrophages. NO
production in adherent, peritoneal macrophages was measured as before
(see Fig. 2
legend) but normalized to cell viability as measured by MTT
assay (left panel). Cultures were also induced with a combination of
TNF- (25 µg/ml) and IFN- (10 U/ml). In addition, parallel
cultures (also plated at 2x105 exudate cells/well) were
harvested after 48 h incubation, and cell lysates were analyzed
for arginase activity (right panel). Presented is one of duplicate
experiments using macrophages pooled from five mice. Error bars denote
one standard deviation in triplicate measurements of one representative
experiment.
|
and LPS
activation. Inhibitors with different specificities for the
constitutive and inducible forms of NOS (cNOS and iNOS, respectively)
were applied. AG, having a specificity for iNOS [49
],
blocked NO production in response to inflammatory activation
(Fig. 6A
). In contrast, use of the cNOS inhibitor L-NAME
[50
] permitted a NO burst following macrophage
activation (Fig. 6A)
. The induction of HIV transcripts (by Northern
analysis) in adherent-elicited macrophages was measured through time
pursuant to stimulation (Fig. 6B)
. In cell populations from the three
HIV transgenic lines, levels of HIV transcription increased with time
following plating and activation. Administration of either NOS
inhibitor failed to block an early induction of viral transcription but
was effective at interfering with transcription after the 8-h time
point. The inhibition achieved with L-NAME (to 29±9% at
24 h) exceeded that achieved with AG (to 55±10% at 24 h)
with marginal significance (P<0.05). This confirmed that
the later, prolonged stage of HIV gene induction was dependent on NOS
function. However, NO production remained high in L-NAME
cultures (22.5–35 nmoles NO/million cells as opposed to 2.4–3.9
nmoles NO/million cells in the presence of AG), implying that
cNOS-derived NO, not antimicrobial NO bursts from iNOS activation, was
important in viral gene regulation.
 View larger version (28K): [in a new window] |
Figure 6. Inhibition of viral gene transcription with inhibitors of NOS. Two
metabolic inhibitors of NOS, L-NAME and AG, were applied at 1 mM to
differentially affect the production of NO in control (FVB/N),
inflammatory peritoneal macrophages activated with LPS (1 µg/ml) and
IFN- (10 U/ml) (A). AG blocked the production of exogenous NO as
detected in culture fluids measured 24 h after stimulation,
whereas L-NAME had little effect on exogenous NO production. (B) The
effect of NOS inhibitors on the induction of HIV gene transcription in
inflammatory peritoneal macrophages is shown. Plated cells
(2x106 exudate cells/well) were activated in the presence
of inhibitors, and total RNA was extracted at the designated times
post-stimulation. RNA yields from each well were subjected to Northern
analysis, and HIV transcription was measured digitally from scanned
autoradiograms and normalized to actin mRNA levels. One of two
experiments with similar results, in which macrophages were pooled from
five mice/group, is shown.
|
 View larger version (56K): [in a new window] |
Figure 7. Effect of exogenous NO on HIV transcription in plated macrophages. (A)
Inflammatory peritoneal macrophages were treated with the NO-releasing
compound SNAP, activated with LPS and IFN- (1 µg/ml and 10 U/ml,
respectively), or left untreated. RNA was extracted 14 h after
treatment, and total well yields were subjected to Northern analysis. A
representative autoradiogram used in quantitating HIV and ß-actin
expression is shown. (B) The cumulative data measuring HIV RNA
expression from two such experiments (each with duplicate samples) are
shown normalized to ß-actin RNA levels (left panel) or to cell
viability as determined by MTT assay (right panel). All values are
normalized to untreated X5 samples (arbitrarily set at 1).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In contrast to our observations with HIV transgenic mice, systemic NO production is often elevated in HIV-positive individuals [9 , 10 ]. Acute infection by HIV [12 , 13 ] and SIV [11 ] in vitro resulted in an induction of NO release in what might be considered a normal, antiviral response [51 ]. In particular, AIDS dementia has been associated with elevated NO production in astrocytes, induced by factors released from adjacent HIV-infected cells [18 , 19 , 26 ]. However, there are studies associating HIV infection with a reduced capacity to produce NO. In a cohort of infected hemophiliacs [52 ], higher-than-normal levels of serum nitrates characterized asymptomatic patients, but lower-than-normal levels characterized advanced patients with HIV-related disease. Considered in conjunction with studies demonstrating reduced NO in exhaled air of HIV-positive subjects [30 ] and reduced NO production associated with toxoplasmic encephalitis [28 ] and pulmonary tuberculosis [29 ] in HIV-positive patients, there is cause to consider the contributions of ineffective NO production in certain aspects of AIDS. These contrasting observations in human patients could reflect situational differences between acute HIV infection (productive infection), where NO production is increased in a normal, antiviral response, and chronic HIV infection (or nonproductive infection), where a viral-induced impairment in NO production may predominate. The latter scenario may be modeled in HIV transgenic mice. In a broader interpretation, demonstration of abnormal NO production in HIV transgenic macrophages reflects the capacity of HIV gene products to perturb macrophage function. For example, given the propensity of human alveolar macrophages to be infected [53 , 54 ], this perturbation may have significant ramifications about the development of pulmonary disease and HIV persistence in this compartment.
Mechanistically, it remains unclear how HIV down-modulates NO
production in transgenic, murine macrophages. The capacity of HIV
proteins Env (gp120) and Tat to induce iNOS expression in human
cell-culture systems [17
, 24
] had no
apparent effect on the murine system, perhaps because of inefficient
viral protein expression, inappropriate host cofactors, or its
counteraction by yet-undefined mechanisms. It should be noted that in
rodent macrophages, Tat did exert a negative effect on iNOS synthesis,
but this was in response to IFN- stimulation, and a negative effect
was not observed with LPS stimulation [55
]. We found no
apparent abnormality in iNOS levels in HIV macrophage cultures that
would implicate a defect in the induction of iNOS synthesis and explain
the down-regulation of NO production in wild-type HIV mice. Thus, it
seems likely that Nef- and Vpr-expressing cells regulate iNOS activity
differentially. This may be unusual because iNOS is principally
regulated at the level of transcription [1
]. However,
NOS monomers require four allosteric regulators for functional
dimerization: heme, tetrahydropterin, calmodulin, and arginine. One
alternative mechanism of iNOS regulation, the depletion of the
substrate arginine by the enzyme arginase [46
,
47
], was explored but dismissed on the basis that, if
anything, lower- and not higher-than-normal levels of arginase activity
were detected in wild-type HIV macrophage cultures. In fact, uniformly
lower levels of arginase activity were observed in HIV-positive
macrophage cultures despite their different abilities to generate NO.
This arginase response may reflect an abnormal cell condition dependent
on a viral gene other than nef or vpr. A
dissociation of the reciprocal relationship between arginase and iNOS
activities has been observed elsewhere [56
]. Another
potential regulatory mechanism, as yet unexplored, involves the
availability of tetrahydrobiopterin (BH4), which is
normally induced in response to TNF- and IFN-
[57
]. Experimentally, at least, a dissociation of
BH4 and iNOS biosynthesis has been demonstrated
[58
, 59
].
The involvement of Nef and Vpr in inducing this altered behavior in murine macrophages suggests that their biological functions overlap. This also may be the case in the development of HIV-associated nephropathy in HIVd1443-transgenic murine models [38 ]. Although their viral-related functions are far from defined, Nef interferes with activation-signal pathways involving protein tyrosine kinases, and Vpr alters cellular differentiation states [60 ]. Hence, the effect of these viral proteins may be to determine a cellular state in which iNOS is differentially regulated. It is reasonable to surmise that reduced NO production by chronic HIV-expressing cells, apart from reflecting an altered differentiation/activation state, may contribute to the survival of that cell under inflammatory conditions. The anti-HIV effects of NO may be a significant component of the host’s response to acute infection. A reduced capacity to produce NO in chronically infected cells may also sustain chronic infections by prohibiting NO-induced apoptosis [47 , 61 , 62 ], by limiting the reactivation of quiescent virus, and by promoting virus transmission to sensitive T cells.
Defective NO production by chronically infected cells could be a basis for a host innate-immune defect. Although NO has proven antimicrobial effects in mice, it is less clear in humans where much lower levels of NO are produced by monocytes/macrophages in response to bacterial activation [1 , 63 ]. We have preliminary evidence demonstrating that HIVd1443 wild-type, transgenic mice, but not their mutant counterparts, are more susceptible to infection with the bacterium Listeria monocytogenes (unpublished results). However, murine host defense against Listeria is not predicated on a NO response alone [64 ]; thus, additional macrophage defects may be pertinent to innate resistance in mice. Definition of the full extent of macrophage dysfunction in murine cells expressing HIV may be pertinent to human immunology, particularly if the observed NO effects reflect a broader perturbation in the expression of immune-modulating cytokines or chemokines.
NO, in turn, was shown to have a role in the induction of HIV gene
transcription. Our results suggested that cNOS-derived NO in particular
was required for the induction of HIV gene expression observed in our
system. Exogenous NO or iNOS-derived NO had a much lesser impact on
viral gene expression. In support of this conclusion, (i) viral gene
transcription increased at a time when NO output was high in activated
macrophages; (ii) the NO donor, SNAP, did not dramatically affect viral
gene expression; and (iii) the selective cNOS inhibitor
L-NAME effectively blocked the time-dependent increase in
HIV transcription without markedly affecting levels of NO production in
macrophage cultures. NO has dual effects on the activation of HIV
transcription in human cells. Two known studies implicate NO in the
activation of HIV in chronically infected human U1 macrophages
[31
, 37
], whereas NO inhibited HIV
transcription in transiently infected cell-culture systems
[35
, 36
]. It is conceivable that NO effects
are a dose-dependent phenomenon. Low levels of NO production may
activate the virus, perhaps through an effect on NF-B activation
[65
] or the capacity of NO to modulate cellular
differentiation [66
, 67
], processes that
favor HIV induction [68
]. At high concentrations of NO,
the effect may be negative through a direct inhibitory effect on
NF-B. Alternatively, we suspect that endogenous NO determined levels
of HIV transcription indirectly through controlling effects on the
differentiation/activation of the host cell. We reiterate that
exogenous NO may have had dual effects: negative effects on
transcription balanced by a positive effect gained through the
induction of cellular activation or differentiation. Importantly,
however, we can conclude that high levels of NO were not inhibitory in
this model of chronic HIV expression.
In conclusion, HIV transgenic macrophages have a reduced capacity to produce NO in response to inflammatory stimuli and tolerate high doses of exogenous NO to sustain HIV transcription. Apart from any advantage that a reduced NO response has on virus survival, this macrophage-related perturbation may identify a more-general, functional condition resulting in immunologic complications. By inference, chronically infected, human macrophage-like cells may adopt a functionally compromised state that promotes cell survival and virus transmission. Modulating NO production to activate quiescent virus in reservoir cells as suggested [31 ] may be a feasible means of exposing these cells to normal, immune mechanisms and drug therapy. A more effective means may be revealed with an improved understanding of viral mechanisms that promote HIV persistence. An alternative approach may be to target viral genes such as nef and/or vpr, which appear to determine the functional suppression of chronically infected cells. Thus, the HIV transgenic mouse system, in its modeling of the chronic infection of macrophages, may be useful for understanding functional deficiencies in this unique cell population and for developing novel anti-HIV approaches.
Received January 13, 2001; revised April 23, 2001; accepted April 24, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, and nitric oxide in glial cultures derived from fetal, neonatal, and adult human brain J. Exp. Med. 182,941-952B and HIV-1 long terminal repeat transcriptional activation by inducible nitric oxide synthase 2 activity J. Biol. Chem. 273,3895-3900B activation Am. J. Pathol. 155,275-284 RII-mediated nitric oxide and cyclic AMP pathways Mol. Med. 2,38-49[Medline]This article has been cited by other articles:
 |
|
 |
|
  E. R. Kline, D. J. Kleinhenz, B. Liang, S. Dikalov, D. M. Guidot, C. M. Hart, D. P. Jones, and R. L. Sutliff Vascular oxidative stress and nitric oxide depletion in HIV-1 transgenic rats are reversed by glutathione restoration Am J Physiol Heart Circ Physiol, June 1, 2008; 294(6): H2792 - H2804. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
