-stimulated nitric oxide production by retinoic acid in RAW 264.7 cells
* Department of Nutrition, The Pennsylvania State University, University Park, Pennsylvania, and
Institute for Nutrition Research, University of Oslo, 0316 Oslo, Norway
Correspondence: A. Catharine Ross, Ph.D., Department of Nutrition, Pennsylvania State University, 126.S. Henderson Building, University Park, PA 16802. E-mail: acr6{at}psu.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(IFN-) was significantly greater in cells cultured in
low-retinoid medium and treated with all-trans-RA
(10-10– 10-6 M, P <0.05),
as well as with 9-cis-RA and several retinoids selective
for the RA receptor subfamily of nuclear retinoid receptors. Similar
results were obtained with lipopolysaccharide and monophosphoryl lipid
A as stimuli. The RA-potentiated production of NO was positively
correlated with inducible NO synthase (iNOS) protein
(r =0.94, P <0.002), although the
expression of iNOS mRNA was not altered. We hypothesize that modulation
of the macrophage response to suboptimal immune stimuli by
physiological concentrations of RA, as observed in these studies, may be important in establishing an optimal balance between T helper
(Th) 1- and Th2-mediated immunity.
Key Words: lipopolysaccharide • monophosphoryl lipid A • immune regulation • T-helper-cell differentiation
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In the early 1900s, vitamin A deficiency was shown to be associated with reduced resistance to certain infections [8 , 9 ]. Vitamin A is now understood to be essential for normal immune system development, maintenance of mucosal barriers, and normal functioning of nonspecific and cell-mediated and humoral specific immunity. A reduction of T-cell precursors [10 ] and an abnormal shift toward the development of Th1-type immune responses at the expense of Th2-type immunity have been proposed as underlying causes of impaired immunity during vitamin A deficiency [11 , 12 ]. The increased severity of several infectious diseases observed in vitamin A-deficient animals and humans may be due, in part, to inappropriate development of Th1 versus Th2 immune responses.
Most physiological processes affected by vitamin A are mediated through the action of two families of nuclear retinoid receptors, retinoid acid (RA) receptor (RAR) and retinoid X receptor (RXR), which function as ligand-activated transcription factors for a large number of genes [reviewed in ref. 13 ]. All-trans RA binds with high affinity to members of the RAR subfamily, and 9-cis-RA binds to both RXR and RAR subfamily members [13 ]. Heterodimers composed of an RAR and an RXR bind to specific RA response elements (RAREs) in target genes; while the RXR may bind either as a homodimer to a different set of response elements, the retinoid X response elements (RXREs), or as a heterodimer with a number of other nuclear receptors, including those for vitamin D3, thyroid hormone, and peroxisome proliferators [14 ]. Numerous retinoids with selectivity for RAR or RXR have been synthesized because of their potential utility as drugs to promote cell differentiation and/or reduce proliferation [15 ].
The effect of immune and inflammatory stimulation on iNOS expression and NO production may differ significantly depending on the microenvironment and context of stimulation, including such factors as the type of cells, their state of maturation and differentiation, their prior exposure to activating or inhibitory stimuli, and the dose of NO-inducing stimuli to which they are exposed. As indicated by previous research, the ability of RA to modulate NO production may also differ within various contexts of infection or inflammation. In several studies conducted with cells that are already mature and activated, RA exhibited anti-inflammatory properties that appeared to be at least partly mediated by inhibition of iNOS gene expression. RA was shown to decrease proinflammatory-cytokine-, CD23-, or lipopolysaccharide (LPS)-induced NO production in vascular smooth muscle cells [16 , 17 ], peritoneal macrophages [18 ], keratinocytes [19 ], Kupffer cells [20 ], and mesangial cells [21 ]. However, in the context of RA-induced cell differentiation, RA may have the opposite effect, that is, enhancement of iNOS expression and NO synthesis. RA-induced enhancement of NO production has been observed in vivo [22 ] and in several cell types, including monocytic cells [23 ], breast cancer cells [24 ], epithelial cells [25 ], and neuroblastoma cells [26 ]. Additionally, RA treatment in vivo significantly enhanced production of iNOS mRNA, protein, plasma nitrate, and nitrate in LPS-injected but not saline-injected rats [22 ].
An understanding of how vitamin A deficiency and retinoid exposure
affect cellular responses to immune stimuli is of both basic and
practical importance. In the present experiments, RAW 264.7 cells were
initially cultured in medium containing a low level of retinol to
provide a model system of vitamin A deficiency. We then investigated
the effects of RA on NO production and iNOS expression during immune
activation by interferon (IFN)- or LPS. The results indicate that
exposure to RA sensitizes RAW 264.7 macrophages, enabling enhanced NO
production in response to levels of immune stimuli that otherwise are
suboptimal. We propose that RA-potentiated NO production may be
important in regulating the differentiation of Th1 and Th2 cells and
thereby in shaping the host’s immune response to infection.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
was purchased from PharMingen (San Diego,
CA), LPS (Pseudomonas aeruginosa Fisher Devlin immunotype 1)
was from List Biological Laboratories (Campbell, CA), and
monophosphoryl lipid A (MPL) was from Ribi (Hamilton, MT).
N-Monomethyl-L-arginine (L-NMA) and
its D-isomer (D-NMA) were from Sigma.
Cell culture and experimental protocol
The RAW 264.7 cells (ATTC TIB-71; American Type Culture
Collection, Manassas, VA) were grown in high-glucose (4.5-g/L)
Dulbecco’s modified Eagle’s medium (Gibco/BRL, Grand Island, NY)
supplemented with 4 mM L-glutamine, penicillin (50 U/mL),
streptomycin (0.05 mg/mL), and 10% heat-inactivated fetal bovine serum
(FBS) (Gibco-BRL) and at 37°C in an environment with 5%
CO2 and humidified air [27
]. Cultures of RAW
264.7 cells used for experiments were replaced from frozen stock
cultures no less than every 2 months. For the experiments, RAW 264.7
cells were seeded at a concentration of 2 x 106
cells/mL in Dulbecco’s modified Eagle’s medium with 1%
heat-inactivated FBS together with a retinoid or vehicle as a control
[0.01% ethanol (final concentration)], as indicated in the figure
legends. After RA pretreatment, the medium was changed to medium
containing the iNOS-inducing stimuli, as indicated in the figure
legends.
Determinations of nitrite in cell culture medium and cellular DNA
Levels of nitrite in cell culture medium were determined by a
colorimetric assay involving the Griess reaction, as previously
described [28
]. To determine cell DNA content as a
measure of cell number, a DNA-intercalating fluorochrome, Hoechst 33258
compound
2-[2-(4-hydroxyphenyl)-6-benzimidazolyl]-6-[1-methyl-4-piperazyl]-benzimidazole-3HCl
(Sigma), was used [29
]. Briefly, cells were grown and
treated in 96-well plates; and at the desired time point the medium was
aspirated, 100 µL of distilled deionized water was added to each
well, and plates were incubated at 37°C for 30 min. After one
freeze-thaw cycle, 100 µL of a solution containing Hoechst 33258 (at
a concentration of 10 µg/mL in 0.05 M phosphate buffer, pH 7.4), 2 M
NaCl, and 2 mM EDTA were added to each well. Herring sperm DNA was used
as a DNA standard. Fluorescence was measured immediately using an
excitation wavelength of 360 nm and an emission wavelength of 465 nm.
Western blot analysis
The RAW 264.7 cells were lysed in phosphate-buffered saline, pH
7.4, containing 1% sodium dodecyl sulfate and a Complete Mini Protease
Inhibitor Cocktail tablet (Roche, Indianapolis, IN), after which the
samples were passed through a 21-gauge needle and placed on ice for 45
min. The protein concentration was determined by a detergent-compatible
protein assay (Bio-Rad, Hercules, CA). Between 20 and 40 µg of
protein were dissolved in Laemmli sample buffer (Bio-Rad), boiled for 5
min, and then loaded on a sodium dodecyl sulfate-7.5% polyacrylamide
gel (Bio-Rad) for size fractionation. The separated proteins were
electroblotted onto a nitrocellulose membrane (Micron Separations,
Westboro, MA). For Western blot analysis, the primary monoclonal
antibody for iNOS (Transduction Laboratories, Lexington, KY) and the
secondary peroxidase-linked anti-mouse immunoglobulin (Ig)G (Amersham
Pharmacia Biotech, Piscataway, NJ) were diluted 1:2,000 and 1:5,000,
respectively, in Tris-buffered saline, pH 7.4, with 0.1% Tween 20
(TBS-T) containing 5% dry milk, which was used as the blocking agent.
The membrane was washed with TBS-T after blocking and after incubations
with the primary antibody and the secondary antibody. Then the iNOS
protein was detected by chemiluminescence using an enhanced
chemiluminescence detection solution and Hyperfilm enhanced
chemiluminescence (ECL; Amersham Pharmacia Biotech) film according to
the manufacturer’s directions. Densitometry was performed using NIH
Image 1.56 software (National Institutes of Health, Bethesda, MD).
Isolation of total and poly(A)+ RNA and Northern blot
analysis
The isolation of total RNA and the Northern blot analysis were
performed essentially as previously described [30
].
Briefly, total RNA was isolated using Trizol (Gibco-BRL), quantified
spectrophotometrically, and examined for integrity by ethidium bromide
staining [30
]. Three micrograms of poly(A)+
RNA that had been isolated by affinity chromatography using oligo(dT)
cellulose were size fractionated on a 1.2% agarose–0.66 M
formaldehyde gel, transferred to a nylon membrane (Schleicher &
Schuell, Keene, NH), and cross-linked with UV light
[30
]. The cDNAs for the RAR
, RARß, RAR, RXR,
RXRß, and RXR [31
] used as probes were kindly
provided by Pierre Chambon (Strasbourg, France). The probes were
labeled with [--32P]dCTP (3,000 Ci/mmol and 10
µCi/µL; Amersham) using the Prime-a-Gene Labeling System (Promega,
Madison, WI). Prehybridization, hybridization, and autoradiography were
performed as previously described [30
].
For reverse transcription (RT)-PCR analysis of RARß and glyceraldehyde-3-phosphate dehydrogenase (GAPDH), used as a control, the Perkin-Elmer Gene Amp PCR system 24000 (PE Biosystems, Foster City, CA) was used. Reverse transcription was performed using 1 µg of RNA in a 25-µL reaction. The cDNA was diluted 1:4 before 5 µL of this solution was used for one PCR in a total volume of 25 µL. The final concentration of each primer was 0.3 µM for analysis of RARß and 0.2 µM for analysis of GAPDH. For detection of RARß, 24 PCR cycles were run, and for GAPDH, 20 cycles were run. Ten microliters of PCR product were separated on a 2% agarose gel in Tris-acetic acid-ethylenediaminetetraacetate buffer, pH 8.0, containing 250 µg of ethidium bromide per mL and photographed. The primers for GAPDH (Clontech Laboratories, Palo Alto, CA) yielded a 982-bp PCR product. The primers for RARß were designed using Primer3 software (http://www.genome.wi.mit.edu/genome_software/other/primer3.html). The forward primer for RARß was 5'-TCC-ACA-CCT-AGA-GGA-TAA-GC-3', the reverse primer was 5'-TGG-TAC-TCT-GTG-TCT-CGA-TG-3', and the PCR product was 227 bp.
Quantitative real-time PCR analysis was used to determine iNOS mRNA expression. Duplicate or triplicate aliquots of RNA were analyzed on 96-well plates using the Perkin-Elmer/Applied Biosystems Division (PE/ABD) 7700 Sequence Detector. The iNOS forward primer was 5'-CAG-CTG-GGC-TGT-ACA-AAC-CTT-3', the probe was 5'-CGG-GCA-GCC-TGT-GAG-ACC-TTT-GA-3', and the reverse primer was 5'-TGA-ATG-TGA-TGT-TTG-CTT-CGG-3'. The 5' label was 6-carboxyfluorescein (FAM) and the 3' -label was 6-carboxytetramethylrhodamine. The primers and probe for iNOS were designed using the PE/ABD Primer Express software based on the mouse cDNA sequence reported by Kone et al. [32 ]. Each value for iNOS mRNA was normalized relative to the 18S RNA analyzed in the same well.
Statistical analysis
One-factor analysis of variance (ANOVA) was performed using the
software SuperANOVA (Abacus Concepts, Berkeley, CA). Differences were
considered statistically significant according to Fisher’s protected
least-significant-difference test (P <0.05).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-, LPS-, or MPL-stimulated NO production that, alone, induced only submaximal NO
production (Fig. 1A
). In concentrations of IFN- that induced maximal NO
production, RA had no additional effect. When the results from
experiments using suboptimal doses of IFN- were calculated as the
nanomoles of nitrite in the cell culture medium per microgram of cell
DNA, iNOS specific activity was significantly greater in RA-treated
cells than in vehicle-treated cells (data not shown). RA also
potentiated the response of RAW 264.7 cells to stimulation with LPS and
MPL (Fig. 1B
and 1C
, respectively).
 View larger version (23K): [in a new window] |
Figure 1. Potentiation of NO production by RA in RAW 264.7 cells stimulated with
IFN-, LPS, or MPL. Cells were pretreated for 24 h with
10-7 M all-trans-RA or vehicle (0.1% ethanol,
EtOH). Then the medium was changed to medium containing IFN- (A),
LPS (B), or MPL (C) at the indicated concentrations, together with
10-7 M all-trans-RA or vehicle. After 24 h, the nitrite assay for NO production was performed immediately on
cell culture medium. The data shown are the means ±
SD, n = 3–4. *, P < 0.05
versus vehicle control.
|
0.7 µM
(A. C. Ross, unpublished data), which would equal 7 nM in culture
medium containing 1% FBS. Therefore, the cell cultures not receiving
RA were deficient in retinoids, similar to a state of vitamin A
deficiency.
The potencies of all-trans-RA and 9-cis-RA were
compared in dose-response experiments. IFN--stimulated NO production
was enhanced when all-trans-RA was added in the range from
10-9 to 10-6 M (P<0.05)
(Fig. 2A
). Potentiation of NO production was greatest in the physiological
range of 10-7 to 10-8 M, whereas at
10-5 M, all-trans-RA abolished NO production
(data not shown). Enhanced NO production was also observed in cells
treated with 9-cis-RA, but 9-cis-RA was less
potent than all-trans-RA (P <0.05) (Fig. 2A)
. RA alone did not elicit NO production (data not shown).
 View larger version (22K): [in a new window] |
Figure 2. Dose-response comparison of all-trans-RA and
9-cis-RA and the effect of RA pretreatment and kinetics of
NO production in RAW 264.7 cells. Cells were pretreated for 24 h
(A and C) or 6–24 h (B) with all-trans- or
9-cis-RA at concentrations from 10-12 to
10-6 M (A), 10-7 M all-trans-RA (B
and C), or vehicle (0.1% ethanol, EtOH). Then the medium was changed
to medium containing IFN- (A, 5 U/mL; B and C, 3 U/mL) and
all-trans- or 9-cis-RA or vehicle at the same
concentrations as for the pretreatment. Immediately after 24 h (A,
B) or at 0, 6, 12, 18, and 24 h (C), the nitrite assay was
performed on cell culture medium. The data shown are the means ±
SD, n = 3–4. *, P < 0.05
versus vehicle control.
|
(Fig. 2B)
, LPS, or MPL (data not shown). Pretreatment with RA
for 12, 18, and 24 h potentiated NO production by 2.0-, 3.2-, and
3.7-fold, respectively (P <0.02).
The kinetics of NO production (nitrite accumulation) in RAW 264.7 cell
culture medium was examined in cells pretreated with RA or vehicle only
and stimulated with a suboptimal dose of IFN- (3 U/mL). Nitrite was
readily detectable in the cell culture medium 12–18 h after
stimulation with IFN- (Fig. 2C)
. Nitrite was not detected any
earlier in RA-treated cells than in vehicle-treated cells, but once
detected the rate of nitrate accumulation was higher in RA-treated
cells.
To ensure that the source of nitrite in the cell culture medium is
L-arginine, L-NMA was added as an inhibitor of
the NOS enzymes [6
] during maximal cell stimulation with
IFN- (20 U/mL) and LPS (40 µg/mL). With equimolar concentrations
(0.4 mM) of L-arginine and L-NMA, the
accumulation of nitrite in the medium was about half of that observed
without L-NMA (Fig. 3
). At an 8:1 molar ratio of L-NMA to
L-arginine, nitrite formation was reduced by 90%. The
D-NMA isomer, added as a negative control, did not affect
either IFN-- or LPS-stimulated NO production.
 View larger version (20K): [in a new window] |
Figure 3. Inhibition of NO production by L-NMA. RAW 264.7 cells were
stimulated with IFN- (20 U/mL) and LPS (40 µg/mL) alone or in the
presence of L-NMA or D-NMA at the indicated
concentrations. Immediately after 24 h, the nitrite assay was
performed on cell culture medium. The data shown are the means ±
SD, n = 4. *, P <
0.001.
|
-induced growth
inhibition and RA separately and
together. An analysis of the results of seven independent experiments
showed that IFN- treatment alone caused a significant
(P <0.05) growth inhibition that was observed after
48 h of culture and that IFN- with or without RA inhibited
growth after 72 h (Fig. 4
). RA treatment alone did not inhibit cell growth.
 View larger version (36K): [in a new window] |
Figure 4. Cell growth is inhibited by IFN- but not by RA. RAW 264.7 cells were
pretreated for 24 h with RA or vehicle (0.1% ethanol, EtOH),
after which the medium was changed to medium containing IFN- and
all-trans-RA or vehicle, all-trans-RA alone, or
vehicle alone. After 24 h, the medium was changed again to medium
containing IFN- and all-trans-RA or vehicle,
all-trans-RA alone, or vehicle alone, and cells were
incubated until 48 h. Cells were harvested at this time or after
another 24-h IFN- stimulation and incubation (noted as 72 h).
Cell DNA content was assayed. The data shown are the means ±
SD of data from 7 independent experiments. * denotes
P < 0.05 compared to ethanol-treated cells at each
time point.
|
-stimulated NO production, and induction of RARß by RA, RARß, RAR, RXR, and RXRß; but RXR
was undetectable. To determine whether the potentiation of NO
production by RA is most likely mediated through the RAR or RXR
pathways, several receptor-selective retinoids were compared for their
ability to enhance IFN--stimulated NO production as well as for
their ability to induce expression of RARß, whose gene contains an
RARE (ßRARE) [34
]. Table 1
shows the reported receptor selectivity for the retinoids tested
[34a–34f], the enhancement by each retinoid of IFN--stimulated NO
production, and their abilities to induce RARß mRNA. As determined by
RT-PCR analysis, all of the RAR-selective retinoids [Ro40-6055
(Am580), Ro19-0645, and all-trans-UAB8] induced RARß
expression, as did all-trans-RA and 9-cis-RA;
these retinoids also stimulated significant enhancement of NO
production. The RXR-selective retinoids (Ro25-6603, Ro25-7386, and
9-cis-UAB30) either had no significant effect compared with
control cells or were less effective than all-trans-RA or
Ro40-6055, an RAR--selective retinoid (Table 1)
. In combination with
an RAR-selective retinoid, Ro25-6603 and Ro25-7386 neither enhanced nor
inhibited activity. With the exception of
9-cis-UAB30—reported to be RXR selective
[35
]—which did increase expression of RARß mRNA but
did not significantly increase NO production, there was a consistent
relationship between the ability of a retinoid to induce RARß and its
ability to potentiate NO production stimulated by IFN-.
 View larger version (70K): [in a new window] |
Figure 5. Retinoid receptor expression in RAW 264.7 cells. Three micrograms of
poly(A)+ RNA was used for Northern blotting for RAR,
RARß, and RAR and for RXR, RXRß, and RXR as described in
Materials and Methods.
|
|
View this table: [in a new window] |
Table 1. Ability of Various Receptor-Selective Retinoids to Potentiate
IFN--Induced Nitric Oxide Production and Induce Expression of
Retinoic Acid Receptor RARß
|
(Fig. 6
). Maximal levels were reached after 6 h, and expression
returned to background levels after 18–24 h (data not shown). However,
in three independent experiments, there was no difference in the iNOS
mRNA level attributable to treatment with RA, but stimulation with
IFN- with and without RA resulted in equal increases in iNOS mRNA.
Unstimulated RAW 264.7 cells expressed a low but detectable level of
iNOS mRNA, and cells that had been maintained in culture for longer
times had a higher basal level of iNOS mRNA. Unstimulated RAW 264.7
cells also expressed a low basal level of tumor necrosis factor
mRNA, indicative of the partially activated state of this cell line
(data not shown).
 View larger version (21K): [in a new window] |
Figure 6. Induction of iNOS mRNA after exposure of RAW 264.7 cells to IFN-
with and without treatment with all-trans-RA. Cells were
pretreated for 24 h with 10-8 M
all-trans-RA or vehicle (0.1% ethanol, EtOH). Then the
medium was changed to medium containing IFN- (3 U/mL) and
all-trans-RA or vehicle, all-trans-RA alone, or
vehicle alone. After 0.5, 2, 4, and 6 h, RNA was extracted and
quantified in triplicate by real-time PCR analysis for iNOS mRNA. A
representative experiment of three independent experiments is shown.
|
- and LPS-treated RAW 264.7 cells (Fig. 7 A
). The concentration of nitrite in the cell culture medium and the
expression of iNOS protein in cell lysates were strongly correlated
(r =0.94, P <0.002). The potentiation by
RA was greatest (nearly 10-fold) in the first 6 h after IFN-
stimulation, during which time iNOS protein increased most rapidly
(Fig. 7B) . iNOS protein was at or below the limit of detection in cells
stimulated with IFN- only (therefore not shown in Fig. 7B insert).
Similarly, iNOS protein was also enhanced by RA in cells stimulated
with LPS in both the presence and absence of RA (data not shown).
 View larger version (29K): [in a new window] |
Figure 7. Expression of iNOS protein after exposure of RAW 264.7 cells to IFN-
or LPS with and without treatment with all-trans-RA. (A)
Cells were pretreated for 24 h with 10-8 M
all-trans-RA or vehicle. Then the medium was changed to
medium containing IFN- (3 or 6 U/mL), LPS (1 ng/mL), or LPS (100
ng/mL) plus IFN- (50 U/mL) for maximal stimulation,
10-7 M all-trans-RA or vehicle (0.1% ethanol,
EtOH), all-trans-RA alone, or vehicle alone. After 24 h, cell culture medium samples were taken for nitrite analysis (black
bars), and the cells were lysed for Western blot analysis for iNOS
protein and quantified by densitometry (open bars). The data are
expressed as the fold difference relative to results for IFN- at 3
U/mL. The Western blot analysis for iNOS protein illustrates the
results for the conditions shown directly above in the bar graph. (B)
Cells were pretreated with 10-8 M all-trans-RA
or vehicle for 24 h. Then the medium was changed to medium
containing IFN- (3 U/mL) and 10-8 M
all-trans-RA or vehicle, all-trans-RA alone, or
vehicle alone. After 3, 6, 18, and 24 h, cells were lysed for
Western blot analysis. Densitometric analysis of the blot was performed
followed by normalization to the Coomassie blue-stained gel. Since iNOS
protein was nearly undetectable without IFN- treatment, the
y-axis represents the fold difference in iNOS protein
relative to the level of iNOS protein after treatment with IFN- at 3
U/mL for 3 h.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
or LPS resulted in enhanced NO production relative to NO
production by cells cultured in a low-retinoid medium. It is important
to note that RA pretreatment did not alter maximum NO production but
rendered RAW 264.7 cells more sensitive to submaximal concentrations of
these iNOS-inducing agents (Fig. 1) . The increase in NO production by
RA was not related to a change in the number of cells, and iNOS
specific activity was higher in RA-treated cells. RAW 264.7 cells
expressed all of the RAR and RXR receptors except the RXR isoform, a
finding similar to a previous study in isolated Kupffer cells
[36
], in which all the RAR and RXR isoforms were present
but RXR was expressed at significantly lower levels. The results of
our studies using several synthetic retinoids with selectivity for the
RAR or RXR receptors are consistent with a model in which the
RA-induced enhancement of NO production is mediated through an RAR
pathway, because all of the RAR-selective retinoids tested enhanced NO
production and because naturally occurring all-trans-RA was
more effective than 9-cis-RA. Of the retinoid analogues
tested, only the RXR-selective retinoids Ro25-7386 and
9-cis-UAB30 did not enhance IFN--stimulated NO
production. The ability of all-trans-RA to enhance NO
production in RAW 264.7 cells was strongly correlated with an increase
in iNOS protein (Fig. 7A)
. This positive correlation suggests that RA
pretreatment did not affect the catalytic activity of iNOS and that a
higher rate of new iNOS protein synthesis was indeed responsible for
the higher level of NO production in RA-treated cells. The level of
iNOS protein expression increased rapidly during the first 6 h of
IFN- stimulation in RA-pretreated cells and remained high for the
next 18 h.
The results of various studies have shown that the ultimate expression
of iNOS activity and NO production is regulated by several mechanisms,
including transcriptional and post-transcriptional control. The
transcriptional response to inflammatory and immunomodulatory factors
such as IFN-, LPS, IL-6, and tumor necrosis factor is mediated
through the binding of specific factors to DNA response elements in the
region up to 1.6 kb 5' of the transcription start site of the iNOS
gene [37
, 38
]. Although binding to these
and several other response elements generally confers positive
regulation of iNOS gene expression [38
], negative
regulation also has been reported [39
], an example being
transcription factor binding to an activator protein 1 site in the iNOS
promoter [37
]. Regulation may also occur via changes in
iNOS mRNA stability, rate of mRNA translation, and protein degradation
[40
41
42
43
44
]. We had anticipated that the effect of RA on NO
production would involve increases in both iNOS mRNA and iNOS protein.
Such changes were observed in rats in vivo after treatment with RA and
LPS [22
]. However, in three independent time-course
experiments, iNOS mRNA was not changed as a function of RA treatment,
although its production was significantly induced by IFN- (Fig. 6)
.
Nevertheless, iNOS protein expression was increased within 6 h of
IFN- stimulation to a much greater extent in RA-treated than in
vehicle-treated cells (Fig. 7)
. These results suggest that RA’s effect
on iNOS expression may involve post-transcriptional mechanisms,
possibly enhanced translational efficiency and stability, owing to the
fact that iNOS protein increased rapidly, within 6 h, in
RA-treated IFN--stimulated cells and, once induced, remained
elevated for several hours. Although this significant increase in iNOS
protein without a prior or concomitant change in iNOS mRNA was
unanticipated, it was not surprising, given the known complexity of
iNOS regulation.
Some previous studies have demonstrated the potential of RA to reduce
iNOS gene expression or activity, implying an anti-inflammatory effect
of RA treatment, while others have demonstrated RA’s potential to
increase NO production [22–26 and this study], suggesting a
proinflammatory action. Although these results seem opposed, we suggest
that several potentially important factors may account for the
differences that have been observed among studies with respect to the
ability of RA to modulate iNOS expression and NO production. The
potentiation of NO production by RA in our RAW 264.7 cells cultured
under low serum retinoid conditions suggests that RA at physiological
doses is necessary for an efficient response to low levels of immune or
proinflammatory stimuli. In several other studies
[16
17
18
, 21
], the culture medium contained
5–10% FBS, which might have supplied sufficient vitamin A (retinol)
to maintain cells in a state of vitamin A adequacy. In our studies with
RAW 264.7 cells cultured in 1% FBS, the potentiation of NO production
by RA occurred within a physiological range of RA concentrations, with
greatest potentiation between 1 and 100 nM all-trans-RA. At
10 µM all-trans-RA, NO production was abolished. In
previous cell studies [16
17
18
19
20
21
], the higher the
concentration of RA added, the greater was the inhibition of NO
synthesis, with the greatest inhibition in the high nanomolar to
micromolar range. These RA concentrations are 10- to 50-fold above
those present in normal plasma, and their use in cell studies is likely
to reveal RA’s pharmacological potential. We also titrated a broad
dose range of the iNOS inducing agents. This experimental design
revealed that the potentiation of NO production by RA occurs in
response to submaximal doses of iNOS-inducing stimuli. In contrast,
several studies employed high doses of the iNOS-inducing agent
[16
17
18
19
20
21
] without dose titration [16
17
18
19
20
21
].
Hence, it is not known whether the NO response to submaximal
iNOS-inducing stimuli would be enhanced by RA in other cell types,
similar to our results in RAW 264.7 cells. The states of cell
differentiation, maturation, and prior activation have also varied
among studies. In some studies that demonstrated RA’s potential to
increase NO production [23
24
25
26
], this was seen as a part
of a program of RA-induced cell differentiation occurring over a course
of several days, during which time there was a concomitant increase in
both iNOS mRNA and protein levels [23
, 25
].
Taking all of these studies together, it is apparent that RA’s effects
on NO production are mediated in a context-dependent manner by several
complex mechanisms that may include changes in iNOS gene expression
[17
, 19
, 21
, 23
,
25
] or altered iNOS protein levels without altered mRNA
levels, as in this report. The variety of responses to RA among cells
is reflective of the overall complexity of iNOS regulation and NO
production. In vivo, differences in the number of iNOS-producing cells
would also be expected to modulate outcome. It is still uncertain how
NO responses in rodent models and humans compare quantitatively with
one another. NO production in rodent macrophages may exceed production
in human cells, but evidence for high-level NO production by human
monocytes/macrophages has been presented [45
];
MacMicking et al. [2
] concluded, based on a review of
studies with patients with infectious or inflammatory diseases, that
human monocyte and macrophage iNOS (NOS2) is functionally expressed.
The results of these studies may have implications for understanding
how vitamin A deficiency versus RA sufficiency shapes the immune
response to infectious diseases. Vitamin A deficiency is known to be
associated with poor cell-mediated and humoral immune responses and
with increased severity of infectious diseases (see ref 9
and 46
for reviews). Vitamin A deficiency is thought to
cause an imbalance in the development of Th1 and Th2 responses, with a
relative expansion of Th1 cell production and Th1-type cytokine
production and with a down-regulation of Th2 cell functions
[11
, 12
, 47
]. NO is reported
to restrict Th1 cell proliferation and inhibit Th1 cell cytokine
production [4
]. Based on the results in the present
study showing that physiological concentrations of RA potentiate the
macrophage NO response to suboptimal concentrations of immune stimuli,
it may be postulated (Fig. 8
) that during the early stage of an infection, when amounts of the
infectious or inflammatory agent are still low (submaximal), a
deficiency of vitamin A (RA) would compromise the macrophage’s ability
to produce NO in response to immune stimuli such as IFN- and LPS. A
low macrophage NO response, favorable to Th1 cell development, could be
an initiating factor in the imbalance between the Th1 and Th2 cell
functions observed in vitamin A deficiency in vivo [11
,
48
]. It is interesting that a requirement for NO for
optimal DNA synthesis by human peripheral-blood lymphocytes has been
reported [49
]. It also has been reported that
antigen-activated Th1 cells, but not Th2 cells, are capable of a high
output of NO on their own [50
]. Therefore, if it is the
case that Th2 cells are dependent on macrophages (or other cells) for
NO production to initiate their differentiation, whereas Th1 cells are
able to produce NO autonomously, the differential in macrophage-derived
NO production due to RA may be especially important in initiating Th2
cell responses. An abnormally strong Th1 response due to vitamin A
deficiency might predict that, later in the immune response, NO
production would increase. However, iNOS expression and NO production
in response to proinflammatory immune stimuli in vitamin A-deficient
animals have not to our knowledge been reported. Further studies are
needed to determine whether the response threshold to proinflammatory
stimuli is elevated in vitamin A deficiency in vivo and whether
decreased NO synthesis contributes to a microenvironment that is
favorable to Th1 cell expansion and its sequelae.
 View larger version (29K): [in a new window] |
Figure 8. Model of the potential effect of vitamin A deficiency on nitric oxide
(NO) production and its sequelae for the development of T helper cell
(Th1/Th2) immunity. Provision of RA prior to macrophage stimulation is
proposed to correct the initial low NO response, helping in turn to
normalize the balance between the development of the Th1 response and
that of the Th2 response.
|
Received September 19, 2000; revised January 29, 2001; accepted January 31, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Biochem. Biophys. Res. Commun. 270,846-851[Medline] Nature 355,359-361[Medline] and lipopolysaccharide Proc. Natl. Acad. Sci. USA 90,9730-9734 and bacterial lipopolysaccharide J. Exp. Med. 177,1779-1784-stimulated RAW 264.7 cells by dexamethasone. Evidence for glucocorticoid-induced degradation of iNOS protein by calpain as a key step in post-transcriptional regulation J. Biol. Chem. 272,16679-16687This article has been cited by other articles:
 |
|
 |
|
  S. K. Rayala, E. Martin, I. G. Sharina, P. R. Molli, X. Wang, R. Jacobson, F. Murad, and R. Kumar Dynamic interplay between nitration and phosphorylation of tubulin cofactor B in the control of microtubule dynamics PNAS, December 4, 2007; 104(49): 19470 - 19475. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. M. Luo and A. C. Ross Physiological and Receptor-selective Retinoids Modulate Interferon {gamma} Signaling by Increasing the Expression, Nuclear Localization, and Functional Activity of Interferon Regulatory Factor-1 J. Biol. Chem., October 28, 2005; 280(43): 36228 - 36236. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Seguin-Devaux, Y. Devaux, V. Latger-Cannard, S. Grosjean, C. Rochette-Egly, F. Zannad, C. Meistelman, P.-M. Mertes, and D. Longrois Enhancement of the inducible NO synthase activation by retinoic acid is mimicked by RARalpha agonist in vivo Am J Physiol Endocrinol Metab, September 1, 2002; 283(3): E525 - E535. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
