Published online before print June 3, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Critical Care Medicine Department, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, Maryland
3Correspondence: Critical Care Medicine Department, Bldg. 10, Rm. 7D43, Warren Grant Magnuson Clinical Center, National Institutes of Health, Bethesda, MD 20892. E-mail: rdanner{at}nih.gov
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) by distinct mechanisms. Dibutyryl-cyclic guanosine 5'-monophosphate (cGMP) failed to simulate NO·-induced increases in TNF- or IL-8 production. In contrast, dibutyryl-cyclic adenosine monophosphate blocked NO·-induced production of TNF- (P=0.009) but not IL-8. NO· increased IL-8 (5.7-fold at 4 h; P=0.04) and TNF- mRNA levels (2.2-fold at 4 h; P=0.037). However, nuclear run-on assays demonstrated that IL-8 transcription was slightly decreased by NO· (P=0.08), and TNF- was increased (P=0.012). Likewise, NO· had no effect on IL-8 promoter activity (P=0.84) as measured by reporter gene assay. In THP-1 cells and human primary monocytes treated with actinomycin D, NO· had no effect on TNF- mRNA stability (P>0.3 for both cell types) but significantly stabilized IL-8 mRNA (P=0.001 for both cell types). Because of its role in mRNA stabilization, the p38 mitogen-activated protein kinase (MAPK) pathway was examined and found to be activated by NO· in LPS-treated THP-1 cells and human monocytes. Further, SB202190, a p38 MAPK inhibitor, was shown to block NO·-induced stabilization of IL-8 mRNA (P<0.02 for both cell types). Thus, NO· regulates IL-8 but not TNF- post-transcriptionally. IL-8 mRNA stabilization by NO· is independent of cGMP and at least partially dependent on p38 MAPK activation.
Key Words: cytokines • protein kinases • gene regulation • mRNA • cAMP • cGMP
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
During septic shock, eNOS is down-regulated in the endothelium, and an inducible NO· synthase (iNOS) isoform is expressed in vascular smooth muscle, greatly increasing NO· output [17
18
19
20
21
]. For a local nidus of infection, this response may help ensure the adequate delivery of neutrophils to the affected site. Vascular dilation increases leukocyte flow to the involved tissue. As NO· acts over short distances, moving NO· production deeper into the vessel wall might permit enhanced production without undue interference with leukocyte attachment to the endothelium [22
], an essential step in migration [23
]. Furthermore, exposure of leukocytes to high concentrations of NO·, as encountered while traversing inflamed vessels, has been shown to boost the production of cytokines such as interleukin-8 (IL-8) [22
, 24
] and tumor necrosis factor (TNF-) [6
, 7
, 25
26
27
], thus reinforcing a strong, chemotactic gradient. Although potentially beneficial in a localized infection, these responses may be harmful in overwhelming, systemic infections and contribute to refractory hypotension, tissue damage, and death [3
, 4
, 13
, 21
]. It is important that noninfectious conditions such as reperfusion injury [28
, 29
] and atherosclerosis [30
] have also been linked to similar maladaptive responses that result in leukocyte-mediated injury. This has led to investigations to better understand the underlying mechanisms by which NO· regulates inflammation and gene expression.
IL-8 is a potent C-X-C chemokine that plays a pivotal role in innate immunity [31
, 32
]. As the primary, endogenous chemoattractant of neutrophils, IL-8 regulates essential host defenses in the early stages of microbial invasion. However, IL-8 is also thought to contribute to neutrophil-mediated tissue injury and organ failure in conditions such as the acute respiratory distress syndrome, reperfusion injury, and septic shock [32
33
34
]. Similar to the effects of NO· on neutrophil adhesion [14
] and TNF- [6
, 7
, 35
, 36
], NO· up-regulation of IL-8 appears to be cGMP-independent [22
]. The TNF- promoter is activated by NO· through reduced binding of Sp protein to a cyclic adenosine monophosphate (cAMP)-sensitive, GC-box repressor element [36
37
38
]. Although NO· has been shown to increase IL-8 protein and mRNA levels in lipopolysaccharide (LPS)-stimulated human neutrophils [22
] and to transcriptionally up-regulate IL-8 in other cell types [39
40
41
], NO· regulation of IL-8 is incompletely understood. The IL-8 promoter lacks known GC box and cAMP-response elements [22
, 42
43
44
], suggesting a regulatory pathway divergent from that of TNF-. Notably, IL-8 and TNF- mRNA harbor AU-rich elements (ARE; AUUUA) in 3'-untranslated regions (3'-UTR), which subject them to post-transcriptional regulation [45
46
47
48
49
50
]. Further, a recent report in human pancreatic adenocarcinoma cells has found that NO· regulation of IL-8 may be, in part, post-transcriptional [41
]. It is important that p38 mitogen-activated protein kinase (MAPK) activation has been associated with the ability of LPS to regulate IL-8 mRNA stability [47
48
49
]. We [51
] and others [24
] have found that NO· activates p38 MAPK in U937 cells, and p38 MAPK inhibition has been shown to decrease NO·-induced IL-8 production in these cells [24
].
In the current investigation, we sought to comprehensively examine NO· regulation of IL-8 in LPS-stimulated THP-1 cells. These cells were chosen as a reproducible, LPS-responsive, human cellular model that has been evaluated extensively in studies of post-transcription regulation [49
]. Further, we wanted to approach this question in a system that was different from our previous studies in phorbol ester-differentiated U937 cells [35
36
37
38
]. Several key experiments were also performed in human primary monoctyes to investigate the biological relevance of results obtained with THP-1 cells. Specific goals were to directly compare the cyclic nucleotide dependence of IL-8 and TNF- gene regulation; test effects of NO· on IL-8 and TNF- mRNA kinetics, transcription, and stability; investigate the IL-8 promoter for NO·-induced changes in activity and transcription factor binding; and finally, examine the role, if any, of p38 MAPK in the regulation of IL-8 by NO·.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
THP-1 cells, a human monocytic line, were obtained from the American Type Culture Collection (Manassas, VA). For all experiments, cells were cultured at 37°C in a 5% CO2 atmosphere using RPMI-1640 complete medium (BioWhittaker, Walkersville, MD) containing 10% fetal calf serum (FCS; Cellgro, Herndon, VA), 100 U/ml penicillin, 100 µg/ml streptomycin, 25 µg/ml amphotericin B (all from Cellgro), and 50 µM ß-mercaptoethanol (Sigma-Aldrich).
Human primary monocytes obtained from healthy donors by standard leukapheresis were purified by counterflow centrifugation (elutriation), which yields cells that are 93–95% pure and >98% viable (Department of Transfusion Medicine, Section of Blood Services and Cell Processing, Clinical Center, National Institutes of Health, Bethesda, MD). Monocytes (1x107) were cultured at 37°C with 5% CO2 in 10 ml RPMI-1640 complete medium in which 10% autoplasma was substituted for 10% FCS. For Bt2cGMP, Bt2cAMP, LPS, SNOG, GSH, and SB202190, the maximum concentrations examined in THP-1 cells or human monocytes did not alter cell viability as determined by trypan blue exclusion.
Cytokine assay
THP-1 cells (5x105/ml/well) were placed in 24-well plates containing RPMI-1640 complete medium with Bt2cGMP (0–50 µM) or Bt2cAMP (0–50 µM) in combination with LPS (1 µg/ml) and SNOG (0–500 µM). GSH, the precursor of SNOG, was used to adjust combined SNOG and GSH concentrations to 500 µM in all wells. SNOG (400 µM) in RPMI-1640 complete medium produced peak NO· concentrations of 
150 µM at 1 h, which decayed rapidly to levels of 50 µM (NO· Analyzer 280, Sievers, Watertown, MA). After 20 h incubation, culture supernatants were collected. IL-8 and TNF- production was measured using an enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions. To examine the effect of p38 MAPK on IL-8 production, THP-1 cells were pretreated with the specific p38 MAPK inhibitor SB202190 (0–0.1 µM) for 1 h and then incubated in complete medium for an additional 8 h with LPS (1 µg/ml) and phosphate-buffered saline (PBS) alone, GSH (400 µM), or SNOG (400 µM). IL-8 concentrations in supernatant were then determined as described above.
Quantitation of cytokine mRNA levels and stability
Total RNA from 1 x 107 THP-1 cells or primary monocytes was isolated using RNeasy kits (Qiagen, Valencia, CA) after incubation in medium with various treatments and time-periods as specified in the corresponding figures. IL-8, TNF-, and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) or ß-actin mRNA levels were measured from 1 to 2 µg total RNA using Quantikine® colorimetric mRNA quantitation kits and specific Quantikine® mRNA probes (R&D Systems) following the manufacturer’s instructions. All data were normalized to the housekeeping genes GAPDH or ß-actin as indicated and expressed in amol/µg total RNA using a standard curve.
In one experiment, IL-8 mRNA was measured in THP-1 cells that were pretreated with the p38 MAPK inhibitor SB202190 (0–0.1 µM) for 1 h followed by LPS (1 µg/ml) stimulation and exposure to PBS, GSH (400 µM), or SNOG (400 µM) for 4 h. For measurement of mRNA stability, THP-1 cells or primary monocytes (1x107) were stimulated with LPS (1 µg/ml) for 4 h, followed by 30 min pretreatment with actinomycin D (2.5 µg/ml). In two experiments measuring mRNA stability, as indicated, pretreatment also included exposure of some cells to SB202190 (0.1 µM). Then PBS, GSH (400 µM), or SNOG (400 µM) was added, followed by further incubation for 0, 1, 2, or 4 h. Total RNA was prepared, and mRNA was measured as above.
Nuclear run-on assay
THP-1 cells (2x107) were incubated in RPMI-1640 complete medium to which PBS or LPS (1 µg/ml) in combination with GSH (400 µM) or SNOG (400 µM) had been added for 2 h. Nuclei were isolated using nuclei isolation kits (Sigma-Aldrich). Nuclear run-on assay was performed according to the method of Eick and Bornkamm [52
] with some modifications. Briefly, nuclei were incubated in reaction buffer [5 mM Tris-CI, pH 8.0, 5.5 mM MgCI2,150 mM KCI, 0.25 mM each adenosine-, cytidine-, and guanosine 5'-triphosphate, and 250 µCi -32P uridine 5'-triphosphate (UTP)] for 30 min at 30°C. RNA was isolated using RNeasy mini kits (Qiagen). Equal amounts of labeled RNA, in 5 ml PerfectHybtm plus hybridization buffer (Sigma-Aldrich), were then hybridized to target DNA immobilized on nitrocellulose filters. After overnight hybridization at 65°C, filters were washed with 2x saline sodium citrate (SSC) containing 0.1% sodium dodecyl sulfate (SDS; 15 min, four times) and then with 0.1x SSC containing 0.1% SDS (30 min, twice). Filters were exposed to X-ray film, and bands were quantitated by densitometry. Results were normalized to the housekeeping gene ß-actin and expressed as relative fold change.
Plasmid construction
A two-plasmid reporter gene system [37
, 38
, 53
] was constructed for analyzing IL-8 promoter activity. The first plasmid was made from plasmid pTet-off (BD Biosciences, Palo Alto, CA), designated here as pCMV-tTA which uses the strong, immediate cytomegalovirus (CMV) promoter to express the tetracycline (tet)-responsive transcriptional activator (tTA). tTA is a fusion protein, which transactivates expression of a reporter gene, chloramphenicol acetyltransferase (CAT), by binding to a tet-responsive element in the promoter of a second plasmid, pUHG10.3CAT. Promoterless ptTA vector was generated from pCMV-tTA by removing the whole CMV promoter region with XhoI/EcoRI. The human IL-8 promoter (nt –1470–+40) was inserted into the XhoI/EcoRI site of ptTA, generating p(IL-8)tTA.
Cell transfection and CAT assay
THP-1 cells (3x106) in 4 ml fresh RPMI-1640 complete medium were planted into six-well plates and cultured for at least 1 h before transfection. Plasmid containing the IL-8 promoter, p(IL-8)tTA, or promoterless plasmid was cotransfected with pUHG10.3CAT and a plasmid containing the simian virus 40 early promoter and ß-galactosidase reporter (pSVß-gal; 0.5 µg, 0.3 µg, 0.1 µg /well, respectively) using Effectene transfection kits (Qiagen). After 5 h, cells were washed three times with PBS, recovered for 5 h in fresh medium, and pretreated without or with LPS (1 µg/ml) for 30 min, followed by 16 h incubation in the presence of PBS, GSH (400 µM), or SNOG (400 µM). CAT concentration was measured using ELISA kits (Roche Applied Science, Indianapolis, IN) and normalized to ß-gal. Promoter activity was expressed as fold induction of CAT compared with p(IL-8)tTA transfection without LPS pretreatment.
Western blotting
Effect of NO· on p38 MAPK phosphorylation in LPS-stimulated THP-1 cells was determined by Western blot. THP-1 cells (5x106) were preincubated with LPS (1 µg/ml) for 90 min, followed by treatment for 30 min with SNOG (0–400 µM). Cells were lysed with ice-cold lysis buffer containing 50 mM HEPES, 5 mM EDTA, 50 mM NaCl, 1% Triton X-100, and protease inhibitor cocktail (Roche Applied Science). Solubilized proteins were collected after centrifugation at 10,000 g in a microcentrifuge at 4°C for 10 min and were then quantitated by bicinchoninic acid assay (Pierce, Rockford, IL). Proteins (20 µg) were then separated on 4–20% Tris-glycine gels and transferred to nitrocellulose membranes (Invitrogen, Carlsbad, CA). The membranes were blocked overnight with Tris-buffered saline containing 5% nonfat dry milk and 0.05% Tween 20 at 4°C and were incubated for 1 h with primary, antiactive p38 MAPK (Promega, Madison, WI) or anti-p38 MAPK antibody (New England Biolabs, Beverly, MA). After incubation with secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA), phosphorylated and nonphosphorylated forms of p38 MAPK were detected by enhanced chemiluminescence (ECL). Results were quantitated using laser densitometry and expressed as relative change compared with control.
Statistical analysis
Data are presented as means ± SEM of at least three independent experiments. All Pvalues are two-sided and considered significant if less than 0.05. Effects of cGMP and cAMP on NO· induction of IL-8 and TNF- production were analyzed by reducing each NO· dose-response curve to the least-squares estimate of their slope using the log of IL-8 or TNF- concentration versus dose. Then, a blocked, exact, nonparametric test for trend was performed across the three ordered cGMP and cAMP doses. To determine the similarity of IL-8 and TNF- responses to cGMP, a three-way ANOVA was used to test for an interaction. For cAMP, the blocked, exact, nonparametric test for trend returned very strong evidence for a dose-related cAMP effect on TNF-, but not IL-8, making testing for an interaction unnecessary.
IL-8 and TNF- mRNA levels over time were analyzed using a repeated measures ANOVA followed by post-hoc tests, adjusted for multiple comparisons using the Bonferroni approach to determine which groups were different. Nuclear run-on and reporter gene results were compared using simple paired t-tests, as only two conditions, GSH and SNOG, were of interest. However, the reporter gene data were also analyzed using a three-way ANOVA to assess the effect of LPS in the experiment. Curves measuring mRNA stability were first converted to their respective area under the curve and then compared using a two-factor ANOVA. If this overall test showed that the groups were different, post-hoc tests using the Bonferroni approach were computed for group-wise comparisons.
NO· activation of p38 MAPK was examined using a two-way ANOVA. The effect of increasing doses of p38 MAPK inhibitor on IL-8 expression was assessed by computing the slope of the dose-response curves after log transformation of dose. Slopes were then compared using two-way ANOVA followed by Bonferroni inequality adjusted post-hoc testing. An identical approach was used to analyze the effect of p38 MAPK inhibitor on IL-8 mRNA degradation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
production (P=0.002) production in LPS-stimulated THP-1 cells (Figs. 1
and 2
). Increasing concentrations (up to 50 µM) of cGMP analog, Bt2cGMP, failed to independently increase IL-8 (Fig. 1A
; P=0.54) production but appeared to have a small but nonsignificant effect on TNF- (Fig. 1B
; P=0.08). Further, Bt2cGMP nonsignificantly augmented NO·-induced IL-8 (P=0.07) and TNF- (P=0.18) production. Examining results across the two cytokines, no evidence was found for a difference in the effect of cGMP analog on IL-8 and TNF- (P=0.93). These experiments suggest that NO· regulation of IL-8 and TNF- in LPS-stimulated THP-1 cells occurs primarily through cGMP-independent signal-transduction pathways. This finding is consistent with previous investigations in other cell types including neutrophils for IL-8 [22
] and neutrophils [7
], peripheral blood mononuclear cells [6
], and differentiated U937 cells [35
, 36
] for TNF-.
 View larger version (15K): [in a new window] |
Figure 1. NO· up-regulation of IL-8 and TNF- is cGMP-independent. THP-1 cells were stimulated with LPS (1 µg/ml). At the same time, Bt2cGMP (0–50 µM), a cell-permeable cGMP analog, and SNOG (0–500 µM), a NO· donor, were added to the cell cultures. Cells exposed to less then 500 µM SNOG were supplemented so that all conditions contained the same concentration of GSH (i.e., 500 µM). IL-8 (A) and TNF- (B) released into the medium were measured by ELISA after 20 h of incubation. Data are presented as the means ± SEM of three independent experiments, each run in duplicate.
|
 View larger version (15K): [in a new window] |
Figure 2. NO· up-regulation of IL-8 but not TNF- is cAMP-independent. THP-1 cells were stimulated with LPS (1 µg/ml). At the same time, Bt2cAMP (0–50 µM), a cell-permeable cAMP analog, and SNOG (0–500 µM), a NO· donor, were added to the cell cultures. Cells exposed to less than 500 µM SNOG were supplemented so that all conditions contained the same concentration of GSH (i.e., 500 µM). IL-8 (A) and TNF- (B) released into the medium were measured by ELISA after 20 h of incubation. Data are presented as the means ± SEM of three independent experiments, each run in duplicate.
|
production were tested for interactions with cAMP. We have previously demonstrated that NO· up-regulates TNF- by decreasing intracellular cAMP in differentiated U937 cells [36
]. It is important that in these cells, NO· up-regulation of TNF- is blocked by Bt2cAMP [36
37
38
], a cell-permeable cAMP analog. However, in LPS-stimulated THP-1 cells, the dose-dependent effect of NO· on IL-8 production was unaffected by Bt2cAMP (Fig. 2A ; P=0.18). Conversely and consistent with its effects in U937 cells, Bt2cAMP dose-dependently suppressed NO·-induced TNF- production (Fig. 2B
; P=0.009). These results suggest that NO· regulation of IL-8 is cGMP- and cAMP-independent, and TNF- is cGMP-independent but cAMP-dependent. The findings support the notion that NO· regulates these two cytokines through distinct signal-transduction pathways.
Effect of NO· on IL-8 and TNF- mRNA expression
To examine whether NO· regulates IL-8 and TNF- mRNA expression, THP-1 cells were incubated with PBS, GSH (400 µM), or SNOG (400 µM) for 2, 4, and 6 h in the presence of LPS (1 µg/ml). As shown in Figure 3A
the NO· donor SNOG significantly increased LPS-induced IL-8 mRNA levels, with a peak effect at 4 h, compared with GSH or PBS (adjusted P=0.004 for both comparisons). Similar to our previous findings in differentiated U937 cells [36
], NO· increased TNF- mRNA in LPS-stimulated THP-1 cells (Fig. 3B
; adjusted P
0.037 compared with GSH or PBS). Peak mRNA levels appeared to occur earlier (at 2 h) and decay faster for TNF- compared with IL-8.
 View larger version (12K): [in a new window] |
Figure 3. Effect of NO· on IL-8 and TNF- mRNA levels. THP-1 cells were stimulated with LPS (1 µg/ml) and incubated with PBS, GSH (400 µM), or SNOG (400 µM). Cells were harvested for total RNA preparation at the indicated time-points. IL-8 (A) and TNF- (B) mRNA levels were quantitated using a colorimetric assay normalized to GAPDH mRNA levels. Results are expressed relative to 1 µg total RNA and presented as the means ± SEM of three independent experiments.
|
transcription gene transcription in differentiated U937 cells by decreasing intracellular levels of cAMP [36
], thereby reducing promoter repression at a proximal GC-box element [37
, 38
]. However, the IL-8 promoter lacks a similar GC-box element [42
]. Further, in LPS-stimulated THP-1 cells, cAMP analog blocked NO· induction of TNF- but not IL-8 (Fig. 2)
. Therefore, we next determined whether NO· transcriptionally regulates IL-8 or TNF- in LPS-stimulated THP-1 cells. As determined with nuclear run-on assays (Fig. 4
), NO· did not increase and tended to slightly decrease IL-8 transcription in LPS-stimulated THP-1 cells (P=0.08, comparing SNOG with GSH). In contrast and as expected from previous results in U937 cells [37
, 38
], NO· significantly increased TNF- transcription (P=0.012) in LPS-stimulated THP-1 cells (Fig. 4)
.
 View larger version (27K): [in a new window] |
Figure 4. Effect of NO· on the transcription of IL-8 and TNF- mRNA as measured by nuclear run-on assay. THP-1 cells (2x107) were stimulated with LPS (1 µg/ml) and incubated with GSH (400 µM) or SNOG (400 µM) for 2 h. Unstimulated cells not exposed to LPS (PBS only) were examined concurrently for comparison. Nuclei were isolated and incubated in reaction buffer containing -P32 UTP. RNA was then isolated and hybridized to target DNA immobilized on nitrocellulose filters. Bands on X-ray film exposed to the filters were quantitated by densitometry. A representative experiment (A) and densitometry results (B) expressed as ratios to concurrently run ß-actin controls are shown. Data presented are the means ± SEM of four independent experiments.
|
 View larger version (17K): [in a new window] |
Figure 5. Effect of NO· on IL-8 promoter activity as measured by gene reporter assay. Plasmids containing the IL-8 promoter, p(IL-8)tTA, or promoterless vector (data not shown) were cotransfected with pUHG10.3CAT and pSVß-gal into THP-1 cells (3x106). Cells were allowed to recover for 5 h and then placed in medium alone or medium with LPS (1 µg/ml). PBS, GSH (400 µM), or SNOG (400 µM) was added 30 min later followed by incubation for 16 h at 37°C. CAT expression was measured and normalized to ß-gal. Promoter activity is expressed as fold induction of CAT relative to the p(IL-8)tTA response to PBS alone in the absence of LPS stimulation. Results are presented as the means ± SEM of five independent experiments.
|
mRNA stability in THP-1 cells mRNA expression. Transcription was then blocked with actinomycin D, and the half-lives of IL-8 and TNF- transcripts were assessed. IL-8 and TNF- are known to undergo substantial post-transcriptional regulation. Although NO· has been shown here and previously [35
36
37
38
] to regulate TNF- transcription, its effects on TNF- mRNA stability have not been reported. As shown in Figure 6A
the NO· donor SNOG significantly protected IL-8 mRNA from degradation (P=0.001) compared with PBS or GSH, which produced nearly identical decay curves (P=1.00). NO· increased the half-life of IL-8 mRNA to more than 4 h compared with 1–1.5 h for GSH (Fig. 6A)
. In contrast, TNF- degradation was unaffected across the three treatments (Fig. 6B
; P=0.63).
 View larger version (12K): [in a new window] |
Figure 6. NO· increases the stability of IL-8 but not TNF- mRNA. THP-1 cells (1x107) were stimulated with LPS (1 µg/ml) for 4 h at 37°C to induce IL-8 and TNF-. After 30 min pretreatment with actinomycin D (2.5 µg/ml), cells were further incubated with PBS, GSH (400 µM), or SNOG (400 µM) for 0–4 h. Cells were then harvested, and total RNA was prepared. IL-8 (A) and TNF- (B) mRNA was quantitated using a colorimetric assay normalized to GAPDH mRNA levels. Results are expressed relative to mRNA levels at 0 h, which were arbitrarily set to 100%. Data presented are the means ± SEM of five independent experiments.
|
 View larger version (24K): [in a new window] |
Figure 7. NO· activates p38 MAPK. THP-cells (5x106) were incubated in medium alone or medium with LPS (1 µg/ml) for 90 min at 37°C. SNOG (0–400 µM) was then added for an additional 30 min, and the cells were lysed for Western blotting. Nonphosphorylated (pp38) and phosphorylated forms (p38) of p38 MAPK were detected by specific antibody and ECL. Results were quantitated with laser densitometry and expressed as ratios relative to control values (SNOG=0 µM). Data presented are the means ± SEM of three independent experiments.
|
 View larger version (13K): [in a new window] |
Figure 8. Effect of p38 MAPK inhibition on IL-8 expression. THP-1 cells were pretreated with SB202190 (0–0.1 µM) for 1 h at 37°C and were then stimulated with LPS (1 µg/ml) and incubated with PBS, GSH (400 µM), or SNOG (400 µM). IL-8 protein (A) released into the medium was measured by ELISA after incubation of 5 x 105 cells for 8 h. Duplicate wells containing THP-1 cells (1x107) were harvested after 4 h of incubation for total RNA preparation. IL-8 mRNA (B) was quantitated using a colorimetric assay normalized to GAPDH mRNA levels, and results are expressed relative to 1 µg total RNA. Data presented are the means ± SEM of five independent experiments.
|
 View larger version (20K): [in a new window] |
Figure 9. Effect of p38 MAPK inhibition on NO·-induced IL-8 mRNA stabilization. THP-1 cells (1x107) were stimulated with LPS (1 µg/ml) for 4 h to induce IL-8. After 30 min pretreatment with actinomycin D (2.5 µg/ml) without and with SB202190 (0.1 µM), cells were further incubated with PBS, GSH (400 µM), or SNOG (400 µM) for 0–4 h. At the specified time-points, cells were harvested for total RNA preparation. IL-8 mRNA was then quantitated using a colorimetric assay normalized to GAPDH mRNA levels. Results are expressed relative to mRNA levels at 0 h, which were arbitrarily set to 100%. Data presented are the means ± SEM of three independent experiments.
|
mRNA stability in human primary monocytes mRNA stabilization in human primary monocytes. As shown in Figure 10A
NO· donor SNOG significantly increased IL-8 mRNA stability compared with the GSH control condition (P=0.001). This effect of NO· was completely blocked by the specific p38 MAPK inhibitor SB202190 (P=0.001), which did not affect the IL-8 mRNA decay curve of the GSH control (Fig. 10A
; P=1.0, adjusted). In contrast to IL-8, TNF- mRNA stability was not altered by NO·; SNOG and GSH resulted in similar decay curves (Fig. 10B
; P=0.32). Furthermore, the p38 MAPK inhibitor SB202190 had no effect on TNF- mRNA degradation in the presence of SNOG or GSH (Fig. 10B
; P=0.32).
 View larger version (12K): [in a new window] |
Figure 10. Effects of NO· and p38 MAPK inhibition on IL-8 and TNF- mRNA stabilization in human primary monocytes. Elutriated monocytes (1x107) were stimulated with LPS (1 µg/ml) for 4 h to induce cytokines. After 30 min pretreatment with actinomycin D (2.5 µg/ml) without and with SB202190 (0.1 µM), cells were further incubated with GSH (400 µM) or SNOG (400 µM) for 0–4 h. At the specific time-points, cells were harvested for total RNA extraction. IL-8 (A) and TNF- (B) mRNA was then quantitated using a colorimetric assay normalized to ß-actin mRNA levels. Results are expressed relative to mRNA levels at 0 h, which were arbitrarily set to 100%. Data presented are the means ± SEM of four independent experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, independent of cGMP. Up to 50 µM Bt2cGMP had only modest effects on IL-8 or TNF-, suggesting that cGMP was not an essential second messenger in NO· up-regulation of these cytokines. This is in accordance with previous findings that NO· increases the production of IL-8 and TNF- in a variety of cell types [6
, 7
, 22
, 35
, 36
] through cGMP-independent mechanisms.
TNF- promoter activity is inhibited by cAMP via activation of cAMP-dependent protein kinase and the subsequent phosphorylation of transcriptional factors [43
, 53
, 58
]. We previously demonstrated that NO· increases TNF- gene transcription in phorbol ester-differentiated U937 cells by decreasing intracellular cAMP concentrations [36
]. Further investigation identified a proximal GC box in the TNF- promoter that transduces NO·/cAMP/cAMP-dependent protein kinase signaling [37
, 38
]. NO· enhances TNF- promoter activity by inhibiting the binding of trans-repressor Sp proteins to this GC box [38
]. Consistent with these results, the present study showed in LPS-stimulated THP-1 cells that Bt2cAMP, a cell-permeable cAMP analog, attenuated NO·-induced TNF- production. Further, nuclear run-on assay, like previous reporter gene experiments in U937 cells [37
, 38
], demonstrated that TNF- gene transcription was up-regulated by NO· in LPS-stimulated THP-1 cells. In contrast, the up-regulatory effect of NO· on IL-8 was not altered by Bt2cAMP. Moreover, NO· did not increase the transcription of IL-8 as measured by nuclear run-on assay and reporter gene analysis. Accordingly, DNA foot-printing in THP-1 cells demonstrated that NO·, in the concentrations examined here, had little or no effect on protein binding to the activated protein 1 (AP-1), nuclear factor (NF)-IL-6, NF-
B, and Oct-1 sites in the IL-8 proximal promoter (data not shown). Finally, we demonstrated that NO· regulates IL-8 mRNA half-life in THP-1 cells and human primary monoctyes. This effect is at least partially dependent on p38 MAPK activation. Collectively, these results show that NO· regulates TNF- and IL-8 by distinct cGMP-independent mechanisms.
In contrast to our results, other studies have shown that NO· regulates IL-8 at the level of transcription [39 40 41 , 55 56 57 ]. Most frequently, NO· has been observed or suggested to increase rather than decrease IL-8 promoter activity or transcription. Work reporting this result has been conducted in a wide variety of cell types including melanocytes [39 ], keratinocytes [40 ], U937 cells [24 ], and pancreatic adenocarcinoma. Conversely, down-regulation of promoter activity has also been seen, particularly in endothelial cells [55 , 56 ]. However, another group of investigators has reported that NO· increases the release of IL-8 protein in the same endothelial cell line [59 ]. Although these studies demonstrate, in some systems and cell types, that changes in transcription can play an important role in NO· regulation of IL-8, the possibility of substantial post-transcriptional effects was not investigated. Closest to our findings, Xiong et al. [41 ] showed in human pancreatic adenocarcinoma cells that NO· stabilized IL-8 mRNA in addition to up-regulating its transcription. Possible mechanisms responsible for mRNA stabilization were not investigated. Our results in LPS-stimulated THP-1 cells and human monocytes suggest that post-transcriptional stabilization of mRNA may be the predominant mechanism by which NO· regulates IL-8. Furthermore, p38 MAPK signaling was identified as a major downstream pathway responsible for this effect of NO·.
To date, NO· regulation of IL-8 and other genes has been attributed to a broad range of possible mechanisms. However, most of these appear to have little impact on the system we investigated. A number of IL-8 promoter elements such as AP-1, NF-IL-6, NF-B, and Oct-1-binding sites [42
43
44
] have been experimentally associated with NO· regulation of gene expression [6
, 8
, 10
, 11
, 39
, 54
55
56
57
]. For example, AP-1 can be activated by NO· through the second messenger cGMP and thereby regulate transcription [10
, 11
]. Although most studies, like ours, have found that IL-8 regulation by NO· is cGMP-independent, one report in a mesangial cell line came to the opposite conclusion [60
]. This result and well-documented examples of IL-8 promoter activation by AP-1 [44
, 48
, 61
] suggest that the NO·/cGMP/AP-1 signaling pathway is likely to be an important transducer of NO· effects on IL-8 under some conditions. A contrasting report in an endothelial cell line found that NO· suppressed IL-8 transcription by blocking C-Jun N-terminal kinase-induced AP-1 activation [56
]. Likewise, it has been shown that NO· can variably activate or inactivate NF-B signal transduction through several mechanisms [6
, 8
, 54
, 55
, 62
, 63
]. Although our results clearly rule out NO· activation of the IL-8 promoter in LPS-stimulated THP-1 cells under other circumstances, NO· effects at these promoter sites might be substantial, as transcriptional regulation is highly context-dependent.
The specificity of post-transcriptional regulation and its overall important contribution to gene expression have been increasingly recognized [64
65
66
67
]. IL-8 and TNF- mRNAs have ARE in their 3'-UTR and have been shown to undergo substantial regulation after transcription [45
46
47
48
49
50
]. However, as shown here, the mRNA of IL-8 but not TNF- was significantly stabilized by NO· through a p38 MAPK-dependent signal-transduction pathway. The 3'-UTR of TNF- tightly controls its degradation and translation [49
, 50
, 68
]. Knockout mice with TNF- mRNA lacking a 3'-UTR are unable to appropriately turn off TNF- and develop an autoimmune illness phenotype [68
]. Notably, Brook et al. [69
] demonstrated in a mouse macrophage cell line that TNF- mRNA was stabilized by p38 MAPK. In contrast, here, we found that TNF- mRNA half-life was not affected by NO·, despite p38 MAPK activation. Our results agree with a recent experiment, also conducted in LPS-stimulated THP-1 cells, showing that the post-transcriptional regulation of IL-8 but not TNF- was p38 MAPK-dependent [49
]. However, compared with IL-8, TNF- mRNA half-life is shorter, suggesting that these two cytokines have different mRNA decay kinetics. Therefore, we cannot exclude the possibility that NO· may affect TNF- mRNA degradation at time-points earlier than 1 h. The 3'-UTR of IL-8 has multiple ARE [46
]. At least two of these, located between nucleotides 972 and 1132, have been associated with p38 MAPK-mediated stabilization of IL-8 mRNA. It is interesting that fewer than 10% of known ARE-containing genes appear to be post-transcriptionally regulated by p38 MAPK [49
].
In summary, the current investigation has identified a cGMP-independent mechanism, distinct from that for TNF-, by which NO· regulates IL-8. NO· substantially stabilizes IL-8 mRNA through a pathway that is dependent on its ability to activate p38 MAPK. The overall importance of this pathway in the regulation of inflammation and leukocyte migration by NO· remains to be determined. However, recent reports indicate that these immune-modulatory effects function in vivo [25
, 26
]. Shanley et al. [25
] and Okamoto et al. [26
] found that the lungs of iNOS knockout mice compared with wild-type animals had reduced proinflammatory cytokine production and neutrophil accumulation in response to LPS challenge. These studies strongly support the concept that iNOS induction contributes to the appropriate trafficking of neutrophils to sites of inflammation. It is interesting that immunologic and gene-regulatory effects of NO· appear to be compartmentalized geographically and mechanistically. For example, NO· inhibition of neutrophil adhesion and induction of TNF- and IL-8 all occur through different pathways. This may permit cross-talk and also maintain the potential for the differential regulation of these dissimilar responses in the microenvironment of the vessel wall. Therefore, NO·-mediated modulation of immunity might be, to some degree, therapeutically dissectible. Interventions may be possible that allow certain pathways to proceed unabated, even as others are selectively blocked.
2 Current address: Columbia University College of Physicians and Surgeons, Department of Anesthesiology, 630 West 168th Street (PH5-505), New York, NY 10032-3784.
Received December 23, 2003; revised March 31, 2004; accepted April 1, 2004.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
production by human neutrophils J. Immunol. 152,4102-4109[Abstract]
by nitric oxide in monocytes/macrophages down-regulates p47phox and attenuates the respiratory burst J. Immunol. 169,2619-2626 production in transfected human U937 cells Blood 90,1160-1167 promoter functions as a nitric oxide response element J. Biol. Chem. 274,33190-33193 induction by LPS is regulated posttranscriptionally via a Tpl2/ERK-dependent pathway Cell 103,1071-1083[CrossRef][Medline]B-mediated transcription in human monocytic cells and endothelial cells J. Biol. Chem. 271,20828-20835 B by S-nitrosylation Biochemistry 40,1688-1693[CrossRef][Medline]B binding to DNA: effects on endothelial function Biochem. Cell Biol. 77,201-208[CrossRef][Medline]B activities and stability of interleukin 8 (IL-8) transcripts are implicated in IL-8 mRNA superinduction in lung epithelial H292 cells Biochem. J. 330,429-435B activation by interfering with phosphorylation and degradation of IB- Arterioscler. Thromb. Vasc. Biol. 18,1796-1802 B-dependent transcription and stabilization by serine 15 phosphorylation J. Biol. Chem. 277,33501-33508 mRNA stability by the mitogen-activated protein kinase p38 signalling cascade FEBS Lett. 483,57-61[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  S. Wang, J. Zhang, Y. Zhang, S. Kern, and R. L. Danner Nitric oxide-p38 MAPK signaling stabilizes mRNA through AU-rich element-dependent and -independent mechanisms J. Leukoc. Biol., April 1, 2008; 83(4): 982 - 990. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Ptasinska, S. Wang, J. Zhang, R. A. Wesley, and R. L. Danner Nitric oxide activation of peroxisome proliferator-activated receptor gamma through a p38 MAPK signaling pathway FASEB J, March 1, 2007; 21(3): 950 - 961. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Zhang, S. Wang, S. Kern, X. Cui, and R. L. Danner Nitric Oxide Down-regulates Polo-like Kinase 1 through a Proximal Promoter Cell Cycle Gene Homology Region J. Biol. Chem., January 12, 2007; 282(2): 1003 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Wang, J. Zhang, S. Theel, J. J. Barb, P. J. Munson, and R. L. Danner Nitric oxide activation of Erk1/2 regulates the stability and translation of mRNA transcripts containing CU-rich elements Nucleic Acids Res., June 6, 2006; 34(10): 3044 - 3056. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Fan, N. M. Heller, M. Gorospe, U. Atasoy, and C. Stellato The role of post-transcriptional regulation in chemokine gene expression in inflammation and allergy Eur. Respir. J., November 1, 2005; 26(5): 933 - 947. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-O. Pae, G.-S. Oh, B.-M. Choi, Y.-M. Kim, and H.-T. Chung A Molecular Cascade Showing Nitric Oxide-Heme Oxygenase-1-Vascular Endothelial Growth Factor-Interleukin-8 Sequence in Human Endothelial Cells Endocrinology, May 1, 2005; 146(5): 2229 - 2238. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
