| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online February 5, 2001 as doi:10.1096/fj.00-0432fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Thoracic Medicine, National Heart & Lung Institute, Imperial College School of Medicine, Dovehouse Street, London SW3 6LY
2Correspondence: Thoracic Medicine, National Heart & Lung Institute, Imperial College School of Medicine, Dovehouse Street, London SW3 6LY. E-mail: ian.adcock{at}ic.ac.uk.
SPECIFIC AIMS
Glucocorticoidsact, at least in part, through recruitment of histone deacetylases (HDACs) to sites of inflammatory gene transcription. In this study we show that cigarette smoke, an oxidative stress, decreased HDAC activity in human biopsies and macrophages in vivo. This reduced activity correlated with enhanced induction of inflammatory cytokines and a reduction in glucocorticoid responsiveness in vitro.
PRINCIPAL FINDINGS
1. HAT and HDAC expression in bronchial biopsies and alveolar
macrophages.
The histone acetyltransferases (HATs) CBP and HDAC1 and HDAC2 are
localized within the airway to all cells with the most intense staining
within the epithelium and inflammatory cells. Smoking did not affect
the site of the expression of any of these proteins. Western blot
analysis detected no difference in the expression of CBP or PCAF
between the two groups. In contrast, there was a decrease in HDAC2
expression (0.63±0.02 vs. 0.37±0.06 OD units, P<0.01),
but not HDAC1 (0.55±0.07 versus 0.55±0.10 OD units) in samples
extracted from cigarette smokers. HDAC activity was also reduced in
smokers (111±11 vs. 146±14 dpm/µg protein,
P <0.01).
Similar reductions in HDAC2 expression were observed in bronchoalveolar
macrophages from smokers, whereas HDAC1 expression was not altered
(Fig. 1
). These changes resulted from reduced HDAC2 mRNA. We
also detected decreased HDAC activity in macrophages isolated from
smokers (59.1±8.4 dpm/µg protein, n=6) compared with
nonsmoking controls (120±11 dpm/µg protein, n=6,
P=0.0012), which correlated with the reduction in HDAC2
expression (Fig. 1
).
|
2. Cytokine production in bronchoalveolar lavage (BAL) macrophages
Reduced HDAC activity should lead to increased inflammatory gene
expression following cell stimulation. Therefore, we measured
interleukin (IL) -1ß-stimulated tumor necrosis factor 
(TNF-)
and IL-8 release. Although basal release of TNF- was unaltered
between groups (0.51±0.09 versus 0.58±0.12 ng/ml), IL-1ß-stimulated
TNF- release was significantly increased in both groups, but with a
greater increase in smokers (n=6) than in nonsmokers (2.06±0.28 vs.
0.83±0.1 ng/ml, n=6, P <0.05). In addition,
there was a significant correlation between IL-1ß-stimulated TNF-
release and HDAC activity (r=0.79,
P=0.0021). Stimulation of cells with TSA, an inhibitor of
HDAC activity, significantly enhanced IL-1ß-stimulated TNF-
release in both groups. This reached the level seen in IL-1ß–
stimulated cells in smokers. TSA and IL-1ß together enhanced TNF-
release in nonsmokers to levels seen in smokers, whereas TSA failed to
further enhance IL-1ß– stimulated TNF- production in smokers.
There was a significant correlation between HDAC activity and the ratio
between IL-1ß– stimulated and maximal TNF- release (IL-1ß plus
TSA, r=0.94, P <0.0001). IL-1ß–
stimulated IL-8 release was greater in smokers (n=6) than in
nonsmokers (1.54±0.2 vs. 1.24±0.14 ng/ml, n=6), although
this failed to reach significance (P=0.073). There was a
significant correlation between the ratio between IL-1ß– stimulated
and maximally stimulated IL-8 expression and HDAC activity
(r =0.92,
P <0.0001).
3. Smoking reduces dexamethasone actions in macrophages
Reduced HDAC activity should also limit glucocorticoid actions in
these cells. We therefore examined the effect of dexamethasone on
IL-1ß-stimulated TNF- release in macrophages from smokers and
nonsmokers. In nonsmokers, dexamethasone (10–6M)
suppressed TNF- release by 77%. In contrast, dexamethasone elicited
only a 34% reduction in TNF- release from smoker’s macrophages.
Similar results were seen for IL-8 release. Dexamethasone suppressed
IL-8 release by 64% in nonsmokers and by 29% in smokers. A
significant correlation appeared between HDAC activity and the
inhibitory effect of dexamethasone on both TNF- release
(r =0.88, P <0.0001) and IL-8 release
(r =0.65, P=0.029).
4. Oxidative stress suppresses dexamethasone repression of cytokine
release
To model whether cigarette smoking-induced reduction in HDAC
activity could increase inflammatory gene expression, we examined the
effect of oxidant stress (100 µM
H2O2) on GM-CSF release and
inhibition by dexamethasone in A549 cells
(Fig. 2
). IL-1ß caused a significant increase in GM-CSF
(23.0±4.4 versus 714±93.6 pg/ml) that was inhibited 96% by
dexamethasone (1µM) (Fig. 2)
. Trichostatin A (TSA, 10 ng/ml)
inhibited the ability of dexamethasone (1 µM) to reduce GM-CSF
release (96% inhibition vs. 49% inhibition). Hydrogen peroxide (100
µM) enhanced IL-1ß–stimulated GM-CSF release (1280.5±103 versus
714.3±93.6 pg/ml) and reduced dexamethasone inhibition to equivalent
levels seen in the presence of TSA (52% inhibition). The potency of
dexamethasone was also altered reflecting a reduced sensitivity
(IC50=2.0x 10–9 M
vs.1.1 x10–7 M).
H2O2 alone had no effect on
GM-CSF production. We also measured HDAC activity in the
monocyte/macrophage cell line U937. In these cells
H2O2 gave a
concentration-dependent inhibition of HDAC activity, which reached a
level of 63% of control at 100 µM
H2O2 (Fig. 2)
. Without
H2O2, a
concentration-dependent inhibition of lipopolysaccaride (LPS)-induced
IL-8 production by dexamethasone (IC50 of
1.0 x10–9 M) was apparent.
Following H2O2 (100 µM)
costimulation for 4 h, there was a significant increase in
LPS-induced IL-8 production and a reduced effectiveness of
dexamethasone to suppress IL-8 release. The maximal suppression reached
was 25% of the maximal stimulus.
H2O2 alone had no effect on
IL-8 production (Fig. 2)
.
|
CONCLUSIONS
We have shown that cigarette smoking is associated with a
reduction in HDAC2 expression and activity in bronchial biopsies and
alveolar macrophages. This condition is correlated with an increase in
basal and IL-1ß–simulated TNF- expression in alveolar
macrophages. The level of HDAC activity is inversely correlated with
maximal TNF- and IL-8 production in these cells. Previous data have
suggested that increased histone acetylation is associated with gene
induction. The overall acetylation status of histones depends on the
dynamic equilibrium between HAT and HDAC activities. Repression of HDAC
activity can lead to enhanced histone acetylation and increased gene
expression. This condition is most noticeable with the HDAC inhibitor
trichostatin A (TSA), which induces increased expression of
inflammatory genes following an inflammatory stimulus. Indeed, we show
that TSA causes the expected enhanced release of the cytokines GM-CSF,
TNF-, and IL-8 in this study. These results provide the first
evidence for a suppressive effect of cigarette smoking on histone
acetylation status. This effect results in increased acetylation and
causes local unwinding of DNA and allows increased inflammatory gene
expression to occur.
Cigarette smoke causes an inflammatory cascade resulting in the
production of many inflammatory mediators in important regulatory cells
within the airway, including epithelial cells, neutrophils, and
alveolar macrophages. Cigarette smoke contains high concentrations of
reactive oxygen species (ROS), such as superoxide and other
free-radicals. Exposure of lungs to these ROS leads to an accumulation
of neutrophils in alveolar walls and BAL fluid. This accumulation is
caused by the induction of chemotactic factors such as IL-8 in the lung
and particularly within alveolar macrophages. In addition to
neutrophils, IL-8 is also chemotactic to eosinophils, and cigarette
smoke has been shown to increase airway eosinophilia in biopsies and
BAL. Ex vivo studies have also shown that cigarette smoke
can enhance TNF- production from human monocytes.
Earlier studies have also reported increased IL-8 expression and neutrophilia associated with cigarette smoking, but in addition found elevated numbers of macrophages and increased expression of the inflammatory mediators IL-1ß, IL-6, and MCP-1. Macrophages play an important role in tissue repair and remodeling after inflammatory insults to the airway. Cigarette smoke increases basal metabolism and LPS-responsiveness of macrophages in vitro. This excessive activation of macrophages may result in release of proteinases and parenchymal destruction that occur in emphysema.
In contrast to our results, previous studies have reported that cigarette smoking causes either no change or a reduction in LPS-induced cytokine release from BAL macrophages. We have examined IL-1ß– stimulated cells and observed an increase in cytokine release. A simplistic explanation for these effects may be that smoking reduces the expression of CD14/Tol or other aspects of its signaling pathway. This finding may account for the observation that other aspects of macrophage activation are increased in cells isolated from smokers.
We have shown previously that glucocorticoid suppression of inflammatory genes requires recruitment of HDACs to the activation complex by the glucocorticoid receptor. This finding implies that reduced HDAC activity will not only increase inflammatory gene expression but will also cause a reduction in glucocorticoid function. Indeed, we show that this is the case in BAL macrophages from smokers and in A549 and U937 cells after oxidative stress (hydrogen peroxide). These data suggest that one potential reason for the failure of glucocorticoids to function effectively in reducing inflammation in COPD is that oxidative stress may reduce HDAC activity and expression.
Reduced HDAC expression may explain the enhanced expression of
inflammatory mediators such as GM-CSF, IL-8, and TNF- by cigarette
smoke seen in biopsy and lavage samples of smokers. Therefore, we
postulate that reduced HDAC expression and activity may also reduce GR
recruitment of HDAC activity to a repressor complex, thus limiting
glucocorticoid effectiveness in suppressing inflammation, as evidenced
in these studies and clinically in patients with
COPD.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.00-0432fje ; to cite this
article, use (February 6, 2001) FASEB J. 10.1096/fj.00-0432fje 
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
