| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 278, Issue 37, 34834-34844, September 12, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Gonda Diabetes Research Center, Beckman Research Institute of the City of Hope, Duarte, California 91010
Received for publication, March 19, 2003 , and in revised form, June 3, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
B (Bay11-7082), oxidant
stress, protein kinase C, ERK, and p38 MAPKs. S100b (4-h treatment)
significantly increased transcription from a human COX-2 promoter-luciferase
construct (4-fold, p < 0.001). Promoter deletion analyses and
inhibition of transcription by an NF-B superrepressor mutant confirmed
NF-B involvement. This was further supported by inhibition of
S100b-induced PGE2 by Bay11-7082. Additionally, S100b-induced
adherence of THP-1 monocytes to vascular smooth muscle cells was blocked by
the COX-2 inhibitor NS-398, Bay11-7082, inhibitors of ERK and p38 MAPK, and
protein kinase C thereby indicating functional relevance. S100b also increased
COX-2 mRNA expression in human peripheral blood monocytes from healthy donors.
Moreover, COX-2 mRNA levels were clearly evident in monocytes obtained from
diabetic patients but not from normal subjects. These results show for the
first time that AGEs can augment inflammatory responses by up-regulating COX-2
via RAGE and multiple signaling pathways, thereby leading to monocyte
activation and vascular cell dysfunction. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
, IL-1
,
and tumor necrosis factor (TNF)-, growth factors, and chemicals
(1–10).
An exception is seen in some tissues
(5), including the pancreatic
islet, that constitutively and dominantly expresses COX-2
(11,
12) and where its products
such as prostaglandin E2 (PGE2) are believed to play a
role in inflammation, islet destruction, and inhibition of insulin secretion
(11–15).
COX-2 and its products are implicated in the pathogenesis of several
inflammatory diseases, and selective inhibition of COX-2 has been shown to be
effective in reversing inflammation in various tissues without gastric side
effects (1,
5,
16,
17). Although COX-2 can lead
to the formation of the vasodilatory and protective prostacyclin
(PGI2), it also produces the potent inflammatory prostaglandin,
PGE2 (3,
5).
The sequence of the human COX-2 gene is known, and several cis-acting
regulatory elements have been identified
(18). Reports show the ability
of stimuli to induce transcription of the COX-2 gene via the involvement of
cis-acting elements such as AP2, STAT-1, STAT-3, NF-B, NF-IL-6,
cAMP-response element, peroxisomal proliferator response element (PPRE), and
CCAAT/enhancer-binding protein (C/EBP) transcription factors
(7,
18–23).
In addition, COX-2 has also been reported to be regulated
post-transcriptionally
(23–26).
COX-2 and its pro-inflammatory products have been implicated in the pathogenesis of atherosclerosis, and it is also induced by oxidized lipids (27–29). COX-2 was shown to promote early atherosclerotic lesion formation in low density lipoprotein receptor-deficient mice (28). Because COX-2 inhibitors also block formation of the protective prostacyclin (PGI2), studies have been performed to determine whether these inhibitors could worsen atherosclerosis (30, 31).
COX-2 has also been shown to play a role in islet dysfunction related to
the development of Type 1 diabetes. Thus COX-2 and PGE2 are
implicated in pancreatic islet -cell destruction and inhibition of
insulin secretion
(11–15).
Nitric oxide, as well as cytokines associated with islet dysfunction such as
IL-1, can induce COX-2 expression
(12). Although a recent study
demonstrated that high glucose can induce IL-1 in human pancreatic
islets (32), the effects on
COX-2 were not examined. High glucose (HG) was shown to enhance
IL-1-induced COX-2 expression in vascular smooth muscle cells
(33), and a very recent report
showed that high glucose treatment of endothelial cells increased COX-2
expression and decreased nitric oxide availability
(34). The administration of
the selective COX-2 inhibitor NS-398 could prevent the onset of diabetes in
mice (35). Elevated levels of
COX-2 protein and its product PGE2 were found in the spinal cord of
diabetic rats (36). However,
very little is known regarding the potential involvement of COX-2 in diabetic
vascular complications, diabetic atherosclerosis, or the regulation of COX-2
in monocytes under diabetic conditions. This is a significant issue,
especially because evidence now indicates the importance of inflammation in
the pathogenesis of atherosclerosis and diabetes. Furthermore, the mechanisms
of altered monocyte prostaglandin production in diabetes are largely unknown.
Our present studies provide new information in this connection.
Studies with inflammatory cells such as monocytes demonstrate that
simulated diabetic conditions in vitro, such as high glucose culture
conditions or treatment with advanced glycation end products (AGEs), can
induce the expression of inflammatory cytokine and related genes via
activation of specific signaling pathways and transcription factors such as
NF-B
(37–41).
This could then result in increased monocyte activation, migration, and
adhesion to the endothelium. In the present study, we evaluated the hypothesis
that AGEs acting via their receptor, RAGE, can lead to increased COX-2 gene
expression and activity and subsequent monocyte activation.
AGEs are products of non-enzymatic glycation/oxidation of proteins/lipids
that accumulate during natural aging and are also greatly augmented in
disorders such as diabetes, renal failure, and Alzheimer's disease
(42–45).
Formation of AGEs or glycooxidation products is related to circulating high
glucose concentrations in diabetes. Several receptors for AGEs have been
identified on vascular, renal, and other cells
(46). The well studied cell
surface receptor for AGE, namely RAGE, is a multiligand member of
immunoglobulin superfamily
(45–48).
Ligands for RAGE include AGEs, EN-RAGE, the S100/calgranulin family of
proteins, amphoterin, amyloid –peptide
(49,
50), and carboxymethyl lysine
adducts of protein (47).
Several short peptides including EN-RAGE and S100b belonging to
S100/calgranulin family signal through RAGE and can produce key
pro-inflammatory mediators in endothelium, mononuclear phagocytes, and
lymphocytes (49,
50) These peptides now serve
as valuable tools in the study of RAGE signaling.
Interaction of these ligands with RAGE can lead to the generation of
oxidant stress, production of growth factors and cytokines, chronic
inflammatory responses, and cellular and vascular dysfunction associated with
diabetic complications (49,
50). Blockade of RAGE can
suppress the inflammatory response in murine models, diabetic vascular
hyperpermeability, and diabetes-induced accelerated atherosclerosis in apoE
null mice
(50–52).
The proximity of cells expressing RAGE to lesional areas rich in AGEs suggest
that AGE-RAGE interaction can trigger key cell signal transduction pathways
and thereby lead to chronic cellular activation and dysfunction associated
with diabetes. AGE ligation of RAGE activates oxidant stress, p21 ras, and
downstream targets such as mitogen-activated protein kinases (MAPKs) and leads
to the activation of transcription factors such as NF-B
(37,
45,
48,
53).
Although AGE-RAGE interaction has been implicated in inflammatory
responses, the effects on COX-2 expression and regulation are not known. In
the present studies, we demonstrate for the first time that RAGE ligation by
exogenous AGEs or by S100b can lead to potent increases in COX-2 expression
and activity in THP-1 monocytic cells, as well as in primary human blood
monocytes. COX-2 induction at 4 h was regulated transcriptionally and involved
key NF-B elements on the COX-2 promoter. These results suggest that
simultaneous production of cytokines and COX-2 products by HG and AGEs in a
diabetic environment can trigger an amplifying inflammatory loop and lead to
accelerated vascular complications.
|
EXPERIMENTAL PROCEDURES |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
-32P]UTP (3000
Ci/mmol) was from PerkinElmer Life Sciences. RT-PCR reagents were from Applied
Biosystems (Foster City, CA) whereas the RPA III kit and Quantum RNA 18 S
internal standards were from Amnion Inc. (Austin, TX). Ficoll-Paque-plus was
from Amersham Biosciences. The luciferase assay system was obtained from
Promega, Inc. (Madison, WI). The prostaglandin E2 EIA kit and NS398
were from Cayman Chemical (Ann Arbor, MI). Anti-phospho-PKC (recognizes
phospho-PKC and -PKC11), antiphospho-ERK, anti-phospho-p38,
anti-p38, anti-IB, anti-phospho-IB, and anti-actin antibodies
were from Cell Signaling (Beverly, MA) whereas anti-COX-2 antibody was from
Santa Cruz Biotechnology, Inc. (Santa Cruz, CA).
2',7'-Bis-(2-carboxyethyl)-5-(and 6)-carboxyfluorescein
acetoxymethyl ester (BCECF/AM) fluorescent label was obtained from A. G.
Scientific, Inc. (San Diego, CA). Preparation of Methylglyoxal Modified Albumin (AGE)—Methylglyoxal-modified albumin (40) (AGE) was prepared as follows: bovine serum albumin (fraction V; fatty acid-free, low endotoxin) (10 mg/ml) was incubated with 100 µM methylglyoxal (Sigma) in PBS containing 0.1% sodium azide and 1 mM phenylmethylsulfonyl fluoride at 37 °C for 7 days. After incubation, the free methylglyoxal was removed by dialysis against PBS at 4 °C for 24 h. The methylglyoxal modified albumin was then vacuum-dried and dissolved in PBS at a concentration of 10 mg/ml. Aliquoted samples were stored at –70 °C.
Cell Culture and Treatments—Human THP-1 monocytic cells were
obtained from American Type Culture Collection and cultured as described
(38,
41) in RPMI 1640 medium
supplemented with 10% fetal calf serum (FCS), glutamine, HEPES,
streptomycin/penicillin (100 mg/ml/100 units/ml), 50 µM
-mercaptoethanol, and 5.5 mM D-glucose (normal glucose; NG)
in a 5% CO2 incubator at 37 °C. Cells were treated with or
without 5–10 µg/ml S100b. In some experiments, THP-1 cells were
pre-treated with actinomycin D (transcription inhibitor; 0.1 µg/ml),
cycloheximide (protein synthesis inhibitor; 0.5 µg/ml), Bay11-7082
(NF-B inhibitor; 10 µM), NS-398 (COX-2 inhibitor; 25
µM), AG-490 (Janus tyrosine kinase (JAK) inhibitor; 100
µM), SB202190 (p38 MAPK inhibitor; 1 µM), or GFX
(PKC inhibitor; 0.5 µM), PD-98059 (MEK pathway inhibitor; 25
µM), or NAC (antioxidant, 100 µM). They were then
incubated alone in control NG medium or with S100b for various time periods.
Porcine vascular smooth muscle cells (PVSMC) were cultured in Dulbecco's
modified Eagle's medium supplemented with 10% FCS, glutamine, HEPES, and
streptomycin/penicillin as described earlier
(54).
Isolation of Human Peripheral Blood Monocytes (PBMC)—50–60 ml of blood from adult volunteers with established Type 1 diabetes and also from normal healthy donors were collected in the presence of anticoagulant in accordance with an approved Institutional Review Board protocol. The blood was diluted with equal volumes of PBS. An equal volume of diluted blood was overlaid on Ficoll-Paque-plus in 1:1 ratio and centrifuged at 400 x g for 20–30 min at 18–20 °C. The leukocyte population was collected from the interface and washed with PBS several times to remove plasma and Ficoll. About 50 million washed cells in 10 ml of RPMI medium containing 10% FCS were plated in 100-mm culture dishes to allow monocytes to adhere on the surface of the dish for 2–3 h. The non-adherent cells (mainly lymphocyte population) were removed, washed with fresh medium, cultured in RPMI medium, and labeled as PBLC. Attached monocytes were washed twice with warm RPMI medium containing 10% FCS and allowed to remain in the dish overnight at 37 °C in 5% CO2. During this period the monocytes detach from the dish. They were collected and washed in fresh RPMI medium and labeled as PBMC. Then, about 1 x 105 cells from normal volunteers per well in 6-well plates were treated with S100b for 4 h, and total RNA was isolated as described below. Monocytes from Type 1 diabetic patients were directly processed for RNA extraction.
RNA Preparation and Relative RT-PCR—THP-1 cells (2 x 106/sample) in 4 ml of medium containing 5.5 mM (NG) with or without 6.5 µg/ml S100b were cultured in duplicate in 6-well dishes for various time intervals. Total RNA was isolated by the RNA-STAT method, and 1 µg was used for the RT reaction using a Gene Amp RNA PCR kit (41). cDNA corresponding to 0.05 µg of RNA was then used in multiplex PCR reactions containing gene-specific primers paired with Quantum RNA 18 S internal standards, and the multiplex PCR reactions were performed for 25–35 cycles in a GeneAmp9700 machine (Applied Biosystems, Inc., Foster City, CA).
The 5' and 3' primers for human COX-2 were
5'-ATCTACCCTCCTCAAGTCCC-3' and
5'-TACCAGAAGGGCAGGATACAG-3'. The 5' and 3' primers for
IL-1 were 5'-CTCTCTCACCTCTCCTACTCAC-3' and
5'-ATCTGCACGCCATCACAGTC-3'. The 5' and 3' primers for
COX-1 were 5'-CCGGATGCCAGTCAGGATGATG-3' and
5'-CTAGACAGCCAGATGCTGACAG-3'. PCR products were fractionated on
2.5% agarose gels and photographed using an AlphaImager 2000 Documentation and
Analysis system. DNA bands corresponding to amplified products and 18 S RNA
were quantitated with Quantity One software (Bio-Rad). Results are expressed
as -fold stimulation over NG after normalizing with paired 18 S RNA
levels.
RNase Protection Assay (RPA)—RPA was performed using RPA IIITM kit according to instructions of the manufacturer (Ambion Inc., Austin, TX) with some modifications. COX-2 antisense riboprobe was prepared by cloning RT-PCR amplified COX-2 fragment into pGEM-T vector (Promega) and verified by sequencing. Antisense template was amplified using SP6 promoter primer as the 5' primer 5'GCTATTTAGGTGACACTATAGAA3' and COX-2 primer 5'CACCAGGCAAATTGCTGGCAG3' as the 3' primer. COX-2 antisense riboprobe was generated using SP6 RNA polymerase, [32P]UTP, and PCR-amplified COX-2 DNA fragments as template. Antisense riboprobe of 18 S was generated using SP6 RNA polymerase and linearized plasmid pTRIS RNA 18 S (Ambion). Total RNA (150 µg) isolated from control and S100b-treated THP-1 was hybridized with 32P-labeled antisense riboprobes of human COX-2 (321 bp) and 18 S (165 bp). The hybridization product was then treated with RNase A and T to remove unhybridized probes. The protected and hybridized mRNAs were resolved in 6%-8 M urea polyacrylamide gel. The gel was dried and exposed on x-ray film at –80 °C. The bands were visualized by autoradiography.
Western Blot Analysis—Cells were washed twice with 5 ml of Hanks' salt solution and then lysed with 0.1 ml of lysis buffer. The cell lysates were assayed for protein by the Bio-Rad Dc protein assay kit (Bio-Rad). Proteins were separated by electrophoresis on SDS-PAGE gels with 10% acrylamide. Proteins were transferred to Immobilon P membranes by semi-dry transfer and then subjected to immunoblotting as described earlier (38) using relevant antibodies. Detection of immunoreactive bands was by chemiluminescence.
PGE2 Enzyme Immunoassay—THP-1 cells (5 x 105cells/ml) were incubated in 6-well tissue culture plates in RPMI 1640 medium with 0.2% BSA. Cells were treated with or without S100b for 8 h. The supernatant conditioned medium was then harvested and assayed for PGE2 levels using a specific EIA kit according to the manufacturer's instructions (Cayman Chemical, Ann Arbor, MI). Medium alone without cells was incubated under the same conditions and used as blank control for the EIA.
Plasmid Construction—Construction of the deletion mutants
containing specific regions of the human COX-2 gene promoter in the luciferase
reporter vector pGL3 Basic (Promega) was accomplished by PCR amplification.
Plasmid hCOX-2 (–1437/+127) was generated by deleting a 5673-bp DNA
fragment from upstream of the recombinant plasmid containing firefly
luciferase gene under the control of the 
7273-bp promoter region of the
human COX-2 gene (generous gift from Dr. Thomas McIntyre, University of Utah,
Salt Lake City, UT) by digesting with restriction enzyme KpnI. The
resulting plasmid was re-ligated using T4-DNA ligase, yielding the luciferase
gene under the control of –1437/+127 human COX-2 promoter region.
Deletion constructs containing various promoter regions of hCOX-2, namely from
–860 to +127, –360 to +127, –218 to +127, –123 to
+127, and from –52 to +127, were generated using the recombinant plasmid
containing an 7-kb promoter region of the human COX-2 gene as template.
The following primers were used: upstream primers, from –860,
5'GGTACCCACATTAACTATTTACAG3'; from –360,
5'GGTACCCCAAGGCGATCAGTCCAG3'; from –218,
5'GGTACCTACCCCCTCTGC TCCCAA3'; from –123,
5'GGTACCTTTTTTAAGGGGAGAGG3'; from –52,
5'GGTACCCATGGGCTTGGTTTTC3'; downstream primer +127,
5'AAGCTTCGGGCAGGGCGCGGCGC 3'. All upstream PCR primers
contained KpnI restriction sites (underlined), and the downstream
primer contained a HindIII recognition site (underlined), which
forced cloning of the fragments in the desired orientation into the pGL3 Basic
vector. Orientation and sequence of all constructs were verified by direct
sequencing using the ABI PRISM 377 DNA sequencer.
DNA Transfection and Luciferase Assays—THP-1 cells plated in
6-well plates (1.2 x 106 per well) were transfected with 1
µg of the indicated COX-2 promoter plasmids, the control pGL3-Luc plasmid
(Invitrogen), or the pCMV-mIB plasmid
(41,
55) (generous gift from Dr. E.
Zandi, University of Southern California) using LipofectAMINE 2000 in RPMI
medium with serum according to manufacturer's protocols. Following the
overnight recovery period, the transfected cells were cultured in medium
containing 5.5 mM (NG) with or without 6.5 µg/ml S100b. Cells
were then washed with PBS, lysed with 100 µl of lysis buffer, and stored
overnight at –70 °C. Samples were thawed, brought to room
temperature, and 20 µl of each lysate was used to analyze luciferase
activity by the luciferase assay system according to the manufacturer's
instructions.
THP-1 Cell Binding Assay—After 4 h of S100b treatment, THP-1 cells (106 cells/ml) were washed with Hanks' balanced salt solution twice and incubated with 10 µg/ml of BCECF/AM fluorescent tag in PBS for 30 min at 37 °C to label the cells and washed twice with serum-free medium. In some experiments, cells were pretreated for 1 h with various inhibitors as indicated before treatment with or without S100b. About 80% confluent PVSMC or human VSMC in 24-well plates were washed with serum-free medium, and 5 x 104 fluorescently labeled THP-1 cells were added to each well and incubated at 37 °C for 1 h. Non-specifically bound cells were removed by carefully washing several times with serum-free medium. Specifically bound cells were lysed with lysis buffer (100 mM Tris-Cl, pH 8, and 0.1% Triton X-100) at room temperature for 1 h. Fluorescence densities in the lysates were determined at 485 nm with a fluorescence multi-well plate reader, f-max (Molecular Devices), and quantitated using SOFT-max PRO-f software.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
mRNAs but Not COX-1 mRNA in
THP-1 Cells—Our recent study showed that high glucose (HG) could
induce the transcriptional regulation of chemokine and cytokine genes in
monocytes via activation of specific signaling pathways
(38,
41). Because AGEs accumulate
under diabetic conditions and contribute to the progression of diabetic
complications, we hypothesized that ligation of RAGE by AGEs or S100b can lead
to the expression of the inflammatory gene, COX-2. This could unravel
additional pro-inflammatory consequences of RAGE ligation. We also evaluated
whether AGE-induced COX-2 expression occurs directly or via increases in the
expression of other proteins or cytokine such as IL-1.
We initially carried out experiments to evaluate whether S100b could lead
to the induction of COX-2 and COX-1 mRNAs in THP-1 cells and the optimal
conditions required for the same. THP-1 cells were treated with or without
S100b protein for various time intervals from 0.5 to 24 h as shown in the
Fig. 1. RNA extracted from
these experiments was subjected to relative RT-PCR analyses. In this method,
specific primers for human COX-2, COX-1, and IL-1 genes were paired with
18 S rRNA primers as internal standards in the multiplex RT-PCR reactions.
Amplification of 18 S rRNA, in addition to the mRNA of interest, allows one to
normalize samples for differences in loading across several samples.
S100b-induced changes in gene expression were evaluated as -fold over control
NG samples after normalizing to 18 S rRNA internal control. Results showed
that S100b treatment led to a marked increase in COX-2 mRNA expression by 1 h,
peaking at 2 to 4 h and declining by 16 to 24 h
(Fig. 1A). In
contrast, COX-1 mRNA levels showed no delectable change during the 24-h time
period (Fig. 1B). This
suggests that S100b specifically up-regulates the inducible COX-2 isoform and
not COX-1 mRNA. (Fig. 1, A and
B).
|
We noted that S100b also increased IL-1 mRNA expression that was
evident by 30 min and peaking at 2 h (Fig.
1C). Because IL-1 is a potent inducer of COX-2, to
determine whether S100b-induced COX-2 mRNA at 4 h is via IL-1
expression, we evaluated the effects of pretreatment with an IL-1
neutralizing antibody or control IgG added to the medium. We noted that the
IL-1 antibody only slightly attenuated the effects of S100b at 4 h,
indicating that the effect of S100b, at least at 4 h, is not primarily
mediated via IL-1. Because cytokines such as IL-1 and IL-1
have been shown in many instances to mediate COX-2 mRNA stability rather than
transcription (24,
26), we hypothesized that
S100b could be inducing COX-2 mRNA via both transcription and mRNA
stabilization via de novo protein synthesis, the latter possibly via
a protein such as IL-1. We therefore examined this further at both 2-h
(time of peak appearance of IL-1) and 4-h time points.
Increased COX-2 Expression by S100b at 4 h, but Not 2 h, Is Because of Transcriptional Regulation—Time course analyses showed that S100b-induced expression of COX-2 mRNA is evident by 1 h, peaking at 2 to 4 h and then declining to control levels by 24 h (Fig. 1). To determine whether S100b-induced COX-2 mRNA expression at 2 and4his because of increases in transcription, THP-1 cells were pretreated for 1 h with actinomycin-D an inhibitor of transcription or with cycloheximide, a inhibitor of de novo protein synthesis, and then treated with S100b for 2 or 4 h. RT-PCR analyses in Fig. 2A showed that at the 2-h time point, actinomycin-D did not block the effects of S100b, but cycloheximide was clearly inhibitory. In contrast, at 4 h, actinomycin-D completely blocked S100b-induced COX-2 mRNA expression whereas cycloheximide had no inhibitory effect, in fact being slightly stimulatory. (Fig. 2B). These interesting results indicate that increased COX-2 expression by S100b at 4 h, but not 2 h, is because of increases in transcription. Furthermore, new protein synthesis is involved at the 2-h time point but not at 4 h. Interestingly, a similar kind of biphasic dual mode of COX-2 regulation involving transcription and de novo protein synthesis at different time periods has been reported recently (56) with another agonist. Our data were further confirmed by luciferase reporter assays as shown later.
|
It is possible that the increased COX-2 mRNA by S100b at 2 h may be because
of cytokines that are initially produced by S100b prior to COX-2 mRNA
induction, for example, by increased IL-1 expression. IL-1 is a
well known inducer of COX-2, and quite often by mRNA stabilization and
post-transcriptional mechanisms
(23–26).
Cycloheximide inhibited S100b-induced COX-2 mRNA expression only at2handnot4h
(Fig. 2). Furthermore the
IL-1 neutralizing antibody had only a slight inhibitory effect at 4 h
(Fig. 1D). These new
results indicate that the effects of S100b could be via both an
IL-1–dependent (early) and a unique (later) IL-1-independent
mechanism involving transcription.
S100b Induces Significant Increases in COX-2 mRNA and Protein Expression and Formation of Its Product PGE2 in THP-1 Cells—Because S100b effects on COX-2 expression were maximal at 4 h and also appeared to be transcriptionally regulated at this time point, we performed additional experiments at this time period. The bar graph in Fig. 3A shows data from multiple experiments normalized to 18 S internal control. This indicates that the effects of S100b at 4 h are statistically significant (p < 0.001). Fig. 3B shows that the actions of S100b were dose-dependent with maximum effects on COX-2 mRNA induction occurring at a concentration of 5 to 7.5 µg/ml.
|
Although S100b is a physiological ligand for RAGE, we wanted to determine whether AGEs prepared in vitro could also elicit similar effects on COX-2 expression as this would also be physiologically significant and relevant to diabetic complications. We therefore examined the effects of in vitro prepared methyl glyoxylated albumin (40). THP-1 cells were treated with this AGE (400 µg/ml) for 1 to 4 h. Fig. 3C shows that this methyl glyoxal AGE could also increase COX-2 expression that reached a maximum by 2 h and to a lesser extent at 4 h. This demonstrates that physiological AGEs can also induce COX-2 in THP-1 cells (Fig. 3C). As a negative control we treated the cells with unmodified BSA (400 µg/ml) for 4 h (Fig. 3D). This BSA treatment, unlike AGE, did not induce COX-2 expression thereby demonstrating that the AGE-induced COX-2 mRNA expression is specific to AGE-modified BSA.
Because both S100b and in vitro prepared AGE could induce COX-2 mRNA expression, we wanted to confirm the involvement of RAGE. THP-1 cells were pretreated with a specific anti-RAGE antibody (generous gift from Dr. David Stern, Medical College of Georgia) for 1 h prior to S100b treatment. RT-PCR analyses of RNA from these cells showed complete blockage of S100b-induced COX-2 mRNA levels in antibody-treated cells (lane 4 in Fig. 3E). This suggests that S100b-induced COX-2 mRNA is via RAGE activation.
To further confirm the RT-PCR data on S100b-induced COX-2 mRNA regulation, we further performed sensitive RPA using 32P-labeled COX-2 and 18 S antisense riboprobes. Fig. 4 shows a representative autoradiograph of the results. A clear protected COX-2 RNA (254-nt) band was seen in the S100b (6.5 µg/ml)-treated sample (Fig. 4, lane 3), whereas no protected band was seen in the untreated control (lane 2). 18 S was used as an internal control. The 18 S protected bands of the predicted sizes are seen in both the control and treated lanes. Lane 1 shows probe whereas lane 4 is experiment run without RNA. These results substantiate the RT-PCR data.
|
The RT-PCR and RPA data clearly showed significant induction of COX-2 mRNA in S100b-treated cells. We therefore evaluated whether the COX-2 protein levels were also regulated in S100b-treated THP-1 cells. Western blot analysis with a specific COX-2 antibody was carried out using total protein prepared from control and S100b-treated cells at various time points as shown in Fig. 5A. It is seen that COX-2 protein appeared by 4 h and peaked at 8 h after S100b treatment. This is consistent with the mRNA data, and it demonstrates that COX-2 protein appears shortly after mRNA induction by S100b, which peaked by 2–4 h. Equal loading of protein in each lane was confirmed by probing with an anti-actin antibody as internal control (lower panel of Fig. 5A).
|
To determine whether the induction of COX-2 mRNA and protein levels were
also associated with increase in COX-2 enzyme activity, we examined the levels
of the COX-2 product PGE2 in S100b-treated versus control
THP-1 cells. PGE2 released into the medium was measured in culture
supernatants by a specific EIA. Fig.
5B shows that S100b (6.5 µg/ml) treatment for 8 h led
to a significant increase in PGE2 levels (340 ± 48 fg/ml,
p < 0.001). In contrast, there were no detectable levels of
PGE2 (Fig.
5B, U.D., undetectable) in supernates of
untreated control cells. Pretreatment of the cells with the specific COX-2
inhibitor (NS-398) led to complete abrogation of S100b-induced PGE2
production confirming the source of PGE2.Itwas also blocked by an
NF-B inhibitor (Bay11-7082), suggesting that COX-2 regulation by S100b
is mediated to a large extent by NF-B activation. Overall, these
results demonstrate that S100b and AGE can induce the expression and activity
of COX-2, and this could be relevant to monocyte activation and dysfunction
associated with atherosclerosis and other diabetic vascular and inflammatory
complications.
Signal Transduction Mechanisms Involved in S100b-induced COX-2 mRNA
Expression—To determine the key signal transduction pathways
involved in S100b-induced COX-2 mRNA in THP-1 cells, we evaluated the effects
of inhibitors of pathways known to be activated by AGEs. We pretreated THP-1
cells with Bay11-7082 (NF-B inhibitor), SB202190 (p38MAPK inhibitor),
PD-98059 (MEK/ERK MAPK inhibitor), AG-490 (JAK inhibitor), GFX (PKC
inhibitor), and NAC (antioxidant). Results in
Fig. 6, A and
B shows that S100b-induced COX-2 mRNA expression was
significantly blocked by PD-98059 (PD), NAC, and Bay11-7082
(Bay) as well as by GFX and SB202190 (SB). However, AG-490,
the JAK inhibitor, had no effect at all on S100b-induced COX-2 mRNA expression
(Fig. 6B). These
results implicate the involvement of multiple pathways including the MAPK
pathway, oxidant stress, PKC, and the transcription factor NF-B in
S100b-induced COX-2 mRNA expression.
|
We next determined whether these pathways were directly activated by S100b.
Thus Fig. 7 shows that S100b
increases the activities of p38MAPK, ERK1/2 MAPK, and PKC as indicated by
increased levels of the phosphorylated forms of these (pPKC, pERK, pp38).
Total p38MAPK (p38) was not altered under these conditions. S100b also led to
IB- phosphorylation and degradation (indicators of NF-B
activation) as shown later in Fig.
10.
|
|
S100b Treatment Activates Transcription from the COX-2 Promoter—To examine whether S100b at 4 h can induce transcription from the COX-2 promoter, we transfected THP-1 cells with promoter reporter construct plasmids, pCOX2(–7146/+127) and pCOX-2(–1430/+127), that expresses the firefly luciferase (Luc) gene under the control of human hCOX-2 gene promoter. Luciferase analyses of 4-h S100b-stimulated THP-1 cells transfected with these plasmids showed significantly elevated luciferase activity in pCOX2(–1430/+127) transfected samples (Fig. 8A) (p < 0.001). However, the longer promoter construct phCOX2 (–7146/+127) did not yield any luciferase activity over that seen by the control pGL3-Luc construct (Fig. 8A). These results suggest that COX-2 up-regulation at 4 h is mediated by key promoter elements present within the 1430 region, but it appears that certain repressive elements could be present more upstream resulting in a silencing effect by the longer –7146 construct of the 5' end of the COX-2 promoter.
|
The human COX-2 promoter region (–1430/+127) contains several
cis-acting elements. We were interested in five cis-acting elements near the
5' end of the coding region of COX-2, namely two NF-B binding
sites, two NF-IL-6 binding sites, and a cAMP-response element binding site
(Fig. 9A), all of
which have been shown to be involved in the regulation of COX-2 gene
transcription by various agonists. Our results in
Fig. 5B and
Fig. 6 showed clear inhibition
of S100b-induced COX-2 activity and mRNA expression by the pharmacological
NF-B inhibitor Bay11-7082, suggesting the involvement of NF-B
transcription factor. To confirm this result at the molecular level and to
further evaluate the key promoter elements involved in S100b transcriptional
activation, we co-transfected the cells with the phCOX2 (–1430/+127) and
an NF-B super-suppressor IB (mutant) plasmid that suppresses
NF-B activation. Fig.
8B shows that co-transfection of this mutant IB
plasmid significantly inhibited S100b-induced COX-2 promoter activity. The
mutant IB plasmid had no significant effect on the control pGL3-Luc
plasmid.
|
Our earlier data indicated that at the 2-h time point, S100b effects on COX-2 mRNA were not transcriptionally regulated. We tested this further with promoter-luciferase assays. Fig. 8C shows that when cells were treated for 2 h with S100b, unlike at 4 h, there was no increase in luciferase activity with the pCOX-2 (–1430/+127) construct. This provides additional support that at 2 h, COX-2 expression is not transcriptionally regulated.
Further Analysis of the Role of Specific NF-B Elements
in the COX-2 Promoter—The COX-2 promoter contains two NF-B
binding elements flanked by two NF-IL-6 elements
(Fig. 9A). Data in
Fig. 8 and earlier figures
demonstrated the involvement of NF-B activation in COX-2 expression but
did not provide information on which cis-acting NF-B elements are
involved in this activation. So in the next step we examined the relative
contribution of the two NF-B sites on the COX-2 promoter. We
constructed several COX-2 promoter luciferase gene fusion constructs
containing luciferase gene under the control of various regions of the hCOX-2
promoter. This was performed by PCR amplification as described under
"Experimental Procedures." As shown in
Fig. 9, A and
B, COX-2 promoter deletion constructs such as
pCOX2(–860/+127), pCOX2(–360/+127), pCOX2(–216/+127),
pCOX2(–123/+127), and pCOX2(–52/+127) were transfected into THP-1
cells. After transfection, the cells were stimulated with S100b, and
luciferase activities were measured. Fig.
9B shows that transfection of THP-1 cells with the
pCOX2(–860/+127) construct, which contains all five cis-acting elements,
showed over a 2-fold increase in luciferase activity over control cells
transfected similarly. In this case, basal activity was also increased
markedly. As noted earlier in Fig.
8, the pCOX2(–1437/+127) construct also showed a significant
increase in activity over control. On the other hand, removal of NF-IL-6
(–530/–543) and the NF-B (–439/–448) (namely
the pCOX2(–360/+127) construct), led to abrogation of the luciferase
activity. Furthermore, co-transfection of the IB mutant plasmid with
pCOX2(–860/+127) blocked luciferase activity induced by this plasmid
(Fig. 9B). These data
confirm the key role of the distal NF-B binding site
(–439/–448) in the COX-2 regulation by S100b.
S100b Induces IB- Phosphorylation and
Degradation in THP-1 Cells—The results in
Fig. 5B,
Fig. 6,
Fig. 8B, and
Fig. 9 all clearly indicate the
involvement of NF-B in S100b-induced COX-2 mRNA expression and its
product PGE2 synthesis. To further confirm that S100b can lead to
the activation of NF-B, we examined whether IB-
phosphorylation and degradation in the cytoplasm is induced by S100b.
Cytoplasmic fractions from S100b treated THP-1 cells were prepared as
described earlier (38) and
analyzed by immunoblotting with a phosphospecific IB- antibody
and an anti-IB- antibody.
Fig. 10 (upper panel)
showed that S100 treatment steadily decreased the levels of the
IB- protein with increasing time. This decrease reached a
maximum at 60 min post-S100b treatment induction after which the levels began
rising again by 2 and 4 h most likely because of new synthesis of the
IB- protein. At the same time,
Fig. 10 (middle
panel) shows that S100b treatment also increased levels of phosphorylated
IB- protein as early as 15 min and remained high up to 240 min.
Equal protein loading was confirmed by stripping the blot and reprobing with
an anti-actin antibody (Fig.
10, lower panel). These data confirm that S100b can
directly lead to IB- phosphorylation and degradation and hence
NF-B activation in THP-1 cells.
S100b Induces the Binding of THP-1 Monocytes to VSMC—To
determine the functional significance of S100b-induced COX-2 expression and
increased PGE2 levels, we evaluated a potential role in monocyte
adherence to PVSMC. Monocyte adhesion to VSMC may play a key role in monocyte
retention in vessel wall, a key step in the pathogenesis of atherosclerosis
that has not been well studied. Results in
Fig. 11A show that
S100b-treated THP-1 cells displayed a 2- to 3-fold increase in adherence to
PVSMC relative to the untreated control cells. Pretreatment with NS-398, a
COX-2-specific inhibitor abolished the S100b-induced adherence, suggesting the
involvement of COX-2 in S100b-induced monocyte activation and adhesion. To
further evaluate the signal transduction pathways involved in S100b-induced
monocyte binding, we performed the binding assays with cells that had been
pretreated with various inhibitors, NS-398 (COX-2 inhibitor), Bay11-7082
(NF-B inhibitor), AG-490 (JAK inhibitor), PD-98059 (MEK inhibitor),
SB202190 (p38 MAPK inhibitor), or GFX (PKC inhibitor).
Fig. 11B shows that
inhibitors of COX-2, NF-B, MEK, p38 MAPK, and PKC significantly
inhibited binding activity induced by S100b, whereas AG-490 had no effect.
These results are similar to their effects on COX-2 mRNA induction and suggest
the involvement of NF-B, MAPK, and PKC signaling pathways in
S100b-induced monocyte binding, as well as COX-2 induction.
|
S100b Induces COX-2 mRNA Expression in PBMC—Because THP-1 cells may not fully represent the phenotype of blood monocytes, we next examined whether S100b can also induce COX-2 mRNA expression in PBMC isolated from normal healthy adult donors. These PBMC were treated with S100b for 4 h, total RNA was isolated, and COX-2 mRNA levels were analyzed by RT-PCR as described in Fig. 1. Fig. 12A shows that S100b could clearly induce COX-2 mRNA expression in the PBMC. However, PBLC obtained from the same donors and treated similarly did not respond to S100b in this manner. COX-2 induction in THP-1 cells is seen in the far right. These results clearly suggest that induction of COX-2 by S100b is evident even in primary blood monocytes and is specific to monocytes, because no induction was observed in lymphocytes.
|
COX-2 mRNA Expression Is Elevated in Peripheral Blood Monocytes Isolated from Diabetic Patients Relative to Normal Healthy Non-diabetic Controls—We next obtained PBMC from Type 1 diabetic patients to determine whether they had increased levels of COX-2 mRNA. Fig. 12B shows that PMBC isolated from two Type 1 diabetic patients (run in duplicate) had markedly elevated expression of COX-2 mRNA unlike the normal controls who did not have any COX-2 mRNA expression. These important results further establish the significance and pathological relevance of our in vitro data.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONEXPERIMENTAL PROCEDURESRESULTSDISCUSSIONREFERENCES |
|---|
and partly through a novel
IL-1–independent mechanism based on transcription. This kind of
biphasic dual regulation of COX-2 based on transcription and de novo
protein synthesis was observed recently
(56) with another agonist.
Importantly, we noted that S100b also increased COX-2 mRNA expression in
normal primary human blood monocytes from non-diabetic volunteers but not in
lymphocytes from the same individuals. Furthermore, we noted high levels of
COX-2 mRNA expression in monocytes isolated from Type 1 diabetic patients,
whereas monocytes from non-diabetic subjects do not have basal expression of
COX-2 mRNA. These data support the physiological and pathological relevance of
our findings.
S100b-induced COX-2 mRNA expression at 4 h was transcriptionally regulated,
because actinomycin D blocked expression, and furthermore, S100b at this time
period could increase transcription from a minimal human COX-2 promoter
luciferase construct. Promoter deletion analyses, as well as pharmacologic and
genetic inhibitors of NF-B, demonstrated the key role of NF-B in
S100b-induced COX-2 gene transcription. Furthermore, the specific
S100b-responsive NF-B site in the hCOX-2 promoter was evaluated using
promoter deletion analyses. The promoter region of the human COX-2 gene has
two NF-B consensus sites
(18), one located within
–455 to –428 bases, and the other located within –232 to
–205 bases from the transcriptional start site
(Fig. 9A). In the
present study, we were able to determine that the site located further away
from the start site (at –439/–448) is essential for S100b-induced
COX-2 transcription.
S100b was also able to phosphorylate and degrade IB-,
indicating NF-B activation. S100b also increased activities of PKC,
p38, and ERK1/2 MAPKs. Furthermore, using signal transduction pathway-specific
inhibitors, we were able to determine that activation of NF-B, ERK1/2,
p38 MAPK, and PKC are all involved in S100b-induced COX-2 mRNA induction, thus
indicating the operation of multiple pathways in COX-2 regulation in monocytes
under diabetic conditions. There are reports that AGEs can activate the
JAK-STAT pathway (59).
However, because the JAK inhibitor, AG490, did block COX-2 mRNA induction, it
appears that the JAK-STAT pathway does not mediate S100b effects on COX-2 mRNA
expression.
COX-2 products have been associated with several cellular properties including vasodilation, vasoconstriction, cellular adhesion, and migration (3–5). Products such as PGI2 are protective and vasodilatory, whereas others including PGE2 are inflammatory and are related to the proinflammatory properties of COX-2 (3–5). Although PGI2 is a major COX-2 product in endothelial cells, PGE2 is the major product in monocytes. Hence, our observation of increased PGE2 production under diabetic conditions in monocytes demonstrates that diabetes induces a proinflammatory environment. The overall beneficial effects of specific COX-2 inhibitors in the treatment of various inflammatory diseases without the gastrointestinal side effects of aspirin indicate the pathogenic role of COX-2 (3–5, 17, 31). However, some reports have indicated potential drawbacks of long term use of COX-2 inhibitors (60). Based on our data, it is possible that COX-2 inhibitors could be beneficial for diabetic vascular complications. This is further supported by the new report showing that high glucose could increase COX-2 expression in endothelial cells and thromboxane production and decreased PGI2 release (34).
To determine the functional relevance of COX-2 induction by S100b in
monocytes, we examined whether COX-2 activation plays a role in the adhesion
of THP-1 cells to PVSMC. This assay models monocyte retention in the vascular
wall, a key step in the pathogenesis of atherosclerosis. Although monocyte
adhesion to endothelial cells is a well documented and studied key early step
in atherogenesis, we are interested in evaluating the less studied mechanisms
of monocyte retention in the vascular wall. We observed that S100b treatment
of THP-1 cells could significantly increase their adhesion to a monolayer of
VSMC. This increased adhesion was significantly inhibited by the COX-2
inhibitor, NS-398, as well as inhibitors of NF-B, PKC, p38, and ERK1/2
MAPK pathways. Because these inhibitors also blocked S100b-induced COX-2 mRNA
expression, our data suggest that the activation of these signaling pathways
by S100b (or AGEs) can lead to monocyte adhesion, retention, and foam cell
formation. Overall these new results indicate that AGEs acting via RAGE can
lead to COX-2 expression and monocyte activation and thereby accelerate the
progression of atherosclerosis.
Several reports have shown that COX-2 is a highly inducible gene. It is
particularly responsive to growth factors and mediators of inflammation such
as TNF-, IL-1, IL-6, and lipopolysaccharide
(1–9).
Regulation by growth factors such as platelet-derived growth factor suggest
that COX-2, like the immediate early genes, c-fos and c-myc,
may also play some general role in mitogenesis. Thus, although COX-2 is
induced rapidly as also seen in the present study, it can have sustained
cellular effects. COX-2 products, once formed, can further propagate cellular
effects via their own receptors.
Recent studies have shown that HG culture of monocytes can lead to the
transcriptional regulation of the inflammatory cytokine, TNF-, and the
chemokine monocyte chemoattractant protein-1 via oxidant stress-dependent
mechanisms, as well as key signaling pathways
(38,
41). Furthermore, NF-B
activation by HG seemed to play a key role in the transcription of the
TNF- and monocyte chemoattractant protein-1 genes in monocytes
(38,
41). HG culture of THP-1
monocytes also significantly increased their adherence to human aortic
endothelial cells (41). Very
recently, gene profiling with DNA arrays demonstrated that key cytokine,
chemokine, and related genes are regulated by HG in THP-1 monocytes
(41). Most of the genes
induced by HG were regulated at least in part by NF-B, thereby
indicating a central role played by this transcription factor in leading to
diabetic vascular and other complications. The present data showing the
importance of NF-B in COX-2 regulation by AGEs, and RAGE ligation
further underscores this and also adds to the growing list of inflammatory
genes regulated under diabetic conditions. Furthermore, it indicates the
operation of a vicious loop in diabetes in which various cells within the
vessel wall and in circulation can respond to HG and AGEs by producing
inflammatory mediators that operate in autocrine and paracrine fashions to
amplify the overall inflammatory response. Thus, although the role of COX-2
and its products in leading to inhibition of insulin secretion and islet
destruction is known, this report describes for the first time the regulation
of COX-2 by diabetic conditions in monocytes that could have far reaching
consequences.
|
FOOTNOTES |
|---|
To whom correspondence should be addressed: Dept. of Diabetes, Beckman
Research Inst. of the City of Hope, 1500 E. Duarte Rd., Duarte, CA 91010.
Tel.: 626-359-8111 (ext. 62289); Fax: 626-301-8136; E-mail:
rnatarajan{at}coh.org
