Molecular Human Reproduction vol.6 no.9 pp. 771–778, 2000
NF-IL6 and CRE elements principally account for both basal and
interleukin-1β-induced transcriptional activity of the proximal
528 bp of the PGHS-2 promoter in amnion-derived AV3 cells:
evidence for involvement of C/EBPβ
S.Potter1, M.D.Mitchell1, W.R.Hansen2 and K.W.Marvin1,3
1Department
of Pharmacology and Clinical Pharmacology, University of Auckland, Faculty of Medical and Health
Sciences, Auckland, New Zealand, and 2Department of Pathology, Yale University, School of Medicine, New Haven,
CT 06520–8023, USA
3To
whom correspondence should be addressed at: Department of Pharmacology and Clinical Pharmacology,
University of Auckland, Faculty of Medical and Health Sciences, 85 Park Road, Grafton, Auckland, New Zealand.
E-mail: [email protected]
Prostaglandin H synthase (PGHS)-2 promoter fragments (–528 to ⍣9 bp and 5⬘ unidirectional deletions
thereof) were cloned upstream of the chloramphenicol acetyl-transferase (CAT) reporter gene. These were
transfected into amnion-derived AV3 cells. The region, –528 to –203, which includes NF-κB sites, had little
influence on CAT expression. The region, –203 and –52, however, was responsible for most of the basal
promoter activity and also conferred responsiveness to interleukin (IL)-1β (>3-times basal). Point mutations
of NF-IL6 and cAMP response element (CRE) in this region reduced both basal and IL-1β-stimulated production
of CAT; dual mutation eliminated IL-1β responsiveness. Factors in nuclear extracts from control or IL-1βstimulated AV3 cells specifically complexed the NF-IL6 and CRE sequences. However, the NF-IL6 and CRE
oligonucleotides cross-competed, suggesting a common factor. C/EBPβ was identified by supershift assay as
interacting with both sequences. To a lesser extent C/EBPα and δ also interacted with the NF-IL6 site.
However, CRE binding protein (CREB), was absent from the complex with the CRE. In conclusion, NF-IL6 and
CRE elements principally account in AV3 amnion cells for basal and IL-1β-inducible transcriptional activity of
the proximal 528 bp of the PGHS-2 promoter, while NF-κB elements play no substantial role. C/EBPs,
particularly C/EBPβ, are implicated in control of PGHS-2 transcription through the NF-IL6 and CRE sites.
Key words: amnion/cytokines/interleukin 1β/prostaglandins/PGHS-2
Introduction
A significant proportion of preterm labour has been associated
with intrauterine infection (Wahbeh et al., 1984; Romero et al.,
1989a). In the presence of infection, values of pro-inflammatory
cytokines, e.g. interleukin (IL)-1β and tumour necrosis factorα (TNF-α) are elevated in the amniotic fluid (Romero et al.,
1989b, 1992). These cytokines, produced locally in gestational
tissues in response to bacterial infection, are capable of
activating fetal membrane and decidual prostaglandin (PG)
production (Mitchell et al., 1991), in part, by increasing
prostaglandin H synthase-2 (PGHS-2) mRNA expression and
activity (Teixeira et al., 1994; Hansen et al., 1998). Furthermore, IL-1β is capable of inducing the expression of other
cytokines, e.g. IL-8 (Dudley et al., 1993), TNF-α (Ikejima
et al., 1990) and IL-6 (Shalaby et al., 1989), or enhancing its
own production (Dinarello, 1987; Gunn et al., 1996). These
observations suggest that IL-1β production of gestational
tissues during bacterial invasion in pregnancy may play an
important role in the events leading to preterm labour.
PGE2 is thought to be a key mediator of the onset of labour
(Keirse and Gravenhorst, 1979; Dudley and Trautman, 1994).
Two isoforms of the PGHS enzyme which catalyses the
committed step in the prostaglandin biosynthetic pathway have
been described (Funk et al., 1991; Kujubu et al., 1991). The
two isoforms, referred to as PGHS-1 and PGHS-2, are encoded
© European Society of Human Reproduction and Embryology
by genes on different chromosomes and share a 60% amino
acid identity. PGHS-1 is constitutively expressed in most cell
types, whereas PGHS-2 is rapidly and transiently expressed in
response to pro-inflammatory cytokines, growth factors and
tumour promoters (Smith and DeWitt, 1995).
The transient nature of PGHS-2 mRNA expression in
response to mitogens shares similarities with the expression
patterns of immediate early genes (Herschman, 1991). Despite
some knowledge concerning its structural and sequence
information, it is unclear how expression of the PGHS-2 gene
is regulated by external stimuli in terms of signalling pathways.
Signals that ultimately trigger the accumulation of PGHS-2
mRNA have been shown to be mediated through various signal
transduction pathways, depending on the cell type and stimulus
used (Blanco et al., 1995; Hamasaki and Eling, 1995). Sequence
analysis of the 5⬘-flanking region of the PGHS-2 gene has
shown the presence of several potential transcription regulatory
sequences, including a TATA box, an NF-IL6 site, a cAMP
response element (CRE) motif and a CCAAT/enhancer binding
protein (C/EBP) motif (Appleby et al., 1994; Kosaka et al.,
1994). Additional cis-acting DNA elements, e.g. NF-κB
elements, which are found in the rat (Sirois and Richards,
1993), mouse and human PGHS-2 promoters (Fletcher et al.,
1992), may also facilitate transcription from other effector
pathways that regulate PGHS-2 transcription. In the present
771
S.Potter et al.
report we have sought to determine the promoter elements in
the proximal 528 bp of the PGHS-2 promoter which play a
role in basal and IL-1β-mediated PGHS-2 gene transcription
in amnion-derived AV3 cells.
Materials and methods
Reagents
Hams F12/Dulbecco’s modified Eagle’s medium (DMEM) was
purchased from Irvine Scientific, CA, USA. Trypsin 1:250 powder
was obtained from Difco (Detroit, MI, USA). Fetal calf serum (FCS),
penicillin G sodium, streptomycin sulphate, Concert ion-exchange
DNA purification columns, gel shift primers, poly dIdC and the
0.24–0.95 Kb RNA ladder were obtained from Life Technologies,
Auckland, New Zealand. Human recombinant IL-1β was obtained
from Immunex, Seattle, Washington, USA. Restriction enzymes,
Fugene 6 transfection reagent, CAT and β-galactosidase (β-gal)
enzyme-linked immunosorbent assay (ELISA) kits were purchased
from Boehringer Mannheim New Zealand Ltd (Auckland, New
Zealand). GeneScreen Plus membrane was obtained from Life Technologies. [α-32P]-dCTP, [γ-32P]-ATP and HyperfilmMP were obtained
from Amersham Pharmacia (Auckland, New Zealand). Polynucleotide
kinase (PNK), PNK buffer and T4 DNA ligase were purchased from
Epicentre Technologies (Madison, WI, USA). Anti-C/EBPα, C/EBPβ
and C/EBPδ antibodies were obtained from Santa Cruz Biotechnology
Inc (Santa Cruz, CA, USA). Anti-CREB antibody was obtained from
Upstate Biotechnology (Lake Placid, NY, USA). The random prime
labelling kit was purchased from Amersham Pharmacia. The DNA
purification columns were purchased from Qiagen (Hilden, Germany)
and the QuikChange Site-Directed Mutagenesis Kit from Stratagene
(La Jolla, CA, USA). Other biochemicals were purchased from Sigma
Chemical Co (St Louis, MO, USA) or Riedel de Haen (Seelze,
Germany).
Cell culture
Amnion-derived AV3 cells (Rivadeneira and Robbins, 1958) were
maintained in Ham’s-F12/DMEM plus 10% heat-inactivated FCS,
6 g/l penicillin, 10 g/l streptomycin and 40 g/l gentamycin, at 37°C
in a humidified 95% air/5% CO2 atmosphere. The medium was
changed every 2 days.
Isolation of total RNA
Cells were treated in triplicate, confluent, 6 cm plates with vehicle
or IL-1β (1 ng/ml). At the times indicated, medium was removed
and total RNA was directly isolated by acid guanidinium thiocyanate–
phenol–chloroform extraction (Chomczynski and Sacchi, 1987). RNA
pellets were washed with 80% ice-cold ethanol, resuspended in
deionized formamide at 55°C for 15 min and stored at –70°C. RNA
concentration was determined by measuring the absorbance at 260 nm.
RNA gel electrophoresis and Northern Blot analysis
Total RNA (15 µg per sample) was fractionated on a 1% agarose
gel and transferred to GeneScreen Plus membrane overnight by
downward capillary transfer, as described previously (Hansen et al.,
1998). The cDNA probe, a 1.8 kb EcoRI/ApaI fragment of the PGHS2 cDNA (Hla and Neilson, 1992), was labelled with [α-32P]-dCTP
(100 mCi/ml) using the random prime labelling kit to a specific
activity ⬎109 cpm/µg. Hybridization and washing were carried
out essentially as described previously (Hansen et al., 1998). The
sodium dodecyl sulphate (SDS) concentration was reduced to 0.2%
in the first two wash solutions and 2% in the final. The final wash
was 50 mmol/l NaH2PO4, pH 7.2, 1 mmol/l EDTA, 2% SDS for a
772
Table I. Oligonucleotides for preparation of chloramphenicol acetyltransferase (CAT) deletion constructs
Construct
Sequence
p528CAT
p321CAT
p203CAT
p52CAT
5⬘-TAC TCG CCC CAG TCT GTC-3⬘
5⬘-CGG GCA AAG ACT GCG AAG-3⬘
5⬘-CCA AAT TGG GGC AGC TTG-3⬘
5⬘-CAT GGG CTT GGT TTT CAG-3⬘
total of 30 min at 53°C. The hybridization signal was quantified
using a Packard InstantImager (Parkland Instruments Co., Mariden,
CT, USA).
Cell transfections
Preparation of CAT reporter constructs
To produce deletion constructs by long template polymerase chain
reaction (PCR), a –891/⫹9 fragment of the PGHS-2 promoter
contained within the plasmid hcox-2luc (Tazawa et al., 1994) was
subcloned into pBLCAT3 (Luckow and Schutz, 1987, Genbank
accession no. X64409). hcox-2luc and pBLCAT3 were linearized
with HindIII and XhoI respectively. The linearized plasmids were
end-filled with Kenow fragment and then digested with BamHI.
Finally, the BamHI/HindIII fragment of hcox-2luc was subcloned
into the BamHI/XhoI-digested pBLCAT3 to produce p891CAT.
Unidirectional deletions from the 5⬘ end of the promoter fragment
were accomplished by PCR. The upstream primers (Table I) defined
the 5⬘ ends of p321CAT, p203CAT and p52CAT as noted. A common
downstream primer (5⬘-GGA TCC TCT AGA TCG AC-3⬘) which
annealed only to pBLCAT3 vector sequence at the upstream
pBLCAT3/PGHS-2 promoter sequence junction, was utilized in
subsequent PCR reactions to amplify both vector and the retained
portion of the promoter. The resulting PCR products were purified
by low-melting point agarose electrophoresis, phosphorylated by
polynucleotide kinase, and re-annealed with T4 DNA ligase.
PGHS-2 promoter constructs containing mutations within the NFIL6 and/or CRE elements were generated by the QuikChange SiteDirected Mutagenesis Kit according to the manufacturers instructions.
Primers for mutations of the CRE and NF-IL6 elements are shown
in Table II.
All constructs and plasmid junctions were verified by restriction
digest and automated dye terminator Sanger dideoxy sequencing
(Auckland DNA Sequencing Facility, School of Biological Sciences,
University of Auckland). DNA purification columns were used to
purify all plasmids for transfection.
Transfection of deletion constructs into AV3 cells
AV3 cells were plated in 6 cm dishes (106 cells/dish) the day
before transfection. On the day of transfection the medium was
exchanged at least 1 h before transfection. A transfection reagent
mix was prepared by diluting Fugene 6 with serum-free medium
(3 µl/100 µl). For each plate to be transfected, 1 µg of one of the
above described CAT reporter constructs, 0.75 µg of pGEM7Zf(–) and
0.25 µg pdx11 (CMV-β-galactosidase) were added to an appropriately
labelled tube. Transfection reagent mix (100 µl per plate) was added
dropwise to each tube containing the CAT constructs and tubes were
incubated for 15 min at room temperature. Each construct (100 µl)
was then added dropwise to appropriately labelled plates and incubated
for 24 h at 37°C. Cells were then treated with IL-1β (1 ng/ml) or
vehicle for 24 h, lysed and amounts of CAT and β-gal enzymes were
measured by ELISA. The results are expressed as a ratio of picogram
CAT/pg β-gal and are indicative of the transcriptional activity of the
indicated regions in the human PGHS-2 promoter in AV3 cells.
PGHS-2 promoter regulation by IL-1β in amnion-derived cells
Table II. Primers for the mutations of cAMP response element (CRE) and NF-IL6 elements
Primer
Sequencea
CRE
5⬘-GGC GGA AAG AAA CAG TCA TTT GAG CTC ATG GGC TTG GTT TTC AGT A-3⬘
3⬘-C CGC CTT TCT TTG TCA GTA AAC TCG AGT ACC CGA ACC AAA AGT CAG-5⬘
5⬘-CCC TGC CCC CAC CGG GCT ACT CAC TTT TTT TAA GGG GAG AGG-3⬘
3⬘-GGG ACG GGG GTG GCC GGA TGA GTG AAA AAA ATT CCC CTC TCC-5⬘
NF-IL6
aUnderlined
regions indicate location of the specified response elements in the wild-type sequence; mutated
bases are in bold. The mutation of the CRE introduces a SacI restriction. All construct sequences were
verified by restriction analysis and automated dye terminator Sanger dideoxy sequencing.
To evaluate responsiveness to IL-1β mediated by cryptic elements
in the vector backbone, the construct pGLCAT4 (Nyborg et al., 1990),
which contains the minimal promoter (–32 to ⫹51) from the herpes
symplex thymidine kinase (TK) gene, was used in place of the
PGHS-2 promoter-CAT constructs. For estimation of the transfection
efficiency, pGFP-C1 (Clontech Laboratories Inc, Palo Alto, CA, USA)
was substituted for the CAT construct and pGEM7Zf(–). Green
fluorescent protein expressing cells resulting from transfection with
this construct were counted using a Leica DM IRB microscope and
fluorescein filter set (Leica Microsystems Wetzlar GmbH, Wetzlar,
Germany) and compared with the number of cells detected by white
light microscopy. An efficiency of 7% was obtained by this method.
Electrophoretic mobility shift assay (EMSA)
Nuclear protein isolation
Confluent cultures in 10 cm plates were treated for 1 h with 1 ng/ml
IL-1β or carrier at 37°C. After incubation the monolayers were
washed twice with ice-cold phosphate-buffered saline (PBS), pH 7.2.
The cells were removed by scraping and recovered by centrifugation in
pre-chilled tubes at 4°C for 5 min at 650 g. Supernatants were
discarded and the cells were resuspended in 5 ml ice-cold lysis buffer
(10 mmol/l Tris–Cl pH 7.4, 10 mmol/l NaCl, 3 MgCl2, 0.5% NP40).
After 5 min on ice the nuclei were recovered by centrifugation at
4°C for 5 min at 650 g. The resuspension of the nuclei, incubation
on ice, and centrifugation were repeated. The volume of the nuclear
pellet was estimated. The pellets were resuspended gently in half of
the packed nuclear volume (pnv) of low salt buffer [20 mmol/l
HEPES, pH 7.9 at 4°C, 25% glycerol, 1.5 mmol/l MgCl2, 0.2 mmol/
l EDTA, 20 mmol/l KCl, 0.2 mmol/l phenyl methyl sulphonyl fluoride
(PMSF), 0.5 mmol/l dithiothreitol (DTT)] and transferred to microfuge
tubes on ice. High salt buffer (20 mmol/l HEPES, pH 7.9 at 4°C,
25% glycerol, 1.5 mmol/l MgCl2, 0.2 mmol/l EDTA, 1600 mmol/l
KCl, 0.2 mmol/l PMSF, 0.5 mmol/l DTT), half pnv, was added
dropwise to the pellet. Following incubation on ice for 30 min,
insolubles were removed from the nuclear extracts by centrifugation
at 23 100 g for 30 min at 4°C. The concentration of nuclear protein
in the extracts was determined by the bicinchoninic acid (BCA)
method calibrated against bovine serum albumin (Redinbaugh and
Turley, 1986).
Double-stranded oligonucleotide probe production
The sense or anti-sense oligonucleotide (5 pmol in each case)
was added to separate tubes containing 10 IU polynucleotide kinase
(PNK) and 15 µCi of [γ32P]-ATP and incubated for 30 min at room
temperature. The reaction was extracted with phenol/chloroform and
the complementary nucleotides were combined and purified using
Sephadex G-50. Reactions were boiled for 5 min, then placed in a
water bath pre-heated to 80°C which was then allowed to cool
overnight.
Formation of complexes and electrophoresis
The EMSA reaction containing 5 µg of nuclear extract, 2 µl 10⫻
reaction buffer (38 mmol/l HEPES pH 7.9, 1.85 mmol/l MgCl2,
0.18 mmol/l EDTA, 27.5% glycerol, 0.95 mmol/l DTT) and 2 µl of
poly dIdC was incubated for 20 min at room temperature. To each
tube 50 fmol of the appropriate double-stranded oligonucleotide probe
(NF-IL6: 5⬘-CGG CTT ACG CAA TTT TT-3⬘; 5⬘-AAA AAT TGC
GTA AGC CG-3⬘; CRE: 5⬘-ACA GTC ATT TCG TCA CAT GGG
CTT G-3⬘; 5⬘-CAA GCC CAT GTG ACG AAA TGA CTG T-3⬘)
was added with the indicated molar excess of unlabelled doublestranded oligonucleotide (C/EBP consensus: 5⬘-TGC AGA TTG CGC
AAT CTG CA-3⬘; 5⬘-TGC AGA TTG CGC AAT CTG CA-3⬘; NFIL6, CRE, the mutated sequence within the CRE site, the mutated
sequence within the NF-IL6 site or E-selectin NF-κB). Tubes were
incubated for 20 min at room temperature. For the supershift experiments, 1 mg/ml of anti-C/EBPα, C/EBPβ, C/EBPδ or CREB antibody
was added to tubes and incubated for a further 20 min. Complexes
were resolved on 6% polyacrylamide gels at 200 V for 2 h. Gels
were vacuum-dried at 80°C for 1 h, then exposed to HyperfilmMP
for 1–2 weeks with an intensifying screen at –70°C, depending on
signal intensity of the bands.
Statistical analysis and data expression
Measurements of CAT and β-gal production are expressed as a ratio
of pg CAT/pg β-gal. Results are displayed as the pooled data from
three experiments performed in triplicate. Statistical significance was
determined by two-way analysis of variation (ANOVA) followed by
least squares analysis or by paired t-test as appropriate. CAT and βgal activity in cells transfected with the TK minimal promoter
construct were expressed as ratios to the mean value for untreated
control cells in each experiment and the significance of effects of
treatment assessed by t-test. P ⬍ 0.05 was considered to be statistically
significant. Errors are expressed as SEM.
Results
Northern analysis
To investigate the changes in mRNA expression in response
to IL-1β, AV3 cells were treated with fresh medium containing 1 ng/ml IL-1β or vehicle and total RNA was isolated at
the times indicated from 0–16 h. Northern blot analysis
revealed a rapid IL-1β-induced increase of PGHS-2 mRNA
levels in AV3 cells within 30 min after cytokine addition,
reaching maximal levels by 2 h before declining gradually
thereafter (Figure 1, lanes 8–14). In control cells, with the
exception of slight expression at the 1 h time point, expression
of PGHS-2 mRNA was not detectable (Figure 1, lanes 1–7).
773
S.Potter et al.
Figure 1. Interleukin-1β (IL-1β) induces prostaglandin H synthase2 (PGHS) mRNA expression in AV3 cells. Cells were treated with
either vehicle or IL-1β (1 ng/ml) and total RNA was isolated at the
times shown. PGHS-2 mRNA expression reached maximal levels at
2 h in response to IL-1β. Blots were re-probed with
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) to confirm
equivalence of mRNA loading. The blot is representative of three
experiments.
Figure 3. The NF-IL6 and cAMP response element (CRE) sites are
both required for increased chloramphenicol acetyl-transferase
(CAT) expression. AV3 cells were transfected with the indicated
CAT prostaglandin H synthase-2 (PGHS) promoter constructs and
treated with 1 ng/ml interleukin-1β (IL-1β) or vehicle for 24 h.
CAT and β-galactosidase (β-gal) enzyme production was measured
by enzyme-linked immunosorbent assay (ELISA). Results are
expressed as pg CAT/pg β-gal and are the pooled data from three
separate experiments. The –203NFM and –203 CREM constructs
reduced the basal and and IL-β-stimulated CAT production, while
the –203NFCREM construct almost completely abolished CAT
production both basally and in the presence of IL-1β (u ⫽ control;
j ⫽ IL-1β-treated). Statistical significance was determined as in
Figure 2.
Figure 2. The promoter elements involved in interleukin-1β
(IL-1β)-stimulated up-regulation of chloramphenicol acetyltransferase (CAT) expression are contained between –52 and
–203 bp of the prostaglandin H synthase-2 (PGHS) gene. AV3 cells
were transiently transfected with the indicated CAT PGHS-2
promoter constructs, then treated with 1 ng/ml IL-1β for 24 h and
CAT and β-galactosidase (β-gal) production were measured by
enzyme-linked immunosorbent assay (ELISA). IL-1β stimulated a
significant increase in transcription in the p203CAT, p321CAT and
p528CAT constructs, but not in the p52CAT construct (u ⫽
control; j ⫽ IL-1β-treated). Results are the pooled data from three
separate experiments. Statistical significance (P ⬍ 0.05) was
determined by two-way analysis of variance (ANOVA) followed by
least squares analysis.
Transcriptional regulation of PGHS-2 expression in
AV3 cells
Deletion analysis of the PGHS-2 promoter in AV3 cells
Expression of CAT by the the p528CAT construct following
treatment with 1 ng/ml IL-1β for 24 h was increased threefold (P ⬍ 0.05) compared with basal expression (Figure 2).
Deletion of the sequences located between –321 and –528
(p321CAT) which contain a NF-κB response element
(–447/–438) did not alter expression. Further deletion of the
sequences between –203 and –321 (p203CAT) which contain
a second NF-κB response element (–222/–213) maintained a
three-fold induction of CAT activity by IL-1β. However,
deletion of the sequences between –203 and –52 (p52CAT), a
region known to contain a NF-IL6 response element (–132/
–124) and a CRE site (–59/–52), had a significant effect
774
on both basal and IL-1β-stimulated CAT production. Basal
expression was decreased to one third of the basal expression
of the p528CAT construct. The ability of IL-1β to stimulate
CAT expression was substantially reduced and was no longer
significant.
The expression of β-galactosidase (β-gal) was not significantly influenced by treatment with IL-1β (1.2 ⫾ 0.1 times
control versus control: 1.0 ⫾ 0.05, P ⫽ 0.16, n ⫽ 18 for each
group). Basal expression from the heterologous minimal TK
promoter was approximately half that from the p528CAT
construct. This was stimulated 2.6 ⫾ 0.4 times by IL-1β
(versus control 1.0 ⫾ 0.1, P ⫽ 0.0003, n ⫽ 18 and n ⫽ 16
for the stimulated and unstimulated cultures respectively).
Site-directed mutagenesis of putative IL-1β regulatory
elements within the PGHS-2 promoter
Mutation of the NF-IL6 response element in p203CAT to
produce a construct referred to as p203NFM resulted in
a 80% decrease in basal CAT production and a 70–75%
decrease in IL-1β-stimulated CAT production compared to
p203CAT (Figure 3). Mutation of this element alone did not
entirely eliminate CAT expression. Mutation of the CRE
site (–203CREM) resulted in a 90% decrease in basal CAT
production and an 80% reduction in IL-1β-stimulated CAT
expression compared to p203CAT, although, once again, did
not completely eliminate CAT expression. Double mutation of
both the NF-IL6 and the CRE sites (–203NFCREM) reduced
basal CAT transcription by 90% and reduced IL-1β-stimulated
expression by 95%. The effect of IL-1β on transcriptional
activity was also essentially negated by the double mutation.
PGHS-2 promoter regulation by IL-1β in amnion-derived cells
Figure 4. A specific complex is formed with the NF-IL6 and cAMP response element (CRE) oligonucleotides using interleukin-1β (IL-1β)stimulated nuclear extracts containing (A) the [32P]-labelled NF-IL6 or (B) the [32P]-labelled CRE probe. An electrophoretic mobility shift
assay (EMSA) was performed using 5 µg nuclear protein isolated from AV3 cells treated with either 1 ng/ml IL-1β (I) or control cells (C)
for 1 h. Reactions contained unlabelled oligonucleotides in 50-fold molar excess: the NF-IL6 oligonucleotide (N); a consensus C/EBP
element (CE); the NF-IL6 site-directed mutation (NM); the CRE59 oligonucleotide sequence (CR); the CRE site-directed mutation (CM); and
an NF-κB-derived sequence from the E selectin gene (ES). The results are representative of data obtained from three separate experiments.
Identification of nuclear factors which regulate PGHS2 transcription by IL-1β
We used EMSA to determine the nuclear factors that bound
the NF-IL6 and CRE elements in AV3 cells treated with 1 ng/ml
IL-1β or vehicle alone. The PGHS-2-derived [32P]-labelled
NF-IL6 oligonucleotide incubated with nuclear extracts from
control (Figure 4A, lane 2) or IL-1β-stimulated cells (Figure
4A, lane 3) produced a high molecular weight DNA–protein
complex. A high molecular weight complex was not observed
in the lane containing probe only (Figure 4A, lane 1). Incubation
with 50-fold molar excess of unlabelled NF-IL6 (N) or a consensus C/EBP oligonucleotide (CE) (Figure 4A, lanes 4 and 5)
diminished the intensity of the complex compared with that seen
without added competitor. The presence of 50-fold molar excess
of the mutated NF-IL6 oligonucleotide (NFM) (Figure 4A, lane
6) or the mutated CRE oligonucleotide (CREM) (Figure 4A,
lane 8) failed to decrease the amount of complex observed. An
additional low mobility band was observed at the top of lane 6,
but is artefactual, as it was not observed upon repetition. Use of
5–, 10–, 25–, 50–, 100–, or 500– fold molar excess of the NFM
oligonucleotide demonstrated no reduction in intensity of the
labelled complex at less than the 100-fold excess and modest
reductions thereafter (data not shown). Incubation with the E
selectin (ES) oligonucleotide, which binds specifically to NFκB, also did not diminish the intensity of the complex (Figure
4A, lane 9). The presence of a 50-fold molar excess of the CRE59
(C) oligonucleotide, however, completely abolished the intensity
of the complex (Figure 4A, lane 7).
This experiment was repeated using a [32P]-labelled CRE
probe (Figure 4B). The control and IL-1β-treated extracts both
produced a high molecular weight complex resulting in a
mobility shift (Figure 4B, lanes 2 and 3). A high molecular
weight complex was not observed in the lane containing probe
only (Figure 4B, lane 1). The intensity of the DNA–protein
complex was reduced with the inclusion of 50-fold molar
excess of the NF-IL6 (N), the C/EBP consensus oligonucleotide
(CE) or the CRE oligonucleotide (C) (Figure 4B, lanes 4, 5
and 7). Again the mutant oligonucleotides (Figure 4B, lanes
6 and 8) and the heterologous, E selectin, oligonucleotide
(Figure 4B, lane 9) failed to compete with the probe for
binding to the nuclear proteins in the complex. As in Figure
4A the additional low mobility band observed at the top of
Figure 4B, lane 6 is artefactual. Use of a 500-fold molar
excess of the CREM oligonucleotide was required to obtain a
reduction in intensity of the labelled complex and even at this
excess the reduction was only slight (data not shown).
Identification of the proteins which bind NF-IL6 and
CRE
Supershift EMSA was used to identify the nuclear proteins
that interact with the PGHS-2-derived NF-IL6 oligonucleotide
(Figure 5) in the presence or absence of IL-1β. The reactions
contained a [32P]-labelled NF-IL6 probe and polyclonal antibodies to C/EBPα, C/EBPβ or C/EBPδ (Figure 5A). Lane 1
contained probe without nuclear extract. A specific protein–
DNA complex was formed with both control and IL-1βstimulated nuclear extracts (Figure 5A, lanes 2 and 6). The
reactions containing isoform specific anti-C/EBP antibodies
(Figure 5A, lanes 3–5 and 7–9) produced supershifted complexes in both the control and IL-1β stimulated extracts as
indicated by the presence of a second lower mobility band
above the main complex. The greatest amount of the supershifted complex was observed with anti-C/EBPβ (Figure 5A,
lanes 4 and 8), followed by anti-C/EBPδ (Figure 5A, lanes 5
and 9) while anti-C/EBPα appeared also to have produced a
faint supershift (Figure 5A, lanes 3 and 7).
Similar experiments were performed with the CRE
oligonucleotide as the probe (Figure 5B). In addition to the
anti-C/EBP antibodies, an anti-CREB antibody was used in
these experiments to probe for involvement of this protein in
the complexing of the PGHS-2 CRE site. In contrast to the
results when the NF-IL6 oligonucleotide was used as the
probe, the complex formed on the CRE probe was supershifted
only by anti-C/EBPβ (Figure 5B, lanes 4 and 9). Notably antiCREB did not produce such a shift (Figure 5B, lanes 6 and 11).
Discussion
This study examined the mechanism by which IL-1β induces
the transcription of the PGHS-2 gene in amnion-derived AV3
775
S.Potter et al.
Figure 5. C/EBPβ appears to be utilized in the regulation of prostaglandin H synthase-2 (PGHS) gene transcription in response to IL-1β. A
supershift electrophoretic mobility shift assay (EMSA) was performed using 5 µg nuclear protein from either stimulated (I) or unstimulated
(C) cells, incubated with either (A) [32P]-labelled NF-IL6 or (B) CRE probe and antibodies to C/EBPα (Cα), C/EBPβ (Cβ), C/EBPδ (Cδ),
or CREB (CR). Results were reproduced in two other experiments.
cells. It has been shown previously that PGHS-2 mRNA
expression can be induced by a variety of stimuli including
agents which act through G protein-mediated mechanisms, the
protein kinase C pathway mediated by phorbol esters and the
tyrosine kinase-mediated pathways activated by growth factor
receptors (Herschman, 1996). However, the signal transduction
mechanisms, the transcription factors, and the regulatory
regions of the PGHS-2 gene necessary for PGHS-2 enzyme
induction are not well understood. A number of PGHS-2
promoter elements have previously been implicated in the
transcriptional regulation of the PGHS-2 gene, including the
NF-IL6 element, the CRE motif, a TATA box and the NF-κB
sites (Appleby et al., 1994). With respect to IL-1β-induced
up-regulation of PGHS-2 mRNA expression, it has been
reported that the NF-κB promoter element is involved in the
transcriptional activation of the PGHS-2 gene (Newton et al.,
1997; Wang and Tai, 1998).
It has been suggested that the NF-IL6 site was involved
in the transcriptional regulation of the PGHS-2 gene by
gonadotrophin in rat granulosa cells (Sirois and Richards,
1993). Recently, Inoue et al. (1995) showed that induction of
PGHS-2 expression in vascular epithelial cells by phorbol
ester or LPS involved the CRE and NF-IL6 sites, and that C/
EBPδ functioned as a trans-acting factor, perhaps in association
with CREB. Other studies have provided evidence for the
involvement of the CRE regulatory element in an oncogeneinduced up-regulation of the murine PGHS-2 gene (Xie and
Herschman, 1995).
Our CAT reporter studies indicate that a region of the
human PGHS-2 promoter responsible for significant basal
transcriptional activity and IL-1β-responsiveness in amnionderived AV3 cells lies between bases –203 and –52 upstream
from the transcription start site. This region contains CRE and
NF-IL6 elements. However, our results appear to exclude a
role for NF-κB elements between –528 and –203 in these
processes.
776
Both the NF-IL6 and CRE response elements seem to
contribute independently to CAT expression in both stimulated
and unstimulated cells. Mutation of either site reduces CAT
expression, substantially. Furthermore, point mutation of both
the NF-IL6 and CRE elements abolished the responsiveness
of the promoter to IL-1β. This is important to note, as cryptic
sites in the backbone of the vector used in making the promoter
constructs confer IL-1β responsiveness on the minimal TK
promoter–reporter construct. Abolition of IL-1β responsiveness
in the PGHS-2 promoter fragment by these mutations of the
promotor sequence suggests that the effects of the cryptic
sites are negligible in the context of the PGHS-2 promoter
constructs, at least in the case of the –203 plasmid. Therefore,
both the CRE and NF-IL6 response elements seem to be
involved in the transcriptional regulation of the PGHS-2 gene
in AV3 cells.
To further define the mechanism of IL-1β-induced PGHS2 mRNA transcription, EMSA assay was utilized. Nuclear
extracts from control or IL-1β-stimulated AV3 cells contain a
factor or factors which specifically bind the PGHS-2 NFIL6 and CRE site-containing oligonucleotides. Interestingly,
however, the two oligonucleotides cross-compete for binding
of the factor or factors involved. These results again suggest
that both the CRE and NF-IL6 promoter elements are involved
in control of transcriptional activity of the PGHS-2 gene in
AV3 cells. They further suggest involvement of a common
factor at the two sites. The inability of the mutated sequence
within the CRE oligonucleotide (CREM) or the mutated
sequence within the NF-IL6 oligonucleotide (NFM) to displace
the EMSA probe oligonucleotides from the DNA–protein
complex indicates that the two elements are specifically
required for the binding of the factors in the nuclear extracts.
In order to evaluate which proteins may be involved in the
complexes formed in the EMSA assay, antibodies to factors
known to bind the response elements identified were added to
the assay. C/EBP was originally described as a rat liver-specific
PGHS-2 promoter regulation by IL-1β in amnion-derived cells
DNA-binding protein with a leucine zipper motif (Johnson
et al., 1987). Several isoforms of this trans-acting factor,
C/EBPα, C/EBPβ and C/EBPδ bind to the NF-IL6 site (Akira
et al., 1990; Pall et al., 1997). Therefore, these C/EBP isoforms
as well as the CRE binding protein, CREB, were evaluated.
Our results indicate that C/EBPβ is involved in the complexes
formed with both the NF-IL6 and CRE response elements and
that CREB does not participate in the complex between the
CRE site and nuclear factors from the AV3 cells. Interestingly,
C/EBP isoform specificity at the NF-IL6 site is more relaxed
than at the CRE site with evidence that C/EBPδ and possibly
C/EBPα also bind the NF-IL6 site. The effectiveness of the
CRE oligonucleotide to compete with the NF-IL6 oligonucleotide probe (Figure 4A, lane 5) for binding of the factors
in the nuclear extracts suggests that C/EBPβ is the major
C/EBP binding the NF-IL6 site.
IL-1β did not induce significant changes in either the CRE
or NF-IL6 binding capacity in AV3 cell nuclear extracts. Rapid
up-regulation, such as that observed for PGHS-2, tends to
involve postranslational mechanisms for activation of the
factors responsible. C/EBP function is regulated extensively
by phosphorylation (Takiguchi, 1998). Of particular note is
that binding of C/EBPβ to a promoter has been reported to be
unaffected by cAMP, while transcriptional activation by the
factor required cAMP-induced phosphorylation (Tae et al.,
1995). Thus the lack of significant change in binding cannot be
taken to contradict the (modest) up-regulation observed in
the promoter-reporter study. The large contrast between the
induction of PGHS-2 expression by IL-1β observed in
the Northern analyses and the analyses of the cloned promoter fragments, nevertheless, suggests that regions of the
promoter outside of the –528 to ⫹9 region may be important
in mediating the up-regulation by IL-1β.
C/EBPβ is thought to be involved in the mediation of
extracellular signalling and its expression is up-regulated by
various agents such as cytokines, calcium, lipopolysaccharide
and phorbol ester, or by glutamate receptor activation (Yano
et al., 1996). Interestingly, NF-κB sites are often involved in
mediation of pro-inflammatory cytokine induction (Tazawa
et al., 1994; Collins et al., 1995). Furthermore, involvement
of both NF-κB sites in the –591 to –203 region of the promoter
in induction of PGHS-2 by IL-1β has been reported in the
amnion-derived WISH cell line (Wang and Tai, 1998). Yet
AV3 cells clearly do not utilize these NF-κB sites in our
experiments. This might result from differing site specificities
of the subsets of the NF-κB/Rel family members expressed in
the cells or activated during IL-1β-induced signalling (Collins
et al., 1995; Perkins, 1997). These matters, as well as the
question of whether the utilization of NF-κB by WISH cells
or AV3 cells is more reflective of the cells of the amnion,
will require further experimentation to resolve. Our evidence
suggests that evaluation of additional regulatory sequences
outside the proximal promoter region examined here may be
required for full understanding of regulation of PGHS-2 gene
transcription by IL-1β. However, the data strongly supports a
role for C/EBPs, particularly C/EBPβ, binding at the –132
NF-IL6 and –59 CRE sites in regulation of transcription of
PGHS-2 in AV3 cells.
Acknowledgements
This research was supported by grants from The Royal Society of
New Zealand Marsden Fund, The Health Research Council of New
Zealand, and The Lottery Grants Board of New Zealand. The authors
would like to thank Dr Larry Chamley (National Womens Hospital,
Auckland, New Zealand) for providing the AV3 cells. We would also
like to thank Allan Drew for technical assistance, and Dr Marion
Blumenstein and Dr Jeffrey Keelan for critical review of the manuscript.
References
Akira, S., Isshiki, H., Sugita, T. et al. (1990) A nuclear factor for IL-6
expression (NF-IL6) is a member of a C/EBP family. EMBO J., 9,
1897–1906.
Appleby, S., Ristimaki, A. and Neilson, K. (1994) Structure of the human
cyclooxygenase-2 gene. Biochem. J., 302, 723–727.
Blanco, A., Habib, A., Levy-Toledano, S. et al. (1995) Involvement of tyrosine
kinases in the induction of cyclooxygenase-2 in human endothelial cells.
Biochem. J., 312, 419–423.
Chomczynski, P. and Sacchi, N. (1987) Single step method of RNA isolation
by acid-guanidinium thiocyanate-phenol-chloroform extraction. Analyt.
Biochem., 162, 156–159.
Collins, T., Read, M.A., Neish, A.S. et al. (1995) Transcriptional regulation
of endothelial cell adhesion molecules: NF-kappa B and cytokine-inducible
enhancers. FASEB J., 9, 899–909.
Dinarello, C. (1987) Interleukin-1: multiple biological properties and
mechanisms of action. Adv. Prost. Thromb. Leukot. Res., 17B, 900–904.
Dudley, D., Trautman, M., Edwin, S. et al. (1993) Inflammatory mediators
regulate IL-8 production by cultured gestational tissues: evidence for a
cytokine network at the chorio-decidual interface. J. Clin. Endocrinol.
Metab., 76, 404–410.
Dudley, D.J. and Trautman, M.S. (1994) Infection, inflammation and
contractions: the role of cytokines in the pathophysiology of preterm labor.
Sem. Reprod. Endocrinol., 12, 263–272.
Fletcher, B., Kujubu, D., Perrin, D. et al. (1992) Structure of the mitogeninducible TIS10 gene and demonstration that the TIS10-encoded protein is
a functional prostaglandin G/H synthase. J. Biol. Chem., 267, 4338–4344.
Funk, C., Funk, L. and Kennedy, M. (1991) Human platelet/erythroleukemia
cell prostaglandin G/H synthase: cDNA cloning, expression, and gene
chromosomal assignment. FASEB., 5, 2304–2312.
Gunn, L., Hardiman, P., Tharmaratman, S. et al. (1996) Measurement of
interleukin-1β and interleukin-6 in pregnancy-associated tissues. Reprod.
Fertil. Dev., 8, 1069–1073.
Hamasaki, Y. and Eling, T. (1995) EGF and TPA stimulate de novo synthesis
of PGHS-1 and PGHS-2 through different signal transduction pathways.
Prost. Leukot. Essent. Fatty Acids, 53, 225–229.
Hansen, W.R., Sato, T., and Mitchell, M.D. (1998) Tumour necrosis factoralpha stimulates increased expression of prostaglandin endoperoxide H
synthase type 2 mRNA in amnion-derived WISH cells. J. Mol. Endocrinol.,
20, 221–231.
Herschman, H. (1991) Primary response genes induced by growth factors and
tumour promoters. Ann. Rev. Biochem., 60, 281–319.
Herschman, H.R. (1996) Prostaglandin synthase 2. Biochim. Biophys. Acta,
1299, 125–140.
Hla, T. and Neilson, K. (1992) Human cyclooxygenase-2 cDNA. Proc. Natl
Acad. Sci. USA, 89, 7384–7388.
Ikejima, T., Okusawa, S., Ghezzi, P. et al. (1990) Interleukin-1 induces tumour
necrosis factor (TNF) in human peripheral blood mononuclear cells in vitro
and a circulating TNF-like activity in rabbits. J. Infect. Dis., 162, 215–223.
Inoue, H., Yokoyama, C., Hara, S. et al. (1995) Transcriptional regulation of
human prostaglandin-endoperoxide synthase-2 gene by lipopolysaccharide
and phorbol ester in vascular endothelial cells. J. Biol. Chem., 270,
24965–24971.
Johnson, P., Landschulz, W., Graves, B. et al. (1987) Identification of a rat
liver nuclear protein that binds to the enhancer core elements of three
animal viruses. Genes Dev., 1, 133–146.
Keirse, M. and Gravenhorst, J. (eds) (1979) Human Parturition. Martinus
Nijhoff Publishers, CZ Zoetermeer, The Netherlands, 101 pp.
Kosaka, T., Miyata, A., Ihara, H. et al. (1994) Characterisation of the human
gene (PTGS2) encoding prostaglandin-endoperoxide synthase-2. Eur. J.
Biochem., 221, 889–897.
Kujubu, D., Fletcher, B., Varnum, B. et al. (1991) TIS10, a phorbol ester
777
S.Potter et al.
tumour promoter-inducible mRNA from swiss 3T3 cells, encodes a novel
prostaglandin synthase/cyclooxygenase homologue. J. Biol. Chem., 266,
12866–12872.
Luckow, B. and Schutz, G. (1987) CAT constructions with multiple unique
restriction sites for the functional analysis of eukaryotic promoters and
regulatory elements. Nucleic Acids Res., 15, 5490.
Mitchell, M.D., Romero, R. and Avila, C. (1991) Prostaglandin production
by amnion and decidual cells in response to bacterial products. Prost.
Leukot. Essent. Fatty Acids, 42, 67–169.
Newton, R., Kuitert, L., Bergmann, M. et al. (1997) Evidence for involvement
of NF-κB in the transcriptional control of COX-2 gene expression by IL1β. Biochem. Biophys. Res. Commun., 237, 28–32.
Nyborg, J.K., Matthews, M.A., Yucel, J. et al. (1990) Interaction of host cell
proteins with the human T-cell leukemia virus type I transcriptional
control region. II. A comprehensive map of protein-binding sites facilitates
construction of a simple chimeric promoter responsive to the viral tax2
gene product. J. Biol. Chem., 264, 8237–8242
Pall, M., Hellberg, P., Brännström, M. et al. (1997) The transcription factor
C/EBP-β and its role in ovarian function; evidence for direct involvement
in the ovulatory process. EMBO J., 16, 5273–79.
Perkins, N.D. (1997) Achieving transcriptional specificity with NF-κB. Int. J.
Biochem. Cell Biol., 29, 1433–1448.
Redinbaugh, M.G. and Turley, R.B. (1986) Adaptation of the bicinchoninic
acid protein assay for use with microtiter plates and sucrose gradient
fractions. Analyt. Biochem., 153, 267–271.
Rivadeneira, J.C. and Robbins, F.C. (1958) On the origin of variants of human
amniotic cells in tissue culture. Am. J. Dis. Child., 96, 43.
Romero, R., Sirtori, M. and Oyarzun, E. (1989a) Prevalence, microbiology
and clinical significance of intraamniotic infection in women with preterm
labor and intact membranes. Am. J. Obstet. Gynecol., 161, 817–824.
Romero, R., Manogue, K. and Murray, M. (1989b) Infection and labor. IV.
Cachectin tumour necrosis factor in the amniotic fluid of women with intraamniotic infection and preterm labor. Am. J. Obstet. Gynecol., 166,
1576–1587.
Romero, R., Mazor, M. and Brandt, F. (1992) Interleukin-1 alpha and
interleukin-1 beta in preterm and term parturition. Am. J. Reprod. Immunol.,
27, 117–123.
Shalaby, M., Waage, A., Aarden, L. et al. (1989) Endotoxin, tumour necrosis
factor-α and interleukin-1 induce interleukin-6 production in vivo. Clin.
Immunol. Immunopathol., 53, 488–498.
Sirois, J. and Richards, J. (1993) Transcriptional regulation of the rat
prostaglandin endoperoxide synthase-2 gene in granulosa cells. J. Biol.
Chem., 268, 21931–21938.
Smith, W. and DeWitt, D.L. (1995) Biochemistry of prostaglandin H synthase1 and synthase-2 and their differential susceptibility to nonsteroidal antiinflammatory drugs. Sem. Nephrol., 15, 179–194.
Tae, H.J., Zhang, S., and Kim, K.H. (1995) cAMP activation of CAAT
enhancer-binding protein-beta gene expression and promoter I of acetylCoA carboxylase. J. Biol. Chem., 207, 21487–21494.
Takiguchi, M. (1998) The C/EBP family of transcription factors in the liver
and other organs. Internat. J. Exp. Pathol., 79, 369–391.
Tazawa, R., Xu, X.-M., Wu, K.K. et al. (1994) Characterisation of the genomic
structure, chromosomal location and promoter of the human prostaglandin
H synthase-2 gene. Biochem. Biophys. Res. Commun., 203, 190–199.
Teixeira, F.J., Zakar, T., Hirst, J.J. et al. (1994) Prostaglandin endoperoxideH synthase (PGHS) activity and immunoreactive PGHS-1 and PGHS-2
levels in human amnion throughout gestation, at term, and during labor. J.
Clin. Endocrinol. Metab., 78, 1396–1402.
Wahbeh, C., Hill, G. and Eden, R. (1984) Intra-amniotic bacterial colonisation
in premature labor. Am. J. Obstet. Gynecol., 148, 739–743.
Wang, Z. and Tai, H.H. (1998) Interleukin-1β and dexamethasone regulate
gene expression of prostaglandin H synthase-2 via the NF-κB pathway in
human amnion-derived WISH cells. Prost. Leukot. Essent. Fatty Acids, 59,
63–69.
Xie, W. and Herschman, H.R. (1995) v-src induces prostaglandin synthase 2
gene expression by activation of the c-jun N-terminal kinase and the c-jun
transcription factor. J. Biol. Chem., 270, 27622–27628.
Yano, S., Fukunaga, K., Takiguchi, M. et al. (1996) Regulation of CCAAT/
enhancer binding protein family members by stimulation of glutamate
receptors in rat cortical astrocytes. J. Biol. Chem., 271, 23520–23527.
Received on November 2, 1999; accepted on June 8, 2000
778