Mol. Hum. Reprod. 2002 Zervou 597 605

8/12/2019 Mol. Hum. Reprod. 2002 Zervou 597 605

1/9

Molecular Human Reproduction Vol.8, No.7 pp. 597605, 2002

Steroids mediate the expression of cytoplasmic andmembrane-linked components in human myometrial cells

S.Zervou1,*, E.Karteris*, E.W.Hillhouse and R.W.Old

Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK

1To whom correspondence should be addressed. E-mail: [email protected]

*These authors have contributed equally to this work

It is well known that the smooth muscle of the human myometrium is a target for the steroid hormones progesterone (P 4) and

estrogen. Progesterone is believed to participate in the maintenance of pregnancy, while estrogen is possibly involved in the

process of parturition by promoting cervical dilatation. We examined the combined effects of P 4 and 17-estradiol (E2) oncomponents of signalling pathways in human myometrial cells in vitro by immunoblotting. Long-term treatment of myometrial

cells with a series of concentrations of P4 and E2 in combination caused a change in the phosphorylation status of p42/44

mitogen-activated protein kinase and of c-Jun N-terminal kinase (SAPK/JNK). P4and E2caused a decrease in protein expressionof Gq, Gz, Gi1/2and, to a lesser extent, G 0. The two steroids caused a decrease in the expression of the two small GSisoforms. Cyclo-oxygenase-2 expression was increased by 2.5-fold after steroid treatment, while proliferating cell nuclear antigen

expression levels remained unchanged. These observations show that the combination of P4 and E2 influences intracellular and

membrane-bound components of signal transduction pathways in human myometrial cells. The implications of the two steroid

hormones on intracellular signalling pathways in the human myometrium merits further investigation.

Key words: G-proteins/kinases/myometrium/steroids

Introduction

The human uterine smooth muscle and the uterine vasculature aretargets for estrogen and progesterone (P4), two known modulators of

the myometrial contractile state (Maggi et al., 1992; Vagnoni et al.,

1998; Rupnow et al., 2001). P4 is considered to be necessary for

pregnancy maintenance and its withdrawal in many species is a key

event in parturition, whereas 17-estradiol (E2) augments the capacityfor uterine contraction and cervical dilatation and therefore promotes

parturition by causing myometrial contractions (Mesiano, 2001).

Existing evidence on the effects of P4 and E2 on processes such as

formation of gap junctions supports the concept that these steroids

play a role in human parturition. While the involvement of estrogens

and P4 in pregnancy is widely acknowledged, little is known about

the steroid-mediated mechanisms that control the expression of key

proteins regulating the excitability and contractility of the human

myometrium. Previous work has shown that P4 increases intracellularcalcium concentrations in myometrial cells in vitro (Fomin et al.,

1999) and that estrogen promotes calcium influx into myometrial

cells (Wehling, 1997).

Mitogen-activated protein (MAP) kinases are ubiquitous serine/

threonine kinases that are activated by a wide variety of extracellular

signals and are essential in triggering cell cycle progression (Seger

and Krebs, 1995). It has been demonstrated that oxytocin acutely

activates MAP kinase through an islet-activating protein-sensitive

G-protein in human uterine myometrial cells, suggesting that MAP

kinase maybe involved in the modulation of humanand ratmyometrial

contractility by oxytocin (Ohmichi et al., 1995, 1997; Nohara et al.,

1996). Another well characterized subfamily of the MAP kinase

European Society of Human Reproduction and Embryology 597

superfamily are the stress-activated protein kinases (SAPK), also

referred to as Jun N-terminal kinases (JNK), that function in a protein

kinase cascade transducing cellular stress signals (Kyriakis and

Avruch, 1990; Hibi et al., 1993; Derijard et al., 1994).

The presence of the rate-limiting enzyme in prostaglandin synthesis,

prostaglandin endoperoxide-H synthase (cyclo-oxygenase; COX), in

the non-pregnant uterus has been previously reported (Moonen et al.,

1984). Two isoforms of COX exist, namely COX-1 and COX-2 (Hla

et al., 1992). COX-2 is localized in human myometrial cells and it

has been shown that human labour is associated with the up-regulation

of prostaglandins within the uterus, synthesised via COX-2 (Moonen

et al., 1985; Slater et al., 1999a,b; Sparey et al., 1999). COX-2

expression can be induced through multiple signalling pathways

involving protein kinase A, protein kinase C, tyrosine kinases and

lipopolysaccharides (Xie and Hershman, 1995). It is known that E2and P4regulate prostaglandin synthesis in bovine endometrium during

the estrus cycle (Xiao et al., 1998), but the effects of these steroidson human myometrial COX-2 are unknown.

Similarly, very little is known about the effects of estrogen and P 4on G-proteins in the human myometrium. In rat myometrium, it has

been demonstrated that Gi1/2 and Gq subunits are physiologicaltargets for both steroids in vivo (Cohen-Tannoudji et al., 1995). It is

known that estrogen and P4 modulate the expression of G proteins

in lactotropes (Livingstoneet al., 1998).

We have previously reported on the synergistic effects of P 4 and

E2 on nitric oxide synthase expression in isolated myometrial cells

(Zervou et al., 1999a). These observations suggest that ovarian

steroids may have a profound influence on the myometrial intracellular

microenvironment with important physiological implications.


2/9

S.Zervouet al.

Figure 1. Detection of p42/44 MAP kinase by immunoblotting in MCF-7 cells (a) and in cultured human myometrial cells (b). An antibody recognizing totalp42/44 MAP kinase was used. (c) and (d): Densitometric analysis of total p42/44 MAP kinase protein levels in MCF-7 (corresponding to a) and in myocytes(corresponding to b). NS no supplement; PE treatment with the combination of progesterone and 17-estradiol, each at 5 mol/l; A23187 treatmentwith calcium ionophore. Data are expressed as mean SE.

Figure 2. Detection of p42/44 MAP kinase by immunoblotting, in MCF-7 cells (a) and in cultured human myometrial cells (b), as in Figure 1. An antibodyrecognizing the phosphorylated p42/44 MAP kinase was used. (c) and (d) Densitometric analysis of phosphorylated p42/44 MAP kinase protein levels inMCF-7 cells (corresponding to a) and in myocytes (b). NS no supplement; PE treatment with the combination of progesterone and 17-estradiol, eachat 5 mol/l; A23187 treatment with calcium ionophore. Data are expressed as mean SE. *P 0.05; **P 0.01.

598


3/9

Effects of steroids on human myometrial cell components

Figure 3. Detection of c-Jun N-terminal kinase (SAPK/JNK) in cultured human myometrial cells. Immunoblotting analysis was carried out using antibodiesagainst total SAPK/JNK protein levels (a) or against phosphorylated SAPK/JNK (b). Two protein bands (54 and 46 kDa) were detected, corresponding tototal levels (a), while one protein band (46 kDa) was detected by the antibody against phosphorylated SAPK/JNK protein. ( c) and (d) densitometric analysisof total and phosphorylated SAPK/JNK in MCF-7 cells and in myocytes respectively. NS no supplement; PE treatment with a combination ofprogesterone and 17-estradiol, each at 5 mol/l. Data are expressed as mean SE. **P 0.01.

Figure 4. Immunoblotting analysis of cyclo-oxygenase-2 (COX-2) protein expression in cultured human myometrial cells ( a). A protein band of 62 kDa wasobtained. (b) Densitometric analysis of COX-2 protein levels in myocyte cultures, corresponding to (a). NS no supplement; PE treatment withcombination of progesterone and 17-estradiol, each at 5 mol/l. Data are expressed as mean SE. **P 0.01.

In our attempt to make progress in identifying further actions of

P4 and E2 upon the human myometrium, we sought to investigate

their combined action on cultured myometrial cells by determining

their effects on the phosphorylation status of the family of p42/44

MAP kinases and SAPK/JNK kinases, on the expression of G-

proteins, and on the expression of the enzyme COX-2. Unlike most

other animal species including sheep,P4withdrawal does not occur

599

in human or primate pregnancies (Csapo and Pinto-Dantas, 1965;

Mazor et al., 1993). P4 and E4 are present in high concentrations

throughout human gestation, and no dramatic changes are seen

towards the end of term. P4production by the placenta reaches ~300

mg per day at the end of pregnancy (Perusquia, 2001), while E2increases by almost 100-fold during pregnancy compared with non-

pregnant levels (Mesiano, 2001). Therefore, we hypothesized that the


4/9

S.Zervouet al.

Figure 5. Immunoblotting analysis of proliferating cell nuclear antigen (PCNA) protein levels in cultured human myometrial cells. A protein band at the sizeof 36 kDa was detected (a). (b) Densitometric analysis of PCNA protein levels in myocyte cultures, corresponding to ( a). NS no supplement; PE treatment with the combination of progesterone and 17-estradiol, each at 5 mol/l. Data are expressed as mean SE.

two steroid hormones modulate the expression of key proteins in the

myometrial smooth muscle cells, causing an alteration to their profile,

with an impact on uterine physiology and activity.

Materials and methods

Materials

Hanks Balanced Salt Solution (HBSS) with and without Ca2/Mg2,

Dulbeccos Modified Eagles Medium (DMEM), penicillin G, streptomycin,

L-glutamine, fatty acid-free bovine serum albumin, P 4, calcium ionophore

A23187 and E2 were provided by Sigma (Poole, UK). Fetal calf serum (FCS)

and non-essential amino acids were products from Labtech (East Sussex, UK)

and Gibco Brl (Paisley, UK) respectively. Anti-phosphorylated p42/44 and

anti-total MAP kinase antibodies were raised in rabbits, with a synthetic

peptide corresponding to residues 345358 of rat p42 MAP kinase (New

England Biolabs, Beverly, MA, USA). The antibody against human prolifer-

ating cell nuclear antigen (PCNA) was provided by Zymed (San Francisco,

CA, USA).

Antisera against the Gz -subunit were purchased from Calbiochem

(Nottingham, UK). Anti-G protein antibodies GC/2, AS/7, RM/1 and QL

directed against the -subunits, were obtained from New England Nuclear-

DuPont (Boston, MA, USA). All primary antibodies were raised in rabbits.

AS/7, RM/1 and QL were polyclonal, whereas GC/2 and Gz antibodies were

monoclonal. The anti-rabbit IgG antibodies were obtained from Sigma. All

electrophoretic reagents were obtained from BioRad (Richmond, CA, USA).

Anti-COX-2 polyclonal antibody was obtained from Santa Cruz Biotech

(Santa Cruz, CA, USA). Polyclonal antibodies against human SAPK/JNK

were produced by immunizing rabbits with a full length p54 SAPK/JNK2

fusion protein. The antibody was raised against a recombinant protein

corresponding to amino acids 50111 mapping near the C-terminus of COX-

2 of human origin, non-cross-reactive with COX-1.

Selection of cells and experimental subjects

All tissue samples were collected from women undergoing a gynaecological

operationat Womens Hospital, University Hospitals of Coventryand Warwick-

shire, NHS Trust, Coventry, UK. The study was approved by the local ethics

committee, and informed consent was obtained from each patient prior to

operation. Myometrial biopsies were taken from the upper third of the uterinebody ~5 mm away from endometrial or serosal surfaces, immediately after

hysterectomy. All women recruited to the study were premenopausal and had

not been exposed to steroid treatment for at least 3 months prior to the

operation, and did not have either an intrauterine contraceptive device (IUD)

in situ or evidence of uterine pathology, such as fibroids or polyps. Human

adenocarcinoma breast cancer MCF-7 cells were obtained from the European

Collection of Animal Cell Cultures (CAMR, Centre for Applied Microbiology

and Research, Salisbury, Wiltshire, UK). Although no positive controls were

available for SAPK/JNK, COX-2, G protein subunits or PCNA, MCF-7

cells were used as a positive control for MAP kinase experiments. This cell

line was chosen because it expresses both estrogen and P4 receptors. MCF-7

cells have been used in the past to characterize steroid effects on p42/44 MAP

kinase. No such controls are available for the rest of the proteins studied here.

600

Establishment of primary myometrial cell cultures and maintenanceof MCF-7 cells

Myometrial biopsies weighing ~3 g were collected from women undergoing

hysterectomy for menorrhagia. Primary myometrial cell cultures were estab-

lished as previously described (Phaneufet al., 1993). Cells were re-suspended

in DMEM, containing 10% FCS, 0.2% L-glutamine, 10 000 IU/ml penicillin G

and 7610 IU/ml streptomycin, supplemented with P4and E2. Ano supplement

culture served as a negative control. Myometrial cells were plated into 25cm2 cultureflasks at a density of 0.52104 cells/cm2 and stored at 37C in

a humidified atmosphere (95% air and 5% CO2) for up to 72 h. The purity

of the myocyte cultures was assessed as previously described (Zervou et al.,

1999a) using the ratio of the number of cells stained for -actin to the total

cell nuclei present. Analysis of large numbers of cells indicated that 95%

of the cultured cells were identified as smooth muscle cells. All primary

cultures were maintained for up to 4 days, and incubated with charcoal-

stripped FCS and phenol red-free media for 24 h prior to steroid treatments.

In this way, endogenous and exogenous steroids were eliminated. MCF-7

cells were maintained in phenol red-free minimum essential medium (Sigma)

with 1% non-essential amino acids, 10% charcoal-stripped fetal bovine serum,

10 000 IU/ml penicillin (Sigma) and 7610 IU/ml streptomycin (Sigma).

Protein extraction from cultured cells

Media were removed and cell monolayers were washed with phosphate-buffered saline (PBS) at room temperature. A buffer containing PBS, 1%

NP40 (Sigma), 0.5% sodium deoxycholate (Gibco), 0.1% sodium dodecyl

sulphate (SDS; Gibco), 10 g/ml polymethyl-sulphonyl fluoride, 30 l/ml

aprotinin and 10 l/ml sodium orthovanadate (100 mmol/l) was added to the

cultures. The cells were scraped off the tissue culture surface with cell

scrapers. The lysate was then centrifuged for 10 min at 13 000 g, at 4C. The

supernatant was removed and stored as the total cell lysate. The protein

concentration was determined in all tissue extracts using the BioRad Reagent,

according to the manufacturers instructions.

Immunoblotting

Myometrial cell lysates (80 g) were solubilized with Laemmli buffer

(5 mol/l urea, 0.17 mol/l SDS, 0.4 mol/l dithiothreitol and 50 mmol/l Tris

HCl, pH 8.0), mixed and placed in a boiling water bath for 5 min and allowed

to cool at room temperature. Samples were separated on an SDS10%polyacrylamide gel and the proteins were electrophoretically transferred to a

nitrocellulose filter at 250 mA for 1618 h in a transfer buffer containing

20 mmol/l Tris, 150 mmol/l glycine and 20% methanol. The filter was then

blocked in PBS containing 0.1% Tween-20 and 5% dried milk powder (w/v)

for 2 h at room temperature. After three washes with PBS0.1% Tween, the

nitrocellulose filters were incubated with each primary antibody against MAP

kinase and JNK total and phospho- forms, PCNA, COX-2 and G-protein

-subunits as described previously (Karteriset al., 2000). All primary antisera

were used at a 1:1000 dilution in PBS0.1% Tween for 1 h at room

temperature. The filters were washed thoroughly for 30 min with PBS0.1%

Tween before incubation with the secondary anti-rabbit HRP-conjugated Ig

(1:2000) for 1 h at room temperature and further washing for 30 min with

PBS0.1% Tween. In order to detect the antibody complexes, solution A


5/9


Figure 6. Immunoblotting analysis of G protein levels in cultured human myometrial cells and corresponding densitometric analysis of expression levels. (a)and (b): Gq at 42.5 kDa; (c) and (d): Gz at 39 kDa; (e) and (f): Gs at 67, 54, 47 and 45 kDa; (g) and (h): Gi1/2at 41 kDa; (i) and (j): G0at 40.5 kDa. NS no supplement; PE treatment with the combination of progesterone and 17-estradiol, each at 5 mol/l. Data are expressed as mean SE. **P 0.01.

601


6/9

S.Zervouet al.

containing 100 mmol/l Tris pH 8.0 and 30% H2O2 was mixed with solution

B containing 100 mmol/l Tris pH 8.0, 90 mmol/l coumaric acid and

250mmol/lluminol,and applied toimmunoblotsfor 2 minat room temperature.

Immunoblots were visualized by exposure on X-ray film (Fuji Photo Film

Co. Ltd, Tokyo,Japan), and were replicated at least twice on each tissue sample.

Statistical analysis

All experiments were performed in triplicate, independently, on myometrial

biopsies from four individuals. The intensities of immunoreactive staining

were measured using a scanning densitometer coupled to a scanning softwarepackage (Image Quant, Molecular Dymanics, Pharmacia Amersham Life

Sciences, Little Chalfont, UK). Data are shown as the mean SEM of each

measurement. In each case results were evaluated between groups by using

the two-tailed Studentst-test. AP-value of0.05 was considered significant.

Results

Effect of steroids on p42144 MAP kinase in myometrial andMCF-7 cells

Primary myocyte cultures were treated for up to 24 h with P4 and E2combined, at a series of concentrations, optimized at 5 mol/l orwith the calcium ionophore A23187 at 10 mol/l for 15 min. SDSpolyacrylamide gel electrophoresis was carried out using myometrial

cell protein extracts from treated and untreated myocyte cultures,

followed by immunodetection with antibodies against total (Figure 1)

and phosphorylated p42/44 MAP kinase (Figure 2). The treatment

decreased the phosphorylated form of p42 and p44 MAP kinase (both

at P 0.05), while total p42/44 MAP kinase remained unchanged.

The ionophore A23187 caused an increase in p44 MAP kinase

(P 0.05).

The effect of steroids on p42/44 MAP kinase in vitro was also

tested in the human adenocarcinoma MCF-7 cells (Figure 1a,c and

Figure 2a,c), which endogenously express estrogen and P4 receptors

and were used as a positive control. P4 and E2 in combination, for

15 min, each at 5 mol/l, caused a slight increase in the expressionof phosphorylated p44 and a more marked increase in phosphorylated

p42 MAP kinase protein levels (P 0.01), while total p42/44 MAP

kinase remained unchanged.

Effects of steroids on SAPK/JNK expression

In an attempt to investigate any changes in the expression of SAPK/

JNK in cultured myocytes, primary myometrial cultures were treated

for up to 24 h with P4and E2combined, at a series of concentrations,

optimized at 5 mol/l. Protein amounts for total SAPK/JNK wereshown to be equal in both untreated and steroid-treated myometrial

cell extracts, as demonstrated with immunoblotting using an antibody

recognizing total SAPK/JNK. Two protein bands were detected, at

46 and 54 kDa, corresponding to p46 and p54 SAPK/JNK (Figure

3). An antibody recognizing the phosphorylated form of SAPK/JNK

was also used. The phospho-SAPK/JNK (Thr183/Tyr185) antibody

detects the phosphorylated isoforms of all SAPKs/JNKs. Thisantibodydid not cross-react with endogenous levels of the corresponding

phosphorylated forms of p42/44 MAP kinase or p38 MAP kinase.

Combined steroidal treatment caused a decrease in phosphorylated

p46 SAPK/JNK (P 0.01; Figure 3b,d). The phosphorylated form

of p54 was detected in neither the untreated, or the hormone-treated

myometrial protein extracts with this antibody.

Effect of steroids on COX-2

Immunoblotting experiments were performed, following treatment of

myocyte cultures for up to 24 h with the combination of P4 and E2,

at a series of concentrations, optimized at 5 mol/l. A single proteinband of 62 kDa was obtained in all samples, corresponding to the

602

expected size band of human COX-2 (Figure 4). This experiment

was carried out using an anti-COX-2 antibody, which does not

recognize COX-1. The COX-2 protein levels were substantially

increased by 2.5-fold in the steroid-treated samples compared with

the untreated control myocyte cultures (P 0.01).

Effect of steroids on PCNA

Cultured myometrial cells were treated with the combination of P 4

and E2. Immunoblotting of protein samples showed that culturedmyometrial cells contained immunoreactive PCNA, with the expected

molecular mass of ~36 kDa. Treatment for up to 24 h with P4 and

E2 combined, at a series of concentrations, optimized at 5 mol/l didnot cause a noticeable change in PCNA protein levels (Figure 5).

P4 and E2 effects on G-protein -subunits

Exposure of myometrial cell cultures for up to 24 h with the

combination of P4 and E2, at a series of concentrations, optimized at

5 mol/l, had profound effects on the expression of the subunitsof G-proteins (Figure 6).

Detection of Gq

Probing with QL, a specific antibody for q, 11(C-terminus), wedetected a single band at 42.5 kDa in myometrial cells. Estrogen and

P4 treatment reduced the expression of Gq by 3-fold compared with

the control (P 0.01; Figure 6a,b).

Detection of Gz

A similar effect was found on the subunit of Gz. Probing with aspecific antibody that does not cross-react with any of the other

G-proteins, we were able to detect z as a 39 kDa protein, whichwas down-regulated by almost 3-fold following treatment with P4and estrogen (P 0.01; Figure 6c,d).

Detection of Gs

Immunoblotting experiments, using a specific antibody against the

RM1 (C-terminus) region ofs, detected four s species of apparentmolecular weight 45, 47, 54 and 67 kDa in both treated and untreated

myometrial cells. P4 and estrogen treatment had no apparent effect

on the two large isoforms (54 and 67 kDa), whereas it reduced theexpression of the small isoforms (47 and 44 kDa) by 2.5- (P 0.01)

and 3-fold (P 0.01) respectively (Figure 6e,f).

Detection of Gi1/2For this study we used an antibody (AS/7) which recognizes both

i1 and i2. This antibody detected a band of 41 kDa in myometrialcells, indicating the presence of the G i1/2subunit. P4 and estrogentreatment reduced its expression by 2-fold when compared with the

control (P 0.01; Figure 6g,h).

Detection of G0Immunoblotting with a specific 0 antibody (GC/2) (N-terminus)detected a band of 40.5 kDa in the myometrial cell extract. P4 and

estrogen treatment caused a slight reduction in G0 expression when

compared with the control (Figure 6i,j).

Discussion

In the present study, we demonstrate that P4 and E2 influence the

expression of important intracellular and membrane-bound signalling

molecules in human myometrial cells in culture. During our experi-

mental design, the authors agreed to use P4 and E2, in combination,

to mimic the hormonal milieu during pregnancy, since the roles of

the two steroids in human pregnancy are not fully explained. Due to

limited evidence on the exact role of E2 in human pregnancy and on

the extensive cross-talk between P4and E2signalling pathways in all

reproductive tissues, the steroids were used in combination, in this


7/9


preliminary survey of potential responses, in an attempt to mimic the

steroid hormone milieu of pregnancy.

Estrogen is required for synthesis of P4 receptors (Aronica and

Katzenellenbogen, 1991) and estrogen receptor expression is estrogen-

dependent in reproductive tissues (Ing and Ott, 1999). Cross-talk

between estrogen and P 4 receptors has been reported, suggesting

interactions between the estrogen and P4 intracellular pathways

(Migliaccio et al., 1998; Katzenellenbogen, 2000).

In our study, P4

and E2

were used at relatively high concentrations

to mimic the hormonal milieu of pregnancy. P 4 production by the

placenta reaches ~300 mg per day at term (Perusquia, 2001), while

E2 increases almost by 100-fold during pregnancy (Mesiano, 2001).

The concentrations of P4 and E2 were optimized at 5 mol/l after aseries of doseresponse experiments (data not shown). Similar effects

were exerted by P4and E2at concentrations up to 10 mol/l, whereas

higher amounts of the two steroids appeared to have toxic effects.

A number of time-point experiments were also performed as part

of this study (data not shown). Myometrial cells were treated with

P4 and E2 for up to 16 h. The effects of these steroids on protein

kinases remained unchanged over the 16 h period, although some

effects were evident before this time-point. A short-term incubation

of up to 20 min could be part of a study on rapid, non-transcriptional

events of steroids. Although myocytes were treated with steroids forup to 24 h, it was decided that steroid treatment for 16 h was

not necessary for the purpose of our study. The change in the

phosphorylation state of p42/44 MAP kinase in MCF-7 cells is rapid

(Migliaccio et al., 1996), and therefore the cell line was treated with

steroids for only 15 min. Similarly, 15 min of treatment with A23187

is the time required to initiate membrane-linked, Ca2-associated

signalling (Chenet al., 1999).

Here we demonstrate that P4and E2 treatment of cultured myometr-

ial cells causes a down-regulation in the phosphorylation status of

p42/44 MAP kinase. This was in direct contrast to the calcium

ionophore A23187, which caused an increase of the two phosphoryl-

ated protein kinases, demonstrating the activation of MAP kinase in

the presence of increased Ca2 concentrations. The effect of P4 and

E2 is surprising in view of the fact that P 4 increases intracellularCa2 concentrations in myometrial cellsin vitro (Fomin et al., 1999)

and that E2 promotes Ca2 influx into myometrial cells (Wehling

et al., 1997) as part of the non-transcriptional events initiated by the

two steroid hormones in myometrial cells. This suggests that the

effects of P4and estrogen on MAP kinase phosphorylation status are

not due to direct effects on Ca2 mobilization. MAP kinase can also

be activated in myometrial cells by G-protein-coupled receptor ligands

such as oxytocin, endothelin and urocortin (Kimura et al., 1999;

Molnar et al., 1999; Grammatopoulos et al., 2000).

We used MCF-7 cells only for MAP kinase experiments due to

the fact that the cell line provides a very appropriate positive control,

as demonstrated in a plethora of studies (Migliaccio et al., 1996;

Castoria et al., 1999, Mougdil et al., 2001). The use of MCF-7 cells

also shows the considerable variety of effects steroids could cause indifferent cell types. The same cell line could not be used as a positive

control for the study of either SAPK/JNK (Caristi et al., 2001) or G

protein -subunit expression. MCF-7 cells express COX-2 (Liu andRose, 1996). However, these cells would not be an appropriate

positive control for the study of steroid effects on COX-2 expression.

In previous studies, we have shown a steroid-mediated increase of

COX-2 mRNA levels in human myometrial cells in culture (Zervou

et al., 1999b). In this study we confirm that this increase in mRNA

is translated into increased protein expression. Our findings support

the pivotal role that COX-2 has in myometrial physiology, as already

demonstrated by others (Slater et al., 1999a,b; Allport et al., 2001).

In our studies the combination of E2 and P4 did not cause a

603

noticeable change in the proliferation events of isolated human

myometrial cells, as revealed by PCNA immunodetection. Our

findings complement the work by Matsuo et al. and Maruo et al. in

terms of showing the combined effect of P4and E2on PCNA (Matsuo

et al., 1999; Maruo et al., 2000).

Protein kinases such as SAPK/JNK and MAP kinase are believed

to interact with a wide range of G protein subunits. Examples arethe activation of SAPK/JNK by Gi in HEK293 cells (Yamauchiet al., 2000). MAP kinase is also activated by Gi (Jo et al., 1997).Gi and Gs are believed to interact with Srcpathways, which areclosely related to MAP kinase pathways (Ram and Lyengar, 2001).

Some G protein-coupled receptors are able to activate JNK in certain

cell types (Naor et al., 2000). The Gi-coupled m2 muscarinic

acetylcholine receptor activates JNK in COS-7 cells (Coso et al.,

1996). G13 has also been shown to stimulate the COX-2 promoterin NIH3T3 cells (Slice et al., 1999), but no data are available on the

link between G subunits and COX-2 expression.In the human myometrium, a steroid-mediated expression of

cytoplasmic and membrane-linked components would cause a shift

in G protein coupling to certain G protein-coupled receptors as part

of tissue remodelling. A number of signal transduction pathways can

be either activated or inactivated, leading to a change in smooth

muscle contractile state.

This study demonstrates for the first time that ovarian steroids

regulate expression of the subunits of G-proteins in primary humanmyocyte cultures. Treatment with P4 and E2 markedly decreased

expression of Gi1/2, Gz, Gq and Gs and to a lesser extent G0.

Interestingly, of the four isoforms of the Gs -subunit, only thesmaller isoforms appeared to be down-regulated by P4and E2. It is

attractive to speculate that one of the mechanisms that might contribute

to their down-regulation would be the effects of P4 and E2. It is

known that the levels of Gs fall at the onset of parturition (Europe-Finneret al., 1994). Moreover the fact that short isoforms are affected

provides further evidence for the importance of Gs alternativesplicing. Europe-Finner and colleagues have shown that alternative

splicing of Gs precursor mRNA has a primary role in regulatingexpression of Gs protein isoforms during pregnancy and labour

(Europe-Finner et al., 1997).

It appears, therefore, that the -subunits of G-proteins in humanmyometrial cells are physiological targets for P4and E2 in vitro. Our

data are in agreement with previous findings on steroid-mediated

expression of G-protein -subunits in other tissues. In lactotropes,combined treatment with P4and E2leads to decreased Gi/G0amounts

(Livingstone et al., 1998). Moreover, in-vivo administration of P4 in

rat myometrium has been shown to significantly reduce the amounts

of Gq (Cohen-Tannoudji et al., 1995).

Classically, steroids exert their effects transcriptionally through

nuclear receptors. However, recent evidence shows that steroids can

influence membrane physicochemical properties (Wehling, 1997). P4treatment has been shown to influence calcium signalling evoked by

ligand stimulation of G-protein coupled receptors expressed in several

cell lines (Burger et al., 1999). Many signalling mechanisms initiatedby peptide hormone receptors can also be activated by membrane

actions of steroid hormones (Watson and Gametchu, 1999). Such

membrane-initiated responses on G-protein activity with subsequent

effects on a number of signal transduction pathways could eventually

influence transcriptional events, critical for the physiological and

biochemical responses of myocytes during pregnancy.

Collectively, our findings indicate a number of novel P4- and

E2-mediated effects on signal transduction pathways involving

intracellular as well as membrane-bound components of human

myometrial cells. These two steroid hormones could potentially

modify the protein levels of these components and may be linked to

myometrial tissue remodelling due to pregnancy.


8/9

S.Zervouet al.

AcknowledgementsThe authors would like to thank consultant gynaecologists and theatre staffat the University Hospitals of Coventry and Warwickshire, NHS Trust, WestMidlands. This work was funded by the Sir Jules Thorn Charitable Trust(96/02A) and by the Wellcome Trust. E.W.H. is the WPH Charitable TrustChair of Medicine.

References

Allport, V.C., Pieber, D., Slater, D.M., Newton, R., White, J.O. and Bennett,P.R. (2001) Human labour is associated with nuclear-kappaB activity whichmediates cyclo-oxygenase-2 expression and is involved with thefunctionalprogesterone withdrawal. Mol. Hum. Reprod., 7, 581586.

Aronica, S.M. and Katzenellenbogen, B.S. (1991) Progesterone receptorregulation in uterine cells: stimulation by estrogen, cyclic adenosine 3,5-monophosphate, and insulin-like growth factor I and suppression byantiestrogens and protein kinase inhibitors. Endocrinology, 128, 20452052.

Burger, K., Fahrenholz, F. and Gimpl, G. (1999) Non-genomic effects ofprogesterone on the signaling function of G protein-coupled receptors.FEBS Lett., 464 , 2529.

Caristi, S., Galera, J.L., Matarese, F., Imai, M., Caporali, S., Cancemi, M.,Altucci, L., Cicatiello, L, Teti, D., Bresciani, F. et al. (2001) Estrogens donot modify MAP kinase-dependent nuclear signaling during stimulation ofearly G(1) progression in human breast cancer cells. Cancer Res., 61,63606366.

Castoria, G., Barone, M.V., Di Domenico, M., Bilancio, A., Ametrano, D.,

Migliaccio, A. and Auricchio, F. (1999) Non-transcriptional action ofestradiol and progestin triggers DNA synthesis. EMBO J., 18 , 25002510.

Chen, Z., Yuhanna, I.S., Galcheva-Gargova, Z., Karas, R.H., Mendelsohn, M.E.and Shaul, P.W. (1999) Estrogen receptor alpha mediates the nongenomicactivation of endothelial nitric oxide synthase by estrogen. J. Clin. Invest.,103, 401406.

Cohen-Tannoudji, J., Mhaouty, S., Elwardy-Merezak, J., Lecrivain, J.L., Robin,M.T., Legrand, C. and Maltier, J.P. (1995) Regulation of myometrial Gi2,Gi3, and Gq expression during pregnancy. Effects of progesterone andestradiol.Biol. Reprod., 53 , 5564.

Coso, O.A., Teramoto, H., Simonds, W.F. and Gutkind, J.S. (1996) Signalingfrom G protein-coupled receptors to c-Jun kinase involves beta gammasubunits of heterotrimeric G proteins acting on a Ras and Rac1-dependentpathway. J. Biol. Chem., 271 , 39633966.

Csapo, A.M. and Pinto-Dantas, C.A. (1965) The effect of progesterone onthe human uterus. Proc. Natl Acad. Sci. USA, 54 , 10691076.

Derijard, B., Hibi, M., Wu, I.H., Barrett, T., Su, B., Deng, T., Karin, M. andDavis, R.J. (1994) JNK1: a protein kinase stimulated by UV light and Ha-Ras that binds and phosphorylates the c-Jun activation domain . Cell, 76,10251037.

Europe-Finner, G.N., Phaneuf, S., Watson, S.P. and Lopez Bernal, A. (1993)Identification and expression of G proteins in human myometrium: up-regulation of G alpha s in pregnancy. Endocrinology, 132 , 24842490.

Europe-Finner, G.N., Phaneuf, S., Tolkovsky, A.M., Watson, S.P. and LopezBernal, A. (1994) Down-regulation of G alpha s in human myometrium interm and preterm labour: a mechanism for parturition. J. Clin. Endocrinol.

Metab., 79 , 18351839.

Europe-Finner, G.N., Phaneuf, S., Cartwright, E. Mardon, H.J. and LopezBernal, A. (1997) Expression of human myometrial G alpha s messengerribonucleic acid transcript during pregnancy and labour: involvement ofalternative splicing pathways. J. Mol. Endocrinol., 18 , 1525.

Fomin, V.P., Cox, B.E. and Word, R.A. (1999) Effect of progesterone onintracellular calcium homeostasis in human myometrial cells. Am. J.

Physiol., 276 , C379C385.Grammatopoulos, D.K., Randeva, H.S., Levine, M.A., Katsanou, E.S. and

Hillhouse, E.W. (2000) Urocortin, but not corticotropin-releasing hormone(CRH), activates the mitogen-activates protein kinase signal transductionpathway in human pregnant myometrium: an effect mediated via R1and R2 CRH receptor subtypes and stimulation of Gq-proteins. Mol.

Endocrinol., 14 , 20762091.

Hibi, M., Lin, A., Smeal, T., Minden, A. and Karin, M. (1993) Identificationof an oncoprotein- and UV-responsive protein kinase that binds andpotentiates the c-Jun activation domain. Genes Dev., 7, 21352148.

Hla, T. and Neilson, K. (1992) Human cyclooxygenase-2 cDNA. Proc. NatlAcad. Sci. USA, 89 , 73847388.

Ing, N.H. and Ott, T.L. (1999) Estradiol up-regulates estrogen receptor-alphamessenger ribonucleic acid in sheep endometrium by increasing its stability.

Biol. Reprod., 60 , 134139.

604

Jo, H., Sipos, K., Go, Y.M., Law, R., Rong, J. and McDonald, J.M. (1997)Differential effect of shear stress on extracellular signal-regulated kinaseand N-terminal Jun kinase in endothelial cells. Gi2- and Gbeta/gamma-dependent signaling pathways. J. Biol. Chem., 272 , 13951401.

Karteris, E., Grammatopoulos, D., Randeva, H. and Hillhouse, E.W. (2000)Signal transduction characteristics of the corticotropin-releasing hormonereceptors in the feto-placental unit. J. Clin. Endocrinol. Metab., 85,19891996.

Katzenellenbogen, B.S. (2000) Mechanisms of action and cross-talk betweenestrogen receptor and progesterone receptor pathways. J. Soc. Gynecol.

Investig., 7, S33S37.Kimura, A., Ohmichi, M., Takeda, T., Kurachi, H., Ikegami, H., Koike, K.,

Masuhara, K., Hayakawa, J., Kanzaki, T., Kobayashi, M. et al. (1999)Mitogen-activated protein kinase cascade is involved in endothelin-1-induced rat puerperal uterine contraction. Endocrinology, 140 , 722731.

Kyriakis, J.M. and Avruch, J. (1990) pp54 microtubule-associated protein 2kinase. A novel serine/threonine protein kinase regulated by phosphorylationand stimulated by poly-L-lysine. J. Biol. Chem., 265 , 1735517363.

Liu, X.H. and Rose, D.P. (1996) Differential expression and regulation ofcyclooxygenase-1 and -2 in two human breast cancer cell lines. Cancer

Res., 56 , 51255127.

Livingstone, J.D., Lerant, A. and Freeman, M.E. (1998) Ovarian steroidsmodulate responsiveness to dopamine and expression of G-proteins inlactotropes. Neuroendocrinology, 68, 172179.

Maggi, M., Magini, A., Fiscella, A., Giannini, S., Fantoni, G., Toffoletti, F.,Massi, G. and Serio, M. (1992) Sex steroid modulation of neurophysialhormone receptors in human nonpregnant myometrium.J. Clin. Endocrinol.

Metab., 74 , 385392.Maruo, T., Matsuo, H., Samoto, T., Shimomura, Y., Kurachi, O., Gao, Z.,

Wang, Y., Spitz, I.M. and Johansson, E. (2000) Effects of progesterone onuterine leiomyoma growth and apoptosis. Steroids, 65, 585592.

Matsuo, H., Kurachi, O., Shimomura, Y., Samoto, T. and Maruo, T. (1999)Molecular bases for the actions of ovarian sex steroids in the regulation ofproliferation and apoptosis of human uterine leiomyoma. Oncology, 57(Suppl. 2), 4958.

Mazor, M., Wiznitzer, A., Levy, J., Sharoni, Y., Meril, Z., Minster, A. andGlezerman, M. (1993) The relationship between estrogen/progesterone ratioand term human parturition. Isr. J. Med. Sci., 29 , 9799.

Mesiano, S. (2001) Roles of estrogen and progesterone in human parturition.In Smith, R. (Ed.) The Endocrinology of Parturition. Basic Science andClinical Application. Front. Horm. Res., Basel, Karger.

Migliaccio, A., Di Domenico, M., Castoria, G., de Falco, A., Bontempo, P.,Nola, E. and Auricchio, F. (1996) Tyrosine kinase/p21ras/MAP-kinase

pathway activation by estradiol-receptor complex in MCF-7 cells. EMBOJ., 15, 12921300.

Migliaccio, A., Piccolo, D., Castoria, G., Di Domenico, M., Bilancio, A.,Lombardi, M., Gong, W., Beato, M. and Auricchio, F. (1998) Activationof the Src/p21ras/Erk pathway by progesterone receptor via cross-talk withestrogen receptor. EMBO J., 17, 20082018.

Molnar, M., Rigo, J. Jr, Romero, R. and Hertelendy, F. (1999) Oxytocinactivatesmitogen-activated protein kinase and up-regulates cyclooxygenase-2 and prostaglandin production in human myometrial cells. Am. J. Obstet.Gynecol., 181 , 4249.

Moonen, P., Klok, G. and Keirse, M.J. (1984) Increase in concentrations ofprostaglandin endoperoxide synthase and prostacyclin synthase in humanmyometrium in late pregnancy. Prostaglandins, 28 , 309321.

Moonen, P., Klok, G. and Keirse, M.J. (1985) Immunohistochemicallocalisation of prostaglandin endoperoxide synthase and prostacyclinsynthase in pregnant human myometrium.Eur. J. Obstet. Gynecol. Reprod.

Biol.,19 , 151158.

Moudgil, V.K., Dinda, S., Khattree, N., Jhanwar, S., Alban, P. and Hurd, C.(2001) Hormonal regulation of tumor suppressor proteins in breast cancercells. J. Steroid Biochem. Mol. Biol.,76 , 105117.

Naor, Z., Benard, O. and Seger, R. (2000) Activation of MAPK cascades byG-protein-coupled receptors: the case of gonadotropin-releasing hormonereceptor. Trends Endocrinol. Metab., 11, 9199.

Nohara, A., Ohmichi, M., Koike, K., Masumoto, N., Kobayashi, M., Akahane,M., Ikegami, H., Hirota, K., Miyake, A. and Murata, Y. (1996) The roleof mitogen-activated protein kinase in oxytocin-induced contraction ofuterine smooth muscle in pregnant rat. Biochem. Biophys Res. Commun.,229, 938944.

Ohmichi, M., Koike, K., Nohara, A., Kanda, Y., Sakamoto, Y., Zhang, Z.X.,Hirota, K. and Miyake, A. (1995) Oxytocin stimulates mitogen-activatedprotein kinase activity in cultured human puerperal uterine myometrialcells.Endocrinology, 136 , 20822087.


9/9


Ohmichi, M., Koike, K., Kimura, A. Masuhara, K., Ikegami, H., Ikebuchi,Y., Kanzaki, T., Touhara, K., Sakaue, M., Kobayashi, Y. et al. (1997) Roleof mitogen-activated protein kinase pathway in prostaglandin F2alpha-induced rat puerperal uterine contraction. Endocrinology, 138 , 31033111.

Perusquia, M. (2001) Nongenomic action of steroids in myometrialcontractility. Endocrine, 15, 6372.

Phaneuf, S., Europe-Finner, G.N., Varney, M., MacKenzie, I.Z., Watson,S.P. and Lopez Bernal, A. (1993) Oxytocin-stimulated phosphoinositidehydrolysis in human myometrial cells: involvement of pertussis toxin-sensitive and -insensitive G proteins. J. Endocrinol., 136 , 497509.

Ram, P.T. and Iyengar, R. (2001) G protein coupled receptor signaling throughthe Src and Stat3 pathway: role in proliferation and transformation.Oncogene, 20 , 16011606.

Rupnow, H.L., Phernetton, T.M., Shaw, C.E., Modrick, M.L., Bird, I.M. andMagness, R.R. (2001) Endothelial vasodilator production by uterine andsystemic arteries. VII. Estrogen and progesterone effects on eNOS. Am. J.Physiol. Heart Circ. Physiol.,280 , H1699H1705.

Seger, R. and Krebs E.G. (1995) The MAPK signaling cascade. FASEB J.,9,726735.

Slater, D.M., Dennes, W., Sawdy, R., Allport, V. and Bennett, P. (1999a)Expression of cyclo-oxygenase types-1 and 2 in human fetal membranesthroughout pregnancy. J. Mol. Endocrinol., 22 , 125130.

Slater, D.M., Dennes, W.J., Campa, J.S., Poston, L. and Bennett, P.R. (1999b)Expression of cyclo-oxygenase types-1 and -2 in human myometriumthroughout pregnancy. Mol. Hum. Reprod., 5, 880884.

Slice, L.W., Walsh, J.H. and Rozengurt, E. (1999) Galpha(13) stimulates Rho-

dependent activation of the cyclooxygenase-2 promoter. J. Biol. Chem.,274, 2756227566.

Sparey C., Robson, S.C., Bailey, J., Lyall, F. and Europe-Finner, G.N. (1999)The differential expression of myometrial connexin-43, cyclooxygenase-1

605

and -2, and Gs alpha proteins in the upper and lower segments of thehuman uterus during pregnancy and labour. J. Clin. Endocrinol. Metab.,84, 17051710.

Vagnoni, K.E., Shaw, C.E., Phernetton, T.M., Meglin, B.M., Bird, I.M. andMagness, R.R. (1998) Endothelial vasodilator production by uterine andsystemic arteries. III. Ovarian and estrogen effects on NO synthase. Am. J.Physiol., 275 , H1845H1856.

Watson, C.S. and Gametchu, B. (1999) Membrane-initiated steroid actionsand the proteins that mediate them. Proc. Soc. Exp. Biol. Med.,220 , 919.

Wehling, M. (1997) Specific, nongenomic actions of steroid hormones. Ann.

Rev. Physiol., 59 , 365393.Xiao, C.W., Liu, J.M., Sirois, J. and Goff, A.K. (1998) Regulation ofcyclooxygenase-2 and prostaglandin F synthase gene expression by steroidhormones and interferon-in bovine endometrial cells. Endocrinology,139,22932299.

Xie, W. and Hershman, H.R. (1995) v-src induces prostaglandin synthase-2gene expression by activation of the c-Jun N-terminal kinase and the c-Juntranscription factor. J. Biol. Chem., 270 , 2762227628.

Yamauchi, J., Kaziro, Y. and Itoh, H. (1995) Carboxyl terminal of G proteinbeta subunit is required for association with gamma subunit. Biochem.

Biophys Res. Commun.,214 , 694700.

Zervou, S., Klentzeris, L.D. and Old, R.W. (1999a) Nitric oxide synthaseexpression and steroid regulation in the uterus of women with menorrhagia.

Mol. Hum. Reprod., 5, 10481054.

Zervou, S., Klentzeris, L.D. and Old, R.W. (1999b) Steroid effects ofmyometrial prostaglandin endoperoxideH-synthase-2 and oxytocin receptor.Abstract, 81st Annual Meeting of the Endocrine Society, The Endocrine

Society, San Diego, USA. p192.

Submitted on June 25, 2001; resubmitted on December 28, 2001; acceptedon April 17, 2002

Documents

Mol. Hum. Reprod. 2002 Zervou 597 605