249 Alteration of prostaglandin production and agonist responsiveness by n-6 polyunsaturated fatty acids in endometrial cells from late-gestation ewes
Z Cheng, M Elmes, S E Kirkup,DREAbayasekara and D C Wathes Reproduction and Development Group, Royal Veterinary College, Hawkshead Lane, North Mymms, Hatfield, Herts AL9 7TA, UK (Requests for offprints should be addressed to D C Wathes; Email [email protected])
Abstract We investigated the effect of n-6 polyunsaturated fatty alterations in responsiveness as a result of PUFA treatment. acids (PUFAs) on prostaglandin (PG) production by In the cells supplemented with 100 µM AA, there was no the uterus. A mixed population of endometrial cells further increase in PGF2 output in the presence of OT or (epthelium and stroma) from late-gestation ewes were LPS and when 100 µM GLA was present neither LPS nor cultured in defined medium containing linoleic acid OT stimulated PGE2 significantly. When LPS was given (LA, 18:2, n-6), -linolenic acid (GLA, 18:3, n-6) or to AA-supplemented cells, the E:F ratio was increased. arachidonic acid (AA, 20:4, n-6) in concentrations of 0 DEX did not change PGE2 production in control or (control), 20 or 100 µM. After 45 h in test medium with or LA-treated cells, but the cells produced significantly less without added PUFAs, cells were challenged with control PGF2, so the E:F ratio was increased. In contrast, in medium (CM), oxytocin (OT, 250 nM), lipopolysaccha- GLA- and AA-treated cells, DEX reduced the production ride (LPS, 0·1 µg/ml) or dexamethasone (DEX, 5 µM) for of both PGF2 and PGE2, so the E:F ratio was unaltered. 22 h in the continued presence of the same concentration In summary, the study showed altered production of PGs of PUFA and the medium was collected for measurement in the presence of different PUFAs according to their of PGF2 and PGE2. Supplementation with LA inhibited position in the n-6 metabolic pathway. The type of PUFA ff the production of PGF2 but did not alter PGE2, whereas present a ected responsiveness to OT, LPS and DEX and GLA and AA increased production of both PGs. All PUFA also changed the ratio of PGE2 to PGF2 produced. supplements thus increased the ratio of PGE2 to PGF2 The possible implications of this work are discussed in (E:F ratio) two- to threefold. In control cells, OT and relation to the effect of diet on term and pre-term labour, LPS challenges stimulated the production of PGF2 and which both require upregulation of the endometrial PG PGE2. In all challenge groups, the concentrations of synthetic pathway. PGF2 in response to PUFAs followed the same pattern – Journal of Endocrinology (2004) 182, 249–256 LA Introduction as the precursors for a variety of signalling molecules and influence many fundamental cell processes. LA is an Cardiovascular disease is the principal cause of death essential fatty acid, which is metabolised in the body to among the human population of developed countries, with GLA, dihomo--linolenic acid (DGLA, 20:3, n-6) and the saturated fat content of the diet identified as one of arachidonic acid (AA, 20:4, n-6) by the position-specific the major risk factors (Department of Health 1995). This 6 and 5 desaturase and coenzyme A (CoA)-dependent finding has promoted a trend to reduce dietary intakes of chain elongases (Sprecher 2000). Cyclo-oxygenases meat and dairy products in favour of a substantial increase (COX) catalyse conversion of DGLA and AA into 1- and in vegetable oils, which are high in n-6 polyunsaturated 2-series prostaglandins (PGs) respectively. fatty acids (PUFAs), notably linoleic acid (LA, 18:2, n-6). In the n-6 PUFA metabolic pathway from LA to AA, Oils containing -linolenic acid (GLA, 18:3, n-6), such as there are two rate-limiting catalytic steps, 6 and 5 primrose oil, are also promoted as health products, because desaturases. The amount of LA converted to GLA is tiny of their potential anti-inflammatory and anti-proliferative and only a small fraction of GLA can be catalysed into AA properties (Fan & Chapkin 1998). Dietary PUFAs provide (Fan & Chapkin 1998). Supplementation with GLA energy and components for cell membranes. They also act bypasses the rate-limiting 6 desaturase and GLA is Journal of Endocrinology (2004) 182, 249–256 Online version via http://www.endocrinology.org 0022–0795/04/0182–249 2004 Society for Endocrinology Printed in Great Britain Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access 250 Z CHENG and others · Effects of PUFAs on endometrial PG production quickly elongated to DGLA by elongase. DGLA can in several agonists known to play an important part in natural turn be converted into 1-series PGs, such as PGF1 and or pre-term labour were also examined by testing the PGE1. Supplementation with AA avoids the limitation of responsiveness of the cells to OT, LPS and DEX. 5 desaturase and AA is directly converted to 2-series PGs, such as PGF2 and PGE2, by COX enzymes (Lapetina 1982). This suggests that supplementary PUFAs Materials and Methods may alter the n-6 PUFA metabolic pathway and that the precise effects of PUFAs on PG production will depend on All reagents were from Sigma (Poole, Dorset, UK) or their position in the pathway. BDH Merck Ltd (Lutterworth, Leics, UK) unless other- Considerable attention has been paid to the effects of wise stated. Solutions were sterilised by passage through a PUFA supplementation on carcinogenesis, because PGs 0·20 µm filter (Nalge Nunc International, Rochester, NY, have an important role in the regulation of tumour growth USA or Millipore Company, Bedford, MA, USA). All (Bishop-Bailey et al. 2002, Dommels et al. 2002). To date, culture media used included 50 000 units/l penicillin and relatively little information is available about the possible 50 mg/l streptomycin. effects of increased n-6 PUFA concentrations on the production of PG in relation to reproductive processes Cell isolation and culture (Abayasekara & Wathes 1999). Previous studies have produced contradictory results, in that feeding dairy cows Gravid uteri were obtained on day 135 of gestation (term a diet high in LA caused either a decreased (Thatcher et al. is approximately 145 days). Ewes were killed using an 1994, Cheng et al. 2001a) or an increased (Robinson overdose of pentabarbitone and the reproductive tract was et al. 2002) production of 2-series PG by the uterine collected. endometrium. Graham et al. (1994) found a reduced Endometrial cells were isolated and cultured under ability of the endometrium to produce PGF2 and PGE2 sterile conditions in a laminar flow hood, following in women who had been taking a dietary supplement of methods described previously (Cheng et al. 2003). From GLA for 6 months. the intercotyledonary region of the uterus, maternal Upregulation of the 2-series PGs, in particular PGE2 endometrium was separated, cut into strips and put into and PGF2, is an essential component of parturition serum-free Dulbecco’s minimal essential medium/F12 1:1 (Challis 2000). In the ewe, activation of the fetal nutrient mix (DMEM/F12 medium, Gibco Life hypothalamic–pituitary–adrenal axis in late gestation leads Technologies, Paisley, Strathclyde, UK) containing to increased production of PGE2 by the placenta and 1·125 g/l fatty-acid-free bovine serum albumin (BSA, subsequent initiation of parturition (Thorburn 1991). In Sigma) and 1·125 g/l sodium bicarbonate at pH 7·3. The phase 1 of labour, the increased cortisol output stimulates strips were then chopped into 1 mm3 cubes using the conversion of androgen to oestrogen in the placenta, a McIlwain mechanical tissue chopper (McIlwain upregulates the expression of COX-2 and downregulates Laboratory Engineering, Guilford, Surry, UK). Two the production of PG dehydrogenase (PGDH), resulting in aliquots of 20 g chopped tissue were placed into two falcon increased production of PGE2 (Challis et al. 2002). In vials and mixed with 30 ml digestive solution. This phase 2 of labour, oxytocin (OT) receptors are upregulated contained 0·05% trypsin III (Roche, Diagnostic GmbH, in the intercotyledonary endometrium (Leung et al. 1999) Mannheim, Germany), 0·05% collagenase II (Roche) and and mediate increased release of PGF2 in response to 0·1% BSA (Sigma) in Hank’s Balanced Saline (HBS, OT (Jenkin 1992), resulting in uterine contractions and Sigma). After incubation for 2 h at 37 C with shaking, delivery (Whittle et al. 2001, Challis et al. 2002). Synthetic the suspensions were filtered through a 100 µm mesh. glucocorticoids such as dexamethesone (DEX) can mimic The filtrates were pooled into a 50 ml falcon vial and the endogenous increase in cortisol and induce parturition re-suspended with washing medium [HBS containing (Challis et al. 2002). Pregnant women given exogenous 10% (v/v) fetal calf serum (FCS, Sigma) and 3 µg/ml corticosteroids to promote fetal maturation have an trypsin inhibitor (Sigma)]. The vial was then centrifuged at increased incidence of pre-term birth (Challis et al. 2002). 100 g and 4 C for 10 min. The supernatant was removed Intrauterine infection is another major risk factor in and the pellet re-suspended with washing medium. After pre-term labour (Goncalves et al. 2002). The bacterial two repetitions of the above washing procedures, the cell endotoxin, lipopolysaccharide (LPS), increased the pellets containing mixed epithelial and stromal cells were activities of phospholipase (PL) A2 and COX-2 and re-suspended in DMEM/F12 medium containing 10% inhibited PGDH (Lamont et al. 1990, Flynn & Hoff 1995, FCS to a concentration of 0·5106 cells/ml. This cell van Meir et al. 1997). Administration of LPS to pregnant suspension was plated (2 ml/well) into 24-well IWAKI mice provoked premature delivery (Romero et al. 1991). microplates (Scitech DIV, Asahi Techno Glass, Japan). The aim of the present study was, therefore, to compare Cells were cultured in a humidified incubator at 39 C ff the e ects of LA, GLA and AA on the generation of PG by with 5% CO2. Culture medium was changed every 48 h the ovine endometrium in late gestation. The influence of for 7–8 days to allow the cells to grow to confluence. Journal of Endocrinology (2004) 182, 249–256 www.endocrinology.org Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access Effects of PUFAs on endometrial PG production · Z CHENG and others 251 PUFA supplementation and cell challenge Statistical analysis LA, GLA and AA (Sigma) were dissolved in 100% ethanol The data are expressed as meansS.E. unless otherwise to a concentration of 100 mM. Further dilution to 10 mM stated. Analysis of variance (ANOVA) was carried out was made in dilution medium (serum-free DMEM/F12 using a mixed-procedure built-in SAS software package containing 0·1125% fatty-acid-free BSA as a carrier). OT (Version 8.0, SAS Institute Inc., Cary, NC, USA). The (Sigma) was dissolved in 0·01 M acetic acid at 5 mM then differences between PUFA treatments, challenges and further diluted to 500 µM in dilution medium. LPS from their interactions were taken as the fixed effects and sheep Escherichia coli 026.B6 (Sigma) was dissolved in distilled as the random effect. When a significant difference water at 500 µg/ml then diluted to 50 µg/ml in dilution (P<0·05) was achieved, the least square means based on medium. DEX (Sigma) was dissolved in 100% ethanol to Fisher’s LSD for multiple comparisons were calculated to 20 mM, then diluted to 2 mM in dilution medium. test the differences between the fixed effects. Before PUFA supplementation, confluent cells were incubated with 2 ml serum-free test medium containing 0·1125% BSA and ITS (0·5 µg/ml insulin, 0·5 µg/ml transferrin and 0·5 ng/ml selenium; Sigma) for 3 h to Results remove any fatty acids present in the FCS. The cells were then grown in 2 ml PUFA-supplemented test medium for Effects of n-6 PUFAs on PG production in 45 h. The PUFA treatments were LA, GLA or AA at 0 non-challenged cells (control), 20 or 100 µM, formulated in the test medium. After this initial culture period, cells cultured in 0 or There were significant differences between the PUFA 100 µM PUFA were challenged with control medium treatments with respect to both PGF2 and PGE2 concen- (CM), OT (250 nM), LPS (0·1 µg/ml) or DEX (5 µM) in trations (Fig. 1). Compared with control, supplementation 1 ml of PUFA-supplemented test medium for 22 h. The with LA dose-dependently inhibited the generation of ff < doses of PUFA were selected to fall within the e ective PGF2 (P 0·05–0·01), whereas GLA and AA increased its range, after preliminary work using a wider dose–response production by up to 14-fold (P<0·001). There were no ff curve. The maximum ethanol concentration in the incu- significant di erences in PGF2 concentrations between bation medium was always less than 0·05%. Spent medium 20 µM and 100 µM GLA or AA supplementation. AA at was stored at 20 C until analysis for PGs. There were 20 and 100 µM stimulated significantly greater generation four replicate wells per treatment for each ewe, and all of PGF2 than did GLA at the same concentrations treatments were repeated in cells isolated from at least (P<0·05–0·01). three separate ewes. LA supplementation did not alter the production of PGE2 significantly, although the PGE2 concentration was slightly lower in cells treated with 100 µM LA. PG RIA Supplementation with GLA increased PGE2 production by up to 15-fold (P<0·01), supplementation with 100 µM PGE and PGF in the spent medium were quantified 2 2 than using charcoal–dextran-coated RIA methods as described producing significantly greater generation of PGE2 supplementation with 20 µM (P<0·01). In the cells sup- previously (Cheng et al. 2001a). The samples were diluted plemented with 20 and 100 µM AA, the PGE (20–500) in 0·05 M Tris buffer containing 0·1% gelatin 2 concentrations were up to 40 times greater than those in and 0·01% sodium azide. At this dilution, the presence of control cells (P<0·001). Compared with the GLA- the control medium had negligible effect on the RIAs supplemented cells, those supplemented with AA (<1%), so no extraction of the samples was required. generated significantly more PGE2 at the same PUFA The tritiated tracers of PGE2 ([5, 6, 8, 11, 12, 14, 15 3 concentrations (P<0·01). (n)- H]-PGE2) and PGF2 ([5, 6, 8, 9, 11, 12, 14, 15 3 In the cells cultured with PUFA-free medium, similar (n)- H]-PGF2) were purchased from Amersham amounts of PGF2 and PGE2 were produced, indicated by International plc (Amersham, Bucks, UK). The standards a PGE2 to PGF2 (E:F) ratio of 1·0 0·11 (Fig. 1). PUFA of PGE2 and PGF2 were supplied by Sigma. The antisera to these two PGs were a kind gift from DrNLPoyser supplementation alone (LA, GLA and AA) significantly (University of Edinburgh, Edinburgh, UK) (Poyser 1987). increased this ratio, two- to threefold. Their cross-reactivities were as follows: PGF2 antiserum, 34% with PGF1 and 25% with PGF3; PGE2 antiserum, 23% with PGE1 and 15% with PGE3. The limit of Effect of challenges on PG production detection was 2 pg/tube for PGE2 and 1 pg/tube for ffi PGF2. The intra- and interassay coe cients of variation Cells stimulated with OT in PUFA-free medium were 3·5% and 6·3% for PGE2 (n=6), and 4·1% and 9·6% increased their release of both PGF2 and PGE2 about for PGF2 (n=6) respectively. twofold in equal proportion, resulting in an E:F ratio of www.endocrinology.org Journal of Endocrinology (2004) 182, 249–256 Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access 252 Z CHENG and others · Effects of PUFAs on endometrial PG production Figure 2 Effects of n-6 PUFAs on the production of PGs by ovine endometrial cells after OT challenges. Confluent cells were preincubated for 45 h in either control (PUFA-free) medium or Figure 1 Theeffectofn-6 PUFA supplementation on the with 100 M LA, GLA or AA. Cells were then treated with control production of PGs by ovine endometrial cells. Cells were medium (M) or with 250 nM OT () for 22 h. The line on the supplemented with no PUFA (0, CONT) or 20 or 100 M LA, GLA bottom graph marks an E:F ratio of 1, indicating that equivalent or AA. There were three ewes per treatment and four replicates amounts of each PG were being produced. There were three per ewe. Values are expressed as meansS.E.M. over a 22 h ewes per treatment and four replicates per ewe. Values are culture period. Data were analysed by split ANOVA. The line on expressed as meansS.E.M. Data were analysed by split ANOVA. the bottom graph marks an E:F ratio of 1, indicating that Significant differences attributable to OT treatment: *P<0·05, equivalent amounts of each PG were being produced. Different **P<0·01; significant differences attributable to PUFA treatment: letters indicate significant difference at P<0·05–0·0001 a 1·20·18 (Fig. 2). This response was attenuated by LA. The combination of LPS in the presence of AA produced The cells produced more prostaglandins in the presence of a further significant increase in the E:F ratio, to both GLA and AA, but the further response of PGF2 to 3·5 0·47 (Fig. 3). OT was no longer significant in cells cultured with AA In cells treated with DEX in PUFA-free medium or and the response of PGE2 to GLA was likewise lost in the with LA, PGF2 was reduced, whereas PGE2 was presence of GLA. There was a trend for all three n-6 unaltered, resulting in a significant increase in the E:F PUFAs tested to increase the E:F ratio, but this was only ratio, to more than 3 (Fig. 4). In the presence of GLA and significant for GLA (2·8 0·27; Fig. 2). OT treatment did AA, DEX reduced the production of both PGF2 and not alter the ratio. PGE2, so the E:F ratio was no longer altered. Cells stimulated with LPS in PUFA-free medium also increased release of both prostaglandins and, again, the E:F ratio remained unaltered, at about 1 (0·90·11; Fig. 3). Discussion The production of PGF2 in response to LPS was decreased in the presence of LA. In the presence of AA, In this study we investigated the effect of n-6 PUFA cells could no longer respond to LPS by further production supplementation on the production of prostaglandin by of PGF2. The response of PGE2 to LPS was likewise lost ovine endometrial cells from late-gestation ewes, and in the presence of GLA. There was a progressive increase examined how PUFA treatment affected the responsive- in the E:F ratio from control medium Journal of Endocrinology (2004) 182, 249–256 www.endocrinology.org Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access Effects of PUFAs on endometrial PG production · Z CHENG and others 253 Figure 3 Effects of n-6 PUFAs on the production of PGs by ovine Figure 4 Effects of n-6 PUFAs on the production of PGs by ovine endometrial cells after LPS challenges. Confluent cells were endometrial cells after DEX challenges. Confluent cells were preincubated for 45 h in either control (PUFA-free) medium or preincubated for 45 h in either control (PUFA-free) medium or with 100 M LA, GLA or AA. Cells were then treated with control with 100 M LA, GLA or AA. Cells were then treated with control medium (M)orwith0·1g/ml LPS () for 22 h. The line on the medium (M)orwith5M DEX () for 22 h. The line on the bottom graph marks an E:F ratio of 1, indicating that equivalent bottom graph marks an E:F ratio of 1, indicating that equivalent amounts of each PG were being produced. There were three amounts of each PG were being produced. There were three ewes per treatment and four replicates per ewe. Values are ewes per treatment and four replicates per ewe. Values are expressed as meansS.E.M. Data were analysed by split ANOVA. expressed as meansS.E.M. Data were analysed by split ANOVA. Significant differences attributable to LPS treatment: *P<0·05, Significant differences attributable to DEX treatment: *P<0·05, **P<0·01; significant differences attributable to PUFA treatment: **P<0·01; significant differences attributable to PUFA treatment: a Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access 254 Z CHENG and others · Effects of PUFAs on endometrial PG production GLA to cultured endometrial cells from non-pregnant isomerases/synthases. The amount and ratio of PGF2 ewes or amnion cells, as assessed by separating PG and PGE2 produced are therefore dependent on fractions on HPLC (Z Cheng, S E Kirkup, D C Wathes & both COX activity and the specific PGF and PGE D R E Abayasekara, unpublished observations), so the isomerase/synthase. It is possible that PUFA supplemen- reported production of PG after supplementation with tation may enhance the activity of PGEs. A previous study GLA is probably an underestimate. In these studies of showed that dispersed bovine placental cells were able to 3 non-pregnant endometrial cells, however, LA did not convert up to 19% of added H-PGF2 into PGE2, increase the production of PGF1, suggesting that 6 whereas the reverse conversion of PGE2 to PGF2 was desaturase is not present in the cultures. The inhibitory minimal (<4%) (Gross & Williams 1988). In that study, ff e ect of LA was in agreement with our previous work the identity of the PGE2 produced was confirmed by gas using uterine endometrial cells isolated from non-pregnant chromatography/mass spectrometry and the results cows and ewes (Cheng et al. 2001a,b). LA has variously implied the presence of the 9-hydroxy-PG dehydrogenase been reported to stimulate or inhibit the production of enzyme in the bovine placenta. This raises another possi- PG in different studies. In rat gastric mucosal cells, for bility, that PUFA supplementation may accelerate the example, supplementation with LA stimulated AA and conversion process, although this would require testing in PGE2 production (Nakaya et al. 2001). Similarly, the this system. enhancement of tumour growth by LA has been reported In the cells cultured with PUFA-free medium, OT and in a number of studies and the mechanisms were attributed LPS challenge both significantly increased the production to the increased production of PGE2 catalysed from the of PGE2 and PGF2. This is in accord with many previous increased accumulation of AA by COX-2 (Connolly et al. studies showing that both OT and LPS can stimulate 1996, Dommels et al. 2002). In contrast, Thatcher et al. cPLA2 and COX-2 expression to increase PG output (Lee (1994) reported that LA inhibited the production of PG by & Silvia 1994, Flynn & Hoff 1995, Asselin et al. 1997). the bovine endometrium. A possible reason for the However, neither OT nor LPS increased PG output to the decreased PG production may be that accumulation of same extent as was achieved by the addition of either GLA excess LA competes with AA both for incorporation to or AA alone. Similar stimulatory effects of OT and LPS the cell membrane and for COX. Although AA is the were seen in the added presence of PUFAs; however, preferred substrate for COX-1 and -2, both isoforms can some interesting interactions were observed. OT and LPS oxygenate a variety of n-3 and n-6 18–22 carbon PUFAs were unable to stimulate PGF2 output further in the (Laneuville et al. 1995, Larsen et al. 1996, Rowlinson presence of AA. GLA similarly prevented the stimulatory ff et al. 1999). e ect of both OT and LPS on the production of PGE2. In the cells cultured in PUFA-free medium, similar The combination of AA with LPS altered the E:F ratio in amounts of PGE2 and PGF2 were produced, but supple- favour of PGE2. This indicates that n-6 PUFAs can not mentation with all three PUFAs tested increased the E:F only alter the basal production of uterine PGs, but also the ratio by about two- to threefold. In the presence of LA, responsiveness of the uterus to these challenges. Several PGE2 was unchanged, whereas the production of PGF2 mechanisms may be involved. OT and LPS increase PG was inhibited. In contrast, when GLA or AA were present output via upregulation of cPLA2 and COX-2 expression. there was a greater increase in PGE2 than in PGF2. The In the presence of a high concentration of precursor, the reasons for this are not clear. In the present study, the availability of either of these enzymes may become rate dispersed endometrial tissue contained a mixed population limiting. Studies involving activation of the OT receptor of epithelial and stromal cells, thus allowing the potential in preformed liposomes showed a strong dependence of for paracrine interactions. In the non-pregnant ewe, OT receptor function on both the amount of cholesterol PGF2 is principally produced by the epithelial cells and present within the membrane and the specific structure of PGE2 by the stromal cells (Cherny & Findlay 1990). the cholesterol molecule (Fahrenholz et al. 1995). LPS Different cell types may differ in their capability for PUFA utilizes the Toll-like receptor 4 (Tlr4) to induce the incorporation and metabolism because of their differences transcription and expression of COX-2 (Poltorak et al. in enzyme systems, and they may also use different AA 1998, Rhee & Hwang 2000), and PUFAs, including LA pools for the production of PG. In the late-gestation ewe, and AA, inhibited the induction of LPS on COX-2 however, in situ hybridisation and immunocytochemical expression via interruption of the Tlr4 signalling pathway studies show that COX-2 is localised exclusively to the (Lee et al. 2001). It is likely that membrane characteristics uterine epithelial cells and that COX-1 mRNA is not are altered by their PUFA content, and this may in turn detectable (Gibb et al. 2000, Gyomorey et al. 2000), influence signalling pathways through changes in receptor suggesting that it is mainly the epithelial cells that are binding characteristics or downstream activation. There involved in PG production at this stage. In the PG could well be independent effects of PUFAs on the OT metabolic pathway, COX-1 and COX-2 catalyse AA into and LPS signalling pathways. ff the intermediate PGH2 that is subsequently converted With regard to the inhibitory e ect of the DEX into one of a variety of bioactive PGs by specific PG challenge on the production of PGF2 in PUFA-free cells, Journal of Endocrinology (2004) 182, 249–256 www.endocrinology.org Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access Effects of PUFAs on endometrial PG production · Z CHENG and others 255 glucocorticoids have been classified as COX-2 inhibitors Acknowledgements in the anti-inflammatory pathway (DeWitt & Meade 1993, Evett et al. 1993). In amnion cells, Blumenstein et al. We thank Mr David Manners and his staff for care of the (2000) also demonstrated that DEX could inhibit the sheep and DrNLPoyser for his generous gift of production of PGE2. It has been well documented, how- antibodies. ever, that parturition in the ewe is initiated by the fetus after activation of the hypothalamic–pituitary–adrenal axis, and that increased intrauterine PG output is caused by the Funding increased cortisol output (Whittle et al. 2001). Previous studies showed that synthetic glucocorticoids, such as The study was funded by the BBSRC. Potential conflicts DEX and -methasone, stimulated intrauterine COX-2 of interest: None. expression and inhibited PGDH in intrauterine tissues (Zakar et al. 1995, Challis et al. 2002). COX-2 has a References glucocorticoid response element in its promoter region (Tazawa et al. 1994) and glucocorticoids may also mediate Abayasekara DR & Wathes DC 1999 Effects of altering dietary fatty non-genomic effects through a G-protein-coupled acid composition on prostaglandin synthesis and fertility. membrane-associated receptor (McKay & Cidlowski Prostaglandins, Leukotrienes and Essential Fatty Acids 61 275–287. 1999). Whittle et al. (2001) reviewed the different mecha- Asselin E, DroletP&Fortier MA 1997 Cellular mechanisms involved during oxytocin-induced prostaglandin F2 alpha production in nisms whereby glucocorticoids might alter COX-2 endometrial epithelial cells in vitro: role of cyclooxygenase-2. expression. A wide variety of factors can clearly influence Endocrinology 138 4798–4805. the result, including cell type, dose, time-course, whether Bishop-Bailey D, Calatayud S, Warner TD, Hla T, Mitchell JA 2002 the cells are dividing and other components of the Prostaglandins and the regulation of tumor growth. Journal of Environmental Pathology Toxicology and Onology 21 93–101. endocrine background such as the presence of oestrogen Blumenstein M, Hansen WR, DevalD&Mitchell MD 2000 and inflammatory interleukins. They suggested that the Differential regulation in human amnion epithelial and fibroblast pro-inflammatory effects occurred when COX-2 expres- cells of prostaglandin E2 production and prostaglandin H synthase-2 sion was basal, whereas the anti-inflammatory effects were mRNA expression by dexamethasone but not tumour necrosis apparent only when the expression of COX-2 had already factor-. Placenta 21 210–217. ff Challis JRG 2000 Endocrine and paracrine regulation of birth at term been upregulated. In our study, an inhibitory e ect of and preterm. Endocrine Reviews 21 514–550. DEX on PGE2 production was seen only in the additional Challis JRG, Sloboda DM, Alfaidy N, Lye SJ, Gibb W, Patel FA, presence of GLA or AA, when PG production was Whittle W & Newnham JP 2002 Prostaglandins and mechanisms of increased, thus shifting the ratio in favour of more PGF . preterm birth. Reproduction 124 1–17. 2 Cheng Z, Robinson RS, Pushpakumara PG, Mansbridge RJ & This adds another dimension to an already complex issue. Wathes DC 2001a Effect of dietary polyunsaturated fatty acids on In the context of labour, such a shift could potentially uterine prostaglandin synthesis in the cow. Journal of Endocrinology promote uterine contractility. 171 463–473. In conclusion, supplementation of n-6 PUFAs in vitro Cheng Z, Elmes M, Bainbridge DRJ, Abayasekara DRE & Wathes ff can alter both the amount and the proportion of the basal DC 2001b E ect of n-6 polyunsaturated fatty acids on prostaglandin production by ovine uterine epithelial cells. Endocrine Abstracts 2 production of uterine PGs and the responsiveness of the OC28. uterus to OT, LPS and DEX. These results suggest that Cheng Z, Elmes M, Abayasekara DRE & Wathes DC 2003 Effects of dietary differences in n-6 PUFAs may affect both the conjugated linoleic acid on prostaglandins produced by cells isolated initiation and progression of parturition. Within the UK, from maternal intercotyledonary endometrium, fetal allantochorion and amnion in late pregnant ewes. Biochimica et Biophysica Acta 1633 2·5% of the female population consume more than 20 g 170–178. n-6 PUFA per day (Office of Population Censuses and Cherny RA & Findlay JK 1990 Separation and culture of ovine Surveys 1990), mostly as LA derived from plant oils and endometrial epithelial and stromal cells: evidence of morphological soya. This is nearly 10 times the amount needed for normal and functional polarity. Biology of Reproduction 43 241–250. bodily function (Kinsella et al. 1990). GLA-rich products, Connolly JM, Liu XH & Rose DP 1996 Dietary linoleic acid-stimulated human breast cancer cell growth and metastasis in such as evening primrose oil, blackcurrant oil and borage nude mice and their suppression by indomethacin, a cyclooxygenase oil, are widely considered to benefit human health (Fan inhibitor. Nutrition and Cancer 25 231–240. Department of Health 1995 Nutritional Aspects of Cardiovascular Disease. & Chapkin 1998). Decreased responsiveness of PGE2 production to LPS challenge caused by GLA supplemen- London: HMSO. DeWitt DL & Meade EA 1993 Serum and glucocorticoid regulation tation might potentially assist the prevention of pre-term of gene transcription and expression of the prostaglandin H labour. The shift in the E:F ratio in response to DEX in synthase-1 and prostaglandin H synthase-2 isozymes. Archives of the presence of AA suggests that diets producing high Biochemistry and Biophysics 306 94–102. tissue concentrations of AA may alter inflammatory Dommels YEM, Alink GM, van Bladeren PJ & van Ommen B 2002 responses. In order to test these interesting possibilities, Dietary n-6 and n-3 polyunsaturated fatty acids and colorectal ff carcinogenesis: results from cultured colon cells, animal models and further studies are required to determine how di erent human studies. Environmental Toxicology and Pharmacology 11 dietary regimes alter PG production in vivo. 297–308. www.endocrinology.org Journal of Endocrinology (2004) 182, 249–256 Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access 256 Z CHENG and others · Effects of PUFAs on endometrial PG production Evett GE, Xie W, Chipman JC, Robertson DL & Simmons DL 1993 Leung ST, Wathes DC, Young IR & Jenkin G 1999 Effect of labor Prostaglandin G/H synthase isoenzyme 2 expression in fibroblasts: induction on the expression of oxytocin receptor, cytochrome P450 regulation by dexamethasone, mitogens and oncogenes. Archives of aromatase and estradiol receptor in the reproductive tract of the Biochemistry and Biophysics 306 169–177. late-pregnant ewe. Biology of Reproduction 60 814–820. FahrenholzF,KleinU&GimplG1995Conversion of the McKay LI & Cidlowski JA 1999 Molecular control of myometrial oxytocin receptor from low to high affinity state by immune/inflammatory responses: interactions between nuclear cholesterol. In Oxytocin, pp 311–319. Eds R Ivell & J Russell. New factor-kappa B and steroid receptor signalling pathways. Endocrine York: Plenum Press. Reviews 20 435–459. Fan YY & Chapkin RS 1998 Importance of dietary gamma-linolenic Nakaya A, Wakabayashi H, Imamura L, Fukuta K, Makimoto S, acid in human health and nutrition. Journal of Nutrition 128 Naganuma K, Orihara T, Minemura M, Shimizu Y, Nagasawa T, 1411–1414. HamazakiT&Watanabe A 2001 Helicobacter pylori alters n-6 FlynnJT&Hoff H 1995 Lipopolysaccharide induces time-dependent fatty acid metabolism and prostaglandin E2 synthesis in rat gastric increases in prostaglandin H synthase-2 and cytosolic phospholipase mucosal cells. Journal of Gastroenterology and Hepatology 16 A2 mRNA in cultured human microvessel-derived endothelial 1197–1205. cells. Shock 4 433–440. Office of Population Censuses and Surveys 1990 The Dietary and Gibb W, Gyomorey S, Lye SJ & Challis JRG 2000 Localization of Nutritional Survey of British Adults. London: HMSO. ff prostaglandin synthase type-1 (PGHS-1) mRNA and prostaglandin Poltorak A, He X, Smirnova I, Liu MY, Hu el CV, Du X, Birdwell synthase type-2 (PGHS-2) mRNA in ovine myometrium and D, Alejos E, Silva M, Galanos C, Freudenberg M, Ricciardi- endometrium throughout gestation. Journal of Endocrinology 165 Castagnoli P, LaytonB&Beutler B 1998 Defective LPS signalling 51–58. in C3H/HeJ and C57BL/10 ScCr mice: mutations in Tlr4 gene. Science 282 2085–2088. Goncalves LF, ChaiworapongsaT&RomeroR2002Intrauterine ff infection and prematurity. Mental Retardation and Develeopment Poyser NL 1987 E ects of various factors on prostaglandin synthesis by Disabilities Research Reviews 8 3–13. the guinea-pig uterus. Journal of Reproduction and Fertility 81 269–276. GrahamJ,FranksS&BonneyRC1994 In vivo and in vitro effects of Rhee SH & Hwang D 2000 Murine TOLL-like receptor 4 confers gamma-linolenic acid and eicosapentaenoic acid on prostaglandin lipopolysaccharide responsiveness as determined by activation of NF production and arachidonic acid uptake by human endometrium. kappa B and expression of the inducible cyclooxygenase. Journal of Prostaglandins, Leukotrienes and Essential Fatty Acids 50 321–329. Biological Chemistry 275 34035–34040. Gross TS, Williams WF 1988 Bovine placental prostaglandin synthesis: Robinson RS, Pushpakumara PGA, Cheng Z, Peters AR, principal cell synthesis as modulated by the binucleate cell. Biology Abayaskekara DRE & Wathes DC 2002 Effects of dietary of Reproduction 38 1027–1034. polyunsaturated fatty acids on ovarian and uterine function in Gyomorey S, Lye SJ, Gibb W & Challis JRG 2000 Fetal-to-maternal lactating dairy cows. Reproduction 124 119–131. progression of prostaglandin H2 synthase-2 expression in ovine Romero R, Avila C, Brekus CA & Marotti R 1991 The role of intra-uterine tissues during the course of labor. Biology of systemic and intrauterine infection in pre-term parturition. Annals of Reproduction 62 797–805. New York Academy of Science 622 355–375. Jenkin G 1992 Oxytocin and prostaglandin interactions in pregnancy Rowlinson SW, Crews BC, Lanzo CA & Marnett LJ 1999 The and at parturition. Journal of Reproduction and Fertility Supplement 45 binding of arachidonic acid in the cyclooxygenase active site of 457–469. mouse prostaglandin endoperoxide synthase-2 (COX-2) a putative Kinsella JE, Broughton KS & Whelan JW 1990 Dietary unsaturated L-shaped binding conformation utilizing the top channel region. fatty acids: interactions and possible needs in relation to eicosanoid Journal of Biological Chemistry 274 23305–23310. synthesis. Journal of Nutritional Biochemistry 1 123–141. Sprecher H 2000 Metabolism of highly unsaturated n-3 and n-6 fatty Lamont RF, Anthony F, Myatt L, Booth L, Farr PM & acids. Biochimica et Biophysica Acta 148 219–231. Taylor-Robinson D 1990 Preterm production of PGE2 by human Tazawa R, Xu XM, Wu KK & Wang LH 1994 Characterization of amnion in vitro in response to addition of media conditioned by the genomic structure, chromosomal location, and promoter of micro-organisms associated with chorionamnionitis and pre-term human prostaglandin H synthase-2 gene. Biochemical and Biophysical labour. American Journal of Obstetrics and Gynecology 162 819–825. Research Communications 203 190–199. Laneuville O, Breuer DK, Xu N, Huang ZH, Gage DA, Watson JT, Thatcher WW, Staples CR, Danet-Desnoyers G, Oldick B & Lagarde M, DeWit DL & Smith WL 1995 Fatty acid substrate Schmidt EP 1994 Embryo health and mortality in sheep and cattle specificities of human prostaglandin-endoperoxide H synthase-1 and Journal of Animal Science 72 (suppl 3) 16–30. -2 formation of 12-hydroxy-(9Z,13E/Z,15Z)-octadecatrienoic acids Thorburn GD 1991 The placenta, prostaglandins and parturition: a from linoleic. Journal of Biological Chemistry 270 19330–19336. review. Reproduction, Fertility and Development 3 277–294. Lapetina EG 1982 Regulation of arachidonic acid production: role of van Meir CA, Matthews SG, Keirse MJ, Ramirez MM, Bocking A & Challis JR 1997 15-Hydroxyprostaglandin dehydrogenase: phospholipases C and A2. Trends in Pharmacological Sciences 4 115–118. implications in preterm labor with and without ascending infection. Larsen LN, DahlE&BremerJ1996 Peroxidative oxidation of Journal of Clinical Endocrinology Metabolism 82 969–976. leuco-dichlorofluorescein by prostaglandin H synthase in Whittle WL, Fatel FA, Alfaiddy N, Holloway AC, Fraser M, prostaglandin biosynthesis from polyunsaturated fatty acids. Gyomorey S, Lye SJ, Gibb W & Challis JRG 2001 Glucocorticoid Biochimica et Biophysica Acta 1299 47–53. regulation of human and ovine parturition: the relationship between fetal hypothalamic–pituitary–adrenal axis activation and intrauterine Lee JS & Silvia WJ 1994 Cellular mechanisms mediating the prostaglandin production. Biology of Reproduction 64 1019–1032. stimulation of ovine endometrial secretion of prostaglandin F in 2 Zakar T, Hirst J, Mijovic JE & Olson DM 1995 Glucocorticoids response to oxytocin: role of phospholipase A . Journal of 2 stimulate the expression of prostaglandin endoperoxide H Endocrinology 141 491–496. synthase-2 in amnion cells. Endocrinology 136 1610–1619. Lee JY, Sohn KH, Rhee SH & Hwang D 2001 Saturated fatty acids, but not unsaturated fatty acids, induce the expression of cyclooxygenase-2 mediated through Toll-like receptor 4. Journal of Received in final form 14 April 2004 Biological Chemistry 276 16683–16689. Accepted 28 April 2004 Journal of Endocrinology (2004) 182, 249–256 www.endocrinology.org Downloaded from Bioscientifica.com at 09/28/2021 09:25:41PM via free access