- Email: [email protected]
Regulation of prostaglandin biosynthesis in dispersed choriodecidual cells in culture
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251 & 2001 Harcourt Publishers Ltd doi:10.1054/plef.2001.0321, available online at http://www.idealibrary.com on
Regulation of prostaglandin biosynthesis in dispersed choriodecidual cells in culture K.W. Marvin, M. D. Mitchell Liggins Institute and Division of Pharmacology and Clinical Pharmacology, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand
Summary We have evaluated the prostaglandin (PG) production and PG biosynthetic gene expression in a choriodecidual dispersed cell culture system. Cells dispersed from human choriodecidual membranes by dispase and trypsin digestion were evaluated after1, 3, 5 and 7 days of culture for basal and tumour necrosis factor a (F-a) stimulated PGE2 production.The highest rates of production (Po0.05) were obtained with cells treated after 3 days of culture, (3.771) 102 pg PGE2 per16 h per mg total cellular protein (mean7SEM), which was 3.9 times basal rate after 3 days culture. In choriodecidual cells treated after 3 days in culture, expression of prostaglandin endoperoxide H synthase-2 (PGHS-2) mRNAwas similarly responsive toTNF-a (3.9 times basal within 3 h of 30 ng/ml TNF-a) while there was little effect on PGHS-1or cytosolic phospholipase A2 expression. Hence, the dispersed choriodecidual cell culture system described retainsTNF-a responsive PG biosynthetic capacity whichis at least in part upregulated via increased expression of PGHS-2 mRNA. & 2001Harcourt Publishers Ltd
INTRODUCTION Prostaglandins are of major importance in the changes within the uterus – including cervical ripening, membrane rupture, and myometrial contractions–that are part of parturition either at term or preterm. There is no storage of prostaglandins, instead they are synthesized and immediately secreted and usually have actions locally. The precursor of prostaglandins, arachidonic acid (AA) is, however, stored in membrane glycerophospholipids within the cell. It is released by various phospholipases and is then converted to prostaglandin (PG) G2 and PGH2 by the action of prostaglandin H synthase (PGHS) enzymes. PGH2 is subsequently metabolized to PGD2, PGE2, PGF2a, prostacyclin (PGI2) or thromboxane via prostaglandin isomerases. PGHS exists in two isoforms. PGHS-1 is the constitutive form in most tissue types while Received 9 July 2001 Accepted 21August 2001 Correspondence to: Keith Marvin, Liggins Institute and Division of Pharmacology and Clinical Pharmacology, University of Auckland, Faculty of Medical and Health Sciences, Private Bag 92019, Auckland, New Zealand.Tel.: 64-9-373-7599 x4647, Fax: 64-9-373-7497, E-mail: [email protected] This work was funded by a grant fromThe Health Research Council of New Zealand.
& 2001Harcourt Publishers Ltd
PGHS-2 is inducible, often dramatically, in response to various cytokines and growth factors. For an excellent review see Reference 1. Human gestational tissues, particularly the placenta and fetal membranes, are a focus of studies involving prostaglandins and parturition. The amnion, an avascular tissue, consists of epithelial cells of fetal origin on a layer of connective tissue containing fibroblasts and macrophages also of fetal origin. The chorion consists of several cell types including trophoblasts and is in contact with the decidua, which is in turn in contact with the myometrium. Like the amnion, the chorion is avascular and of fetal origin whereas the decidua is vascularized and of maternal origin. Both the amnion and decidua have been shown to have increased prostaglandin output during labour.1 The amnion and chorion have an increase in PGHS activity during labour. In women the increase in PGHS-2 expression occurs in the fetal amnion and the chorion laeve tissues, rather than the maternal decidua.2,3 It is known that various cytokines, including TNF-a, stimulate prostaglandin production in cell cultures of human chorion and decidua primarily via induction of PGHS activity.4–6 It has been shown that cytokine concentrations are increased in amniotic fluid of women in preterm labour in the presence of intrauterine infection
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
248
Marvin and Mitchell
and these cytokines can in turn increase the expression of both cPLA2 and PGHS-2 in cell cultures.7,8 Type II PLA2 has been shown to be in higher concentrations in both preterm (o37 weeks gestation) as compared to term and premature rupture of membranes as compared to intact membranes in choriodecidua tissues.9 PGHS-1 and -2 mRNA expression increase in the chorion before term labour10 and PGHS activity also increases in chorion at term. This increase correlates with the induction of PGHS-2.2 In choriodecidua however, Northern blot analysis has shown no statistically significant increase before, during or after labour for PGHS-1.11 Likewise studies using PCR have shown that PGHS-2 but not PGHS-1 expression increases in the choriodecidua with labour.12 Treatments with some cytokines, epidermal growth factor or phorbol 12-myristate 13-acetate increase PGHS-2 levels in both chorion and decidua cells in culture.13 A choriodecidual inflammatory response syndrome (CoDIS) has recently been proposed as the leading and under-recognised cause of preterm delivery and second trimester miscarriage.14 Briefly, focal inflammatory responses in the choriodecidua may be more directly associated with preterm labour than are chorioamnionitis or intra-amniotic infection.14 Choriodecidual cell cultures which are likely more representative of the fetal–maternal boundary than individual cultures of either chorion or decidua thus have a renewed value in in vitro investigation of the inflammatory mechanisms of preterm labour. We have, therefore, determined whether the cytokine, TNF-a, affects the levels of mRNA for the enzymes responsible for prostaglandin synthesis in choriodecidual cell cultures. Specifically, we have evaluated PGHS-1, PGHS-2 and cytosolic PLA2 (cPLA2) mRNA expression in such cultures. We also evaluated the effects of TNF-a on PGE2 production in these choriodecidual cell cultures. MATERIALS AND METHODS Cell culture dishes and plasticware were obtained from Nunc, Naperville, IL, USA and Becton Dickinson, Lincoln Park, NJ, USA. Cell culture medium F12/DMEM (Irvine Scientific, Irvine, CA, USA), and fetal calf serum (FCS) was purchased from Life Technologies, New Zealand. Trypsin 1:250 was a product of DIFCO, Detroit, MI, USA. Gene Screen Plus membranes were obtained from NEN, Life Sciences Products (Boston, MA, USA).[a-32P] dCTP and random primed synthesis kits were obtained from Amersham Pharmacia, Auckland, NZ. Products of random primed synthesis were separated using push columns from Stratagene, La Jolla, CA, USA. Plasmid purification columns were a product of Qiagen GmbH, Hilden, Germany. [3H]PGE2 tracer was obtained from Amersham Pharmacia Auckland, New Zealand. All other biochem-
icals were from Sigma, St Louis, MO, USA and Riedel de Haehn, Seelze, Germany. TNF-a was supplied by Dr John Fraser, University of Auckland, New Zealand.
Cell Culture Primary choriodecidual cell cultures were prepared from placentae obtained after caesarean section at term prior to the start of labour (indications: previous caesarean or malpresentation) as previously described. Maternal informed consent for the collection of tissue was obtained and the work was performed under approval of the Auckland Human Ethics Committee. Briefly, the reflected amnion was manually separated from the underlying choriodecidua. The latter was dissected into strips and washed in phosphate buffered saline (PBS). The cells were dispersed by incubating the tissue at 371C for 1 h with 0.012% collagenase followed by 1 h with 0.25% dispase in Ham’s F12/DMEM (50:50) medium. DNase I (final concentration 4 mg/L) was included during the last 15 min of incubation. The cells were fractionated on a 60%/40%/ 20%/5% discontinuous percoll gradient at 600 g for 20 minutes. Cells lying between the 60–20% percoll layers were washed and plated at 3 105 cells/cm2 in medium M199 with 10% FCS, penicillin and streptomycin and cultured in an humidified 5% CO2/95% air incubator for the times specified. This method has been described and validated elsewhere.15
RNA isolation and Northern blotting Total RNA was isolated by the acid guanidinium thiocynate-phenol-chloroform method.16 and was resuspended in formamide with heating at 551C.17 Northern analysis was performed as previously described.18 Briefly, total RNA (30 mg/lane) was electrophoresed through 1% ag arose, 3-(N-morpholino) propanesulfonic acid (MOPS)formaldehyde gels, transferred to Gene Screen Plus membranes and cross-linked in a Stratagene Stratalinker (Stratagene, La Jolla, CA, USA). Membranes were hybridized overnight at 421C in a solution containing 50% formamide and 100 mg/mL salmon sperm DNA with [a-32P] d-CTP labeled random primed cDNA and washed to moderate stringency. Human cPLA2, PGHS-1, PGHS-2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNAs were used in preparation of the probes. The blots were stripped for 1 hour at 801C in 2% glycerol between probings. Detection was accomplished by exposing membranes to phosphor screens (Molecular Dynamics, Sunnyvale, CA, USA) and imaging with a Molecular Dynamics STORM 860 instrument. Bands of the expected sizes were obtained for cPLA2, PGHS-1, PGHS-2 and GAPDH. Data were analyzed using
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
& 2001Harcourt Publishers Ltd
Prostaglandin biosynthesis in choriodecidual cells 249
ImageQuant software (Molecular Dynamics, Sunnyvale, CA, USA).
PGE2 radioimmunoassay (RIA) PGE2 was measured using a sensitive and specific RIA on unextracted media samples as described previously.19 The bicinchoninic acid (BCA) method20 calibrated against BSA was used to determine cellular protein concentration. PGE2 production in 16 h normalized to total cellular protein in the same well was expressed as % control. Results shown are the mean of three independent experiments performed in replicates of three or four as noted in the figure legends. Statistical variation was evaluated by ANOVA and post hoc Student–Newman– Kuels test with differences resulting in Po0.05 being considered significant. The assay had a limit of detection of approximately 7 pg/ml.
RESULTS Mean normalized basal rates of production of PGE2 of 171730 pg per 16 h per mg total cellular protein were observed for dispersed choriodecidual cell cultures after one day in culture (Fig. 1). Basal production rates declined with extended culture, reaching one-sixteenth
Fig. 1 Basal and TNF-a stimulated PGE2 production rates by choriodecidual cell cultures following different intervals between plating and treatment.Choriodecidual cells were plated in 24-well culture plates and were cultured for1, 3, 5 and 7 day intervals.The media were then exchanged to serum free media containing 0.1% bovine gamma globulin and either 50 ng/ml TNF-a or vehicle.The cultures were incubated a further16 h after which the media were harvested and assayed for PGE2.Total protein content of the cells was also assayed and the PGE2 production rates for each well have been normalized to total cellular protein.The data are plotted and mean (7SEM) rates of production over three experimentsby vehicle (white) and TNF-a (black) treated cultures. Replicates of 3 were used in one experiment and 4 in the remaining two.*Po0.05 by ANOVA and post hoc Student^Newman^Kuels test when compared to the corresponding control and also to all other bars.
& 2001Harcourt Publishers Ltd
the original rate by the 7th day of culture (Po0.05). The highest rates of production of PGE2 (Po0.05 compared to all other treatment conditions) were found following treatment of cells on the third day in culture with TNF-a (50 ng/ml). The rate was 3.9 times (Po0.05) that of the corresponding control. Therefore, subsequent experiments were performed on cells which had been cultured for 3 days. In order to evaluate whether upregulation of mRNA expression of genes involved in prostaglandin synthesis might play a role in the TNF-a stimulated increase in production of PGE2 by choriodecidual cell cultures, we have evaluated the responsiveness of the expression of PGHS-1 and -2 and cPLA2 to treatment with TNF-a (Fig. 2). Both PGHS-1 and PGHS-2 mRNAs were expressed basally (Fig. 2A lanes 1 and 6). PGHS-1 appears to be constitutively expressed, while PGHS-2 clearly increased in expression after either 3 h or 6 h of incubation with TNF-a at various concentrations. Furthermore, the increased PGHS-2 expression was concentration dependent (lanes 6–11). Enhanced expression was 3.9 and 4 times basal with 30 ng/ml TNF-a for 3 h and 3 ng/ml TNFa for 6 h respectively (data not shown). Expression of cPLA2 was also detectable but the signals were very weak (Fig. 2A). Its expression does not appear to be affected by TNF-a. (The increased density in the cPLA2 band in lane 11 is artifactual.) Fig. 2B shows the results of treatment of choriodecidual cell cultures for various times with
Fig. 2 Northern blot analysis of the regulation of mRNA expression of prostaglandin biosynthetic enzymes in choriodecidual primary cell cultures by tumour necrosis factor a (TNF-a).Choriodecidual cultures were prepared as described in the methods. On day 3 of culture they were treated with (A) 0^100 ng/ml TNF-a for 3 h, lanes 1^5, or 6 h, lanes 6^11or (B) for 0^16 hwithvehicle or 50 ng/ml TNF-a. The specific lane assignments are, A: lane1, vehicle for 3 h; lanes 2^5, TNF-a for 3 h at 3,10, 30 and100 ng/ml, respectively; lane 6, vehicle 6 h; lanes 7^11,TNF-a for 6 h at1,3,10, 30 and100 ng/ml; and B: lanes1, 3, 5, 7, 9,11, vehicle for 0.5,1, 2, 4, 8,16 h, respectively; lanes 2, 4, 6, 8,10, and12, 50 ng/ml TNF-a for 0.5,1, 2, 4, 8, and16 h, respectively. Expression of the mRNAs for the enzymes prostaglandin prostaglandin H synthase (PGHS) 1and 2, which catalyse the first committed step of prostaglandin biosynthesis, and cytosolic phospholipase A2 (cPLA2), which specifically releases arachidonic acid from membrane phospolipids, were evaluated. Expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a constitutively expressed housekeeping gene, was also evaluated to assess equivalence of loading.
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
250
Marvin and Mitchell
50 ng/ml TNF-a. Elevated PGHS-2 mRNA expression is evident starting with the 2 h time point. Constitutive expression was again evident for PGHS-1 and cPLA2. DISCUSSION We have previously demonstrated the presence not only of PGHS-1 mRNA but also of PGHS-2 mRNA in choriodecidual tissue.21 In the present study we have shown that cell cultures derived from choriodecidua express PGHS-2 mRNA and that levels of this mRNA are regulated by TNF-a. There is increasing evidence that preterm labour and premature rupture of membranes are associated with intrauterine infections.22,23 One of the proposed mechanisms by which bacterial infection may initiate preterm labour is via bacterial endotoxin-induced formation of proinflammatory mediators that may act directly or indirectly to stimulate myometrial contractions, cervical ripening and/or rupture of fetal membranes. Bacteria and/or bacterial products may elicit inflammatory reactions in gestational tissues resulting in increased release of proinflammatory mediators such as TNF-a and prostaglandins.24–26 The use of inhibitors of prostaglandin biosynthesis in the treatment of preterm labour has had mixed results with many and extreme side-effects noted. Our data suggest that increased expression of PGHS-2 may have a role in the mechanism of TNF-a induced (and probably intrauterine infectionassociated) preterm labour. Hence, development of specific inhibitors of PGHS-227 may have clinical utility in the treatment of preterm labour and greatly reduce the unwanted side-effects associated with the use of nonspecific inhibitors. The action of TNF-a to enhance choriodecidual PGHS2 but not cPLA2 expression suggests a mechanism for TNF-a stimulation of prostaglandin production28 in this system. Similar to separate chorion or decidual cell cultures, upregulation of PGHS-2 mRNA expression in choriodecidual cultures is at least in part if not substantially responsible for the observed upregulation of PGE2 production. Because TNF-a may play a part in the mechanism of term labour the question arises as to the degree of involvement of PGHS-2 in that process. It is known that PGHS-2 levels in amnion are significantly increased during labour at term.29 This, however, leaves the question of causality open. If it is crucial that PGHS-2 levels increase for term labour to proceed in a timely fashion, then it would be of value to determine intrauterine PGHS-2 levels in post-term pregnancies and term labour that requires acceleration/enhancement with oxytocin analogs or prostaglandin preparations. With respect to preterm labour, the present results again emphasise the importance of upregulation of PGHS-2
to the inflammatory process in the gestational membranes. However, trophoblasts are known to be active in prostaglandin metabolism and degradative activity is responsive to TNF-a.30 Thus, measurement of PGE2 production does not fully assess effects of TNF-a on overall prostaglandin production. The increases in PGHS2 levels induced by TNF-a may, therefore, not fully account for the stimulation of prostaglandin production provoked by this cytokine. It is important, therefore, to consider the stimulatory effects of this substance (and probably others) on phospholipase activities that lead to provision of the substrate arachidonic acid for conversion to prostaglandins. It has generally been held that such activities are rate limiting in the prostaglandin biosynthetic pathways and TNF-a has well-described actions in this regard.
ACKNOWLEDGEMENTS The assistance of the nursing and theatre staff at National Women’s Hospital with the collection of the placental tissues is gratefully acknowledged. We also acknowledge Steven Skinner for his technical support and Roberta Eykhot for assistance in preparation of the manuscript. This work was supported by the Health Research Council of New Zealand.
REFERENCE 1. Gibb W. The role of prostaglandins in human parturition. Ann Med 1998; 30: 235–241. 2. Mijovic J. E., Zakar T., Angelova J., Olson D. M. Prostaglandin endoperoxide H synthase mRNA expression in the human amnion and decidua during pregnancy and in the amnion at preterm labour. Mol Hum Reprod 1999; 5: 182–187. 3. Gibb W., Sun M. Localization of prostaglandin H synthase type 2 protein and mRNA in term human fetal membranes and decidua. J Endocrinol 1996; 150: 497–503. 4. Pollard J. K., Mitchell M. D. Tumor necrosis factor alpha stimulates amnion prostaglandin biosynthesis primarily via an action on fatty acid cyclooxygenase. Prostaglandins 1993; 46: 499–510. 5. Pollard J. K., Thai D., Mitchell M. D. Evidence for a common mechanism of action of interleukin-1 beta, tumor necrosis factor-alpha, and epidermal growth factor on prostaglandin production in human chorion cells. Am J Reprod Immunol 1993; 30: 146–153. 6. Pollard J. K., Thai D., Mitchell M. D. Mechanism of cytokine stimulation of prostaglandin biosynthesis in human decidua. J Soc Gynecol Invest 1994; 1: 31–36. 7. Mitchell M. D., Branch D. W., Lundin-Schiller S., Romero R. J., Daynes R. A., Dudley D.J. Immunologic aspects of preterm labor. Semin Perinatol 1991; 15: 210–224. 8. Mitchell M. D., Edwin S. S., Lundin-Schiller S., Silver R.M., Smotkim D., Trautman M. S. Mechanism of interleukin-1 beta stimulation of human amnion prostaglandin biosynthesis: mediation via a novel inducible cyclooxygenase. Placenta 1993; 14: 615–625.
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
& 2001Harcourt Publishers Ltd
Prostaglandin biosynthesis in choriodecidual cells
9. Lappas M., Permezel M., Georgiou H. M., Rice G. E. Type II phospholipase A2 in preterm human gestational tissues. Placenta 2001; 22: 64–69. 10. Mijovic J. E., Zakar T., Nairn T. K., Olson D. M. Prostaglandin endoperoxide H synthase (PGHS) activity and PGHS-1 and -2 messenger ribonucleic acid abundance in human chorion throughout gestation and with preterm labor. J Clin Endocrinol Metab 1998; 83: 1358–1367. 11. Freed K. A., Aitken M. A., Brennecke S. P., Rice G. E. Prostaglandin G/H synthase-1 messenger RNA relative abundance in human amnion, choriodecidua and placenta before, during and after spontaneous-onset labour at term. Gynecol Obstet Invest 1995; 39: 73–78. 12. Slater D., Allport V., Bennett P. Changes in the expression of the type-2 but not the type-1 cyclo-oxygenase enzyme in choriondecidua with the onset of labour. Br J Obstet Gynaecol 1998; 105: 745–748. 13. Edwin S., Trautman M. S., Mitchell M. D. Regulation of prostaglandin H synthase-2 in chorion and decidual cells. Prostaglandins Leukot Essent Fatty Acids 1996; 55: 211–216. 14. Sebire N. J. Choriodecidual inflammatory syndrome (CoDIS) is the leading, and under recognised, cause of early preterm delivery and second trimester miscarriage. Med Hypothes 2001; 56: 497–500. 15. Keelan J. A., Mitchell M. D. The effects of a ceramide analogue, N-acetylsphingosine, on basal and tumor necrosis factor-alphastimulated production of prostaglandins and cytokines in gestational tissues. Prostaglandins Other Lipid Mediat 1998; 55: 141–158. 16. Chomczynski P., Sacchi N. Single-Step Method of RNA Isolation By Acid Guanidinium Thiocyanate-Phenol-Chloroform Extraction. Anal Biochem 1987; 162: 156–159. 17. Chomczynski P. Solubilization in formamide protects RNA from degradation. Nucleic Acids Res 1992; 20: 3791–3792. 18. Eykholt R. L., Mitchell M. D., Marvin K. W. Direct imaging of Northern blots on an optical scanner using ethidium bromide. Biotechniques 2000; 28: 864–866, 868, 870. 19. Simpson K. L., Keelan J. A., Mitchell M. D. Labor-associated changes in interleukin-10 production and its regulation by immunomodulators in human choriodecidua. J Clin Endocrinol Metab 1998; 83: 4332–4337.
& 2001Harcourt Publishers Ltd
251
20. Redinbaugh M. G., Turley R. B. Adaptation of the bicinchoninic acid protein assay for use with microtiter plates and sucrose gradient fractions. Anal Biochem 1986; 153: 267–271. 21. Trautman M. S., Edwin S. S., Collmer D., Dudley D. J., Simmons D., Mitchell M. D. Prostaglandin H synthase-2 in human gestational tissues: regulation in amnion. Placenta 1996; 17: 239–245. 22. Romero R., Mazor M., Wu Y. K. et al. Infection in the pathogenesis of preterm labor. Semin Perinatol 1988; 12: 262–279. 23. Romero R., Baumann P., Gomez R. et al. The relationship between spontaneous rupture of membranes, labor, and microbial invasion of the amniotic cavity and amniotic fluid concentrations of prostaglandins and thromboxane B2 in term pregnancy. Am J Obstet Gynecol 1993; 168: 1654–1664; discussion 1664–1658. 24. Romero R., Wu Y. K., Brody D. T., Oyarzun E., Duff G. W., Durum S. K. Human decidua: a source of interleukin-1. Obstet Gynecol 1989; 73: 31–34. 25. Nicholls L., Farrugia W., Rice G. E. The effect of monophosphoryl lipid A on lipopolysaccharide-induced prostaglandin E2 release in human choriodecidua. Placenta 1997; 18: 243–248. 26. Farrugia W., Nicholls L., Rice G. E. Effect of bacterial endotoxin on the in vitro release of Type II phospholipase-A2 and prostaglandin E2 from human placenta. J Endocrinol 1999; 160: 291–296. 27. Meade E. A., Smith W. L., DeWitt D. L. Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-steroidal anti-inflammatory drugs. J Biol Chem 1993; 268: 6610–6614. 28. Subbarayan V., Sabichi A. L., Llansa N., Lippman S. M., Menter D. G. Differential expression of cyclooxygenase-2 and its regulation by tumor necrosis factor-alpha in normal and malignant prostate cells. Cancer Res 2001; 61: 2720–2726. 29. Teixeira F. J., Zakar T., Hirst J. J. et al. 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 1994; 78: 1396–1402. 30. Mitchell M. D., Goodwin V., Mesnage S., Keelan J. A. Cytokineinduced coordinate expression of enzymes of prostaglandin biosynthesis and metabolism: 15-hydroxyprostaglandin dehydrogenase. Prostaglandins Leukot Essent Fatty Acids 2000; 62: 1–5. doi: 10.1054/plef. 1999.0117
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
Regulation of prostaglandin biosynthesis in dispersed choriodecidual cells in culture K.W. Marvin, M. D. Mitchell Liggins Institute and Division of Pharmacology and Clinical Pharmacology, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand
Summary We have evaluated the prostaglandin (PG) production and PG biosynthetic gene expression in a choriodecidual dispersed cell culture system. Cells dispersed from human choriodecidual membranes by dispase and trypsin digestion were evaluated after1, 3, 5 and 7 days of culture for basal and tumour necrosis factor a (F-a) stimulated PGE2 production.The highest rates of production (Po0.05) were obtained with cells treated after 3 days of culture, (3.771) 102 pg PGE2 per16 h per mg total cellular protein (mean7SEM), which was 3.9 times basal rate after 3 days culture. In choriodecidual cells treated after 3 days in culture, expression of prostaglandin endoperoxide H synthase-2 (PGHS-2) mRNAwas similarly responsive toTNF-a (3.9 times basal within 3 h of 30 ng/ml TNF-a) while there was little effect on PGHS-1or cytosolic phospholipase A2 expression. Hence, the dispersed choriodecidual cell culture system described retainsTNF-a responsive PG biosynthetic capacity whichis at least in part upregulated via increased expression of PGHS-2 mRNA. & 2001Harcourt Publishers Ltd
INTRODUCTION Prostaglandins are of major importance in the changes within the uterus – including cervical ripening, membrane rupture, and myometrial contractions–that are part of parturition either at term or preterm. There is no storage of prostaglandins, instead they are synthesized and immediately secreted and usually have actions locally. The precursor of prostaglandins, arachidonic acid (AA) is, however, stored in membrane glycerophospholipids within the cell. It is released by various phospholipases and is then converted to prostaglandin (PG) G2 and PGH2 by the action of prostaglandin H synthase (PGHS) enzymes. PGH2 is subsequently metabolized to PGD2, PGE2, PGF2a, prostacyclin (PGI2) or thromboxane via prostaglandin isomerases. PGHS exists in two isoforms. PGHS-1 is the constitutive form in most tissue types while Received 9 July 2001 Accepted 21August 2001 Correspondence to: Keith Marvin, Liggins Institute and Division of Pharmacology and Clinical Pharmacology, University of Auckland, Faculty of Medical and Health Sciences, Private Bag 92019, Auckland, New Zealand.Tel.: 64-9-373-7599 x4647, Fax: 64-9-373-7497, E-mail: [email protected] This work was funded by a grant fromThe Health Research Council of New Zealand.
& 2001Harcourt Publishers Ltd
PGHS-2 is inducible, often dramatically, in response to various cytokines and growth factors. For an excellent review see Reference 1. Human gestational tissues, particularly the placenta and fetal membranes, are a focus of studies involving prostaglandins and parturition. The amnion, an avascular tissue, consists of epithelial cells of fetal origin on a layer of connective tissue containing fibroblasts and macrophages also of fetal origin. The chorion consists of several cell types including trophoblasts and is in contact with the decidua, which is in turn in contact with the myometrium. Like the amnion, the chorion is avascular and of fetal origin whereas the decidua is vascularized and of maternal origin. Both the amnion and decidua have been shown to have increased prostaglandin output during labour.1 The amnion and chorion have an increase in PGHS activity during labour. In women the increase in PGHS-2 expression occurs in the fetal amnion and the chorion laeve tissues, rather than the maternal decidua.2,3 It is known that various cytokines, including TNF-a, stimulate prostaglandin production in cell cultures of human chorion and decidua primarily via induction of PGHS activity.4–6 It has been shown that cytokine concentrations are increased in amniotic fluid of women in preterm labour in the presence of intrauterine infection
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
248
Marvin and Mitchell
and these cytokines can in turn increase the expression of both cPLA2 and PGHS-2 in cell cultures.7,8 Type II PLA2 has been shown to be in higher concentrations in both preterm (o37 weeks gestation) as compared to term and premature rupture of membranes as compared to intact membranes in choriodecidua tissues.9 PGHS-1 and -2 mRNA expression increase in the chorion before term labour10 and PGHS activity also increases in chorion at term. This increase correlates with the induction of PGHS-2.2 In choriodecidua however, Northern blot analysis has shown no statistically significant increase before, during or after labour for PGHS-1.11 Likewise studies using PCR have shown that PGHS-2 but not PGHS-1 expression increases in the choriodecidua with labour.12 Treatments with some cytokines, epidermal growth factor or phorbol 12-myristate 13-acetate increase PGHS-2 levels in both chorion and decidua cells in culture.13 A choriodecidual inflammatory response syndrome (CoDIS) has recently been proposed as the leading and under-recognised cause of preterm delivery and second trimester miscarriage.14 Briefly, focal inflammatory responses in the choriodecidua may be more directly associated with preterm labour than are chorioamnionitis or intra-amniotic infection.14 Choriodecidual cell cultures which are likely more representative of the fetal–maternal boundary than individual cultures of either chorion or decidua thus have a renewed value in in vitro investigation of the inflammatory mechanisms of preterm labour. We have, therefore, determined whether the cytokine, TNF-a, affects the levels of mRNA for the enzymes responsible for prostaglandin synthesis in choriodecidual cell cultures. Specifically, we have evaluated PGHS-1, PGHS-2 and cytosolic PLA2 (cPLA2) mRNA expression in such cultures. We also evaluated the effects of TNF-a on PGE2 production in these choriodecidual cell cultures. MATERIALS AND METHODS Cell culture dishes and plasticware were obtained from Nunc, Naperville, IL, USA and Becton Dickinson, Lincoln Park, NJ, USA. Cell culture medium F12/DMEM (Irvine Scientific, Irvine, CA, USA), and fetal calf serum (FCS) was purchased from Life Technologies, New Zealand. Trypsin 1:250 was a product of DIFCO, Detroit, MI, USA. Gene Screen Plus membranes were obtained from NEN, Life Sciences Products (Boston, MA, USA).[a-32P] dCTP and random primed synthesis kits were obtained from Amersham Pharmacia, Auckland, NZ. Products of random primed synthesis were separated using push columns from Stratagene, La Jolla, CA, USA. Plasmid purification columns were a product of Qiagen GmbH, Hilden, Germany. [3H]PGE2 tracer was obtained from Amersham Pharmacia Auckland, New Zealand. All other biochem-
icals were from Sigma, St Louis, MO, USA and Riedel de Haehn, Seelze, Germany. TNF-a was supplied by Dr John Fraser, University of Auckland, New Zealand.
Cell Culture Primary choriodecidual cell cultures were prepared from placentae obtained after caesarean section at term prior to the start of labour (indications: previous caesarean or malpresentation) as previously described. Maternal informed consent for the collection of tissue was obtained and the work was performed under approval of the Auckland Human Ethics Committee. Briefly, the reflected amnion was manually separated from the underlying choriodecidua. The latter was dissected into strips and washed in phosphate buffered saline (PBS). The cells were dispersed by incubating the tissue at 371C for 1 h with 0.012% collagenase followed by 1 h with 0.25% dispase in Ham’s F12/DMEM (50:50) medium. DNase I (final concentration 4 mg/L) was included during the last 15 min of incubation. The cells were fractionated on a 60%/40%/ 20%/5% discontinuous percoll gradient at 600 g for 20 minutes. Cells lying between the 60–20% percoll layers were washed and plated at 3 105 cells/cm2 in medium M199 with 10% FCS, penicillin and streptomycin and cultured in an humidified 5% CO2/95% air incubator for the times specified. This method has been described and validated elsewhere.15
RNA isolation and Northern blotting Total RNA was isolated by the acid guanidinium thiocynate-phenol-chloroform method.16 and was resuspended in formamide with heating at 551C.17 Northern analysis was performed as previously described.18 Briefly, total RNA (30 mg/lane) was electrophoresed through 1% ag arose, 3-(N-morpholino) propanesulfonic acid (MOPS)formaldehyde gels, transferred to Gene Screen Plus membranes and cross-linked in a Stratagene Stratalinker (Stratagene, La Jolla, CA, USA). Membranes were hybridized overnight at 421C in a solution containing 50% formamide and 100 mg/mL salmon sperm DNA with [a-32P] d-CTP labeled random primed cDNA and washed to moderate stringency. Human cPLA2, PGHS-1, PGHS-2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNAs were used in preparation of the probes. The blots were stripped for 1 hour at 801C in 2% glycerol between probings. Detection was accomplished by exposing membranes to phosphor screens (Molecular Dynamics, Sunnyvale, CA, USA) and imaging with a Molecular Dynamics STORM 860 instrument. Bands of the expected sizes were obtained for cPLA2, PGHS-1, PGHS-2 and GAPDH. Data were analyzed using
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
& 2001Harcourt Publishers Ltd
Prostaglandin biosynthesis in choriodecidual cells 249
ImageQuant software (Molecular Dynamics, Sunnyvale, CA, USA).
PGE2 radioimmunoassay (RIA) PGE2 was measured using a sensitive and specific RIA on unextracted media samples as described previously.19 The bicinchoninic acid (BCA) method20 calibrated against BSA was used to determine cellular protein concentration. PGE2 production in 16 h normalized to total cellular protein in the same well was expressed as % control. Results shown are the mean of three independent experiments performed in replicates of three or four as noted in the figure legends. Statistical variation was evaluated by ANOVA and post hoc Student–Newman– Kuels test with differences resulting in Po0.05 being considered significant. The assay had a limit of detection of approximately 7 pg/ml.
RESULTS Mean normalized basal rates of production of PGE2 of 171730 pg per 16 h per mg total cellular protein were observed for dispersed choriodecidual cell cultures after one day in culture (Fig. 1). Basal production rates declined with extended culture, reaching one-sixteenth
Fig. 1 Basal and TNF-a stimulated PGE2 production rates by choriodecidual cell cultures following different intervals between plating and treatment.Choriodecidual cells were plated in 24-well culture plates and were cultured for1, 3, 5 and 7 day intervals.The media were then exchanged to serum free media containing 0.1% bovine gamma globulin and either 50 ng/ml TNF-a or vehicle.The cultures were incubated a further16 h after which the media were harvested and assayed for PGE2.Total protein content of the cells was also assayed and the PGE2 production rates for each well have been normalized to total cellular protein.The data are plotted and mean (7SEM) rates of production over three experimentsby vehicle (white) and TNF-a (black) treated cultures. Replicates of 3 were used in one experiment and 4 in the remaining two.*Po0.05 by ANOVA and post hoc Student^Newman^Kuels test when compared to the corresponding control and also to all other bars.
& 2001Harcourt Publishers Ltd
the original rate by the 7th day of culture (Po0.05). The highest rates of production of PGE2 (Po0.05 compared to all other treatment conditions) were found following treatment of cells on the third day in culture with TNF-a (50 ng/ml). The rate was 3.9 times (Po0.05) that of the corresponding control. Therefore, subsequent experiments were performed on cells which had been cultured for 3 days. In order to evaluate whether upregulation of mRNA expression of genes involved in prostaglandin synthesis might play a role in the TNF-a stimulated increase in production of PGE2 by choriodecidual cell cultures, we have evaluated the responsiveness of the expression of PGHS-1 and -2 and cPLA2 to treatment with TNF-a (Fig. 2). Both PGHS-1 and PGHS-2 mRNAs were expressed basally (Fig. 2A lanes 1 and 6). PGHS-1 appears to be constitutively expressed, while PGHS-2 clearly increased in expression after either 3 h or 6 h of incubation with TNF-a at various concentrations. Furthermore, the increased PGHS-2 expression was concentration dependent (lanes 6–11). Enhanced expression was 3.9 and 4 times basal with 30 ng/ml TNF-a for 3 h and 3 ng/ml TNFa for 6 h respectively (data not shown). Expression of cPLA2 was also detectable but the signals were very weak (Fig. 2A). Its expression does not appear to be affected by TNF-a. (The increased density in the cPLA2 band in lane 11 is artifactual.) Fig. 2B shows the results of treatment of choriodecidual cell cultures for various times with
Fig. 2 Northern blot analysis of the regulation of mRNA expression of prostaglandin biosynthetic enzymes in choriodecidual primary cell cultures by tumour necrosis factor a (TNF-a).Choriodecidual cultures were prepared as described in the methods. On day 3 of culture they were treated with (A) 0^100 ng/ml TNF-a for 3 h, lanes 1^5, or 6 h, lanes 6^11or (B) for 0^16 hwithvehicle or 50 ng/ml TNF-a. The specific lane assignments are, A: lane1, vehicle for 3 h; lanes 2^5, TNF-a for 3 h at 3,10, 30 and100 ng/ml, respectively; lane 6, vehicle 6 h; lanes 7^11,TNF-a for 6 h at1,3,10, 30 and100 ng/ml; and B: lanes1, 3, 5, 7, 9,11, vehicle for 0.5,1, 2, 4, 8,16 h, respectively; lanes 2, 4, 6, 8,10, and12, 50 ng/ml TNF-a for 0.5,1, 2, 4, 8, and16 h, respectively. Expression of the mRNAs for the enzymes prostaglandin prostaglandin H synthase (PGHS) 1and 2, which catalyse the first committed step of prostaglandin biosynthesis, and cytosolic phospholipase A2 (cPLA2), which specifically releases arachidonic acid from membrane phospolipids, were evaluated. Expression of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a constitutively expressed housekeeping gene, was also evaluated to assess equivalence of loading.
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
250
Marvin and Mitchell
50 ng/ml TNF-a. Elevated PGHS-2 mRNA expression is evident starting with the 2 h time point. Constitutive expression was again evident for PGHS-1 and cPLA2. DISCUSSION We have previously demonstrated the presence not only of PGHS-1 mRNA but also of PGHS-2 mRNA in choriodecidual tissue.21 In the present study we have shown that cell cultures derived from choriodecidua express PGHS-2 mRNA and that levels of this mRNA are regulated by TNF-a. There is increasing evidence that preterm labour and premature rupture of membranes are associated with intrauterine infections.22,23 One of the proposed mechanisms by which bacterial infection may initiate preterm labour is via bacterial endotoxin-induced formation of proinflammatory mediators that may act directly or indirectly to stimulate myometrial contractions, cervical ripening and/or rupture of fetal membranes. Bacteria and/or bacterial products may elicit inflammatory reactions in gestational tissues resulting in increased release of proinflammatory mediators such as TNF-a and prostaglandins.24–26 The use of inhibitors of prostaglandin biosynthesis in the treatment of preterm labour has had mixed results with many and extreme side-effects noted. Our data suggest that increased expression of PGHS-2 may have a role in the mechanism of TNF-a induced (and probably intrauterine infectionassociated) preterm labour. Hence, development of specific inhibitors of PGHS-227 may have clinical utility in the treatment of preterm labour and greatly reduce the unwanted side-effects associated with the use of nonspecific inhibitors. The action of TNF-a to enhance choriodecidual PGHS2 but not cPLA2 expression suggests a mechanism for TNF-a stimulation of prostaglandin production28 in this system. Similar to separate chorion or decidual cell cultures, upregulation of PGHS-2 mRNA expression in choriodecidual cultures is at least in part if not substantially responsible for the observed upregulation of PGE2 production. Because TNF-a may play a part in the mechanism of term labour the question arises as to the degree of involvement of PGHS-2 in that process. It is known that PGHS-2 levels in amnion are significantly increased during labour at term.29 This, however, leaves the question of causality open. If it is crucial that PGHS-2 levels increase for term labour to proceed in a timely fashion, then it would be of value to determine intrauterine PGHS-2 levels in post-term pregnancies and term labour that requires acceleration/enhancement with oxytocin analogs or prostaglandin preparations. With respect to preterm labour, the present results again emphasise the importance of upregulation of PGHS-2
to the inflammatory process in the gestational membranes. However, trophoblasts are known to be active in prostaglandin metabolism and degradative activity is responsive to TNF-a.30 Thus, measurement of PGE2 production does not fully assess effects of TNF-a on overall prostaglandin production. The increases in PGHS2 levels induced by TNF-a may, therefore, not fully account for the stimulation of prostaglandin production provoked by this cytokine. It is important, therefore, to consider the stimulatory effects of this substance (and probably others) on phospholipase activities that lead to provision of the substrate arachidonic acid for conversion to prostaglandins. It has generally been held that such activities are rate limiting in the prostaglandin biosynthetic pathways and TNF-a has well-described actions in this regard.
ACKNOWLEDGEMENTS The assistance of the nursing and theatre staff at National Women’s Hospital with the collection of the placental tissues is gratefully acknowledged. We also acknowledge Steven Skinner for his technical support and Roberta Eykhot for assistance in preparation of the manuscript. This work was supported by the Health Research Council of New Zealand.
REFERENCE 1. Gibb W. The role of prostaglandins in human parturition. Ann Med 1998; 30: 235–241. 2. Mijovic J. E., Zakar T., Angelova J., Olson D. M. Prostaglandin endoperoxide H synthase mRNA expression in the human amnion and decidua during pregnancy and in the amnion at preterm labour. Mol Hum Reprod 1999; 5: 182–187. 3. Gibb W., Sun M. Localization of prostaglandin H synthase type 2 protein and mRNA in term human fetal membranes and decidua. J Endocrinol 1996; 150: 497–503. 4. Pollard J. K., Mitchell M. D. Tumor necrosis factor alpha stimulates amnion prostaglandin biosynthesis primarily via an action on fatty acid cyclooxygenase. Prostaglandins 1993; 46: 499–510. 5. Pollard J. K., Thai D., Mitchell M. D. Evidence for a common mechanism of action of interleukin-1 beta, tumor necrosis factor-alpha, and epidermal growth factor on prostaglandin production in human chorion cells. Am J Reprod Immunol 1993; 30: 146–153. 6. Pollard J. K., Thai D., Mitchell M. D. Mechanism of cytokine stimulation of prostaglandin biosynthesis in human decidua. J Soc Gynecol Invest 1994; 1: 31–36. 7. Mitchell M. D., Branch D. W., Lundin-Schiller S., Romero R. J., Daynes R. A., Dudley D.J. Immunologic aspects of preterm labor. Semin Perinatol 1991; 15: 210–224. 8. Mitchell M. D., Edwin S. S., Lundin-Schiller S., Silver R.M., Smotkim D., Trautman M. S. Mechanism of interleukin-1 beta stimulation of human amnion prostaglandin biosynthesis: mediation via a novel inducible cyclooxygenase. Placenta 1993; 14: 615–625.
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251
& 2001Harcourt Publishers Ltd
Prostaglandin biosynthesis in choriodecidual cells
9. Lappas M., Permezel M., Georgiou H. M., Rice G. E. Type II phospholipase A2 in preterm human gestational tissues. Placenta 2001; 22: 64–69. 10. Mijovic J. E., Zakar T., Nairn T. K., Olson D. M. Prostaglandin endoperoxide H synthase (PGHS) activity and PGHS-1 and -2 messenger ribonucleic acid abundance in human chorion throughout gestation and with preterm labor. J Clin Endocrinol Metab 1998; 83: 1358–1367. 11. Freed K. A., Aitken M. A., Brennecke S. P., Rice G. E. Prostaglandin G/H synthase-1 messenger RNA relative abundance in human amnion, choriodecidua and placenta before, during and after spontaneous-onset labour at term. Gynecol Obstet Invest 1995; 39: 73–78. 12. Slater D., Allport V., Bennett P. Changes in the expression of the type-2 but not the type-1 cyclo-oxygenase enzyme in choriondecidua with the onset of labour. Br J Obstet Gynaecol 1998; 105: 745–748. 13. Edwin S., Trautman M. S., Mitchell M. D. Regulation of prostaglandin H synthase-2 in chorion and decidual cells. Prostaglandins Leukot Essent Fatty Acids 1996; 55: 211–216. 14. Sebire N. J. Choriodecidual inflammatory syndrome (CoDIS) is the leading, and under recognised, cause of early preterm delivery and second trimester miscarriage. Med Hypothes 2001; 56: 497–500. 15. Keelan J. A., Mitchell M. D. The effects of a ceramide analogue, N-acetylsphingosine, on basal and tumor necrosis factor-alphastimulated production of prostaglandins and cytokines in gestational tissues. Prostaglandins Other Lipid Mediat 1998; 55: 141–158. 16. Chomczynski P., Sacchi N. Single-Step Method of RNA Isolation By Acid Guanidinium Thiocyanate-Phenol-Chloroform Extraction. Anal Biochem 1987; 162: 156–159. 17. Chomczynski P. Solubilization in formamide protects RNA from degradation. Nucleic Acids Res 1992; 20: 3791–3792. 18. Eykholt R. L., Mitchell M. D., Marvin K. W. Direct imaging of Northern blots on an optical scanner using ethidium bromide. Biotechniques 2000; 28: 864–866, 868, 870. 19. Simpson K. L., Keelan J. A., Mitchell M. D. Labor-associated changes in interleukin-10 production and its regulation by immunomodulators in human choriodecidua. J Clin Endocrinol Metab 1998; 83: 4332–4337.
& 2001Harcourt Publishers Ltd
251
20. Redinbaugh M. G., Turley R. B. Adaptation of the bicinchoninic acid protein assay for use with microtiter plates and sucrose gradient fractions. Anal Biochem 1986; 153: 267–271. 21. Trautman M. S., Edwin S. S., Collmer D., Dudley D. J., Simmons D., Mitchell M. D. Prostaglandin H synthase-2 in human gestational tissues: regulation in amnion. Placenta 1996; 17: 239–245. 22. Romero R., Mazor M., Wu Y. K. et al. Infection in the pathogenesis of preterm labor. Semin Perinatol 1988; 12: 262–279. 23. Romero R., Baumann P., Gomez R. et al. The relationship between spontaneous rupture of membranes, labor, and microbial invasion of the amniotic cavity and amniotic fluid concentrations of prostaglandins and thromboxane B2 in term pregnancy. Am J Obstet Gynecol 1993; 168: 1654–1664; discussion 1664–1658. 24. Romero R., Wu Y. K., Brody D. T., Oyarzun E., Duff G. W., Durum S. K. Human decidua: a source of interleukin-1. Obstet Gynecol 1989; 73: 31–34. 25. Nicholls L., Farrugia W., Rice G. E. The effect of monophosphoryl lipid A on lipopolysaccharide-induced prostaglandin E2 release in human choriodecidua. Placenta 1997; 18: 243–248. 26. Farrugia W., Nicholls L., Rice G. E. Effect of bacterial endotoxin on the in vitro release of Type II phospholipase-A2 and prostaglandin E2 from human placenta. J Endocrinol 1999; 160: 291–296. 27. Meade E. A., Smith W. L., DeWitt D. L. Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-steroidal anti-inflammatory drugs. J Biol Chem 1993; 268: 6610–6614. 28. Subbarayan V., Sabichi A. L., Llansa N., Lippman S. M., Menter D. G. Differential expression of cyclooxygenase-2 and its regulation by tumor necrosis factor-alpha in normal and malignant prostate cells. Cancer Res 2001; 61: 2720–2726. 29. Teixeira F. J., Zakar T., Hirst J. J. et al. 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 1994; 78: 1396–1402. 30. Mitchell M. D., Goodwin V., Mesnage S., Keelan J. A. Cytokineinduced coordinate expression of enzymes of prostaglandin biosynthesis and metabolism: 15-hydroxyprostaglandin dehydrogenase. Prostaglandins Leukot Essent Fatty Acids 2000; 62: 1–5. doi: 10.1054/plef. 1999.0117
Prostaglandins, Leukotrienes and Essential FattyAcids (2001) 65(5&6), 247^251