Increased expression of alpha-enolase in cervico-vaginal fluid during labour

Study design Preterm labour was induced in sheep ( n = 5) via fetal dexamethasone infusion (1 mg/24 h). CVF samples were taken prior to dexamethasone infusion (0 h), 28 h after the start of dexamethasone infusion, and immediately prior to delivery. Two-dimensional gel electrophoresis (2-DE) and mass spectrometry (MS) were used to identify differentially expressed proteins. For the human studies, paired CVF samples were taken 5–9 days before labour and during spontaneous labour onset ( n = 7). Results There was a 4.2-fold increase in α-enolase protein expression in sheep CVF during labour. Likewise, α-enolase protein expression was significantly increased during spontaneous human labour at term. Conclusions Alpha-enolase is known to be bound to neutrophils and interact in the immune response, and thus may play a role in inflammation associated with human labour. Keywords Ovine Human labour Cervico-vaginal fluid Alpha-enolase Two-dimensional gel electrophoresis Mass spectrometry 1 Introduction One of the single most important factors contributing to neonatal and perinatal morbidity and mortality is preterm birth [1] . Further, there is a significant financial, emotional and societal burden associated with preterm birth. Methods currently available for predicting preterm labour, including cervical sonography [2,3] and fetal fibronectin [4] are unreliable and with low sensitivity. Thus, there is an obvious need for effective and reliable diagnostic and prognostic indicators of imminent preterm delivery. The protein composition of cervico-vaginal fluid (CVF) has the potential to provide unique insights into the biochemical pathways initiating human pregnancy and labour. Labour-associated changes in CVF proteins with imminent labour may reflect the biochemical changes occurring within the cervix and fetal membranes overlying the internal os . Studies have identified changes in cytokines, hormones, enzymes and inflammatory proteins in CVF during human pregnancy [5,6] and in association with labour at term [7] and preterm [5,8] . Further investigation of the CVF proteins implicated in the specific biochemical pathways of preterm labour using two-dimensional gel electrophoresis (2-DE) is warranted and may enhance our understanding of the complex biology involved in human labour. Thus, the aim of this study was to further characterise labour-associated differentially expressed CVF proteins using 2-DE protein display and matrix assisted laser desorption ionisation (MALDI) mass spectrometry and liquid chromatography–electrospray ionisation mass spectrometry (LC–ESI-MS). We will use a sheep model of labour as we have previously described [10] . Identification of differentially expressed proteins will then be confirmed using paired CVF samples taken from women during spontaneous labour onset (before rupture of human fetal membranes) and approximately 1 week before active labour. 2 Materials and methods 2.1 Experimental animals All animal experimental procedures were approved by the Physiology Animal Ethics Committee of Monash University. Five pregnant Border Leicester-Merino crossbred ewes underwent surgery at 126 days gestational age (GA) to insert maternal and fetal carotid artery and jugular vein catheters. An electromyographic (EMG) lead was sewn onto the uterus to measure contractile activity as labour progressed. A fetal intravenous dose of dexamethasone (1 mg/24 h) was administered at 135 days GA to induce preterm labour and continued until delivery. Progression of labour was monitored as previously described by Grigsby et al. [9] . A significant increase in uterine activity occurred 38 h post-infusion [8] . CVF samples were collected at three time points for each animal: immediately prior to dexamethasone infusion (0 h), 28 h later at the predicted midpoint of induced labour [10] and immediately before predicted delivery, but prior to membrane rupture (58.7 ± 1.9 h after dexamethasone infusion). Ewes were sacrificed via the administration of a barbiturate overdose and the occurrence of cervical ripening was confirmed by oedematous and flaccid cervices at postmortem. 2.2 CVF sample preparation for 2-DE CVF samples were obtained via the insertion of a lubricated proctoscope into the vaginal lumen as previously described [8] . Briefly, two cotton swabs were held in the high vaginal/cervical region and, provided there was an increase in wet weight greater than 20 mg, they were then added to 1 ml of sample buffer (0.1% SDS, 50 mM HEPES, 15 mM NaCl, 1 mM EDTA, 1 mM PMSF, pH 7.5). Samples were then vortexed at maximum speed for 30 s, centrifuged (20 min, 3000 rpm at 4 °C) and aliquoted for storage at −80 °C until analysed. Proteins present in CVF samples were precipitated using 20% trichloroacetic acid (TCA). Equal volumes of TCA and CVF supernatant were mixed and incubated on ice for 20 min. Proteins were sedimented (15 min, 13,000 × g at 4 °C) and washed with ice-cold ethanol. Pellets were air dried and resuspended in 40 μl of rehydration buffer (3 M urea, 10 mM DTT, 4% CHAPS, 40 mM Tris). Total protein content was measured using the Coomassie Plus Protein Assay Reagent Kit (Pierce Biotechnology, Rockford, IL) based on the modified Bradford dye-binding method [11] . 2.3 Two-dimensional gel electrophoresis First dimension electrophoresis was performed as previously described [8] . Prior to second dimension separation, IPG strips were equilibrated as described by Vorum et al. [12] . Separation was performed on 8–16% Tris–HCl Criterion Precast Gels (Bio-Rad Laboratories, Hercules, CA, USA). Gels were stained with SYPRO ruby protein gel stain and scanned at 532 nm using a FX Molecular Imager (Bio-Rad Laboratories, Hercules, CA, USA). Images were analysed with PDQuest software version 6.2.0 (Bio-Rad Laboratories, Hercules, CA, USA). Protein intensity values were determined and proteins differentially expressed in association with preterm labour were identified. Colloidal Coomassie (17% ammonium sulphate, 3% phosphoric acid solution (85%), 0.1% Coomassie G-250, 34% methanol) was used to stain the gels for visualisation and subsequent protein spot excision. Statistical significance of protein intensity between samples obtained at different stages of the labour induction process was established by paired Student's t -test, p < 0.05 (GraphPad Prism 5.0). 2.4 Mass spectrometry and protein identification Protein spots were manually excised using a 1.5 mm diameter spot picker (The Gel Company, San Francisco, CA, USA). Excised gel plugs were added to 100 μl Milli-Q water in a 96-well plate (Greiner Bio-One BioScience, Frickenhausen, Germany). An Ettan Spot Handling workstation (GE Healthcare Bio-Sciences, Uppsala, Sweden) robotically performed the tryptic digestion and spotted onto a target plate (Bruker Biosciences, Bremen, Germany) as described by Oliva et al. [13] . An Autoflex II MALDI-TOF/TOF was used to analyse peptides as described by Oliva et al. [13] . All peak list data were submitted to the Swiss-Prot database (version Swiss-Prot.2006.02.21) using the MASCOT search algorithm (Matrix Science, www.matrixscience.com ). MASCOT search parameters used were mass tolerance, 0.15 Da; missed cleavages, 2; enzyme, trypsin; global modifications, carboxymethyl (C); variable modifications, deamidation (NQ) and oxidation (M); taxonomy, mammalia. MASCOT scores of above 59 were considered as a positive identification. Confirmation of MALDI-TOF data was achieved with capillary LC–ESI-MS/MS. Samples were characterised and identified using an 1100 series HPLC coupled to a LC/MSD Ion Trap XCT Plus Mass Spectrometer fitted with a HPLC chip cube (Agilent, Palo Alto, CA, USA). The HPLC chip was comprised of a 40 nl enrichment column and a 75 μm × 43 mm separation column both packed with reversed phase resin (Zorbax 300SB-C18, 5 μm). Samples (8 μl) were loaded onto the enrichment column in acetonitrile and formic acid (4:0.1%, v/v, 4 μl/min). A linear gradient (19 min, flow rate 0.5 μl/min, acetonitrile:formic acid, 4–50:0.1%, v/v) was applied to the column to sequentially elute bound peptides. A final gradient step was applied (19–20 min, acetonitrile:formic acid, 50–80:0.1%, v/v) to strip remaining proteins. All MS/MS spectra were collected using data dependent acquisition. Briefly, after the acquisition of a full MS scan ( m / z 300–1800 at 8100 m / z /s) in the first scan event, the three most intense ions (precursor ions) present above a threshold intensity of 10,000 were subsequently selected for fragmentation (MS/MS scan m / z 100–2000 at 26,000 m / z /s). The collision energy for the MS/MS scan events was ramped from 30 to 200% of 1.3 V, for acquisition of the MS/MS scan, 3 spectra were averaged for each event. General instrument parameters were capillary voltage, 2000 V; skimmer, 40 V; capillary exit, 105.3 V; trap drive 77.8; dry gas 5.0 l/min; dry temperature, 350 °C. All peptide fragmentation spectra were extracted from the base peak chromatogram and exported as a mascot generic file (.mgf) using DataAnalysis software (Bruker Biosciences). Extracted data were submitted to the Swiss-Prot database using MASCOT (version 2.1, Matrix Science). The search was performed using the following typical search parameters: mammalian taxonomy; peptide tolerance, 2.0; charge, 1+, 2+ and 3+; MS/MS tolerance, 1.0; enzyme, trypsin; fixed modification, cysteine modified by carbadiomethyl; variable modification, oxidation (M). MASCOT scores of above 45 were considered as a positive identification. 2.5 Analysis of human CVF Patient recruitment : The study was approved by the Mercy Health and Aged Care Research Ethics Committee (R06/56). Primigravidas and women who had pregnancy complications such as gestational diabetes, hypertension, and diagnosed bacterial vaginosis were not recruited into the study. The demographic and obstetric characteristics of the recruited women are displayed in Table 3 . CVF samples were collected from seven women over two time periods: 5–9 days before labour and at the time of spontaneous onset of labour (before the rupture of fetal membranes). Labour was defined as regular painful uterine contractions leading to effacement and dilatation of the cervix (3–7 cm) at time of sampling. Western blotting of human CVF samples : Human CVF was collected as we have previously described [6,7] . Assessment of α-enolase protein expression was analysed by Western blotting. Mouse monoclonal anti-α-enolase (cat. #ab54979) was purchased from Abcam (Cambridge, UK). Twenty micrograms of protein was separated on 10% polyacrylamide gels and transferred to nitrocellulose as previously described [14] . Quantitative analysis of the relative density of the bands in Western blots was performed using Quantity One 4.2.1 image analysis software (Bio-Rad). Data were corrected for background and expressed as optical density (OD/mm 2 ). Statistical analyses were performed using a commercially available statistical software package (Statgraphics Plus version 3.1, Statistical Graphics Corp., Rockville, MD, USA). Student's t -test was used to the means, and statistical difference was indicated by a p value of less than 0.05. Data are expressed as mean density ± standard error of the mean (SEM). 3 Results 3.1 Two-dimensional gel electrophoresis Protein expression was compared in CVF samples collected from ewes ( n = 5) at each of the three time points: 0 h, 28 h and approximately 59 h after dexamethasone infusion. Differential expression of proteins was detected between all time points. Eight proteins were upregulated at delivery after preterm labour onset compared to the pre-induction samples ( Fig. 1 A , Table 1 ). In a previously published study, spot number 5004 was identified as bactenecin-1 [8] , however, seven additional proteins could not be unambiguously identified using in-gel digestion and MALDI-TOF MS. In this study, in-gel digestion coupled with MALDI-TOF peptide mass fingerprinting and LC–ESI-MS peptide sequence analysis identified a protein exhibiting a 4.2-fold expression increase between non-labour ( Fig. 1 B) and 59 h labour groups ( Fig. 1 C). This protein had an estimated molecular weight of 48 kDa and pI of 6.4. 3.2 Identification of α-enolase Using MALDI-TOF MS, a spectrum of peptides ( Fig. 2 A ) generated from a tryptic in-gel digestion was submitted to the Swiss-Prot database using the MASCOT search engine. Trypsin autolytic peaks were detected at 842.509 and 2211.104 m / z and used as internal calibrants. A positive identification was retrieved for α-enolase with a Mowse score of 141 and 50% sequence coverage ( Table 2 A ). Peptides matched to α-enolase and the amino acid sequences for these peptides are shown in Table 2 B. Confirmation of α-enolase identification was determined using capillary LC–ESI-MS/MS ( Fig. 2 B, Table 2 C). As an example, peptide at m / z 1390.67 was fragmented and the fragment ion series submitted to the Swiss-Prot database using the MASCOT search engine, which resulted in a positive identification of α-enolase with a MASCOT score of 66 . 3.3 Evaluation of α-enolase in preterm labour The protein expression of α-enolase was compared in the non-labouring group (0 h, sample taken prior to dexamethasone infusion) and labouring groups (28 and 58.7 ± 1.9 h after dexamethasone infusion). Although three out of the five samples exhibited increased levels of α-enolase (28 h post-dexamethasone infusion) when compared to pre-infusion samples, this failed to reach significance ( p = 0.15). In contrast, α-enolase was significantly increased 58.7 h after dexamethasone infusion ( Fig. 3 ). 3.4 Western blotting of human CVF samples The next aim was to confirm this observation in humans. Paired CVF samples were obtained from women during spontaneous labour onset and approximately 1 week before active labour ( n = 7 paired samples). Fig. 4 A is a Western blot demonstrating the protein expression of α-enolase in human CVF samples taken before labour and during spontaneous labour onset (but before rupture of human fetal membranes). There was a statistically significant increase in CVF α-enolase protein expression with human labour ( Fig. 4 B). 4 Comment Labour-associated changes in CVF proteins were characterised using 2-DE protein separation and mass spectrometry based peptide analysis. The data obtained establish that dexamethasone increased protein relative abundance of α-enolase, a neutrophil-derived glycolytic protein. Additionally, we used paired human CVF samples obtained from women before labour and during spontaneous labour onset to demonstrate increased expression of α-enolase during active labour. Mammalian enolases are composed of three isozyme subunits, and are cytoplasmic glycolytic enzymes. Alpha-enolase, also called non-neuronal enolase, is highly conserved among species, ubiquitous, and expressed in the early stage of embryonic development. It plays an important role in various pathophysiological processes [15] . From these studies, we cannot elucidate what role α-enolase has during labour. However, studies from non-gestational tissues demonstrate a role for α-enolase in remodelling and glycolysis; two processes that are increased in human labour. During the labour process there is an influx of inflammatory cells, including neutrophils, into the cervix [16–19] . Alpha-enolase bound to the surface of neutrophils has been found to act as a receptor for and activator of plasminogen [15] . Plasminogen is a known activator of collagenases that are responsible for the remodelling processes in the cervix required as labour progresses [20–22] . Alpha-enolase is considered a glycolytic enzyme. Labour is associated with increased glycolysis [23] , which plays an important role in the maintenance of uterine contractions by providing an energy source [24] . Hypoxia is a physiological feature in the uterine and cervical tissues during gestation that escalate with labour [25] . Recent studies have shown that α-enolase is upregulated in response to hypoxia in the placenta [26] . It is thus tempting to speculate that increased expression may play a role in two important processes of labour, cervical remodelling and uterine contractions. Although delivery was not allowed to occur in the sheep in this study, the EMG profiles were consistent with our previous studies using this method of induction [8,10] . This suggests that the samples were obtained at similar time points during the induction process in the five animals used. The finding that α-enolase was increased in only three of the five animals at the 28 h time point suggests that this may correspond to the commencement of processes that upregulate the expression of this protein. However, 59 h after dexamethasone treatment, α-enolase was increased in four of the five animals, with one animal demonstrating no change in expression between 0 and 59 h. Further studies of expression and protein abundance of α-enolase in the period from 28 h until the evolution of uterine activity are required to determine their timing in relation to labour onset. Similarly, in the human studies, we showed that α-enolase was significantly increased in CVF samples during spontaneous labour compared to samples taken approximately 1 week before labour. The terminal events of both preterm labour and term labour are the same; cervical ripening, increased uterine contracts and ECM remodelling. Thus, in the sheep model, preterm labour was induced and mimicked the initiation process of normal labour, i.e. increase in corticosteroids, while the comparison human cohort is from women in spontaneous term labour. In summary, in this study we utilised proteomics techniques to identify increased expression of α-enolase in CVF from an ovine model of preterm labour. This observation was confirmed in human studies using CVF samples taken from women approximately 1 week before labour and from the same women during spontaneous labour onset. The exact physiological role and function of α-enolase in the process of cervical ripening, myometrial activation and rupture of the fetal membranes overlying the cervix remain unknown and warrant further investigation. Alpha-enolase, in combination with bactenecin-1 [8] , may form a component of a panel of biomarkers for predicting preterm labour. The use of proteomic approaches to study global changes of protein expression provides new insights into the mechanism of human preterm birth. Funding The work described in this manuscript was partly funded by a National Health and Medical Research Council (NHMRC) Project grant. Professor Greg Rice is a Research Fellow supported by the NHMRC. Dr. Martha Lappas is in recipient of a NHMRC RD Wright Fellowship (grant no. 454777 ). Acknowledgements We thank Alex Satragno from the Department of Physiology, Monash University for his surgical expertise. The authors also express their thanks to Gabrielle Fleming and Mardi Reeves (Mercy Hospital for Women) for obtaining study samples, and Dr. Harry Georgiou and Yujing Heng (Department of Obstetrics and Gynaecology, University of Melbourne) for providing four of the CVF samples. References [1] S. Saigal L.W. Doyle An overview of mortality and sequelae of preterm birth from infancy to adulthood Lancet 371 2008 261 269 [2] W.W. Andrews R. Copper J.C. Hauth R.L. Goldenberg C. Neely M. Dubard Second-trimester cervical ultrasound: associations with increased risk for recurrent early spontaneous delivery Obstet Gynecol 95 2000 222 226 [3] J.D. Iams R.L. Goldenberg P.J. Meis B.M. Mercer A. Moawad A. Das The length of the cervix and the risk of spontaneous premature delivery, National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network N Engl J Med 334 1996 567 572 [4] R.L. Goldenberg B.M. Mercer P.J. Meis R.L. Copper A. Das D. McNellis The preterm prediction study: fetal fibronectin testing and spontaneous preterm birth, NICHD Maternal Fetal Medicine Units Network Obstet Gynecol 87 1996 643 648 [5] L. Pereira A.P. Reddy T. Jacob A. Thomas K.A. Schneider S. Dasari Identification of novel protein biomarkers of preterm birth in human cervical–vaginal fluid J Proteome Res 6 2007 1269 1276 [6] M.K. Di Quinzio K. Oliva S.J. Holdsworth M. Ayhan S.P. Walker G.E. Rice Proteomic analysis and characterisation of human cervico-vaginal fluid proteins Aust N Z J Obstet Gynaecol 47 2007 9 15 [7] M.K. Di Quinzio H.M. Georgiou S.J. Holdsworth-Carson M. Ayhan Y.J. Heng S.P. Walker Proteomic analysis of human cervico-vaginal fluid displays differential protein expression in association with labor onset at term J Proteome Res 7 2008 1916 1921 [8] I.R. Young G.E. Rice H.K. Palliser M. Ayhan N.L. Dellios J.J. Hirst Identification of bactenecin-1 in cervicovaginal fluid by two-dimensional electrophoresis in an ovine model of preterm labour Proteomics 7 2007 281 288 [9] P.L. Grigsby K.R. Poore J.J. Hirst G. Jenkin Inhibition of premature labor in sheep by a combined treatment of nimesulide, a prostaglandin synthase type 2 inhibitor, and atosiban, an oxytocin receptor antagonist Am J Obstet Gynecol 183 2000 649 657 [10] H.K. Palliser G.T. Ooi J.J. Hirst G. Rice N.L. Dellios R.M. Escalona Changes in the expression of prostaglandin E and F synthases at induced and spontaneous labour onset in the sheep J Endocrinol 180 2004 469 477 [11] M.M. Bradford A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein–dye binding Anal Biochem 72 1976 248 254 [12] H. Vorum M. Ostergaard P. Hensechke J.J. Enghild M. Riazati G.E. Rice Proteomic analysis of hyperoxia-induced responses in the human choriocarcinoma cell line JEG-3 Proteomics 4 2004 861 867 [13] K.T. Oliva M. Ayhan G. Barker N.L. Dellios M.A. Quinn G.E. Rice Proteomic profiling of ovarian cancer plasma using immunoaffinity depleted plasma and two-dimensional PAGE Clin Proteomics 3 2007 22 29 [14] M. Lappas M. Permezel G.E. Rice N-Acetyl-cysteine inhibits phospholipid metabolism, proinflammatory cytokine release, protease activity, and nuclear factor-kappaB deoxyribonucleic acid-binding activity in human fetal membranes in vitro J Clin Endocrinol Metab 88 2003 1723 1729 [15] V. Pancholi Multifunctional alpha-enolase: its role in diseases Cell Mol Life Sci 58 2001 902 920 [16] I. Osman A. Young M.A. Ledingham A.J. Thomson F. Jordan I.A. Greer Leukocyte density and pro-inflammatory cytokine expression in human fetal membranes, decidua, cervix and myometrium before and during labour at term Mol Hum Reprod 9 2003 41 45 [17] A.J. Thomson J.F. Telfer A. Young S. Campbell C.J. Stewart I.T. Cameron Leukocytes infiltrate the myometrium during human parturition: further evidence that labour is an inflammatory process Hum Reprod 14 1999 229 236 [18] A.J. Fosang C.J. Handley Connective tissue remodelling in the ovine cervix during pregnancy and at term Connect. Tissue Res 17 1988 277 285 [19] A.J. Fosang C.J. Handley V. Santer D.A. Lowther G.D. Thorburn Pregnancy-related changes in the connective tissue of the ovine cervix Biol Reprod 30 1984 1223 1235 [20] R.W. Kelly Inflammatory mediators and cervical ripening J Reprod Immunol 57 2002 217 224 [21] P.C. Leppert Anatomy and physiology of cervical ripening Clin Obstet Gynecol 38 1995 267 279 [22] M. Munoz-de-Torre J. Varayoud J.G. Ramos H.A. Rodriguez E.H. Luque Collagen remodeling is a key event in cervical ripening for a successful vaginal delivery Braz J Morphol Sci 20 2003 75 84 [23] C.S. Chew G.A. Rinard Glycogen levels in the rat myometrium at the end of pregnancy and immediately postpartum Biol Reprod 20 1979 1111 1114 [24] M.L. Kashyap R. Sivasamboo S.P. Sothy J.S. Cheah P.S. Gartside Carbohydrate and lipid metabolism during human labor: free fatty acids, glucose, insulin, and lactic acid metabolism during normal and oxytocin-induced labor for postmaturity Metabolism 25 1976 865 875 [25] S. Wray Hypoxia in the uterus News Physiol Sci 9 1994 88 92 [26] R. Hu H. Jin S. Zhou P. Yang X. Li Proteomic analysis of hypoxia-induced responses in the syncytialization of human placental cell line BeWo Placenta 28 2007 399 407