the recovery of pro-CRH125â151, the synthetic peptide (Peninsula Labo- ratories, Inc. .... (19 kDa confirmed by Western blotting with M2; data not shown); the ...
0021-972X/00/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2000 by The Endocrine Society
Vol. 85, No. 2 Printed in U.S.A.
Processing of Procorticotropin-Releasing Hormone (Pro-CRH): Molecular Forms of CRH in Normal and Preeclamptic Pregnancy* I. AHMED, B. P. GLYNN, A. V. PERKINS, M. G. CASTRO, J. ROWE, E. MORRISON, AND E. A. LINTON Nuffield Department of Obstetrics and Gynecology, University of Oxford, John Radcliffe Hospital (I.A., B.P.G., A.V.P., E.A.L.), Headington, Oxford, United Kingdom OX3 9DU; the Molecular Medicine Unit, Department of Medicine, University of Manchester (M.G.C., J.R., E.M.), Manchester, United Kingdom M13 9PT ABSTRACT This study examined the different molecular forms of CRH in normal and preeclampsia maternal plasma and protease-blocked placental extracts using antibodies to different regions of the CRH precursor, pro-CRH. In the absence of protease inhibitors, chromatographed normal placental extracts contained four peaks of immunoreactivity corresponding to unprocessed approximately 19kDa pro-CRH, its approximately 8-kDa intermediate metabolite, proCRH125–194, its approximately 2.8-kDa midportion fragment, proCRH125–151, and 4.75-kDa CRH1– 41. However, if protease inhibitors were included in the extraction medium, only pro-CRH and proCRH125–194 were found. Pro-CRH processing was more extensive in protease-blocked preeclampsia placentas than in those from normal pregnancy, with three peaks corresponding to pro-CRH, proCRH125–194, and mature CRH1– 41 peptide found. Using quantitative competitive PCR, the messenger ribonucleic acid levels of CRH precursor in preeclampsia placentas were 1.7-fold higher than those in normal placentas (37.83 ⫾ 3.48 vs. 21.83 ⫾ 2.59 attomoles/g total ribonucleic acid, respectively; P ⬍ 0.005). Preeclampsia placentas contained significantly more CRH1– 41 cross-reactivity (4.72 ⫾ 1.22 pmol/g) than normal term placentas (1.52 ⫾ 0.39 pmol/g; P ⬍ 0.048) extracted in medium containing protease inhibitors. The content of pro-CRH125–151-reactive species in these extracts followed the same
pattern, with more immunoreactivity detected in preeclampsia placentas (4.23 ⫾ 1.39 pmol/g) than in those from normal term pregnancies (1.44 ⫾ 0.32 pmol/g; P ⬍ 0.01). Sequential plasma samples from 10 women with normal pregnancy and 5 women with preeclampsia were assayed for pro-CRH125–151- and CRH1– 41-immunoreactive species. In normal pregnancy, maternal plasma CRH1– 41 immunoreactivity rose with increasing gestational age, reaching 460 ⫾ 48 pmol/L at term. In women with preeclampsia, CRH1– 41 levels at each gestational age point were higher than those at the equivalent stage of normal pregnancy. In contrast, the levels of pro-CRH125–151-immunoreactive species remained barely detectable throughout normal and preeclamptic pregnancy. Both pro-CRH and CRH1– 41, but not pro-CRH125–151, were shown to bind to the plasma CRH-binding protein. Our findings highlight the importance of protection of placental tissue from degrading enzymes during extraction and show that most of the CRH in the human placenta exists as unprocessed pro-CRH, with very little in the form of CRH1– 41 except in preeclampsia. Our studies using maternal plasma indicate that CRH1– 41 is the only one of the pro-CRH fragments studied to be maintained in significant amounts in the maternal circulation and also the only fragment studied for which a specific plasma binding protein exists. (J Clin Endocrinol Metab 85: 755–764, 2000)
M
OST PEPTIDE hormones are synthesized as large prohormones from which they are released posttranslationally by the action of endopeptidases, usually at pairs of basic amino acid residues flanking the biologically active peptide (1, 2). The human placenta produces substantial quantities of CRH messenger ribonucleic acid (mRNA) and peptide, both identical to their hypothalamic counterparts (3). CRH is a 41-amino acid peptide, produced as the Cterminal portion of a 196-amino acid CRH precursor (proCRH). After removal of the signal peptide (4, 5) and Cterminal amidation, this precursor, pro-CRH27–194, has a molecular size of about 19 kDa (6, 7). Pro-CRH contains two potential cleavage sites, CS1124 –125 and CS2151–152 (6, 7). Endoproteolytic cleavage at the pair of basic amino acids at CS2
would give rise to pro-CRH27–151 and mature CRH1– 41 peptide, whereas cleavage at CS1 would result in two other peptides, an N-terminal fragment, pro-CRH27–124 and the approximately 8-kDa intermediate metabolite, pro-CRH125– 194. Cleavage at both sites would give rise to three products: the N-terminal fragment, CRH27–124; the midportion fragment, pro-CRH125–151; and the C-terminal fragment, CRH1– 41. Although several reports exist documenting the presence of CRH1– 41 in the human placenta, little is known about the existence of the other potential cleavage products of proCRH. Previous chromatographic studies found that term chorionic villous extracts contained predominantly mature CRH1– 41 peptide (3, 8 –13), although smaller quantities of various other molecular weight species were also identified in some of these reports (10 –13). These findings imply that most of the pro-CRH in the placenta undergoes immediate posttranslational processing with subsequent storage of mature CRH1– 41 peptide, which is the same fate as that of the CRH precursor in the hypothalamus. However, extraction was carried out in the absence of protease inhibitors in these
Received February 10, 1999. Revision received May 17, 1999. Rerevision received October 21, 1999. Accepted October 28, 1999. Address all correspondence and requests for reprints to: Dr. E. A. Linton, Nuffield Department of Obstetrics and Gynecology, University of Oxford, John Radcliffe Hospital, Headington, Oxford, United Kingdom OX3 9DU. * This work was supported by the Wellcome Trust.
755
756
AHMED ET AL.
earlier studies, and in many cases, placentas delivered vaginally after labor were used without consideration for the artifactual processing that may accompany tissue deterioration associated with lengthy labor. The present study was designed to identify which molecular forms of CRH are present in normal and preeclampsia placentas protected from collection and extraction artifacts, using antibodies raised against distinct regions of the CRH precursor. These antibodies have recently been used to identify CRH1– 41, proCRH125–151, and pro-CRH125–194 as the main processed products of stably transfected AtT20 cells expressing the human prepro-CRH gene (7), demonstrating the ability of these antibodies to detect the pro-CRH peptides. Circulating maternal CRH1– 41 levels rise from very low picomolar amounts in the first half of human pregnancy to reach concentrations several hundred-fold higher at term. Within hours of parturition, plasma CRH1– 41 falls rapidly to baseline levels, suggesting that the placenta is the source of this CRH1– 41 (14, 15). The syncytiotrophoblasts of the chorionic villi are the main placental cell type producing CRH (16, 17), although the fetal membranes also contain CRH immunoreactivity (9, 16, 17). As the amnion and chorion are poorly vascularized, it is unlikely that these latter layers contribute significantly to the high levels of the peptide in the circulation. A 37-kDa binding protein (CRH-BP) specific for human CRH1– 41 (18) and other CRH-related peptides, such as urocortin (19), is found in human plasma and can modulate the bioactivity of its ligands (20). In the final weeks of pregnancy, plasma levels of CRH-BP fall by approximately 50% (21–24). In pregnancies complicated by preeclampsia, plasma CRH levels rise earlier and to a greater extent than in normal pregnancy (14, 25, 26), and CRH-BP levels are lower (22). Several potential roles have been proposed for placental CRH, notably the control of the length of human pregnancy (23). In contrast to the now well documented profiles of plasma CRH1– 41 during pregnancy, little is known about the circulating levels of any other cleavage products of pro-CRH, or indeed whether they also bind to the CRH-BP. In this study investigating the molecular forms of placental CRH immunoreactivity, we also sought to determine whether the potential cleavage fragments, pro-CRH125–194 and pro-CRH125–151 are present in the maternal circulation in the third trimester of normal and preeclamptic pregnancies when CRH1– 41 is high. The ability of pro-CRH and proCRH125–151 to bind to native CRH-BP was also explored. Materials and Methods Experimental subjects Placentas from normal term pregnancies (n ⫽ 6; gestational age, 38 – 40 weeks) were collected immediately after elective cesarean section and processed rapidly as described below. Placentas from preeclamptic pregnancies (n ⫽ 6; gestational age, 29 –32 weeks) were collected similarly. Preeclampsia is defined by new hypertension (⬎140/90 mm Hg) and new proteinuria greater than 500 mg in 24 h, remitting after delivery. Sequential plasma samples were collected fortnightly from 10 women throughout the second half of normal pregnancy. Similar samples from 5 women with preeclampsia, taken from the study described previously (21) and stored in the intervening period at ⫺80 C, were also used for this study. This investigation had the approval of the central Oxfordshire research ethics committee, and all patients gave their informed consent.
JCE & M • 2000 Vol 85 • No 2
Tissue extraction Placental chorionic villi. Villous tissue was rinsed with phosphate-buffered saline [PBS; 0.02 mol/L sodium phosphate and 0.9% (wt/vol) NaCl] and homogenized (4 mL/g) in the following buffers 1) RIPA buffer [150 mmol/L NaCl, 50 mmol/L Tris (pH 8), 1% (vol/vol) Nonidet P-40, and 0.1% (wt/vol) SDS] containing the following protease inhibitors: 1 mmol/L phenylmethylsulfonylfluoride; 1 g/mL each of antipain, chymostatin, and pepstain A; and 2 g/mL aprotinin and leupeptin (Roche Molecular Biochemicals, Indianapolis, IN); 2) RIPA buffer alone; and 3) (for Western blotting only) PBS alone. The extracts were centrifuged at 4000 ⫻ g for 30 min at 4 C, and the supernatants aliquoted into 1-mL volumes and stored at ⫺70 C. CHO cells. To provide a source of CRH precursor for control studies, cell pellets were prepared from stably transfected Chinese hamster ovary (CHO-K1) cells expressing the rat CRH precursor as described previously (27). The cell pellets were homogenized in 3 mL of each of the following extraction media: 1) RIPA containing the cocktail of protease inhibitors listed above, 2) RIPA buffer alone, and 3) (for Western blotting only) PBS alone. After holding on ice for 30 min, the homogenates were sonicated and centrifuged at 10,000 ⫻ g for 10 min at 4 C; supernatants were aliquoted into 1-mL volumes and stored at ⫺70 C.
Plasma extraction As it is known that the CRH-BP interferes with CRH estimation, maternal plasma samples were extracted using ice-cold methanol as described previously (28), reconstituted to their original volume with assay buffer [PBS containing 0.5% (wt/vol) BSA (Sigma, St. Louis, MO) and 0.01% (wt/vol) sodium azide] and then assayed for CRH1– 41 and the midportion fragment, pro-CRH125–151, as described below. To estimate the recovery of pro-CRH125–151, the synthetic peptide (Peninsula Laboratories, Inc., Belmont, CA) was added to normal human plasma (Oxford Regional Transfusion Service, Oxford, UK) at three concentrations (3.57, 1.78, and 0.357 nmol/L; n ⫽ 6 at each concentration), mixed, and extracted with methanol as described above. One milliliter of CHO-K1 cell extract prepared in RIPA buffer with protease inhibitors was also subjected to methanol extraction as described above.
Antibodies The rabbit polyclonal antibodies used in these studies were coded M2 for CRH1– 41 (7, 28) raised against synthetic human CRH1– 41, SJ2 for the midportion fragment, raised against synthetic human pro-CRH125–151 (7), and 781 for CS2 in the CRH precursor, raised against the peptide fragment Gly-Ala-Leu-Glu-Arg-Glu-Arg-Arg-Ser-Glu-Glu-Pro-Pro-IleTyr, the first 14 amino acids of which correspond to rat pro-CRH137–150 (6); all but the first 3 amino acids of this fragment are identical in the equivalent CS2 region of the human CRH precursor, human pro-CRH149 –159.
Western blotting PAGE (12%) was carried out in the presence of 0.1% SDS as described by Laemmli (29). Immunoblotting was carried out by transferring protein onto a 0.2-m nitrocellulose membrane (Bio-Rad Laboratories, Inc., Richmond, CA) using a semidry blotting apparatus (Bio-Rad Laboratories, Inc.) at 15 V for 1 h. The membrane was incubated in a solution of 5% milk powder (Marvel, Adbaston, Stafford, UK) for 2 h at room temperature, followed by overnight incubation with primary antibody, M2 (1:500). After six 100-mL washes in PBS, a secondary antibody, goat antirabbit IgG conjugated to horseradish peroxidase (1:1000; DAKO Corp. A/S, Glostrup, Denmark) was added and incubated for 3 h at room temperature. The membrane was washed with PBS as described for the primary antibody incubation step. The antibody complex was visualized using an ECL chemiluminescence kit (Pharmacia Biotech, Uppsala, Sweden) and exposure to ECL x-ray film (Pharmacia Biotech) for approximately 1 minute.
Gel filtration chromatography Columns were first calibrated with the following: BSA (68 kDa; Sigma), ovalbumin (45 kDa; Sigma), soybean trypsin inhibitor (20 kDa;
PRO-CRH PROCESSING IN PREGNANCY Sigma), cytochrome c (13 kDa; Sigma), 1 ⫻ 106 cpm of both 125I-labeled CRH1– 41 (4.75 kDa) and 125I-labeled pro-CRH125–151 (⬃2.8 kDa). Sephadex G-50 chromatography. One-milliliter aliquots of normal term placental chorionic villous extract (n ⫽ 3) or CHO-K1 cell extract (n ⫽ 3), prepared in RIPA buffer with and without protease inhibitors, were chromatographed on Sephadex G-50 (fine; Pharmacia Biotech; column size, 95 ⫻ 1.6 cm). One-milliliter aliquots of chorionic villous extracts from preeclampsia placentas prepared in RIPA buffer containing the cocktail of enzyme inhibitors were similarly chromatographed. Samples were eluted in PBS containing 0.5% BSA at a flow rate of 10 mL/h, and 1-mL fractions were collected. All fractions were assayed for CRH1– 41, pro-CRH125–151, and intact CS2 using the RIAs described below. Sephadex G-100 chromatography. To investigate the binding of the CRH precursor, CRH1– 41, and pro-CRH125–151 to purified CRH-BP and other binding proteins that may be present in human plasma, aliquots of each of the following mixtures were chromatographed on Sephadex G-100 (Pharmacia Biotech; column size, 52 ⫻ 1.5 cm): 1) 1 mL purified human CRH-BP (20 pmol) preincubated with approximately 1 ⫻ 106 cpm (5.2 pmol in 5 L) 125I-labeled CRH1– 41 for 30 min at 37 C; 2) 1 mL purified human CRH-BP (20 pmol) preincubated with cold CRH1– 41 (0.6 nmol in 5 L) for 30 min at 37 C; 3) 0.5 mL CHO-K1 cell extract (prepared in RIPA buffer containing protease inhibitors) preincubated with 0.5 mL purified human CRH-BP (10 pmol) for 30 min at 37 C; 4) 1 mL purified human CRH-BP (20 pmol) preincubated with approximately 1 ⫻ 106 cpm (5.2 pmol in 10 L) 125I-labeled midportion fragment, pro-CRH125–151, for 30 min at 37 C; 5) 0.6 nmol in 5 L cold pro-CRH125–151 preincubated with 1 mL purified human CRH-BP for 30 min at 37 C; and 6) 1 mL normal human plasma preincubated with approximately 1 ⫻ 106 cpm (5.2 pmol in 10 L) 125I-labeled pro-CRH125–151 for 30 min at 37 C. All samples were eluted in PBS at 10 mL/h, and 1 mL fractions were collected. In 1, 4, and 5, radioactivity in each fraction was counted directly in a ␥-counter. In 2, 3, and 6 the fractions were first methanol extracted, then assayed using the appropriate RIAs as described below.
Iodination Synthetic pro-CRH125–151, CRH1– 41, and CS2 peptides (0.6 nmol) were iodinated using Iodogen (30). For CRH1– 41 and pro-CRH125–151, iodination of the histidine residues was carried out at pH 8.4 in sodium bicarbonate (200 mmol/L), whereas iodination of the tyrosine residue in the CS2 peptide was carried out at neutral pH, with sodium phosphate (200 mmol/L; pH 7.4) replacing sodium bicarbonate. Labeled peptide was purified on a C4 column (Macherey-Nagel, Duren, Germany) for CRH1– 41 and pro-CRH125–151, but on a Sep-Pak C18 cartridge (Water Corp., Nantwich, UK) for the CS2 peptide.
RIA protocols RIAs for CRH1– 41, pro-CRH125–151, and CS2 were carried out as described previously (7) using antibodies M2 at 1:10,000, SJ2 at 1:1,500, and 781 at 1:150 initial dilutions, respectively. To test for plasma protein and/or CRH-BP interference with binding of pro-CRH125–151 tracer to SJ2 antibody, two additional pro-CRH125–151 standard curves were prepared in which the dilutions were carried out in normal human plasma and assay buffer containing CRH-BP (20 nmol/L), respectively. CRH-BP was purified from human plasma as described previously (18) with modifications detailed by Perkins et al. (22).
Total RNA preparation and complementary DNA (cDNA) synthesis Total RNA was prepared from individual placentas using Trizol reagent (Life Technologies, Inc., Paisley, UK), according to the manufacturer’s instructions. cDNA was synthesized from RNA samples using the SuperScript Preamplification System (Life Technologies, Inc., Paisley, UK). The primer was annealed to the RNA from the tissue of interest at 70 C for 10 min, and synthesis was carried out at 37 C for 60 min.
Quantitative competitive PCR PCR was carried out using a PCR Master Mix Kit (QIAGEN, West Sussex, UK). The PCR program consisted of preincubation at 95 C for
757
5 min, denaturation at 95 C for 1 min, primer annealing at 56 C for 1 min, and extension at 72 C for 3 min, (40 cycles), with a final extension at 72 C for 5 min in a thermal cycler (model 480, PE Applied Biosystems, Cheshire, UK). Our quantitative competitive PCR assay for pro-CRH was based on the method of Celi and colleagues (31). The primers for pro-CRH (Life Technologies, Inc.) were as follows: forward, 481 5⬘-TGGATCTCACCTTCCACCTC-3⬘ 500; and reverse, 788 5⬘-CATTGTGTTGCTGCTGCAC-3⬘ 770, generating a fragment of 308 bp. A 227-bp competitor fragment was generated using the following primers: the 20-mer proCRH forward primer (bases 481–500) and a composite 40-mer reverse primer made up of the 19-mer pro-CRH reverse primer described above (bases 788 –770) with an additional 21 bases (688 – 668 of pro-CRH) at its 3⬘-end. Quantitative competitive PCR was performed as described by Rodriguez-Linares and colleagues for the CRH R1 receptor (32). All PCR reaction products were resolved on 1.5% agarose gels and visualized under UV light with ethidium bromide. The variability in RNA recovery was normalized by reference to a housekeeping gene, human glucocerebrosidase, using a primer set designed to amplify a 572-bp fragment as previously described (32). The amount of cDNA competitor fragment required to produce an equimolar amount of target cDNA was calculated as described (32) and was expressed as attomoles per g total RNA.
Statistics The concentrations of CRH1– 41 and pro-CRH125–151 immunoreactivities and pro-CRH mRNA in normal and preeclampsia placentas are given as the mean of the indicated number of samples ⫾ sem. The unpaired t test was used to compare levels in these placentas, with P ⬍ 0.05 considered significant.
Results Placenta
Western blotting. A CRH-immunoreactive band running with the 19.5-kDa standard corresponding to pro-CRH was detected by the M2 antibody in both chorionic villous and CHO-K1 cell extracts subjected to SDS-PAGE (Fig. 1A). The intensity of the band was dependent on the type of extraction medium used. The most intense band was observed in lanes corresponding to extracts prepared in RIPA buffer containing protease inhibitors, whereas the least intense band was observed in lanes with extracts prepared in PBS alone. Sephadex G-50 chromatography. RIAs indicated that the control CHO-K1 cells extracted in RIPA buffer without protease inhibitors gave a single peak of immunoreactivity which eluted in the void region (Vo) of the Sephadex G-50 chromatogram (Fig. 2A), as detected with CRH1– 41, proCRH125–151, and CS2 antibodies. An identical profile was obtained when the CHO-K1 cells were extracted in RIPA buffer containing protease inhibitors (data not shown). Western blotting using M2 antibody (Fig. 1B) was performed on the fractions from the Vo region of the chromatograph (Fig. 2A) to assess the molecular size of CRH-immunoreactive material. The results confirmed the presence in these fractions of an immunoreactive band of about 19 kDa, representing pro-CRH27–194. The amount of immunoreactivity measured at Vo in the three immunoassays differed depending on the antibody used, indicating varying degrees of cross-reactivity of each antibody with the CRH precursor molecule. The lack of pro-CRH cleavage products confirms that these CHO fibroblast cells possess negligible processing enzyme activity (27). The normal term placental extracts displayed different
758
AHMED ET AL.
JCE & M • 2000 Vol 85 • No 2
FIG. 1. A, SDS-PAGE and Western blotting using M2 antibody of extracts of 4 g each of CHO-K1 cells expressing rat pro-CRH and normal term chorionic villi prepared in various extraction media. Lane 1, Molecular weight markers [phosphorylase B (111.7 kDa), BSA (84.4 kDa), ovalbumin (53.2 kDa), carbonic anhydrase (34.2 kDa), soybean trypsin inhibitor (27.6 kDa), and lysozyme (19.5 kDa)]. Lanes 2– 4 and 5–7 correspond to CHO-K1 cell extract and term chorionic villous extract, respectively, prepared in RIPA buffer containing protease inhibitors, RIPA buffer alone, and PBS alone. Lanes 8 and 9 correspond to unextracted maternal plasma (20 L) at 38 and 40 weeks gestation, respectively. B, Sephadex G-50 fractions (Vo) of CHO-K1 cell extract prepared in RIPA buffer containing protease inhibitors (see Fig. 2A). Lane 1, Molecular weight markers. Lanes 2–7, Chromatography fractions 70 –75 from Fig. 2A. C, Sephadex G-50 fractions (Vo) of normal term chorionic villous extract prepared in RIPA buffer containing protease inhibitors (see Fig. 2B). Lane 1, Molecular weight markers. Lanes 2–5, Chromatography fractions 70 –73 from Fig. 2B. D, Recovery of pro-CRH after methanol extraction of CHO-K1 cell extract. Lane 1, Molecular weight markers. Lanes 2 and 3, after and before methanol extraction, respectively.
degrees of processing of pro-CRH depending on the extraction medium used. The profile in Fig. 2B shows that minimal processing occurred when protease inhibitors were included in the extraction buffer. Most of the CRH immunoreactivity was found in the Vo region, and this was detected by all three RIAs. There was an additional small peak detected by M2 and SJ2 RIAs eluting between 13-kDa cytochrome c and 4.75-kDa synthetic CRH1– 41 markers, which most likely corresponds to the intermediate metabolite, pro-CRH125–194, of approximately 8 kDa. Western blots of the Vo region confirmed the presence of approximately 19-kDa pro-CRH (Fig. 1C). When protease inhibitors were omitted from the extraction medium (Fig. 2C), pro-CRH processing was more pronounced, and four distinct immunoreactive peaks were resolved by chromatography. The first (Vo region) peak, again detected with all three RIAs, corresponds to pro-CRH27–194 (⬃19 kDa confirmed by Western blotting with M2; data not shown); the second peak (fraction 110) eluting between the 13- and 4.75-kDa markers, detected by M2 and SJ2 RIAs, correlates to CRH125–194; the third peak (fraction 119), crossreacting in the M2 RIA only, coeluted with the CRH1– 41 marker; the fourth peak (fraction 130), detected by the SJ2 RIA only, coeluted with the pro-CRH125–151 marker. Figure 2D shows the typical profile obtained when preeclampsia chorionic villous tissue was extracted in RIPA buffer containing enzyme inhibitors and subjected to gel permeation chromatography. Most of the CRH immunoreactivity eluted immediately after the Vo, together with two additional small peaks, one eluting between cytochrome c and the synthetic CRH1– 41 marker, corresponding to the approximately 8-kDa intermediate metabolite, pro-CRH125–194,
and the other coeluting with synthetic CRH1– 41. The assay for the midportion fragment confirmed the identity of the first two peaks as pro-CRH and pro-CRH125–194, but no proCRH125–151 was eluted from the column. The assay for CS2 also confirmed the first peak as pro-CRH. Placental content of pro-CRH mRNA. As demonstrated in Fig. 3A, pro-CRH mRNA was quantitated by using doubling quantities of competitor fragment cDNA (1.56 –200 attomoles) in the PCR reactions. Quantitation of target DNA in human placenta is possible when the competitor and target DNA bands are of equal intensity, by reference to the quantity of competitor fragment included in that PCR reaction. In this way, the mean level of mRNA for pro-CRH in preeclampsia placentas was calculated to be 1.7-fold higher than that in normal term placentas (37.83 ⫾3.48 vs. 21.83 ⫾ 2.59 attomoles/g total RNA, respectively; P ⬍ 0.005; Fig. 3B). Placental content of CRH1– 41 and pro-CRH125–151-like immunoreactivities. Chorionic villous tissue from preeclampsia and normal term placentas extracted in RIPA buffer containing protease inhibitors were assayed using CRH1– 41 and proCRH125–151 RIAs, with the resulting peptide levels shown in Fig. 4. Preeclampsia placentas contained significantly more CRH1– 41 immunoreactivity (4.72 ⫾ 1.22 pmol/g) than normal term placentas (1.52 ⫾ 0.39 pmol/g; P ⬍ 0.048). Immunoassay of pro-CRH125–151-reactive species in the same extracts indicated a similar content on a molar basis, with greater tissue content in preeclampsia placentas (4.23 ⫾1.39 pmol/g) than in normal term placentas (1.44 ⫾ 0.32 pmol/g; P ⬍ 0.01).
PRO-CRH PROCESSING IN PREGNANCY
759
FIG. 3. A, Quantitation of the pro-CRH gene in human placenta. Lanes 3–10, Placental RNA was reverse transcribed, and the resulting cDNA was used as a template for PCR amplification in the presence of doubling quantities (1.56 –200 attomoles) of the homologous cDNA competitor fragment. Lanes 1 and 11, Molecular weight markers; lane 2, blank. B, Pro-CRH mRNA levels in preeclampsia and normal term placentas (n ⫽ 6 in each group). After image analysis of the bands in A and linear regression analysis, data were expressed as attomoles per g total RNA.
FIG. 2. Sephadex G-50 chromatography of CHO-K1 cells extracted in RIPA buffer (A), normal term chorionic villi extracted in RIPA buffer containing protease inhibitors (B), normal term chorionic villous extract prepared in RIPA buffer without protease inhibitors (C), and preeclampsia chorionic villous extract prepared in RIPA buffer containing protease inhibitors (D). All fractions were assayed using CRH1– 41 (⽧), proCRH125–151 (E), and CS2 (䡺) RIAs. The arrows indicate the Vo; the elution positions of cytochrome c (13 kDa), CRH1– 41 (4.75 kDa), and pro-CRH125–151 (⬃2.8 kDa); and the total column volume (Vt). The column size was 95 ⫻ 1.6 cm; the eluant was PBS containing 0.5% BSA; the flow rate was 10 mL/h; the fraction volume was 1 mL.
Maternal plasma
To determine whether the midportion cleavage product of the CRH precursor would be bound by the CRH-BP, 125Ilabeled pro-CRH125–151 was incubated with purified human
CRH-BP and subjected to Sephadex G-100 chromatography. The elution profile obtained (Fig. 5A) shows a peak of radioactivity eluting at a volume corresponding to free 125Ilabeled pro-CRH125–151 (fraction 110). For comparison, Fig. 5A also shows the profile obtained using purified CRH-BP preincubated with 125I-labeled CRH1– 41; the majority of labeled CRH1– 41 eluted at fraction 56, confirming its high affinity binding to CRH-BP. A smaller quantity of unincorporated 125I-labeled CRH1– 41 eluted at fraction 110. To rule out the possibility that the pro-CRH125–151-binding site for CRH-BP had been denatured during iodination, the above experiment was repeated using cold pro-CRH125–151 in place of the radioiodinated peptide, and the chromatographic fractions were assayed with the pro-CRH125–151 RIA. A similar profile was obtained (Fig. 5B), with all midportion fragment immunoreactivity eluting as the free peptide in fraction 110. It is unlikely, then, that iodination of pro-CRH125–151 damaged the peptide. For further comparison, Fig. 5B illustrates the profile obtained when cold CRH1– 41 was incubated with CRH-BP; the large peak at fraction 56 corresponds to the CRH1– 41:CRH-BP complex. To ascertain whether pro-CRH binds to CRH-BP, CHO-K1 cell extract was preincubated with purified CRH-BP and chromatographed, and the resulting fractions were assayed by CRH1– 41 RIA. The profile is also illustrated in Fig. 5B. In addition to a small peak at Vo,
760
AHMED ET AL.
JCE & M • 2000 Vol 85 • No 2
FIG. 4. Levels of CRH1– 41-immunoreactive species (solid bars) and pro-CRH125–151-immunoreactive species (open bars) in chorionic villous extracts prepared in RIPA buffer containing protease inhibitors from preeclampsia and normal term placentas (n ⫽ 6 in each group). Extracts were assayed using M2 and SJ2 antibodies, respectively.
two peaks were observed, the larger peak at fraction 76 eluting in the position of free pro-CRH of about 19 kDa and the smaller peak eluting at fraction 47, consistent with the expected elution position of a pro-CRH:CRH-BP complex with a molecular mass of approximately 19 ⫹ 37 ⫽ 56 kDa. This latter peak contained approximately one quarter of the total CRH immunoreactivity measured in the 42-kDa CRH1– 41:CRH-BP complex peak, although only half the quantity of CRH-BP had been used for incubation. As the amount of pro-CRH immunoreactivity bound was proportionately half of that in the CRH1– 41:CRH-BP peak, the affinity of CRH-BP for pro-CRH is less than that for CRH1– 41, if it is assumed that the cross-reactivities of these two CRH species are equivalent. The small Vo peak of immunoreactivity in this profile, as in the other profiles in Fig. 5, probably indicates peptide binding nonspecifically to large molecular weight serum proteins in the purified CRH-BP preparation, which is stored in 5% sheep serum for stability. These results demonstrate that CRH-BP has a high affinity for CRH1– 41, lower affinity for pro-CRH, and negligible affinity for pro-CRH125–151. To confirm that the CRH-BP has minimal binding for the midportion fragment, two pro-CRH125–151 standard curves were prepared, one in assay buffer and the other in assay buffer containing purified CRH-BP. The pro-CRH125–151 immunoassay (Fig. 6) demonstrated that the curves were almost superimposable, indicating that the CRH-BP present in the assay buffer did not effect the binding of pro-CRH125–151 to SJ2 antibody. To determine whether proteins present in human plasma interfere with the pro-CRH125–151 RIA, a third pro-CRH125–151 standard curve was prepared in normal human plasma. As shown in Fig. 6, the degree of binding of SJ2 antibody to pro-CRH125–151 tracer was somewhat enhanced. To investigate the specificity of this interference, normal human plasma that had been preincubated with 125I-labeled pro-CRH125–151 was subjected to G-100 chromatography. The profile (Fig. 5A) demonstrates that the majority of proCRH125–151 eluted as the free peptide with only a small peak
FIG. 5. A, Sephadex G-100 chromatography of purified human CRH-BP preincubated with 1 ⫻ 106 cpm [125I]CRH1– 41 (〫) or [125I]pro-CRH125–151 (E) and of normal human plasma preincubated with 1 ⫻ 106 cpm [125I]pro-CRH125–151 (F). B, Sephadex G-100 chromatography of purified human CRH-BP preincubated with proCRH125–151 (E), CRH1– 41 (〫), and CHO-K1 cell extract (F). All incubations were carried out for 30 min at 37 C. Radioactivity in each fraction in A was counted directly, whereas binding to CRH-BP in B was measured by CRH1– 41 and pro-CRH125–151 RIAs. The arrows indicate the Vo; the elution positions of CRH-BP/pro-CRH complex (56 kDa), CRH-BP:CRH1– 41 complex (42 kDa), and pro-CRH (19 kDa); and the total column volume (Vt). The column size was 52 ⫻ 1.5 cm; the eluant was PBS; the flow rate was 10 mL/h; the fraction volume was 1 mL.
close to the void volume (fraction 33), probably due to proCRH125–151 binding to albumin present in the plasma. These data suggest that there is no specific binding protein for pro-CRH125–151 present in human plasma. As human plasma appeared to interfere in the proCRH125–151 RIA by enhancing SJ2 antibody binding to proCRH125–151 tracer, there was a need for extraction of proCRH125–151 before its plasma measurement. The suitability of methanol for this purpose was investigated, as this method had previously been used successfully for CRH1– 41 (28). ProCRH125–151 was used to spike human plasma and was extracted with methanol; the efficiency of the recovery of this synthetic peptide varied between 102–105% (data not shown). Western blotting using M2 antibody (Fig. 1D) shows
PRO-CRH PROCESSING IN PREGNANCY
FIG. 6. The effects of CRH-BP and normal human plasma on the binding of pro-CRH125–151 to SJ2 antibody. Pro-CRH125–151 standard curves were prepared in assay buffer (F), in assay buffer containing purified human CRH-BP (䡺), and in normal human plasma (E).
the successful recovery of pro-CRH in RIPA buffer containing protease inhibitors after methanol treatment, demonstrating the suitability of methanol as an extraction solvent for this CRH species also. A similar investigation for proCRH125–194 was not possible due to the lack of a supply of the peptide, but the good recovery of recombinant pro-CRH, a larger molecule than pro-CRH125–194, suggests that the intermediate metabolite is unlikely to be denatured during exposure to methanol. The concentrations of CRH1– 41 and pro-CRH125–151 in plasma from women sampled at fortnightly intervals throughout normal and preeclamptic pregnancy were determined. The plasma CRH1– 41 profiles in the women with normal pregnancy (Fig. 7) were similar to those reported previously; CRH1– 41 levels rose from week 28, increased steadily to week 34, then rose more rapidly to term. A similar profile was obtained in the women who developed preeclampsia, although their CRH levels were elevated precociously, with higher levels of CRH immunoreactivity detected at each gestational age point than at the equivalent stage of normal pregnancy. For example, at 38 weeks gestation, the women with preeclampsia displayed mean plasma levels of 585 ⫾ 72 pmol/L, whereas the level was 387 ⫾ 48 pmol/L in those with normal pregnancy. No such pattern was seen for pro-CRH125–151, levels of which were negligible throughout gestation in women with normal pregnancies and in those with preeclampsia. No pro-CRH band was detected in Western blots of unextracted maternal plasma at 38 and 40 weeks of normal pregnancy (Fig. 1A, lanes 7 and 8). Discussion
Our study confirms the presence of pro-CRH mRNA and its approximately 19-kDa protein product in chorionic villi of the human placenta, in keeping with earlier observations that
761
FIG. 7. Levels of CRH1– 41-immunoreactive species (solid lines) and pro-CRH125–151-immunoreactive species (dotted lines) present in maternal plasma during normal (F; n ⫽ 10) and preeclamptic (E; n ⫽ 5) pregnancies. Sequential samples were collected at fortnightly intervals from women throughout the second and third trimesters of pregnancy and assayed using M2 and SJ2 antibodies, respectively.
the gene encoding pro-CRH is expressed in the villous placenta (33). As neither CRH1– 41 nor pro-CRH125–151 was detected in the chromatograms of protease-inhibited placental extracts, our results indicate that the CRH precursor remains largely unprocessed. Some processing does occur, however, as the approximately 8-kDa intermediate metabolite, proCRH125–194, was found in all placental extracts. Further processing of placental pro-CRH to produce mature CRH1– 41 and pro-CRH125–151 in addition to the intermediate metabolite occurred only when protease inhibitors were omitted. This is the first study to show that endoproteolytic enzymes within the human placenta have the capacity to produce other cleavage products (pro-CRH125–151 and proCRH125–194) besides CRH1– 41 from the CRH precursor. So long as protease activity is not blocked, cleavage at each of the two potential cleavage sites, CS1 and CS2, can occur. Both of these are furin-like cleavage sites and therefore potentially subject to cleavage by furin-like endopeptidases, prohormone convertases (such as PC1 and PC2) (34), or other precursor-converting enzymes belonging to the family of subtilisin/Kex2 serine proteases (35). As yet, the specific endoproteases responsible for cleaving placental pro-CRH at each cleavage site have not been identified; for this reason, a cocktail of enzyme inhibitors protecting against the major classes of proteases was used for extraction. It is not known whether the presence of the intermediate metabolite, proCRH125–194, in all placental chromatograms reflects pro-CRH processing in vivo, processing in vitro during tissue collection, or the failure to inactivate the placental pro-CRH CS1-specific processing enzyme by the protease inhibitors employed. Even in the absence of protease inhibitors, conversion of pro-CRH to peptide fragments was not complete, with appreciable pro-CRH immunoreactivity remaining in normal term placental extracts. There was also no evidence of the CRH1– 41 metabolite, CRH36 – 41, in these extracts, as found in
762
AHMED ET AL.
another study (10), even though antibody M2 is known to cross-react with this CRH species (14). The minimal processing found in normal term placentas is in contradiction to other reports that CRH1– 41 is the predominant form of CRH in the placenta (3, 8 –13). Of utmost importance for this type of study is the collection of tissue and preparation of extracts in a manner that avoids extraction artifacts. Although our study clearly shows that the placenta contains enzymes capable of processing pro-CRH, under normal conditions these enzymes will be compartmentalized within the cell. This compartmentalization will become disrupted during tissue extraction, and unless protease inhibitors are present in extraction media, the enzymes released may come into contact with proteins that would normally not be accessible, giving rise to cleavage products that are not representative of the situation in vivo. All of the earlier studies used extraction media without protease inhibitors, which may have resulted in pro-CRH processing in vitro, although it is noted that in some of these reports, extraction was carried out at acidic pH or with boiling, both of which would be expected to minimize protein degradation. However, many groups have used placentas delivered vaginally without considering the possibility of tissue deterioration during lengthy labor and the disrupting effect that this may have on endoproteolytic enzyme compartmentalization in trophoblast. Precursor hormones may be targeted to either the constitutive or the regulatory secretory pathway (36). Pro-CRH synthesized in the cells of the hypothalamic paraventricular nucleus is directed into the regulatory secretory pathway from which secretion is triggered episodically by specific signals; in the hypothalamus, the precursor is processed intracellularly to produce biologically active CRH1– 41, which is stored in large dense core granules. In contrast, precursor hormones entering the constitutive secretory pathway are usually secreted in the unprocessed form in a continuous, rather than episodic, fashion (for reviews, see Refs. 1 and 2). Our finding that pro-CRH is the major form of CRH in normal human placental tissue suggests that pro-CRH and endoproteolytic enzymes remain in separate compartments within trophoblast in vivo such that the precursor is the major species secreted. Consequently, most of the CRH immunoreactivity in syncytiotrophoblast cells must be directed toward the constitutive secretory pathway. This implies that modulation of placental CRH release occurs predominantly at the level of gene transcription, rather than the regulated release that is seen with hypothalamic CRH. However, our results do not rule out the possibility that a small proportion of placental CRH, which may be below the detection limit of the chromatographic and immunoassay system used here, may also be directed toward the regulatory pathway during normal pregnancy. In contrast to this report, a previous chromatographic study of preeclampsia placentas by Goland and colleagues (10) found CRH1– 41 to be the exclusive form present, whereas placental homogenates from uncomplicated pregnancies contained lower quantities of CRH1– 41 and a CRH1– 41 metabolite, possibly CRH36 – 41. Although acidic extracts were used in the earlier study, the placentas were minced before exposure to acid, providing ample opportunity for in vitro
JCE & M • 2000 Vol 85 • No 2
pro-CRH processing by enzymes released from disrupted cells. Also, many of the placentas used had been delivered vaginally, again providing opportunity for protease activation in vitro. The present study does, however, confirm Goland’s report of greater CRH immunoreactivity in preeclampsia placentas compared to those from normal pregnancy, and this is reflected in the higher levels of placental pro-CRH mRNA found in preeclampsia. This is a paradoxical finding in view of the smaller size of the preeclampsia placenta and suggests a fundamental disturbance in the control of the synthesis and subsequent release of the hormone. Although placental CRH immunoreactivity measured here fell within the same range as that reported in Goland’s study, it was twice as high. This may be due to the cross-reactivity of our CRH antibody with additional CRH species (pro-CRH and its ⬃8-kDa intermediate metabolite), which were not detected by the CRH antibody used in the earlier report. We found evidence of increased processing of pro-CRH in preeclampsia placentas compared to normal term placentas, with the appearance of a small peak of CRH1– 41 in chromatographed extracts. This may reflect tissue damage in vivo associated with preeclampsia placental pathology despite sampling away from areas of ischemia and infarction. As every effort was made to handle tissue from normal and abnormal pregnancies equivalently with respect to speed of handling and the extraction and chromatographic conditions employed, the increased processing observed may, however, reflect pro-CRH processing in vivo in preeclampsia placentas. Further experiments are required to ascertain whether compartmentalization of endoproteolytic enzymes is altered in preeclampsia to enable CRH1– 41 to be cleaved from pro-CRH within trophoblast. Normal and preeclampsia placental extracts were measured by immunoassays using two different antibodies, one against mature CRH1– 41 peptide and the other against proCRH125–151. As demonstrated chromatographically, both antibodies recognize the peptide against which they were raised as well as other pro-CRH species containing their respective sequences, i.e. intact pro-CRH and the intermediate metabolite pro-CRH125–194. The molar levels of proCRH125–151-reactive and CRH1– 41-reactive species in protease-protected extracts were not significantly different from each other, consistent with our chromatographic finding that the major form of CRH-like immunoreactivity in the normal and preeclampsia placenta is unprocessed pro-CRH. Having shown that the human placenta contains substantial amounts of pro-CRH and has the enzymatic ability to produce the cleavage products pro-CRH125–194, proCRH125–151, and CRH1– 41, we explored which of these molecular forms circulates in maternal blood. Our plasma studies showed that the level of SJ2-immunoreactive peptides (i.e. pro-CRH, CRH125–194, and pro-CRH125–151) were negligible throughout pregnancy, although there was a gradual increase in levels of M2-immunoreactive peptide(s) in the maternal circulation throughout the third trimester of pregnancy. This CRH-like immunoreactivity must therefore be due predominantly to CRH1– 41, rather than to pro-CRH and/or pro-CRH125–194. This confirms our earlier chromatographic evidence (37) that the major form of CRH immunoreactivity in late pregnancy plasma is CRH1– 41, although
PRO-CRH PROCESSING IN PREGNANCY
under physiological conditions the CRH1– 41 is bound to CRH-BP and circulates largely as a CRH-BP:CRH1– 41 complex. Pro-CRH and the processed products, pro-CRH125–194 and pro-CRH125–151, do not circulate in substantial and increasing quantities in the latter half of pregnancy as does CRH1– 41. This is the case even in preeclampsia, when placental and plasma CRH1– 41 is higher than in normal pregnancy. In the present study, Western blotting of unextracted late gestational maternal plasma confirmed that intact proCRH was undetectable. Whether the same is true for the potential N-terminal fragment of pro-CRH, pro-CRH28 –122, is not yet known due to the lack of an appropriate antibody. Chromatographic studies showed that pro-CRH125–151 did not bind to native human CRH-BP even though the same preparations of this protein bound CRH1– 41 with high affinity as expected. Furthermore, no other binding protein specific for pro-CRH125–151 was found in human plasma. The enhancement of binding of SJ2 antibody to pro-CRH125–151 tracer during plasma immunoassay must therefore result from a weak interference by other plasma components, possibly albumin. Binding was, however, observed between CRH-BP and pro-CRH, although the absolute amount of CRH-like immunoreactivity bound was half of that bound when pro-CRH125–194 was used, suggesting that the CRH-BP has a lower affinity for the CRH precursor than for CRH1– 41. However, this lower affinity may also be due to the use of recombinant rat pro-CRH in the absence of the native human protein. How CRH1– 41 is maintained in the circulation at such high levels when pro-CRH125–151 is cleared rapidly is not yet understood. The lack of detectable pro-CRH and its processed products apart from CRH1– 41 in the maternal circulation together with our finding that the CRH precursor is the major form of CRH in the placenta suggest that precursor processing occurs immediately at the syncytiotrophoblast surface. However, it is not known whether the enzymes involved are those released from the placenta or others endogenous to the circulation. That endoproteolytic enzymes can be released from the cell surface in an active form has been demonstrated in bovine intermediate lobe pituitary cells (38). Further studies are now underway to determine the exact site of placental pro-CRH processing, whether this occurs immediately after release into the bloodstream or, perhaps less likely, as the precursor is in transit through the outer syncytiotrophoblast surface. The lack of circulating pro-CRH, pro-CRH125–194, and proCRH125–151 also suggests that the CRH precursor and the cleavage products studied here do not have endocrine actions in pregnancy, as has been proposed for CRH1– 41. This does not preclude the possibility that at least some of these forms may exert paracrine and/or intracrine actions within the placenta. As placental pro-CRH secreted into the maternal circulation must be rapidly processed here, it is unlikely to have any opportunity to exert paracrine effects within the placenta. However, the fetal circulation also contains CRH1– 41 thought to originate from the placenta (15), presumably by diffusion from the basal syncytiotrophoblast surface across the villous core into the fetal vasculature. If the intact precursor is released from the basal syncytial surface, it could exert paracrine actions on neighboring cells, i.e. cytotrophoblast, stromal, fibroblast, endothelial, and Hofbauer
763
cells, within the villous core. It is also possible that pro-CRH exerts intracrine effects within the syncytiotrophoblast where it is produced. Relevant here is the nuclear localization of pro-CRH previously observed in human T lymphocytes (39) and transfected CHO-K1 cells (40). This phenomenon has also been observed with growth factors such as fibroblast growth factor, and although the function of this nuclear presence is not known, it may be that such molecules are capable of genomic actions (41). Recent studies have shown that intact pro-CRH has several biological activities, including the inhibition of interleukin-6 release by human mononuclear cells (42), the stimulation of ACTH release from primary cultures of rat anterior pituitary cells (40), and the stimulation of DNA synthesis and cell proliferation in CHO cells (27). By analogy with these reports, it is possible that the placental CRH precursor may also possess mitogenic and/or hormone-stimulating activity, thereby affecting placental development and maturation. References 1. Perone MJ, Castro MG. 1997 Prohormone and proneuropeptide synthesis and secretion. Histol Histopathol. 12:1179 –1188. 2. Perone MJ, Windeatt S, Castro MG. 1997 Intracellular trafficking of prohormones and proneuropeptides: cell type-specific sorting and targeting. Exp Physiol. 82:609 – 628. 3. Shibasaki T, Odagiri E, Shizume K, Ling N. 1982 Corticotrophin-releasing factor-like activity in human placental extracts. J Clin Endocrinol Metab. 55:384 –386. 4. Shibahara S, Morimoto Y, Furutani Y, et al. 1983 Isolation and sequence of the human corticotrophin releasing factor precursor gene. EMBO J. 2:775–779. 5. Robinson BG, D’Angio LA, Pasieka KB, Majzoub JA. 1989 Preprocorticotrophin releasing hormone: cDNA sequence and in vitro processing. Mol Cell Endocrinol. 61:175–180. 6. Castro M, Rowe J, Murray C, Tomasec P, Shering A, Linton E, Ahmed I, Lowenstein P. 1995 Generation and characterisation of an antiserum reactive with a proteolytic processing site within rat procorticotrophin releasing hormone. Neuropeptides. 29:183–192. 7. Perone MJ, Murray CA, Brown OA, et al. 1998 Procorticotrophin-releasing hormone: endoproteolytic processing and differential release of its derived peptides within AtT20 cells. Mol Cell Endocrinol. 142:191–202. 8. Thomson M, Chan EC, Falconer J, Madsen G, Smith R. 1988 Secretion of corticotrophin releasing hormone by superfused human placental fragments. Gynecol Endocrinol. 2:87–93. 9. Cooper ES, Brooks AN, Miller MR, Greer IA. 1994 Corticotrophin releasing factor immunostaining is present in placenta and fetal membranes from the first trimester onwards and is not affected by labour or administration of mifepristone. Clin Endocrinol (Oxf). 41:677– 683. 10. Goland RS, Conwell IM, Jozak S. 1995 The effect of pre-eclampsia on human placental corticotrophin releasing hormone content and processing. Placenta.16:375–382. 11. Sasaki A, Tempst P, Liotta A, et al. 1988 Isolation and characterisation of a corticotrophin releasing hormone-like peptide from human placenta. J Clin Endocrinol Metab. 67:768 –773. 12. Chan EC, Thomson M, Madsen G, Falconer J, Smith R. 1988 Differential processing of corticotrophin releasing hormone by the human placenta and hypothalamus. Biochem Biophys Res Commun. 153:1229 –1235. 13. Chan EC, Thomson M, Madsen G, Boettcher B, Falconer J, Smith R. 1990 Large molecular weight immunoreactive corticotrophin releasing hormone has bioactivity on superfused ovine pituitary cells. J Neuroendocrinol. 2:95–101. 14. Campbell EA, Linton EA, Wolfe CDA, Scraggs PR, Jones MT, Lowry PJ. 1987 Plasma corticotrophin releasing hormone concentrations during pregnancy and parturition. J Clin Endocrinol Metab. 64:1054 –1059. 15. Goland RS, Wardlaw SL, Blum M, Tropper PJ, Stark RI. 1988 Biologically active corticotrophin releasing hormone in maternal and fetal plasma during pregnancy. Am J Obstet Gynecol. 159:884 – 890. 16. Riley SC, Challis JR. 1991 Corticotrophin-releasing hormone production by the placenta and fetal membranes. Placenta. 12:105–119. 17. Perkins AV, Linton EA. 1995 Identification and isolation of corticotrophinreleasing hormone-positive cells from the human placenta. Placenta.16:233– 243. 18. Behan DP, Linton EA, Lowry PJ. 1989 Isolation of the human plasma corticotrophin releasing factor binding protein. J Endocrinol. 122:23–31. 19. Vaughan J, Donaldson C, Bittencourt J, et al. 1995 Urocortin, a mammalian
764
20. 21. 22. 23. 24.
25. 26. 27. 28. 29. 30.
31.
AHMED ET AL.
neuropeptide related to fish urotensin I and to corticotrophin-releasing factor. Nature. 378:287–292. Behan DP, De Souza EB, Potter E, Sawchenko P, Lowry PJ, Vale WW. 1996 Modulatory actions of corticotropin-releasing factor-binding protein. Ann NY Acad Sci. 780:81–95. Linton EA, Perkins AV, Wood RJ, et al. 1993 Corticotrophin releasing hormone binding protein CRHBP: plasma levels decrease during the third trimester of normal pregnancy. J Clin Endocrinol Metab. 76:260 –262. Perkins AV, Eben F, Wolfe CDA, Shulte HM, Linton EA. 1993 Plasma measurements of corticotrophin-releasing hormone-binding protein in normal and abnormal human pregnancy. J Endocrinol. 138:149 –157. McLean M, Bisits A, Davies J, Woods R, Lowry P, Smith R. 1995 A placental clock controlling the length of human pregnancy. Nat Med.1:460 – 463. Berkowitz GS, Lapinski RH, Lockwood CJ, Florio P, Blackmore-Prince C, Petraglia F. 1996 Corticotropin-releasing factor and its binding protein: maternal serum levels in term and preterm deliveries. Am J Obstet Gynecol, 174:1477–1483. Wolfe CDA, Patel S, Linton EA, et al. 1988 Plasma corticotrophin releasing factor in abnormal pregnancy. Br J Obstet Gynaecol. 95:1003–1006. Laatikinen T, Virtanen T, Kaaja R, Salminen-Lappalainen K. 1991 Corticotropin releasing hormone in maternal and cord plasma in pre-eclampsia. Eur J Obstet Gynaecol Reprod Biol. 39:19 –24. Castro MG, Tomasec P, Morrison E, et al. 1995 Mitogenic effects and nuclear localisation of procorticotrophin-releasing hormone expressed within stably transfected fibroblast cells CHO-K1. Mol Cell Endocrinol. 107:17–27. Linton EA, Perkins AV, Hagan P, et al. 1995 Corticotrophin-releasing hormone CRH-binding protein interference with CRH antibody binding: implications for direct CRH immunoassay. J Endocrinol.146:45–53. Laemmli DK. 1971 Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 227:680 – 685. Salacinski P, Mclean C, Skyes J, Clement-Jones V, Lowry PJ. 1981 Iodination of proteins, glycoproteins and peptides using a solid-phase oxidising agent 1,3,4,6-tetrachloro-3-a-6-a-diphenyl glucoril Iodogen. Anal Biochem. 117:136 –140. Celi FS, Zenilman ME, Shuldiner AR. 1993 A rapid and versatile method
32.
33.
34.
35. 36. 37.
38.
39.
40.
41. 42.
JCE & M • 2000 Vol 85 • No 2
to synthesise internal standards for competitive PCR. Nucleic Acids Res. 21:1047. Rodriguez-Linares B, Phaneuf S, Lopez-Bernal A, Linton EA. 1998 Levels of corticotrophin-releasing hormone receptor subtype 1 mRNA in pregnancy and during labour in human myometrium measured by quantitative competitive PCR. J Mol Endocrinol. 21:201–208. Frim D, Emanuel RL, Robinson BG, Smas CM, Adler GK, Majzoub JA. 1988 Characterisation and gestational regulation of corticotrophin-releasing hormone messenger RNA in human placenta. J Clin Invest. 82:287–292. Breslin MB, Lindberg I, Benjannet S, Mathis JP, Lazure C, Seidah NG. 1993 Differential processing of proenkephalin by prohormone convertases 13 and 2 and furin. J Biol Chem. 268:27084 –27093. Smeekens SP. 1993 Processing of protein precursors by a novel family of subtilisin-related mammalian endoproteases. BioTechnology. 11:182–186. Kelly RB. 1985 Pathways of secretion in eukaryotes. Science. 230:25–32. Linton EA, Wolfe CDA, Behan DP, Lowry PJ. 1988 A specific carrier substance for human corticotrophin releasing factor in late gestational maternal plasma which could mask the ACTH-releasing activity. Clin Endocrinol (Oxf). 28:315–324. Castro MG, Birch NP, Loh YP. 1989 Regulated secretion of POMC converting enzyme and an aminopeptidase B-like enzyme from dispersed bovine intermediate lobe pituitary cells. J Neurochem. 52:1619 –1628. Ekman R, Servenius B, Castro MG, Lowry PJ, Cederlund AS, Bergman O, Sjogren HO. 1993 Biosynthesis of corticotropin-releasing hormone in human T-lymphocytes. J Neuroimmunol. 44:7–13. Morrison E, Tomasec P, Linton EA, Lowry PJ, Lowenstein PR, Castro MG. 1995 Expression of biologically active procorticotrophin-releasing hormone (proCRH) in stably transfected CHO-K1 cells: characterization of nuclear proCRH. J Neuroendocrinol. 7:263–72. Castro MG, and Morrison E. 1995 Nuclear translocation of neuropeptides: possible nuclear roles. J Endocrinol. 146:9 –13. Salas MA, Brown OA, Perone MJ, Castro MG, Goya RG. 1997 Effect of the corticotrophin releasing hormone precursor on interleukin-6 release by human mononuclear cells. Clin Immunol Immunopathol. 85:35–39.
