Relationship between early pregnancy factor, mouse embryo-conditioned medium and platelet-activating factor.

The effects of synthetic platelet-activating factor (PAF-acether) and mouse embryo-conditioned medium (a source of embryo-derived PAF (EPAF)) on production of early pregnancy factor (EPF) were compared. Embryo-conditioned medium, itself inactive in the EPF bioassay, stimulated ovarian production of EPF in vitro but PAF-acether did not. In vivo, embryo-conditioned medium induced EPF activity in serum of oestrous female, but not in male, mice in contrast to PAF-acether, which induced activity in serum of both male and female mice. This PAF-induced activity was transitory, declining significantly by 2 h and disappearing by 3 h after injection. Activity induced by embryo-conditioned medium was first evident at 2 h after injection, serum concentrations increasing up to 6 h after injection. By discriminating between the behaviour of PAF-acether and EPAF, these studies reinforce the conclusions of other workers that the molecule produced by the embryo is not PAF. Further investigations into the mechanism of action of PAF-acether revealed that it is a potent inducer of activity in the EPF bioassay, with an absolute requirement for platelets in the spleen cell suspension used in the assay. This platelet-derived active species was bound specifically by an anti-EPF monoclonal antibody, indicating that it is EPF-like. This is consistent with parallel studies showing that platelets are not required for induction of activity by either pregnancy serum or purified EPF. These studies were applied to the PAF-induced leukotriene-like species, which had been found by others to be active in the EPF bioassay. Pregnancy serum induced the appearance of this substance from the spleen cell suspension used in the assay; thus the leukotriene-like substance may be regarded as an effector molecule in vitro or mediator of the initiating stimulus of EPF in the bioassay.     

In-vivo and in-vitro determination of components of rabbit early pregnancy factors

Two types of rabbit early pregnancy factor (EPF) components, prepared from the in-vitro perfused ovary and oviduct and from serum ammonium sulphate fractions, were investigated to elucidate the source of EPF production. The state of the animals, i.e. pregnant, pseudopregnant, unstimulated or platelet activating factor (PAF)-treated, and order of addition of the components to the assay lymphocytes were varied to characterize conditions of production and expression of EPF activity. Although each component alone had no EPF activity, combination of two components, i.e. ovary and oviduct components, or serum precipitate and supernatant components, expressed EPF activity. The oviduct and serum supernatant components were produced in pregnancy and pseudopregnancy but the ovary and serum precipitate components were produced only in pregnancy. The similarity of the production pattern and ability of the perfusate and serum components to yield EPF activity when combined suggests that they are similar or the same. The sources of stimulation of EPF production did not appear to affect component production because the activity produced by the perfused ovary and oviduct in pregnancy or in response to PAF stimulation appeared similar. Oviduct and supernatant components apparently bound directly to lymphocytes. These results suggest that EPF components are produced in the ovary and oviduct individually and that the combination of the two components expresses EPF activity in the rabbit.     

Identification of a putative inhibitor of early pregnancy factor in mice

Previous studies have indicated that early pregnancy factor (EPF) produced in the pre- and peri-implantation stage of pregnancy appears to consist of inactive components which combine to produce the active species. This is in contrast with EPF produced later in gestation which appears to consist of a single active species. The original studies on ammonium sulphate fractionation of mouse serum and in-vitro culture of mouse ovaries and oviducts have been repeated but tested in the bioassay for EPF, the rosette inhibition test, over an extended range of dilutions. This revealed that the two components in early pregnancy can be understood as EPF and an inhibitor(s). Once this inhibitor is removed, the active fractions in both early and late pregnancy sera exhibit similar behaviour in the above assay. It was shown also that the ovary alone is the source of activity but that this is modulated by an inhibitory substance(s) from the oviduct. Reversed-phase HPLC studies on purified 'early' EPF confirm that active and inhibitory components are present and demonstrate that the active component exhibits an identical elution pattern to 'late' EPF. Thus as pregnancy proceeds, it is not EPF that alters but rather the inhibitor(s), which disappears from the circulation soon after implantation. This substance(s) is under hormonal control, being present during oestrus as well as the early stages of pregnancy; it may be an important biological regulator of EPF. Its action in the rosette inhibition test has profound implications for further study using this bioassay.     

Identification of molecules involved in the 'early pregnancy factor' phenomenon.

An isolated preparation from ovine placental extracts which was active in the rosette inhibition assay mimicking the activity of the so-called 'early pregnancy factor' (EPF) has been shown to contain a 12 kDa polypeptide which could be partially resolved from low-molecular-weight active moieties. N-terminal amino acid sequence analysis of the polypeptide indicated that it was ovine thioredoxin, an identification confirmed by isolation and complete sequence analysis of the corresponding cDNA. The cDNA for human thioredoxin was expressed in Escherichia coli and the recombinant protein isolated and purified. Pure recombinant thioredoxin alone did not induce the expression of increased rosette inhibition titres (RITs) when tested in the rosette inhibition assay; but, when tested in combination with cell stimuli such as platelet-activating factor (PAF) or serum, it allowed the expression of increased RITs where none was achieved in its absence. Thioredoxin acted in the assay to reverse a refractory state normally induced by these stimuli, allowing lipoxygenase-dependent moieties also induced by the stimuli to exert their effects, resulting in the expression of increased RITs. Antibodies to recombinant thioredoxin removed from pregnancy sera the capacity to induce increased RITs, i.e. to express EPF activity, thus establishing a role for thioredoxin or thioredoxin-like proteins and associated molecules in the mechanisms which allow pregnancy sera to induce increased RITs. Based on a consideration of these and other results, a new model for the study of the EPF phenomenon is presented and discussed.     

Failure to detect platelet-activating factor using the splenectomized mouse bioassay

The ability of purified preparations of platelet-activating factor (PAF), from three different suppliers, to induce thrombocytopaenia in mice after splenectomy and to activate mouse platelets in vitro was examined. Although the PAF preparations were potent activators of horse and cow platelets in vitro, injections of up to 1 microgram PAF failed to elicit thrombocytopaenia responses in either CD1 or Swiss Webster random-bred mice. However, when thrombin was injected into Swiss Webster mice, a dose-dependent decrease in the concentration of platelets was observed. Furthermore, isolated platelets from these strains and from 3 inbred lines (C3H/He, BALB/c, C57BL/6) of mice, were not aggregated by PAF in vitro but were sensitive to adenosine diphosphate and thrombin. No change in circulating platelet concentrations was observed over the initial 7 days of gestation in intact Swiss Webster and C57BL/6 or splenectomized C57BL/6 mice, suggesting either an absence of PAF production during early pregnancy in these strains or insensitivity of their platelets to PAF. These results suggest that many mouse strains are unsuitable for the bioassay of PAF.     

Effect of monoclonal antibodies to early pregnancy factor (EPF) on the in vivo growth of transplantable murine tumours

Summary Neutralisation studies with monoclonal antibodies (mAbs) specific for early pregnancy factor (EPF) have shown it to be essential for the continuation of pregnancy in mice and the growth of some tumour cells in vitro. These studies report that the mAbs are also able to limit the growth of two murine tumour lines transplanted s. c. The development of MCA-2 tumours in CBA mice was unaffected by the injection of 1 mg anti-EPF IgM at the time of tumour cell inoculation. However, four doses of 500 µg anti-EPF, injected one dose per day for 4 days after tumour cell inoculation, significantly retarded tumour development such that no tumours were palpable on day 13. A similar dose regimen of control IgM had no effect on tumour size. Dose/response studies revealed that lower doses of anti-EPF administered after tumour cell inoculation were effective in retarding the growth of the MCA-2 tumours. The effect of anti-EPF mAb administration on the growth rate of palpable B16 tumours established s. c. in C57BL/6 mice was also determined. Tumours injected with 6 mg anti-EPF 5/341 or anti-EPF 5/333 mAbs showed significant decrease in the uptake of [³H]thymidine into tumour tissue, measured 16 h after injection. Furthermore, titration of sera for active EPF showed that a significant reduction in the EPF titre was associated with a significant inhibition of tumour DNA synthesis. Thus it appears that neutralisation of EPF retards tumour growth both in vitro and in vivo. In vitro the effects must be due to anti-EPF mAb interfering with a direct mechanism that contributes to the maintenance of cells in the active growing phase. However, in vivo host immunological mechanism that are modified to allow tumour survival may also be affected. The presence of EPF-induced suppressor factors curculating in the serum of tumour-bearing mice has been confirmed and the contribution of such factors to tumour progression must now be investigated.     

Platelet-activating factor in the rabbit uterus during early pregnancy

Platelet-activating factor (PAF) concentrations were low in the non-pregnant, oestrous uterus (mean +/- s.e.m.: 2.2 +/- 1.2 pmol/g, n = 3). However, uterine PAF increased dramatically during pregnancy to a maximum of 37.8 +/- 4.90 pmol/g (n = 7) on Day 5. By Day 7, PAF concentrations in the uteri of pregnant rabbits had returned to levels similar to those found at oestrus. In contrast, uterine PAF in pseudopregnant rabbits peaked at 30.6 +/- 2.8 pmol/g (n = 8) on Day 4, declined to 20.5 +/- 2.4 pmol/g (n = 8) on Day 5 and then remained at that concentration through Day 7. Uterine PAF co-migrated with synthetic PAF (1-O-hexadecyl-2-acetyl-sn-glycero-phosphocholine) in both thin-layer and normal-phase high-performance liquid chromatography. PAF activity in the uterus during pregnancy and pseudopregnancy was found almost exclusively in the endometrium; little or no PAF was found in myometrium, uterine flushings or blastocysts. While no PAF was detected in blastocysts on Days 5 and 6 of pregnancy, the presence of the embryo appears to modulate biosynthesis and/or degradation of PAF by the uterus, since PAF decreased significantly in uterine tissue apposed to the implanting embryo (but not in similar areas between such attachment sites). Increased concentrations of PAF in the preimplantation rabbit uterus followed by a dramatic decrease on the day of blastocyst attachment suggest that this potent inflammatory autacoid may play a vital role in implantation.     

Purification and characterization of early pregnancy factor from human pregnancy sera

Early pregnancy factor (EPF) is a pregnancy-associated protein detected in the maternal serum by using the rosette inhibition assay and by evaluating the suppression of adoptive transfer of contact sensitivity. Because of its inhibitory effect on the functional reactivity of immunocompetent cells, EPF is thought to be involved in immunoregulation of the maternal immune system during early pregnancy. EPF was purified six million-fold from the serum of pregnant women between 5 and 12 weeks of gestation. The specific activity of purified EPF was approximately 8 x 10(8) units/mg. The purification scheme involved sequential DEAE-cellulose chromatography, S-Sepharose chromatography, concanavalin A-Sepharose chromatography, heparin-Sepharose chromatography, Mono S fast protein liquid chromatography, and preparative sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The purified protein has an apparent molecular weight of 21,500 as analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and 28,000 by gel permeation high pressure liquid chromatography. The isoelectric point of purified EPF moiety is 6.5. The biological activity was susceptible to the proteolytic enzyme trypsin, acidic pH conditions, organic solvents, and sodium dodecyl sulfate, but stable to heat treatment at 56 degrees C for 30 min and the reducing agent dithiothreitol. The biological and physicochemical properties of EPF appear to be distinct from other pregnancy-associated and immunoregulatory proteins.     

Measurement of early pregnancy factor activity for monitoring the viability of the equine embryo

The viability of embryos before flushing from donor mares (n = 5) and after transfer to recipient mares (n = 7) was monitored in mare serum by detecting early pregnancy factor (EPF) using the rosette inhibition test (RIT). The EPF activity was measured in donor mares before and after natural mating at natural estrus; after ovulation on Days 2, 5 and 8; and after embryo flushing (Day 8) on Days 8, 9, 10 and 13 after ovulation. The collected embryos were transferred immediately after flushing. The EPF activity in recipient mares were measured on the day of transfer and after embryo transfer on Days 1, 2, 3 and 5. Pregnancy was confirmed on Day 12 to 14 after embryo transfer. The mean EPF activity of donor mares was increased to the pregnant level (> an RI titer score of 10) on Day 2 after ovulation. Two days after flushing the embryos, the EPF activity of donor mares had decreased to the nonpregnant level. Among the 7 recipient mares, 3 mares were diagnosed pregnant on Day 12 after embryo transfer with ultrasound. The EPF activity of the pregnant recipient mares was increased above the minimum level observed in pregnant mares on Days 2 to 3 after transfer. However, among the nonpregnant recipient mares after embryo transfer, the EPF activity of 3 mares remained at the pregnant level only 2 to 3 d and then declined to the nonpregnant level. In one recipient mare, EPF activity did not reach the pregnant level throughout the sample collection. The results of this study indicated that equine EPF can be detected in serum of pregnant mares as early as Day 2 after ovulation. From our observation, we conclude that the measurement of EPF activity is useful for monitoring the in vivo viability of equine embryos and early detection of embryonic death.     

Investigation of the immunocompetent cells that bind early pregnancy factor and preliminary studies of the early pregnancy factor target molecule

Early pregnancy factor (EPF) is a secreted protein with immunosuppressive and growth factor properties. It has been shown to suppress the delayed-type hypersensitivity response in mice as well as acute and chronic forms of experimental autoimmune encephalomyelitis in rats and mice, respectively. In previous studies, we have demonstrated that EPF binds to a population of lymphocytes and we hypothesized that it mediates its suppressive effects by binding to CD4+ T cells. In the present study, we isolated monocytes and subpopulations of lymphocytes and labelled them with fluoresceinated EPF in order to determine which populations bind EPF. We demonstrated that EPF binds specifically to CD4+, CD8+, CD14+ (monocytes) and CD56+ NK cells but not to CD19+ B cells. The identity of the molecule(s) on the cell surface that is targeted by EPF is unknown, but as EPF is an extracellular homologue of the intracellular protein chaperonin 10 (Cpn10), we examined the possibility that the EPF receptor is a membrane-associated form of chaperonin 60 (Cpn60), the functional associate of Cpn10 within the cell. The EPF target molecule on lymphocytes was visualized by chemical cross-linking of exogenous iodinated Cpn10 to cells and probed with anti-Cpn60. The effect of anti-Cpn60 on activity in the EPF bioassay, the rosette inhibition test, was also examined. In both instances, no specific interaction of this antibody and the putative receptor was observed. It was concluded that the cell surface molecule targeted by EPF is unlikely to be a homologue of Cpn60.     

Prefertilization treatment of mice with platelet activating factor affects pregnancy.

Embryo-derived platelet activating factor (EPAF) is thought to be either biologically similar to platelet activating factor (PAF) or responsible for PAF liberation in vivo. We have previously shown that premating PAF treatment in the mouse renders the platelets nonresponsive to EPAF, leading to a reduced implantation rate in these animals. In this study, we have shown that females, injected with PAF before mating, show altered embryo development invivo on day 4 postfertilization. This is manifested as an interruption of compaction, a reduced cell number per embryo, and reduced embryo number per mouse. Results suggest that EPAF represents an early pregnancy signal that supports embryo development. The most likely mechanism is via platelet activation, since only those mice that showed thrombocytopenia after fertilization were found to have normal embryos on day 4 postmating.     

Role of platelet-activating factor (PAF) in the ovoimplantation in the rat: effect of the specific PAF-acether antagonist, BN 52021

The role of PAF (platelet-activating factor) in early pregnancy has been recently postulated. Indeed, platelet count is markedly reduced immediately prior to and returns to normal following ovoimplantation. Using a pharmacological approach, we therefore further investigated the possible involvement of PAF in ovoimplantation. BN 52021 (10 nmol), a PAF antagonist, was administered after fecondation in the lumen of the left uterine horn. A group of animals was injected with the vehicle alone. The animals were sacrificed 1 week latter and the number of implanted embryos in the treated horn and in the untreated contralateral one was assessed. When injected on day 4 of pregnancy, the PAF antagonist, BN 52021, inhibited in a dose-dependent fashion ovoimplantation suggesting a role for this lipid mediator in this process. Total inhibition of ovoimplantation was observed in the horns treated with indomethacin, NDGA or BW 755 C. In addition, a significant inhibition of ovoimplantation was also produced by the lipoxygenase inhibitor, EP 10045 (10nmol). Since both PAF and leukotrienes trigger the generation of prostaglandins in various tissues, our results indicate that these two lipid mediators may be implicated in early stages of the inflammatory reaction accompanying ovoimplantation and may contribute to the local generation of cyclooxygenase metabolites.     

Production of a recombinant form of early pregnancy factor that can prolong allogeneic skin graft survival time in rats

Early pregnancy factor (EPF), an extracellular chaperonin 10 homologue, has immunosuppressive and growth factor properties. In order to carry out more extensive studies on the in vivo characteristics of EPF, a recombinant form of the molecule has been prepared. Recombinant human EPF (rEPF) was expressed in Escherichia coli using the plasmid pGEX-2T expression system. Potency of rEPF in vitro in the rosette inhibition test, the bioassay for EPF, was equivalent to that of native EPF (nEPF), purified from human platelets, and synthetic EPF (sEPF). However, the half-life of activity (50% decrease in the log value) in serum, following i.p. injection, was significantly decreased (3.2 h, compared with nEPF 6.2 days, sEPF 5.8 days). This was thought to be due to modification of the N-terminus of the recombinant molecule inhibiting binding to serum carrier proteins. Because EPF can modify Th1 responses, the ability of the recombinant molecule to suppress allogeneic graft rejection was investigated. Following skin grafts from Lewis rats to DA rats and vice versa, rEPF was delivered locally at the graft site and the effect on survival time of the allografts noted. Results demonstrated that rEPF treatment significantly prolonged skin graft survival time by as much as 55% in stringent models of transplantation across major histocompatibility barriers.     

Failings of the rosette-inhibition test for early pregnancy factor

Attempts were made to reproduce work applying the rosette-inhibition test for early pregnancy factor (EPF) to early detection of pregnancy in sheep and swine. Twenty-one antisera, raised in rabbits, goats and sheep against sheep or pig lymphocytes isolated from spleen, thymus or peripheral blood, were tested using the original method of Morton et al. (1979) but none provided positive results. Procedures were then modified to incorporate changes claimed to eliminate possible interferences with the test, but all results remained negative, even when using an antiserum found elsewhere to allow detection of EPF. Success in applying the test, reported by some laboratories, is countered by reports of failure in other laboratories, suggesting that a critical elusive factor in the test has yet to be identified.     

Estrogen effects on platelet-activating factor and platelet-activating factor acetylhydrolase activity in rat uterus during the late stages of pregnancy.

The platelet-activating factor (PAF) concentration of the uterus spontaneously increased during pregnancy. When 17alpha-ethynylestradiol (0.25 mg/kg) was administered subcutaneously to pregnant rats for 3 days starting on Day 17 of pregnancy, some rats delivered prematurely on Day 20. However, none of the vehicle-treated (80% dimethylsulfoxide and 20% ethanol) pregnant rats delivered prematurely. The PAF concentration of the uterus in pregnant rats treated with 17alpha-ethynylestradiol was significantly higher than in those treated with vehicle on Days 19 and 20. On the other hand, the specific activity of uterine PAF-acetylhydrolase (PAF-AH) in pregnant rats treated with 17alpha-ethynylestradiol was significantly lower than in those treated with vehicle on Days 19 and 20, and the plasma PAF-AH activity in pregnant rats treated with estrogen was also significantly lower than in treated with vehicle on Days 18, 19, and 20. These findings indicate that estrogen increases PAF concentrations in the rat uterus, and this was correlated with a decrease in PAF-AH in the uterus and plasma. The increase in PAF concentrations in the uterus may be related to premature delivery and labor caused by PAF's known effect on myometrial contraction.     

Platelet-activating factor and serum components from oestrous mice co-operate to mimic the activity of 'early pregnancy factor' in the rosette inhibition assay

When male mouse spleen cells were incubated with a combination of platelet activating factor (PAF, 1-0-alkyl-2-acetyl-sn-glycero-3-phosphocholine) and sera from female mice in oestrus, the cells displayed a markedly increased rosette inhibition titre (RIT) when subsequently tested in the rosette inhibition assay. Neither PAF nor oestrous mouse sera alone could induce this effect, the combined action was required. Lyso-PAF could not substitute for the PAF, nor could male mouse sera nor the sera from females in dioestrus or metoestrus substitute for the oestrous mouse serum requirement. Pro-oestrous mouse sera could replace oestrous mouse sera but were less effective in their dose-responses. Studies on the mechanism of action of the PAF and oestrous mouse serum components suggested that the PAF stimulated the production and release of soluble factors (termed S2 factors) which by themselves could induce increased RIT values when applied to fresh spleen cells. The PAF-stimulated cell populations were rendered refractory to the action of these S2 factors and did not display increased RIT values, unless oestrous mouse serum was added. This serum acted to reverse the refractory state, allowing the S2 factors to exert their effect, and so cells treated with PAF and oestrous mouse serum displayed increased RIT values.     

Detection of early pregnancy factor (EPF) using the rabbit ovary and oviduct perfused in vitro

Two peaks of rabbit serum EPF activity were seen over the course of pregnancy. Rabbit ovaries with or without attached oviducts were perfused in vitro for 5 h beginning 12, 16, 24, 48, 72 and 120 h after mating. Perfused isolated ovaries did not produce EPF in vitro, but significant EPF activity was detected in the perfusate of the ovary together with oviduct. Pseudopregnant animals and those rabbits that did not ovulate exhibited no perfusate EPF activity. Perfusate EPF activity was highest at the time embryos were at the pronuclear stage and continued through the morula stage. Although the location of embryos at 72 h after mating varied between oviduct and uterus, EPF activity was maintained over the perfusion period. The results suggest that EPF release occurs within 3 h of fertilization and that the presence of the preblastocyst embryo is crucial for EPF release.     

Purification and characterisation of functional early pregnancy factor expressed in Sf9 insect cells and in Escherichia coli

Early pregnancy factor (EPF) is a secreted protein with growth regulatory and immunomodulatory properties. It is an extracellular form of the mitochondrial matrix protein chaperonin 10 (Cpn10), a molecular chaperone. An understanding of the mechanism of action of EPF and an exploration of therapeutic potential has been limited by availability of purified material. The present study was undertaken to develop a simple high-yielding procedure for preparation of material for structure/function studies, which could be scaled up for therapeutic application. Human EPF was expressed in Sf9 insect cells by baculovirus infection and in Escherichia coli using a heat inducible vector. A modified molecule with an additional N-terminal alanine was also expressed in E. coli. The soluble protein was purified from cell lysates via anion exchange (negative-binding mode), cation exchange, and hydrophobic interaction chromatography, yielding approximately 42 and 36mg EPF from 300ml bacterial and 1L Sf9 cultures, respectively. The preparations were highly purified (#10878;99% purity on SDS-PAGE for the bacterial products and #10878;97% for that of insect cells) and had the expected mass and heptameric structure under native conditions, as determined by mass spectrometry and gel permeation chromatography, respectively. All recombinant preparations exhibited activity in the EPF bioassay, the rosette inhibition test, with similar potency both to each other and to the native molecule. In two in vivo assays of immunosuppressive activity, the delayed-type hypersensitivity reaction and experimental autoimmune encephalomyelitis, the insect cell and modified bacterial products, both with N-terminal additions (acetylation or amino acid), exhibited similar levels of suppressive activity, but the bacterial product with no N-terminal modification had no effect in either assay. Studies by others have shown that N-terminal addition is not necessary for Cpn10 activity. By defining techniques for facile production of molecules with and without immunosuppressive properties, the present studies make it possible to explore mechanisms underlying the distinction between EPF and Cpn10 activity.