| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Female Reproductive Tract |
Stimulates Secretion of Macrophage Migration Inhibitory Factor from Bovine Endometrial Epithelial Cells1
Centre de Recherche en Reproduction Animale, Faculté de Médecine Vétérinaire, Université de Montréal, St. Hyacinthe, Québec J2S 7C6, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
release, but it also modifies protein synthesis in the endometrium. This is accomplished by the secretion of interferon-tau (IFN-
) from the embryo. The objective of this study was to identify and characterize specific proteins secreted from endometrial epithelial cells in response to IFN- that could be important for endometrial function and/or embryo development. The epithelial cells were prepared and cultured to confluence and then incubated with or without 100 ng/ml IFN-. At the end of the incubation, the proteins in the medium were analyzed by two-dimensional PAGE. The result showed that two major protein spots were induced by IFN-. One has a molecular mass of approximately 12 kDa and an isoelectric point (pI) of 6.7; the other has a molecular mass of 76 kDa and pI of 4.8. Protein sequence analysis showed that the 12-kDa protein contained a partial amino acid sequence that corresponded to macrophage migration inhibitory factor (MIF). To determine whether MIF is expressed in endometrial cells, isolated stromal or epithelial cells were incubated with or without 100 ng/ml IFN- for 0, 3, 6, 12, 24, and 48 h. After incubation, the MIF protein in cells was examined by Western blotting analysis, and the steady-state mRNA for MIF was examined by Northern analysis. Results showed that MIF protein and mRNA were present in the epithelial cells but not the stromal cells. The presence of MIF in the luminal epithelium of endometrial tissue was confirmed by immunohistochemistry. However, there was no effect of IFN- on MIF expression in the epithelial cells. The concentration of MIF in the medium was quantified by Western blotting analysis to determine if IFN- altered MIF protein secretion from the epithelial cells. The results showed that IFN- significantly stimulated the secretion of MIF protein from the cells. These data show that MIF is expressed in the epithelial, but not the stromal, cells of the endometrium and that MIF secretion from the epithelial cells is stimulated by IFN-. It is therefore likely that MIF plays a role in early embryo development, and further characterization of MIF expression and its regulation in the endometrium will add significantly to our understanding of early embryo-uterine interactions.
conceptus, embryo, female reproductive tract, pregnancy, uterus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(PGF2) release, but it also modifies protein secretion from the endometrium [1, 2]. Type I trophoblast interferon-tau (IFN-) is the antiluteolytic protein secreted by conceptuses of ruminants during maternal recognition of pregnancy. IFN- is secreted specifically by the conceptus and is released by the bovine conceptus as early as Day 9 of pregnancy [2]. Secretion of IFN- increases until about Day 17 when it attenuates the release of PGF2 and indirectly rescues the corpus luteum from regression [3] and then secretion declines. Cytokines appear to play critical roles in the establishment of pregnancy and IFN-, apart from preventing the luteolytic secretion of PGF2, also induces the alpha chemokine, granulocyte chemotactic protein-2 [4, 5], and several other uterine proteins such as 2',5'-oligoadenylate synthetase [6], acidic secretory protein [7], and ubiquitin cross-reactive protein [8, 9]. These uterine proteins probably mediate distal responses to IFN- in the endometrium during early pregnancy. The exact role that these proteins play in early pregnancy is still not fully understood, but they are probably involved in growth and development of the embryo, remodeling of the endometrium during attachment, and modulating the immune response.
Under normal conditions, the conceptus (the fetus and its membranes) expresses paternal antigens but is not rejected by the maternal immune system. The conceptus escapes destruction by immune effector cells by expressing atypical major histocompatibility complex molecules [10, 11], which protect the invasive trophoblast against attack by cytotoxic lymphocytes [12] and natural killer (NK) cells [13], and by producing factors that stimulate the production of beneficial cytokines [14, 15]. Few studies have concentrated on immunomodulatory events during the peri-implantation period in ruminants[16] but there is evidence that local immunity plays an important role in the fate of the ruminant conceptus. Immune cells have been reported in bovine [17] and ovine uteri during pregnancy [17, 18], and lymphokine-activated killer cells exert lytic damage on preattachment conceptuses [19]. Furthermore, uterine milk proteins, which are secreted by the endometrium of ruminants in response to IFN-, can inhibit NK cells [20].
Because the embryo interacts predominantly with the epithelial cells of the endometrium, the objectives of this study were: 1) to use high resolution two-dimensional (2D) electrophoresis to characterize the IFN--induced changes in the secretion of newly synthesized proteins from isolated endometrial epithelial cells and 2) to identify specific proteins induced by IFN- that are involved in embryo-uterine interactions during early pregnancy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell culture medium (RPMI 1640), Hanks buffered saline solution (HBSS, calcium and magnesium free), newborn calf serum (NBCS), antibiotics, and trypan blue were purchased from Gibco (Grand Island, NY). Collagenase (type II), trypsin (type III, from bovine pancreas), DNase I (type I, from bovine pancreas), gentamicin, BSA, and progesterone were purchased from Sigma Chemical Co. (St. Louis, MO). A stock solution of progesterone was prepared by dissolving the steroid in ethanol. Matrigel was obtained from VWR Canlab (Montreal, QC, Canada). Protein assay-dye reagent concentrate and electrophoresis reagents were obtained from Bio-Rad Laboratories (Hercules, CA). Mouse anti-human migration inhibitory factor (MIF) antibody and mouse IgG were purchased from Cedarlane Laboratories Limited (Hornby, ON, Canada). Biotrans nylon membranes (0.2 µm) were obtained from ICN Pharmaceuticals (Montreal, PQ, Canada). [-32P]dCTP and [35S]ATP were obtained from Mandel Scientific NEN Life Science Products (Mississauga, ON, Canada). RNA ladder (0.24–9.5 kb), 1 kb DNA ladder, and tissue culture plates were obtained from Corning-Costar (Fisher Scientific, Montreal, PQ, Canada). X-OMAT AR film was obtained from Eastman Kodak Company (Rochester, NY). The recombinant bovine IFN- was a generous gift from Dr. R. Michael Roberts (University of Missouri). The MIF cDNA was a generous gift from Dr. A. Meinhardt (Philipps-University, Marburg, Germany).
Preparation and Culture of Cells
The epithelial cells were prepared as previously described [21]. Uteri from cows at Days 1–3 of the estrous cycle (ovaries with a corpus hemorrhagicum) were collected at the slaughterhouse and transported on ice to the laboratory. Cells prepared from endometrium at this stage respond to IFN- in a physiological manner, i.e., IFN- inhibits the oxytocin stimulation of PGF2 secretion [22]. Briefly, the two horns of the uteri were placed in sterile HBSS containing 100 U penicillin, 100 µg streptomycin, and 0.25 µg amphotericin ml-1. The myometrial layers were dissected from the two horns, and the horns were then everted to expose the epithelium. The everted horns were digested for 2 h in HBSS with 0.3% (w/v) trypsin at 37°C to obtain epithelial cells. At the end of incubation, the digested horns were scraped lightly with forceps, washed twice in HBSS, and then further digested to obtain stromal cells by incubating in HBSS with 0.016% (w/v) trypsin III, 0.016% (w/v) collagenase II, and 0.008% (w/v) DNase I for 45 min at 37°C. Immediately after each cell suspension was collected, 10% NBCS was added to inhibit the trypsin.
For epithelial cells, the cell suspension was centrifuged at 60 x g for 5 min and then the pellet was washed 3 more times with HBSS. For further purification, the epithelial cell pellet was suspended in 20 ml RPMI-1640 medium supplemented with 5% NBCS and 50 mg/ml of gentamicin and plated onto 100 x 20 mm Nunclon Petri dishes (Grand Island, NY) and incubated at 37°C with 5% CO2, 95% air for 3 h. At the end of incubation, contaminating stromal cells adhered to the dish and the floating epithelial cells were collected. After cell counting and viability determination by trypan-blue exclusion, viable cells were plated onto Matrigel-coated culture dishes. One hundred microliters of 11% Matrigel were added to each well of 24-well plates and 200 µl was added to each well of 6-well plates and the plates were dried overnight. For stromal cells, the cell suspension was centrifuged at 60 x g for 5 min to remove clumps of cells and then the supernatant was centrifuged at 1000 x g for 10 min. The pelleted cells were washed twice with HBSS. The stromal cell suspension was plated onto dishes at a concentration of 1 x 107 cells per dish, and after a 3-h incubation, the floating cells were washed away by gentle pipetting.
The cells were cultured at 37°C with 5% CO2, 95% air until they were confluent (about 7 days) in RPMI-medium supplemented with 10% NBCS and 100 ng/ml progesterone. The culture medium was changed every 2 days. The homogeneity of the cell populations was examined by immunocytochemistry. Epithelial cell contamination of stromal cells was about 3%, and stromal cell contamination of epithelial cells was less than 1% [21].
Radioactive Labeling of Secreted Proteins and 2D-PAGE
The confluent epithelial cells were incubated in the presence or absence of 100 ng/ml IFN- for 24 h at 37°C. Cells were washed and incubated with methionine-free RPMI-1640 medium for 30 min. The medium was then replaced with 500 µl of methionine-free RPMI-1640 medium containing 5 µl (50 µCi) of 35S-labeled methionine (specific activity >1200 Ci/mmol) and the cells were incubated in the presence or absence of 100 ng/ml IFN- for a further 24 h. The medium was removed and stored at -70°C until protein extraction.
Before separation and analysis of the proteins by means of 2D SDS-PAGE (2D PAGE), the culture medium was centrifuged (500 x g for 10 min) to remove cell debris prior to protein extraction. The proteins were concentrated to 50 µl using Ultrafree-15 concentrators (5000 MW cutoff; Millipore, Bedford, MA) and added to IPG buffer (8 M urea, 2% CHAPS, 0.5% IPG buffer [pH 3–10], bromophenol blue, 65 mM dithiothreitol [DTT]). Prior to loading, a 5-µl aliquot of each sample was removed for radioactive counts, and 5 µl internal protein standards were added to each sample so that the molecular mass and isoelectric points (pI) of found proteins could be estimated. The separation in first dimension was carried out using Immobiline DryStrips (pH 3–10), which had been rehydrated for at least 10 h in an Immobilin Drystrip reswelling tray (Amersham Pharmacia Biotech AB, Baie d'Urfé, QC, Canada). The samples were then separated on a MultiPhore II flatbed system (Amersham) for 16 h at 15°C. The voltage was 300 V for the first 3 h, from 300 to 2000 V during the following 5 h, and finally 8 h at 2000 V. Before the second dimension was performed, the dry strips were first equilibrated for 10 min in equilibration solution 1 (0.5 M Tris/HCl pH 6.8, 3.6 g urea, 3 ml glycerol, 0.1 g SDS, 25 mg DDT, and distilled water up to 10 ml) and another 10 min in equilibration solution 2 (0.5 M Tris/HCl pH 6.8, 3.6 g urea, 3 ml glycerol, 0.1 g SDS, 0.45 g iodoacetamide, and distilled water up to 10 ml). The second dimension was performed after placing the strips on Pharmacia ExcelGel XL SDS 8–18 using the MultiPhore II flatbed system at 15°C. After running the gels, they were immediately immersed in fixing solution (50% methanol, 10% acetic acid in water) and stained with Coomassie blue, destained, and incubated in a radiographic enhancer and then in a preserving solution. The gels were wrapped in cellophane, air dried, and exposed with Kodak radiographic film for various times. Protein spots on control and IFN- gels were compared, and molecular weight and pI estimated using the computer program Phoretix 2D (version 4.00, Nonlinear Dynamics Ltd., Newcastle upon Tyne, UK).
Protein Sequencing
To obtain proteins for sequencing, the procedure for culture and incubation of the cells was the same as described above except that the proteins were not labeled with 35S-methionine. The 2D gels were run, each loaded with 50 µg of unlabeled proteins. After 2D PAGE the proteins were stained with silver nitrate (silver staining kit, Amersham), and the spots of interest were excised. The sequence analysis was performed at the Harvard Microchemistry Facility by Microcapillary reverse-phase HPLC nanoelectrospray tandem mass spectrometry on a Finnigan LCQDECA quadruple ion-trap mass spectrometer.
Isolation of Total Cellular RNA and Northern Blot Analysis
Bovine epithelial cells were cultured as described above and were then incubated with or without 100 ng of IFN-/ml for different times in 6-well flat-bottom plates. After the incubation the total cellular RNA was isolated according to the manufacturer's specifications (Rneasy Mini Kit (50), Qiagen, Mississauga, ON, Canada). For Northern analysis, RNA samples (10 µg) were denatured at 70°C for 5 min in denaturing buffer, electrophoresed on a 1.2% agarose gel, and transferred overnight by capillarity to a nylon membrane, as previously described [23]. The membranes were UV treated (150 mJ) and prehybridized as described by Johnson et al. [24]. A ladder of RNA standards was run with each gel, and ethidium bromide (10 µg) was added to each sample prior to electrophoresis to compare RNA loading and determine migration of standards. The membrane was first hybridized to MIF cDNA probe overnight at 42°C, which was randomly primed with 50 µCi -32P-dCTP using QuikHyb solution (Stratagene, La Jolla, CA). Blots were washed as described previously [24]. After stripping the radioactivity with 0.1% saline-sodium citrate (0.15 M NaCl and 0.015 M sodium citrate)-0.1% SDS for 30 min at 100°C, the same blot was subsequently hybridized with a pig glyceraldehyde-3-phosphate dehydrogenase.
(GAPDH) cDNA as a control gene for RNA loading and transfer [25]. Probes were labeled with [-32P]deoxy-CTP using the Prime-a-Gene labeling system (Promega, Madison, WI) to a final specific activity greater than 1 x 108 cpm/µg DNA. The membranes were then scanned using a Storm 840 PhosphorImager scanner and quantified by densitometry using ImageQuant software (version 1.2), (both from Molecular Dynamics, Inc., Sunnyvale, CA).
Protein Extracts and Immunoblot Analysis
Bovine epithelial cells were cultured as described above and then incubated with or without 100 ng of IFN-/ml for different times in 6-well flat-bottom plates. After incubation the medium was removed and solubilized protein extracts were prepared from the cells as previously described [23], with minor modifications. Briefly, after treatment, uterine cells were rinsed with HBSS and detached from the dish with 250 µl TED sonification buffer (20 mM TRIS, 50 mM EDTA, 0.1 mM diethyldithiocarbamic acid [DEDTC], pH 8.0) containing 32 mM octyl glucoside and then sonicated (8 sec/cycle; three cycles). The sonicates were centrifuged at 13 000 x g for 25 min at 4°C. The supernatants (solubilized cytoplasmic and cell extracts) were stored at -70°C until immunoblotting analysis. The medium (removed as described above) was centrifuged (500 x g for 10 min) to remove cell debris prior to protein extraction. The proteins in the medium were concentrated using Ultrafree-15 concentrators (5000 MW cutoff; Millipore). The protein concentration was determined by the method of Bradford.
Proteins (25 µg) of cell and medium extracts were resolved by one-dimensional SDS-PAGE and electrophoretically transferred onto nitrocellulose membranes (Hybond-ECL; Amersham Life Science, Inc, Buckinghamshire, UK). The blots were incubated for 18 h at 4°C in the presence of mouse anti-human MIF monoclonal immunoglobulin (Ig) G, diluted to 1:3000. Blots were washed, incubated with second antibody (anti-mouse IgG), and exposed to chemiluminescence detection substrates as described. The membranes were scanned as described above. The molecular size of immunoreactive bands was determined by comigration of a ladder of biotinylated SDS-PAGE molecular weight standards (Bio-Rad Laboratories) applied to a lane in each gel. Prestained standards were also applied to gels to assess the transfer efficiency of samples.
Immunohistochemistry
Uteri (Day 1–3 of the estrous cycle) were obtained from a slaughterhouse. Pieces of tissue were carefully excised and transferred to 10% neutral buffered formalin fixing solution for 24 h and then embedded in paraffin. Paraffin tissue sections (5 µm) were dewaxed, hydrated, and endogenous peroxidase activity quenched. They were then rinsed twice with PBS and subjected to immunohistochemical labeling with an avidin-biotin-peroxidase complex (ABC) method to examine MIF expression and secretion. All primary antibodies were labeled with the peroxidase Vectastain elite ABC Kit (Vector Laboratories, Burlingame, CA). Diaminobenzidine tetrahydrochloride (DAB substrate kit; Vector Laboratories) was used as the substrate-chromogen solution. Negative controls were obtained by omitting primary antibodies. Bovine corpus luteum was used as positive control. After development of the immunoreaction, the slides were counterstained with hematoxylin.
Statistical Analysis
Each experiment was carried out using the cells from one uterus and was repeated with three different uteri. Northern and Western blots were run in triplicate for each uterus. For MIF expression, the data were analyzed by two-way ANOVA, which included the main effects of time and treatment (control, IFN-) and the time x treatment interaction. Differences between individual means were determined by Tukey honestly significant difference test. A probability of P < 0.05 was considered to be statistically significant. The data were analyzed using the computer program JMP (SAS Institute Inc., Cary, NC).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
on Secretion of Newly Synthesized Proteins from Endometrial Epithelial Cells
To determine the effect of IFN- on protein secretion, the confluent cultures of bovine endometrial epithelial cells were treated with or without IFN- and the 35S-methionine labeled proteins were analyzed by 2D PAGE. Figure 1 shows representative 2D gel autoradiographs of labeled proteins from control and IFN- treated cells. A comparison of control and IFN--treated cells showed that two protein spots, one with the estimated pI of 4.8 and molecular mass of 76 kDa (designated P76) and the other with a pI of 6.7 and molecular mass of 12 kDa (designated P12), were present in IFN- but not in control cells.
|
Identification of P12
In-gel digestion and sequence analysis was performed at the Harvard Microchemistry Facility by microcapillary reverse-phase HPLC nanoelectrospray tandem mass spectrometry on a Finnigan LCQDECA quadruple ion trap mass spectrometer. The peptide sequence (LLCGLLTER) that resulted in positive identification of P12 was located in the N-terminal part of bovine MIF (amino acid sequence of bovine MIF from Genbank Database [accession number AAB32021]): PMFVVNTNVP RASVPDGLLS ELTQQLAQAT GKPAQYIAVH VVPDQLMTFG GSSEPCALCS LHSIGKIGGA QNRSYSKLLC GLLTERLRIS PDRIYINYYD MNAANVGWNG STFA.
Effect of IFN- on MIF mRNA and Protein Expression in Endometrial Cells
A single band at approximately 600 bp (the predicted size for the MIF mRNA) was observed in all epithelial RNA samples (Fig. 2). Densitometric analysis of the 600-bp bands, and normalization of these values to those of the respective GAPDH band, revealed that expression did not change with time of incubation and that there was no effect of IFN- on MIF mRNA levels at any of the time points. No band corresponding to MIF mRNA was observed in samples from stromal cells (data not shown).
|
To determine whether MIF protein was expressed in the endometrial cells, cellular extracts were analyzed by Western blotting using mouse anti-human MIF antibody. Results show a strong immunoreactive signal in all samples of epithelial cells. The amount of MIF did not change with time, and there was no effect of IFN- (Fig. 3). MIF protein was not detectable in the stromal cells (data not shown).
|
Analysis of MIF Protein Secretion from the Endometrial Epithelial Cells
To determine whether IFN- altered the secretion of MIF protein from the epithelial cells, proteins present in the culture medium were analyzed by Western blotting using mouse anti-human MIF antibody. The results showed a strong immunoreactive signal in the medium from bovine endometrial epithelial cells cultured in the presence of IFN-, compared with control samples (Fig. 4). In the cells treated with IFN-, the amount of MIF secreted into the medium increased with time (P < 0.01) and was maximum at 12 h.
|
Expression of MIF in Bovine Endometrium
To determine whether MIF expression could be detected in vivo, immunohistochemistry was performed on formalin-fixed sections of bovine uteri taken at Days 1–3 of the cycle. Results showed MIF staining in the endometrium but not the myometrium. The staining was particularly intense on the apical side of the luminal epithelium and in the superficial glandular epithelium (Fig. 5, left image). No staining was observed in the epithelium of the deeper glands or in the subepithelial compact stroma.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
stimulates the synthesis of a variety of proteins as well as preventing the luteolytic secretion of PGF2. The present study has shown that IFN- induces the secretion of two newly synthesized proteins, one of which has been identified as MIF. This demonstrates for the first time that MIF is expressed in the ruminant endometrium in vivo. From the in vitro experiments, it appears that MIF is located specifically in the epithelial cells because it was not detected in the stromal cells in vitro. This agrees with the immunohistochemical data showing MIF in the luminal epithelium and the epithelium of the superficial glands. In our cell preparations, the epithelial cells are from luminal and superficial glands because they are obtained by enzymatic digestion of endometrial tissue from the side of the surface epithelium. There do appear to be differences between epithelial cell types because no staining was observed in the epithelium of the deep glands. The stromal cells used for the in vitro studies are likely to be from the just under the surface epithelium and not from the deep stromal layer. Again, this agrees with the immunohistochemical data because no staining was observed in the subepithelial stroma, although MIF staining was observed in the deep stroma. MIF was discovered as an activated T-lymphocyte-derived protein that inhibits the random migration of macrophages in vitro and is secreted by macrophages in response to cytokine stimulation. To date, its role as a proinflammatory cytokine involved in several aspects of the immune response has been demonstrated extensively [26]. Although there is information emerging concerning the presence of MIF in the uterus, its role in uterine function is not known. As in other animals, the endometrium of the cow contains immune cells [27]. Macrophages are widely distributed in female reproductive tissues and, in the endometrium, these cells represent an important mechanism of defense of its integrity and function. In the nonpregnant uterus, macrophage degradation of cellular debris and foreign material may be important in endometrial shedding and repair and in providing protection against infections.
It is now becoming clear, however, that MIF exerts a variety of biologic functions and MIF expression is found in cells other than activated T cells, indicating its involvement beyond the immune system in different pathophysiologic states [26, 28]. MIF mRNA is expressed in human ovary [29] and ovulated oocytes, zygotes, two-cell embryos, eight-cell embryos, and blastocysts of mice [30] which, together with other findings, suggests a possible role of MIF in different aspects of reproduction, such as ovulation, blastocyst implantation, and embryogenesis [31]. In human pregnancy, a sarcolectin-binding protein whose properties corresponded to those of MIF has been described in term placenta [32]. Recent reports have shown that MIF is expressed in the human endometrium [33] and by first-trimester trophoblasts, in which it can play a role during implantation and early embryonic development [34]. MIF mRNA is also expressed in the mouse ovary, oviduct, and uterus during the preimplantation period and all stages of the estrous cycle [30]. There is, however, no previous report in the literature on the presence of MIF in bovine endometrium.
The other important finding of this study is that secretion of MIF from the endometrial epithelial cells is stimulated by IFN-. Although IFN- can alter prostaglandin secretion from stromal cells of the bovine endometrium [35–37], it had no effect on the expression of MIF in these cells. IFN- did not appear to stimulate either MIF mRNA or protein in the cells but did increase the secretion of MIF. A possible explanation for this is that either IFN- stimulates the synthesis of MIF protein, and this is preferentially secreted, or that IFN- stimulates only the secretion of stored MIF. The anterior pituitary cells and macrophages of mice contain a significant amount of preformed MIF within intracellular pools that can be rapidly released on stimulation. This is in contrast to other proinflammatory cytokines, such as interleukin-1ß and tumor necrosis factor-, that require de novo mRNA generation and protein synthesis before secretion is observed. There is little information on the role of MIF in early pregnancy in any species, and the present results suggest that the bovine conceptus has the capacity for local modulation of the production of cytokines that, in turn, may sustain development and maintain pregnancy.
During pregnancy in mice, macrophages are present at the fetomaternal interface, suggesting an involvement in both the response to infection and the immune interactions between fetal and maternal tissues [38]. Despite such important physiological functions, the mechanisms involved in recruiting, maintaining, and activating macrophages in the uterus are not fully defined. However, several studies emphasize the central role of cytokines in these processes. It has been shown that uterine cells are a potent source of cytokines with well-defined functions in promoting monocyte migration and activation, such as colony-stimulating factor (CSF)-1, granulocyte-macrophage-CSF, monocyte chemotactic protein-1, and regulated upon activation, normal T cell expressed, and secreted [38]. Macrophages are the main target of MIF, which acts on these cells by inhibiting migration and increasing their scavenger activity [39, 40]. Based on this study, it is speculated that MIF may be involved in the macrophage accumulation and activation in the bovine endometrium. The endometrium of pregnant and nonpregnant uterus is also populated by NK cells. It has been proposed that in the nonpregnant uterus, they could affect growth, differentiation, breakdown, and regeneration of the uterine mucosa. In addition, in the endometrium during early pregnancy, these cells could influence implantation by controlling trophoblast invasion of the decidua and downregulating the immune response to the semiallogenic fetus [41, 42]. Several studies have been directed toward comprehension of the mechanisms regulating the activity of NK cells in human endometrium. In this context, the observation that the cytolytic activity of uterine NK cells can be regulated by cytokines, acting either as stimulatory or suppressive factors, is relevant [43, 44]. A remarkable feature of MIF is its immunosuppressive activity, as demonstrated by the recent study of Apte et al. [45]. These authors showed that MIF is able to inhibit NK cell-mediated cytolysis of both neoplastic and normal target cells, indicating that this cytokine can contribute to preserving immune privilege. In this respect MIF is similar to uterine milk proteins, which are also secreted by the endometrium of ruminants and inhibit NK cells [20]. This inhibitory action may be important because activated NK cells can lyse trophoblast cells [16]. Thus, there is the potential involvement of MIF in the control of macrophage and NK cell activity in the bovine endometrium, which could be important for normal embryo development.
It is also possible that MIF can directly affect the nonimmune cells of the endometrium. For example, MIF could affect the growth and differentiation of the endometrial tissue and/or the embryo. It has been shown that MIF can function as an autocrine mediator of growth factor-dependent extracellularly regulated kinase, mitogen-activated protein kinase activation, and cell cycle progression [46]. In this way MIF is able to regulate proliferative and oncogenic processes. In light of the expression of MIF in the bovine endometrial epithelial cells, and its stimulation by IFN-, further work is needed to examine the effects of MIF on endometrial and embryonic cell function.
In conclusion, this study shows MIF mRNA and protein are expressed in cultured bovine endometrial epithelial cells and that the secretion of MIF is stimulated in response to IFN- in vitro. MIF expression is observed in vivo, particularly on the apical surface of the luminal epithelium and superficial glands of the endometrium. Taken together, our results suggest that MIF is likely a factor contributing to the establishment of early pregnancy; however, the functional significance of MIF remains to be determined. Understanding the regulation of MIF secretion and its site of action in the reproductive tract will add significantly to our understanding of early embryo-uterine interactions.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence. FAX: 450 778 8103; goffak{at}medvet.umontreal.ca
Received: 21 October 2002.
First decision: 12 November 2002.
Accepted: 11 June 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E.A. Campbell, L. O'Hara, R.D. Catalano, A.M. Sharkey, T.C. Freeman, and M. H. Johnson Temporal expression profiling of the uterine luminal epithelium of the pseudo-pregnant mouse suggests receptivity to the fertilized egg is associated with complex transcriptional changes Hum. Reprod., October 1, 2006; 21(10): 2495 - 2513. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Akoum, C. N. Metz, and M. Morin Marked Increase in Macrophage Migration Inhibitory Factor Synthesis and Secretion in Human Endometrial Cells in Response to Human Chorionic Gonadotropin Hormone J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2904 - 2910. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Paulesu, C. Cateni, R. Romagnoli, F. Ietta, and V. Dantzer Variation in Macrophage-Migration-Inhibitory-Factor Immunoreactivity During Porcine Gestation Biol Reprod, April 1, 2005; 72(4): 949 - 953. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
