| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Unité d'Endocrinologie de la Reproduction, Centre de Recherche, Hôpital Saint-François d'Assise, Centre Hospitalier Universitaire de Québec, Université Laval, Québec, Québec, Canada G1L 3L5
b Department of Biotechnology, Research Center Dompé SpA, 67100 L'Aquila, Italy
ABSTRACT
Cytokines such as interleukin-1 (IL-1) play a major role in the reparative and inflammatory-like processes that occur in human endometrium during every menstrual cycle, but they also seem to be implicated in critical reproductive events such as ovulation and implantation. Interleukin-1 is tightly regulated in the body by a complex network of control systems. In the present study, we examined the expression of IL-1RII, a natural specific inhibitor of IL-1, in the human endometrium and found an interesting distribution and temporal pattern of expression throughout the menstrual cycle. Immunoreactive IL-1RII was found in stromal as well as epithelial cells, but it was predominant within the lumen of the glands and the apical side of surface epithelium. In situ hybridization and reverse transcription-polymerase chain reaction (RT-PCR) analyses showed higher levels of mRNA in epithelial than in stromal cells. The IL-1RII cellular and luminal secretion followed a regulated cycle phase-dependent pattern of expression. Although elevated in the late proliferative/early secretory phase of the menstrual cycle, IL-1RII luminal secretion significantly decreased in the midsecretory phase, reaching its lowest levels at Day 21, before augmenting markedly again during the late secretory phase. This pattern of expression was less obvious at the level of cellular staining, as examined by immunohistochemistry, but it was corroborated by Western blot analysis of IL-1RII protein and semiquantitative RT-PCR of IL-1RII mRNA in the whole endometrial tissue and separated glandular epithelial cells. The reduced expression of IL-1RII within the implantation window suggests the existence of accurate regulatory mechanisms that, by down-regulating IL-1RII expression, alleviate IL-1 inhibition during this crucial period and facilitate IL-1 proimplantation actions. The elevated expression of IL-1RII observed during the late secretory phase suggests an involvement of IL-1RII in control of the proinflammatory state that takes place in the endometrium during the premenstrual and menstrual periods.
cytokines, female reproductive tract, menstrual cycle
INTRODUCTION
Interleukin-1 (IL-1) is the term used to describe two polypeptides (IL-1
and IL-1ß) that play a key role in immune and inflammatory reactions [1]. Three receptors for IL-1, type I (IL-1RI), type II (IL-1RII), and type III (IL-1RIII), have been identified in different cell types [1, 2]. Cell activation in response to IL-1 appears to be mediated exclusively via the IL-1RI [3, 4], with coexpression of a receptor accessory protein (IL-1R-AcP or IL-1RIII) being crucial to IL-1-mediated signaling events [5–7]. In itself, IL-1RII is not a signaling molecule but, in fact, is reported to be a decoy target of IL-1 [8, 9]. Additionally, IL-RII could be shed from the cell surface as a soluble molecule that would then capture IL-1 and inhibit its binding to IL-1RI [10–12], thus suggesting an important role for IL-RII in regulating the biological activities of IL-1.
The available data indicate that IL-1 is involved in the regulation of endometrial functions [13]. It has been shown that IL-1 is secreted by human blastocysts, and it is thought to act as an embryonic signal [14, 15]. The cytokine was also detected locally in the endometrial tissue during the late secretory phase of the menstrual cycle [16]. This suggests a role in the tissue necrosis and disintegration occurring in the endometrium at the end of the menstrual cycle in the absence of implantation, which is not surprising considering the similarity between these processes and those occurring during inflammation. Expression of the functional receptor of IL-1 (i.e., IL-1RI) has been detected in the human endometrium as well [17, 18], where it appears to play a key role in the implantation process [19].
Due to its pleiotropic activity and potent proinflammatory effects, IL-1 is tightly regulated in the body by complex control systems. In particular, two inhibitors participate in these regulatory mechanisms: the receptor antagonist (IL-1ra), which binds avidly to IL-1RI and prevents IL-1 binding and signal transduction; and IL-1RII, which is considered to be a natural scavenger for IL-1. The IL-1RII can very efficiently bind IL-1ß, whereas its affinity for IL-1 and IL-1ra is 10- to 100-fold lower [20].
In view of the major role of IL-1 in the regulation of various endometrial and reproductive functions, knowledge regarding the local availability and accurate production of specific inhibitors for IL-1 in the endometrium throughout the menstrual cycle becomes essential. The expression of IL-1ra in the human endometrium has been previously reported [21]: IL-1ra immunoreactivity was elevated during the proliferative phase of the menstrual cycle, whereas endometrial cells appeared to express intracellular IL-1ra (icIL-1ra). The objective of the present study was to assess the local availability and expression of IL-1RII throughout the normal menstrual cycle to further elucidate the mechanisms controlling IL-1 activity in the endometrium.
MATERIALS AND METHODS
Subjects
Women who participated in the study provided informed consent for a protocol approved by the Saint-François d'Assise Hospital Ethics Committee on Human Research. These women (n = 42) were aged between 23 and 47 yr (mean ± SD, 34.6 ± 5.0 yr). They were fertile, requested tubal ligation, and had a normal and regular menstrual cycle. None had visible endometrial hyperplasia or neoplasia, inflammatory disease, or endometriosis at the time of clinical examination or laparoscopy. Women were not receiving any anti-inflammatory or hormonal medication at least 3 mo before laparoscopy. The cycle day was determined according to the cycle history and histological criteria [22]. Eighteen women were in the proliferative phase and 24 in the secretory phase.
Collection of Endometrial Biopsy Specimens
Endometrial biopsy specimens were obtained using sterile pipelle (Unimar, Inc., Neuilly-en-Thelle, France). Samples were placed at 4°C in sterile Hank balanced salt solution (HBSS; Gibco BRL, Burlington, ON, Canada) containing 100 U/ml of penicillin, 100 µg/ml of streptomycin, and 0.25 µg/ml of amphotericin. Samples were then immediately transported to the laboratory, washed twice in HBSS at 4°C, then snap-frozen on dry ice and kept at -80°C in Eppendorf tubes for Western blot and reverse transcription-polymerase chain reaction (RT-PCR) analyses or in Tissue-Tek OCT compound (Miles, Inc., Elkhart, IN) for immunohistochemical studies.
Immunohistochemistry
Serial 4-µm cryosections were placed on poly-L-lysine-coated glass microscope slides and fixed for 20 min in formaldehyde (4% [v/v] in PBS; Fisher Scientific, Montreal, PQ, Canada). All incubations were performed at room temperature in a humidified chamber. Sections were rinsed in PBS, immersed in PBS/1% (v/v) Triton X-100 for 20 min at room temperature, rinsed again in PBS, and then treated for 20 min with hydrogen peroxide (H2O2, 0.3% [v/v] in absolute methanol) to eliminate endogenous peroxidase. After a PBS rinse, immunostaining was performed using a mouse monoclonal anti-human IL-1RII antibody (primary antibody; R&D Systems, Minneapolis, MN) at 15 µg/ml in PBS containing 1% (w/v) BSA with a Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) and diaminobenzidine (Sigma Chemical Co., St. Louis, MO) as chromogen.
Sections incubated without the primary antibody or with nonimmune mouse serum were included as negative controls in all experiments. Slides were viewed using a Leica microscope (model DMRB; Leica Mikroskopie und Systeme GmbH, Postfach, Wetzlar, Germany), and photomicrographs were taken with Kodak 100 ASA film (Kodak, Toronto, ON, Canada). The IL-1RII immunostaining was evaluated in a blinded fashion by two independent observers having no knowledge of laparoscopic findings. The intensity of staining was evaluated three times in three different areas that were randomly selected in the section, and a mean score was given using an arbitrary scale (0 = absent, 1 = light; 2 = moderate, and 3 = intense). High concordance between the two observers was found as determined by the kappa (
) measure of agreement ( = 0.89) [23].
Western Blot Analysis
Frozen endometrial tissues were directly homogenized with the use of a microscale tissue grinder (Kontes, Vineland, NJ) in a buffer containing 0.5% (v/v) Triton X-100, 10 mM Hepes (pH 7.4), 150 mM NaCl, 2 mM EGTA, 2 mM EDTA, 0.02% (w/v) NaN3 [15], and a mixture of antiproteases composed of 5 µM aprotinin, 63 µM leupeptin, and 3 mM PMSF. Tissue homogenate was then incubated at 4°C for 45 min under gentle shaking and centrifuged at 11 000 x g for 30 min to recover the soluble extract. Total protein concentration was determined using the Bio-Rad DC Protein Assay (Bio-Rad Laboratories Ltd., Mississauga, ON, Canada). One-hundred micrograms of protein from each extract were heated in a boiling bath for 5 min in 5x SDS sample buffer (1.25 M Tris-HCl [pH 6.8], 50% [v/v] glycerol, 25% ß-mercaptoethanol, 10% [w/v] SDS, and 0.01% [w/v] bromophenol blue), separated by SDS-PAGE in 10% (w/v) acrylamide linear-gradient slab gels, and transferred onto 0.45-µm nitrocellulose membranes (Schleicher & Schuell, Keene, NH) using an electrophoretic transfer cell (Bio-Rad). Equal loading in each lane was confirmed by staining the blots with Ponceau S (2% [w/v]). Nitrocellulose membranes were then immersed in PBS containing 5% (w/v) skimmed milk and 0.1% (v/v) Tween 20 (blocking solution) for 1 h at 37°C, cut into strips, and incubated overnight at 4°C with a monoclonal mouse anti-human IL-1RII antibody (2 µg/ml of blocking solution; R&D Systems) or with normal mouse immunoglobulins (Ig) of the same immunoglobulin class and concentration as the primary antibody (R&D Systems). The specificity of the immunoreaction was also verified by preabsorption of the antibody with an excess of IL-1RII (20 µg/ml). Thereafter, the strips were incubated for 1 h at 37°C with Fc-specific peroxidase-labeled goat anti-mouse antibody (1:3000 dilution in the blocking solution; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA), washed three times in PBS/0.1% (v/v) Tween 20, incubated for 1 min with an enhanced chemiluminescence system (BM chemiluminescence blotting substrate [POD]; Roche Diagnostics, Laval, PQ, Canada), and exposed to BioMax film (Eastman Kodak, Rochester, NY) for 5–30 sec for an optimal detection (all bands visible but not overexposed).
In Situ Hybridization
In situ hybridization was performed as described in our previous studies [24]. Briefly, cDNA for human IL-1RII, a 1.3-kilobase fragment, was subcloned into the plasmid vector pcDNA3 [12]. Biotin-labeled cDNA probes were prepared by nick-translation from the entire plasmid vector with the IL-1RII cDNA [25] or from the plasmid vector alone (negative control) using a BioNick Labeling System (Gibco BRL). Serial cryosections were prepared and fixed in formaldehyde as described earlier, then progressively dehydrated in alcohol baths (50%–100% [v/v]). Sections were prehybridized with the hybridization buffer (50% [v/v] formamide, 10% [v/v] dextran sulfate, 0.1% SDS [w/v], 2x SSC [single strength: 0.15 M sodium chloride and 0.015 M sodium citrate], and 1x Denhardt solution [0.02% (v/v) Ficoll (Amersham Pharmacia Biotech, Inc., Baie d'Urfé, QC, Canada), 0.02% (w/v) human serum albumin (HSA), 0.02% (w/v) polyvinylpyrolidone (Sigma), and 40 mM monosodium phosphate (pH 7)]), then hybridized with 5 ng/µl of biotinylated probe in the hybridization buffer. Biotin was detected by a series of 45-min incubations at 37°C with a rabbit polyclonal antibiotin antibody (1% [v/v] dilution in PBS/0.25% [w/v] HSA; Enzo Diagnostics, Inc., Farmingdale, NY), a biotinylated goat anti-rabbit polyclonal antibody (1% [v/v] dilution in PBS/0.25% [w/v] HSA; Vector Laboratories), and fluorescein isothiocyanate-conjugated streptavidin (0.5% [v/v] in PBS/0.25% [w/v] HSA; Roche Diagnostics, Montreal, PQ, Canada), respectively. Slides were then treated with propidium iodine (1.5 µg/ml of distilled water; Sigma) which makes the nucleus visible in yellow-orange on ultraviolet excitation, and mounted with Mowiol (Calbiochem, San Diego, CA), to which p-phenylenediamine (Sigma), an antifading agent, was added at a final concentration of 1 mg/ml. Sections were finally observed under the Leica microscope equipped for fluorescence and connected to an image analysis system (ISIS; Metasystems, Altlussheim, Germany). As negative controls, sections from each tissue were incubated without specific cDNA probes or with nonspecific DNA probes prepared by nick-translation from the plasmid vector alone (i.e., without the IL-1RII cDNA).
Reverse Transcription-Polymerase Chain Reaction
Total RNA was extracted from endometrial homogenized tissue with Trizol reagent according to the manufacturer's instructions (Gibco BRL). The cDNA was synthesized using 500 ng of total cellular RNA and 2.5 µM random hexamers in 20 µl of a solution containing 50 mM KCl, 10 mM Tris-HCl, 5 mM MgCl2, 1 mM of each deoxyribonucleotide triphosphate (dNTP), 20 U of RNase inhibitor, and 50 U of reverse transcriptase using Gene Amp PCR Core Kit (Perkin-Elmer, Foster City, CA). The reaction was incubated at 25°C for 15 min, 42°C for 30 min, and 99°C for 5 min.
For PCR analysis, 2-µl aliquots of each cDNA were amplified in a final volume of 50 µl containing PCR buffer (10 mmol/L of Tris, 50 mmol/L of KCl, and 1.5 mmol/L of MgCl), 0.2 mmol/L of dNTPs, 2.5 U of Taq DNA Polymerase (New England Biolabs, Beverly, MA), and 100 pmol of each IL-1RII primer (forward primer, 5'-TCC ATG TGC AAA TCC TCT CTT-3'; reverse primer, 5'-TCC TGC CGT TCA TCT CAT ACC-3'; amplimer size, 576 base pairs [bp]). To quantify the PCR products comparatively and to confirm the integrity of the RNA, we coamplified a housekeeping gene, glyceraldehyde-phosphate dehydrogenase (GAPDH), in a companion tube (forward primer, 5'-TGA TGA CAT CAA GAA GGT GGT GAA G-3'; reverse primer, 5'-TCC TTG GAG GCC ATG TGG GCC AT-3'; amplimer size, 240 bp). Amplification of IL-1RII was achieved with 30 cycles of 1 min of denaturation at 95°C, 1 min of annealing at 60°C, and 1 min of primer extension at 72°C. Amplification of GAPDH was achieved with 30 cycles of 30 sec of denaturation at 95°C, 30 sec of annealing at 60°C, and 1 min of primer extension at 72°C. These optimal conditions were determined following linearity tests using 1, 2, 5, and 10 µl of the RT reaction volume and 25, 30, and 35 amplification cycles. Amplification of genomic DNA with these primers did not produce a signal, suggesting that the amplification sites crossed at least one intron/exon boundary.
Twenty percent of the PCR volume was then analyzed on a 1% (w/v) agarose gel in the presence of ethidium bromide and transferred to Qiabrane Nylon Plus membranes (Qiagen, Santa Clarita, CA). Membranes were dehydrated at 37°C for 30 min, prehybridized with a hybridization buffer comprised of 5x SSC, 5x Denhardt solution, 50 mM NaH2PO4, 0.5% SDS, 200 µg/ml of salmon sperm DNA, and 50% (v/v) formamide; hybridized in the same buffer, but without Denhardt solution and with 32P-labeled IL-1RII or GAPDH cDNA; and washed with 1x SSC, 0.2x SSC, and 0.1% (w/v) SDS, respectively, before being exposed to x-ray film (Eastman Kodak) for approximately 1 h.
Specificity of the amplification process was verified by Southern blot hybridization. A negative control (PCR in the absence of cDNA) as well as a positive control (cDNA preparation from human endometrial tissue expressing IL-1RII) were included in each series of IL-1RII or GAPDH amplification.
For each endometrial biopsy, PCR was performed three times. The quantity of the PCR products was determined by densitometric analysis of the intensity of the hybridization signal. The relative level of IL-1RII mRNA normalized to GAPDH mRNA was calculated, and the results were expressed as a percentage of the control value (positive control).
Cell Separation
Endometrial tissue was minced into small pieces and dissociated with collagenase (Sigma) to separate epithelial glands from fibroblast-like cells as previously reported [26]. These two cell populations were further purified using Percoll density gradients (Amersham Pharmacia Biotech, Inc.). The purity of epithelial or fibroblast-like stromal cells was verified morphologically; immunocytochemically on coverslip cultures using antibodies specific to cytokeratins (epithelial cell marker), vimentin (stromal cell marker), smooth muscle -actin, and factor VIII (endothelial cell marker); and by flow cytometry for the presence of leukocytes using anti-CD45 monoclonal antibodies as previously described [26]. Cells were kept at -80°C until use.
Statistical Analyses
The IL-1RII immunostaining scores followed an ordinal scale. Therefore, statistical analyses were performed using nonparametric methods. The association between immunostaining scores and the periods of IL-1RII expression in the menstrual cycle as well as intergroup comparisons of immunostaining scores were analyzed using the Fisher exact test, and the Bonferroni procedure was applied when more than two groups were compared. The correlation between the day of menstrual cycle and the immunohistochemistry scores was evaluated using the Spearman correlation coefficient. The threshold days between the different levels of staining (0, 1, 2, and 3) were determined using the best combination of sensitivity and specificity values for a series of cut-off days within the menstrual cycle. These threshold days allowed us to define or to delimit different expression periods. Analysis of IL-1RII mRNA levels as determined by semiquantitative RT-PCR was performed using one-way ANOVA and the Tukey test for post-hoc multiple comparisons. All analyses were performed using Statistical Analysis System (SAS Institute, Inc., Cary, NC). Differences were considered to be statistically significant at P < 0.05.
RESULTS
Immunohistochemistry
A monoclonal antibody was used to detect IL-1RII protein in endometrial tissue sections. Different concentrations of the antibody (5, 10, 15, and 20 µg/ml) were tested to determine the optimal concentration to use (data not shown). This experiment was performed on three different series of biopsy specimens from different phases of the menstrual cycle. A concentration of 15 µg/ml was selected, because it allowed an optimal ratio of specificity (minimal background) and sensitivity (detectable positive signal). Examples of positive immunostaining with anti-IL-1RII antibody are shown in Figure 1A. Immunoglobulins of the same isotype and species, when used at an equivalent concentration as that of the antibody (15 µg/ml), did not display any detectable immunoreactivity (Fig. 1B). Immunoreactive IL-1RII was detectable throughout endometrial tissue, both in the stroma and the glands. Brown immunostaining could be seen around cells (cellular staining) and along the apical border of luminal and glandular epithelium (luminal staining). This was also observed, although less markedly, in microvessels and isolated aggregates throughout the stroma in sections from late secretory endometrial tissues (Fig. 1A). It is noteworthy that luminal secretion was not uniform for all endometrial sections examined.
|
The IL-1RII immunostaining was assessed semiquantitatively by two independent observers in a double-blind manner, taking into account the intensity as well as the distribution of the immunostaining as described above. Cellular and extracellular staining were scored independently in the stroma and in the glands and surface epithelium. Score distributions according to the day of the menstrual cycle are shown graphically in Figure 2.
|
To better understand IL-1RII cyclic variations, expression was defined throughout the menstrual cycle using the best combination of sensitivity and specificity values for different cut-off days within the cycle. Our analysis revealed that after Day 8, both cellular (sensitivity = 100.0%, specificity = 94.9%, P < 0.001) and luminal (sensitivity = 55.6%, specificity = 100.0%, P < 0.001) immunostaining became significantly detectable in epithelial cells. A further increase in the intensity of staining occurred after Days 21 and 22, but this increase was more obvious at the cellular (Day 21, sensitivity = 88.6%, specificity = 71.4%, P < 0.01) than at the luminal (Day 22, sensitivity = 86.1%, specificity = 50.0%, P = 0.07) level. In stromal cells, cellular staining remained weak to absent throughout the whole proliferative phase of the menstrual cycle, but it increased significantly after Day 15 at the beginning of the secretory phase (P < 0.001). Extracellular staining was virtually undetectable in the stroma, except weakly after Day 21 in tissues from mid to late secretory endometria (sensitivity = 91.2%, specificity = 85.7%, P < 0.001).
Statistical analysis of the data, using the Spearman correlation coefficient, revealed a significant, positive correlation between cellular staining scores and day of the menstrual cycle, both in epithelial (R = 0.59, P < 0.001) and stromal (R = 0.46, P < 0.01) cells. However, no positive correlation between epithelial luminal staining and day of the menstrual cycle was found (R = 0.17, P = 0.29), most probably because of a more fluctuating expression pattern. To better delineate these relationships, we determined the mean values of immunostaining scores for each cycle day, and we found that they follow a third-order polynomial curve (Y = A + BX + CX2 + DX3) (Fig. 2). This curve shows that after a maximal increase at approximately Day 12 in the proliferative phase of the menstrual cycle, luminal staining of IL-1RII declined gradually in the secretory phase, reaching its minimal level at approximately Day 22 before increasing again progressively until the end of the cycle. The midsecretory drop in the intensity of IL-1RII luminal immunostaining (Days 19–22) was statistically significant as compared to late proliferative/early secretory (Days 9–18; P < 0.05) and late secretory (Days 23–28; P < 0.01) immunostaining levels.
Representative examples of IL-1RII immunostaining in the endometrium throughout the menstrual cycle are shown in Figure 1 for Days 6 (Fig. 1C), 13 (Fig. 1D), 18 (Fig. 1E), and 23 (Fig. 1F). Note the reduction of IL-1RII luminal secretion in the glands and surface epithelium of specimens at Day 18 as compared to Days 13 and 23.
Western Blot Analysis
To further examine IL-1RII protein expression throughout the menstrual cycle, total endometrial proteins were extracted and equivalent amounts were subjected to Western blot analysis. Our results showed that the antibody specifically recognized several major and minor bands (Fig. 3). From these, the 68- and 45-kDa bands are consistent with the commonly reported molecular weights of the membrane-bound and the soluble receptors, respectively. The immunoreactive bands were absent when the primary mouse monoclonal anti-IL-1RII antibody was replaced by an equal concentration of normal mouse IgGs (Fig. 3A). Low molecular weight bands (<45 kDa) disappeared when the antibody was preabsorbed with an excess of recombinant IL-1RII (20 µg/ml) before being incubated with nitrocellulose membrane-transferred proteins, whereas the intensity of major higher-molecular-weight bands was considerably reduced (Fig. 3B), suggesting specific interaction with the anti-IL-1RII antibody. As shown in Figure 3A, all IL-1RII bands revealed by the antibody were markedly intense at the approach of ovulation. The intensity of these bands clearly decreased during the midsecretory phase but increased again thereafter in tissues from late secretory endometria.
|
In Situ Hybridization
Expression of IL-1RII mRNA in the endometrium was first studied by in situ hybridization to localize the site of synthesis. Figure 4 shows the appearance of endometrial glands and stroma at 167x (A1, B1, and C1) and 666x (A2, B2, and C2) magnifications following hybridization and staining with propidium iodine (late proliferative endometrial tissue, Day 13). The hybridization signal (green-yellow) could only be visualized at higher magnification (1665x), as illustrated in the same figure, and was more pronounced in glandular (A3) and surface (B3) epithelial than in stromal (C3) cells. No hybridization was observed in negative controls including the omission of biotinylated DNA probes prepared from the plasmid containing IL-1RII cDNA insert or the use of biotinylated DNA probes obtained from the plasmid alone (data not shown).
|
Reverse Transcription-Polymerase Chain Reaction
Expression of IL-1RII mRNA throughout the menstrual cycle was studied by semiquantitative RT-PCR. This was achieved by normalizing the IL-1RII mRNA to the mRNA of the coamplified housekeeping gene GAPDH and by including an equal amount of the same preparation of positive control (RT preparation of cDNA from human endometrial RNA) in every series of amplifications. The control, which was subjected to the same experimental conditions from the amplification reaction until Southern blot analysis and autoradiography, allowed for monitoring of the interassay variation. Results were expressed as a percentage of the control value (i.e., the amount of IL-1RII mRNA relative to that of the corresponding GAPDH divided by the amount of IL-1RII mRNA relative to that of GAPDH in the control x 100). Results from 20 endometrial biopsies across the menstrual cycle (Fig. 5A) show that IL-1RII mRNA levels are low in early proliferative endometrial tissues and follow a kinetics of expression comparable to that found for the protein by immunohistochemistry (luminal secretion) and Western blot analysis. In fact, mRNA levels were elevated in late proliferative, early secretory, and late secretory endometria, but they significantly decreased in tissues from the midsecretory phase. Representative examples of RT-PCR and Southern blot analyses of IL-1RII mRNA in tissues from different cycle phases are shown in Figure 5B. The RT-PCR analysis of IL-1RII mRNA in separated cells showing higher levels of expression in epithelial than in stromal cells confirmed the in situ hybridization data (Fig. 5C).
|
DISCUSSION
Human endometrium is an active site of cytokine production and action. Complex interactions of epithelial, stromal, endothelial, and lymphoid cells occurring in the human endometrium as well as dynamic changes orchestrated in cyclic events of cell proliferation, differentiation, and shedding require a well-elaborated network of intercellular communication signals such as cytokines. Many of these events are reminiscent of those associated with the inflammatory and the reparative processes, which make plausible the involvement of proinflammatory cytokines. Furthermore, cytokines such as IL-1, which have been considered as important immune mediators, also seem to be implicated in critical reproductive events such as ovulation and implantation, and evidence indicates that they act as endocrine and local regulators of many endometrial functions [13, 27].
The control of cytokine action in the endometrium may require the local availability of specific regulatory mechanisms. For IL-1, this is illustrated by local expression of IL-1ra, a specific inhibitor that binds to functional IL-1 receptor type I (i.e., IL-1RI) and that blocks IL-1 binding and cell signaling [20]. Our present study revealed the expression in the endometrium of another natural inhibitor for IL-1, the decoy IL-1RII, which acts differently, by sequestrating active as well as inactive IL-1ß and, thereby, restricting the availability of the ligand for the functional receptor and inhibiting even its maturation [11].
Immunoreactive IL-1RII was localized throughout endometrial tissue both in epithelial and stromal compartments, but it was more obvious in endometrial glands and surface epithelium. This was confirmed by in situ hybridization showing high levels of IL-1RII mRNA in endometrial glands and surface epithelium. The RT-PCR of IL-1RII mRNA expression in separate glandular epithelial and fibroblast-like cells isolated from endometrial tissue corroborated these findings. Immunolocalization further revealed two levels of staining. The first, which was located all around the cells, may correspond to the cellular membrane-bound IL-1RII receptor. The second, which was more intense and located predominantly within the lumen of endometrial glands and at the apical side of surface epithelial cells, is most probably a luminal secretion corresponding to the soluble and secreted form of the receptor (sIL-1RII). In fact, Western blot analysis of IL-1RII protein in endometrial tissue shows the presence of a 68-kDa band, which corresponds to the membrane-bound IL-1RII [9, 20], and a number of lower molecular weight bands, which may correspond to different soluble forms. It is well known that sIL-1RII is released from the membrane-bound receptor following proteolysis, and that the released soluble molecules keep their ability to bind and neutralize IL-1, particularly the circulating form (IL-1ß) [9, 11, 12, 28]. To our knowledge, a full characterization of such molecules has not been performed, but it could be proposed that other forms of soluble IL-1RII could exist, possibly due to the presence of different, unidentified cleavage sites. Neumann et al. [29] reported a 34-kDa soluble form of IL-1RII in the culture medium of IL-1RII-transfected keratinocytes. According to Orlando et al. [28], lipopolysaccharide-treated monocytes release a 60-kDa IL-1-binding molecule, which was identified as the IL-1 decoy receptor.
Both cellular and extracellular/luminal IL-1RII immunoreactivity varied within the menstrual cycle and followed a regulated cycle phase-dependent pattern. In stromal cells, the intensity of staining remained weak to absent during the whole proliferative phase but increased moderately after ovulation. In epithelial cells, the regulation appeared to be different, as immunoreactive cellular IL-1RII first increased moderately after Day 8 in the midproliferative phase, then more markedly after Day 21 in the secretory phase. The IL-1RII luminal secretion followed a more complex kinetics. Being undetectable until Day 8 of the cycle, luminal secretion reached a maximum at the end of the proliferative phase, then declined progressively until Day 22 before undertaking a final, gradual augmentation during the late secretory phase.
The pattern of IL-1RII immunostaining in endometrial epithelial cells was quite unusual, and the temporal expression for this natural inhibitor of IL-1 is rather interesting. That the receptor expression is down-regulated in the midsecretory phase, especially during the implantation window, and up-regulated at the end of the menstrual cycle suggests that IL-1RII may have multiple functions in human endometrium. Western blot analysis appears to confirm the results of immunohistochemical staining showing that endometrial epithelial cells possess low amounts of IL-1RII in the midluteal phase. Epithelial IL-1RII expression and distribution in endometrial tissue across the menstrual cycle is remarkable for several reasons.
First, IL-1RII appeared to be expressed and released predominantly by glandular and luminal epithelial cells rather than by stromal cells. This is of additional interest, because first interactions between the embryo and the endometrium occur at the level of the luminal epithelial cells during the adhesion process.
Second, the significant decline in IL-1RII luminal staining, which started at the beginning of the secretory phase and reached its minimum at approximately Day 22, in the midsecretory phase, is rather interesting and suggests that IL-1RII secretion is subjected to subtle chronological regulation in endometrial epithelial cells. This is remarkable, because the phase of reduced secretion correlates with a putative "implantation window" thought to exist within the midsecretory phase between Days 18 and 22 [30]. It is important to point out that the physiologic basis for this window has not yet been clearly established. Also, no general agreement exists regarding the dates of the endometrial receptivity period or the implantation window during normal menstrual cycles [30]. According to Simon et al. [31], the implantation process starts at Day LH + 5, which is consistent with IL-1RII decreased expression. According to Bergh and Navot [32], first embryonic signal detection (presumed window of implantation) extends between Days 20 and 24. However, others have suggested that the implantation window is confined to postovulatory Days 5–7 [33]. Nonetheless, the decreased IL-1RII expression at that specific time of the cycle, at which embryonic attachment and implantation may occur, is suggestive of a possible role for IL-RII in the initial interactions between maternal and embryonic cells and the establishment of an endometrial period of receptivity.
In fact, numerous studies indicate that IL-1 acts as an embryonic signal and is secreted early by embryonic cells [14, 15]. Interleukin 1 appears to be crucial for successful implantation, because the blockade of its functional receptor (i.e., IL-1RI) in vivo prevents implantation by interfering with embryonic attachment [19, 34]. Interestingly, the IL-1 system appears to mediate the regulation of integrin expression in human endometrium. Simon et al. have reported that binding of embryonic IL-1 and IL-1ß to endometrial epithelial IL-1RI up-regulates endometrial epithelial ß3 subunit, which is considered to be a marker of endometrial receptivity [35], and facilitates adhesion of human blastocyst [36]. Serum IL-1 levels increase at ovulation [37]; the down-regulation of IL-1RII release in endometrial glands and surface epithelium suggests, therefore, the existence of accurate regulatory mechanisms that alleviate IL-1 inhibition during this crucial period, thereby facilitating IL-1 proimplantation actions. Interestingly, our previous studies have shown that the expression of IL-1RI, the functional and signaling receptor for IL-1, is moderately up-regulated during the same period [18]. This a remarkable example of fine-tuned and well-orchestrated endometrial events, the object of which is to preserve the reproductive function of this tissue.
The decreased expression of IL-1RII protein, which was more perceptible at the level of luminal secretion by immunohistochemistry, was detected by Western blot analysis and semiquantitative RT-PCR. This makes an inhibition at the level of IL-1RII mRNA more likely than a translational or posttranslational proteolysis-dependent mechanism. Whether this is due to inhibition of IL-1RII gene transcription or to decreased IL-1RII mRNA stability remains to be further elucidated. However, this requires the identification of the regulatory mechanisms underlying IL-1RII down-regulation.
Third, IL-1RII expression in endometrial tissue markedly increased in the late secretory phase of the menstrual cycle, both at the level of the protein and of the mRNA. This was quite obvious in the glands and luminal epithelium, but it was also visible in the stroma of some late secretory-phase biopsy specimens. This may have a considerable significance, because in the absence of implantation, the endometrial tissue undergoes a process of cell necrosis and disintegration at the end of the menstrual cycle [13]. The elevated expression of IL-1RII observed in the late secretory phase may, therefore, play a key role in the control of such an inflammatory-like process during the premenstrual and menstrual periods.
The potential involvement of IL-1RII in implantation and regulation of local inflammatory-like processes is supported by our recent data showing a marked deficiency in the expression of this specific inhibitor of IL-1 in the endometrium of women with endometriosis, especially those suffering from infertility [38].
Another natural inhibitor for IL-1, IL-1ra, has been found in the endometrium, but more in the proliferative than in the secretory phase of the menstrual cycle. Furthermore, according to Simon et al. [21], endometrial cells rather express the icIL-1ra. In view of the localization and the temporal pattern of expression of IL-1RII revealed in our study, this is quite interesting, and it suggests that IL-1RII and IL-1ra have complementary roles in the control of IL-1 action in endometrial tissues, and that they exert their inhibitory effects at different and complementary levels. This is all the more plausible because, according to Symons et al. [11], IL-1RII does not interfere with IL-1ra-mediated inhibition of IL-1, and the two molecules have an additive effect in inhibiting the binding of IL-1ß to cell-surface IL-1 receptors.
The process whereby IL-1RII expression is regulated in the endometrium is yet established, and it remains unknown at present whether estradiol, progesterone, or other hormones of the reproductive cycle can directly or indirectly affect IL-1RII expression in endometrial tissue. On the other hand, it is quite possible that proinflammatory cytokines such as IL-1 and tumor necrosis factor (TNF), which have been found to be predominant in the endometrium during the late secretory phase [16, 39], can be involved in the up-regulation of IL-1RII expression. In fact, both IL-1 and TNF have been shown to up-regulate IL-1RII expression or release in other types of cells [40–42].
In summary, IL-1RII expression in the endometrium is an interesting and dynamic process. It is tempting to hypothesize that aberrant expression of such a natural inhibitor of IL-1 may be related or associated with pathologic states or infertility. Additionally, IL-1RII appears to be predictable, based on the time in the menstrual cycle, thereby allowing diagnostic use of IL-1RII as a stage-specific marker in this tissue. Finally, an understanding of the function of IL-1RII throughout the menstrual cycle and the means to regulate its expression may prove to be of future therapeutic use.
ACKNOWLEDGMENTS
The authors wish to thank Dr. Yves Labelle from the Unité de Génétique Humaine et Moléculaire, Centre de Recherche, Hôpital Saint-François d'Assise, Centre Hospitalier Universitaire de Québec, Université Laval, Québec, Québec, Canada, for his precious help in RT-PCR analyses; the group of investigation in gynecology (Drs. François Belhumeur, Jacques Bergeron, Jean Blanchet, Marc Bureau, Simon Carrier, Elphège Cyr, Jean-Louis Dubé, Jean-Yves Fontaine, Céline Huot, Johanne Hurtubise, Philippe Laberge, André Lemay, Rodolphe Maheux, Jacques Mailloux, and Marc Villeneuve) for patient evaluation and endometrial biopsies; and Madeleine Desaulniers, Christine Jolicoeur, Monique Longpré, Johanne Pelletier, and Sylvie Pleau for technical assistance. A.A. is a Chercheur-Boursier Senior of the Fonds de la Recherche en Santé du Québec.
FOOTNOTES
First decision: 5 December 2000.
1 Supported by grant MT-14638 to A.A. from the Medical Research Council of Canada. 
2 Correspondence: Ali Akoum, Laboratoire d'Endocrinologie de la Reproduction,
Centre de Recherche, Hôpital Saint-François d'Assise, 10 Rue de
l'Espinay, Local D0-711, Québec, PQ, Canada G1L 3L5. FAX: 418 525 4195; ali.akoum{at}crsfa.ulaval.ca
Accepted: April 24, 2001.
Received: November 1, 2000.
REFERENCES
in human endometrium. Hum Reprod 1995; 10:277-286This article has been cited by other articles:
|
|
 |
|
  S Guay and A Akoum Stable inhibition of interleukin 1 receptor type II in Ishikawa cells augments secretion of matrix metalloproteinases: possible role in endometriosis pathophysiology Reproduction, September 1, 2007; 134(3): 525 - 534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Achache and A. Revel Endometrial receptivity markers, the journey to successful embryo implantation Hum. Reprod. Update, November 1, 2006; 12(6): 731 - 746. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W.G. Cao, M. Morin, V. Sengers, C. Metz, T. Roger, R. Maheux, and A. Akoum Tumour necrosis factor-{alpha} up-regulates macrophage migration inhibitory factor expression in endometrial stromal cells via the nuclear transcription factor NF-{kappa}B Hum. Reprod., February 1, 2006; 21(2): 421 - 428. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Kats, M. Al-Akoum, S. Guay, C. Metz, and A. Akoum Cycle-dependent expression of macrophage migration inhibitory factor in the human endometrium Hum. Reprod., December 1, 2005; 20(12): 3518 - 3525. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rossi, A. M Sharkey, P. Vigano, G. Fiore, R. Furlong, P. Florio, G. Ambrosini, S. K Smith, and F. Petraglia Identification of genes regulated by interleukin-1{beta} in human endometrial stromal cells Reproduction, November 1, 2005; 130(5): 721 - 729. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
