| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
(1-2)Fucosyltransferase Messenger Ribonucleic Acid in the Mouse Uterus1
a School of Biological Sciences, University of Manchester, Manchester, M13 9PT, United Kingdom
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(1–2)fucosylated carbohydrate structure, has been implicated in initial attachment of the murine blastocyst to luminal uterine epithelial cells in vitro. In this study, the expression of the H-type (1–2)fucosyltransferase (FUT1) gene was examined in endometrium of mice. Northern blotting of luminal epithelial RNA identified a single 6.2-kilobase transcript. In situ hybridization studies showed a signal for FUT1 mRNA on Days 1–3 of pregnancy in glands and luminal epithelium. The signal diminished by Day 4 and could not be detected on Day 5 of pregnancy. The in situ signal in endometrial epithelia was highest at estrus and metestrus and was absent at diestrus. Estrogen treatment after ovariectomy gave strong FUT1 mRNA expression in epithelia, but with progesterone, progesterone + estrogen, or vehicle, no message could be detected. A semiquantitative reverse transcription-polymerase chain reaction (PCR) analysis of FUT1 mRNA from luminal epithelium generated large amounts of PCR product on Day 1 of pregnancy; this diminished on Days 2, 3, and 4, and the product was barely detectable on Day 5. A kinetic analysis of FUT1 activity on Day 1 of pregnancy suggested a single enzyme with a Michaelis-Menten constant (Km) of 0.29 mM towards phenyl-ß-D-galactoside and of 1.75 mM towards Galß(1–3)GalNAc. These results suggest that expression of the H epitope is regulated at the level of FUT1 transcription and that transcription is stimulated by estrogen in the endometrial epithelium.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Many different molecules have been implicated in the regulation and execution of the cell biological changes occurring during attachment and adhesion at implantation. Among the possible nonsteroidal regulating factors are cytokines (e.g., leukemia inhibitory factor [LIF], interleukin [IL]-1, IL-6, and colony-stimulating factor [7–10] and growth factors (e.g., the epidermal growth factor [EGF] family: EGF, transforming growth factor- [TGF-], amphiregulin, and heparin-binding EGF-like growth factor [11–15]. Structural molecules of importance in the interactions between trophoblast and uterine cells include integrins [16–18] and glycoconjugates [19–21]. The precise nature and order of molecular interactions, especially for initial attachment, remain unclear.
Likely candidates for the initial attachment reaction are cell-surface glycoconjugates. They extend far above the apical surface of cells and exhibit changes in their expression on the luminal endometrial epithelium at the time of implantation. Surface carbohydrates have also been shown to be involved in other cell adhesion processes [22], such as selectin-ligand binding of leukocytes [23, 24], attachment of viruses and bacteria [25, 26], and sperm-egg interactions [27, 28]. Furthermore, glycosylation in uterine cells has been shown to be regulated by ovarian steroids [29].
The H-type-1 carbohydrate epitope (Fuc(1–2)Galß(1–3)GlcNAc-) has been implicated in initial blastocyst-epithelial attachment in mice [30, 31]. A monoclonal antibody (mAb/6679E9; Monocarb, Lund, Sweden) that recognizes the H-type-1 epitope (although it may cross-react with H-type-2 in some contexts) was found to stain the apical membranes of the luminal uterine epithelium as well as glands during early pregnancy and up to the time of implantation [19]. In an in vitro model, lacto-N-fucopentaose-1 (LNF-1), a pentasaccharide that contains the H-type-1 epitope, was found to specifically inhibit the attachment of the trophoblast to luminal uterine epithelial cell monolayers, while other related oligosaccharides had no effect [30]. One possible explanation of these results is that soluble H-type-1 structures may competitively inhibit binding between H-type structures on the luminal epithelium and a trophectoderm receptor. Indeed, a neoglycoprotein carrying only LNF-1 carbohydrate moieties and labeled with fluorescein isothiocyanate (FITC) stained the abembryonic trophectoderm only, suggesting the presence of a receptor for the H-type-1 epitope [32–34]. From this it was proposed that initial blastocyst attachment may be dependent on the interactions between a receptor on the trophectoderm and the H-type-1 epitope on the luminal epithelium.
Expression of this epitope may reflect transcriptional or posttranscriptional control of glycoproteins carrying it, but one level of control could be that of expression of (1–2)fucosyltransferases (FUT). These enzymes control the final step in the formation of the H-type epitopes, by the addition of terminal fucose in (1–2) linkage. Alpha(1–2)FUT activity in luminal uterine epithelium has been shown to be maximal at Day 1 of pregnancy and to drop through Days 2–6 of pregnancy until it is undetectable. This correlates with H epitope expression [35]. Both (1–2)FUT activity and maximal H epitope expression in luminal epithelium correlate with the estrogen surge at estrus. That estrogen stimulates enzyme activity has been confirmed using ovariectomized mice treated with ovarian steroids [20, 35]. From these experiments, it is thought that (1–2)FUT in the luminal epithelium is a major factor responding to hormonal changes and is responsible for controlling H epitope expression here.
In order to determine whether the (1–2)FUT enzyme is controlled by the level of mRNA or by translational or other posttranscriptional means, we have examined relative H-type (1–2)fucosyltransferase (FUT1) mRNA expression in the murine uterus through Days 1–5 of pregnancy and the estrous cycle, and in ovariectomized mice treated with ovarian steroids, by in situ hybridization, reverse transcription (RT)-polymerase chain reaction (PCR), and Northern blotting. A kinetic analysis of FUT1 enzyme activity was also carried out on Day 1 of pregnancy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Mature female MF-1 mice (6 wk old) obtained from OLAC (Bicester, UK) were used throughout this study. Mice were kept under a controlled photoperiod of 12L:12D, and food and water were provided ad libitum. This work was authorized and carried out under a UK Home Office project license.
Early pregnancy Mature females were mated with males of their own strain, and the day of vaginal plug formation was considered Day 1 of pregnancy. Mice were killed on Days 1–6 of pregnancy at midday by cervical dislocation.
Estrous cycle Female mice were checked by vaginal smears to determine their stage of estrus. This was done by dipping a flamed pipette tip into a 0.9% NaCl solution and dropping the liquid into the vagina, then retracting it back into the pipette. The fluid was transferred to a slide and stained with methylene blue dye. Staging was carried out by analyzing the cell types obtained from the smear [36]. Mice were checked at 1000 h and killed at midday. Three to four mice at each stage of the estrous cycle were used.
Ovariectomy and hormonal replacement Ovariectomy was performed before noon under Hypnorm (Janssen, Beerse, Belgium) anesthetic. Eleven days were allowed for endogenous hormones to subside, and mice were then given daily s.c. priming injections of 100 ng of estradiol benzoate (E2) for 3 days. All hormone injections were given at 1600 h in a 0.1-ml volume of corn oil. After a 2-day wait, the mice were started on the following injection regime. Injections were given daily for 4 days: a) controls, corn oil only; b) 100 ng E2; c) 500 µg P4; d) Days 1–3, 500 µg P4, and Day 4, 500 µg P4 and 10 ng E2. Mice were killed 18 h after the final injection.
Preparation of Murine Uterine Epithelial Cells for RNA Extraction or Cell Lysates for (1–2)FUT Enzyme Assay
The method used for isolating the lining epithelium from the uterine horns was adapted from the method of Bigsby and colleagues [37]. Three to four mice were used for each preparation. Mice were killed, and their uteri were dissected out and placed into Hanks' Balanced Salt Solution (HBSS; Gibco BRL, Life Technologies, Paisley, UK). Horns were cut into 4-mm lengths and incubated in HBSS (Gibco BRL) containing 0.5% dispase (Boehringer, East Sussex, UK) for 2 h at 25°C. After incubation, the tissue was placed into HBSS containing 0.4% BSA (Sigma, Dorset, UK), and the sections were gently squeezed, using a bent syringe needle and forceps, to release the lining of uterine epithelial cells as a tube. Cells were pooled and washed with HBSS and then with sterile H2O. After each wash, cells were spun at 1000 x g for 4 min.
Cells for RNA extraction After the final wash, the pelleted cells were transferred to a cryotube, weighed, and stored in liquid nitrogen until required for RNA extraction.
Cell supernatant for (1–2)FUT assay
Epithelial cells that were to be used for the (1–2)FUT assay were given a further wash in sterile H2O and spun at 1000 x g for 4 min. The cell pellet was resuspended in 3 volumes of 0.25 M sucrose, sonicated for 2 x 15-sec bursts (W385 Ultrasonic Processor; Heat Systems Ultrasonics Inc., Farmingdale, NY) at 60% power, and then centrifuged at 2500 x g for 10 min. The supernatant was collected and stored at -80°C until required for the (1–2)FUT assays.
RNA Extraction
Total RNA was extracted from uterine epithelial cells taken from mice on Days 1–5 of pregnancy using the QIAGEN RNeasy total RNA kit (Qiagen, Surrey, UK). RNA was quantified by measuring the absorbance at 260 nm in a UV spectrophotometer (GeneQuant DNA/RNA Calculator; Pharmacia, Herts, UK). Purity of RNA was assessed by measuring the ratio of the absorbance at 260 nm to that at 280 nm.
Northern Analysis
All reagents used in the Northern analysis were purchased from Boehringer, unless otherwise stated. Digoxigenin (DIG)-labeled riboprobes (antisense or sense) were generated from a 650-base pair (bp) fragment of the mouse FUT1 cDNA (gift from Drs. Hiraiwa and Lowe, Howard Hughes Medical Centre, Ann Arbor, MI), which was subcloned into high-expression vector pGEM4Z (Promega, Southampton, UK).
Twenty micrograms of RNA was denatured and electrophoresed on a 1% agarose, 6.6% formaldehyde gel. RNA was transferred via capillary action to positively charged nylon membranes and fixed by UV cross-linking (UV crosslinker RPN2500; Amersham, Bucks, UK). RNA molecular weight markers type 1 (0.3–7.4 kilobases [kb]; Boehringer) were used to assess the band size. Northern blots were hybridized with DIG-labeled FUT1 antisense or sense (negative control) riboprobes in Easyhyb (Boehringer, Mannheim, Germany) hybridization buffer at 65°C overnight. Blots were washed under stringent conditions with double-strength SSC (single-strength SSC is 0.15 M sodium chloride, 0.015 M sodium citrate)/0.1% SDS, 2 x 5 min at room temperature followed by 0.1-strength SSC/0.1% SDS, 3 x 15 min at 68°C. The membranes were then rinsed in buffer 1 (150 mM NaCl/100 mM maleic acid, pH 7.5) and blocked in buffer 2 (1% [w:v] solution of Boehringer blocking powder in buffer 1) for 1 h at room temperature. Detection was carried out with an anti-DIG antibody conjugated to alkaline phosphatase (1 in 20 000) in buffer 2 for 30 min at room temperature. The antibody was visualized by using a chemiluminescent reagent CDPstar (Boehringer Mannheim) (1 in 100) in buffer 3 (100 mM NaCl/50 mM MgCl2/100 mM Tris HCl, pH 7.5) and exposing the blot to x-ray film (Hyperfilm MP; Amersham).
Semiquantitative RT-PCR
Semiquantitative RT-PCR was used to examine relative changes in FUT1 mRNA during Days 1–5 of pregnancy. Changes in PCR product obtained for FUT-1 were compared to those for an endogenous gene (an internal standard) [38–40], glyceraldehyde-3-phosphate dehydrogenase (GAPDH), expression levels of which remain fairly constant in the uterus [41]. Although exact copy numbers for the target mRNA cannot be established, relative changes in mRNA levels can be determined.
To minimize differences in RNA yield, an equal concentration of RNA (2 µg total RNA quantified using an UV spectrophotometer) was used for each cDNA synthesis reaction. To further minimize differences between samples, master mixes were made up for both the RT and PCR steps.
First-strand cDNA synthesis All reagents used in the first-strand cDNA synthesis were purchased from Gibco BRL. Moloney Murine Leukemia Virus Reverse Transcriptase (M-MLV RT), in the presence of oligo(dT) primers, was used on single-stranded RNA to synthesize cDNA strands. For first-strand cDNA synthesis, the following mixtures were assembled; 2 µg template RNA, 1 µl oligo(dT)12–18 primer (0.5 µg/µl), and sterile H2O to make a total volume of 10 µl. This mixture was heated to 70°C for 10 min and then chilled on ice. To this mixture were added 4 µl 5-strength first-strand buffer, 2 µl 0.1 M dithiothreitol, 1 µl dNTPs (10 mM each dNTP), 1 µl M-MLV RT (200 U), and 2 µl sterile H2O. Five-strength first-strand buffer contained 250 mM Tris HCl (pH 8.3), 375 mM KCl, and 15 mM MgCl2. This mixture was made up to a total volume of 20 µl and incubated at 37°C for 1 h. This was followed by RT inactivation by heating samples to 95°C for 5 min and then cooling on ice. Two microliters (10%) of each cDNA synthesis reaction were used in subsequent amplification reactions.
To confirm that there was no genomic DNA contamination of the total RNA, the RT-PCR reactions were performed under identical conditions with the exception that the M-MLV RT was omitted from the cDNA synthesis step. This mixture was used as a control in subsequent PCR amplification steps.
PCR Two-microliter samples of the first-strand cDNA mix were added to a mixture containing 5 µl 10-strength PCR buffer (Boehringer), 1 µl Taq DNA polymerase (5 U; Boehringer), 50 pM appropriate primers (Table 1) and sterile H2O to make a total volume of 50 µl. This mixture was incubated in a thermal cycler under the following conditions: denaturation at 95°C for 30 sec, annealing at 62°C for 30 sec, and elongation at 72°C for 1 min, for a total of 30 cycles. Ten-strength PCR buffer contained 100 mM Tris HCl, 500 mM KCl, and 15 mM MgCl2 (pH 8.3).
|
RT-PCR products were examined on a 1.5% (w:v) agarose gel using a 100-bp DNA ladder (Gibco BRL) for size assessment of PCR products. FUT1 RT-PCR products were initially identified by restriction enzyme digestion and then subcloned into a plasmid TA cloning vector (pCR II; Invitrogen, NV Leek, Netherlands) from which they were sequenced by an automated ABI DNA sequencer (ABI Analytical, Ramsey, NJ).
In Situ Hybridization
A standard nonradioactive in situ protocol [42] was used to examine 4% paraformaldehyde (PFA; Sigma) fixed, wax-embedded sections of mouse uterus taken from Days 1 to 6 pregnancy, from the estrous cycle, and from ovariectomized mice treated with ovarian steroids.
Tissue preparation Uterine tissue taken from mice in early pregnancy and the estrous cycle, and after ovariectomy and hormonal replacement was used in this study. Mice were killed, and their uterine horns were dissected out. Each horn was cut into 3 segments and fixed in 4% PFA in PBS overnight at 4°C.
Uterine horns were then removed from fixative and dehydrated through a series of ethanols: 50%, 70%, and 90%, 1 h each; 100%, 30 min and 1 h. Tissues were placed in a rotator and transferred to chloroform:ethanol (1:1) for 30 min, then 100% chloroform for 30 min and 1 h. Tissue was then placed into a 63°C wax bath under a vacuum for 1 h, then vacuum-embedded in fresh wax for 1 h. The tissue was mounted in molten wax and left to harden at 4°C. Seven-micrometer-thick transverse sections were cut on a rotary microtome (Reichert-Jung, Milton Keynes, UK). Sections were floated out on diethyl pyrocarbonate (DEPC)-treated, warmed H2O to remove wrinkles, and mounted onto slides. Slides were dried overnight at 42°C.
Pretreatments Sections were dewaxed in xylene, 2 x 10 min, and rehydrated through a series of ethanols: 100%, 90%, 70%, and 30%. Slides were incubated in PBS for 2 x 5 min, fixed in 4% PFA in PBS for 20 min, washed with PBS for 2 x 5 min, and then treated with proteinase K (20 µg/ml; Sigma) in Tris-EDTA buffer for 10 min. Slides were washed again in PBS for 2 x 5 min and incubated in 0.1 M triethanolamine HCl (pH 8.0; Sigma) containing fresh acetic anhydride (0.25% v:v; Sigma). Slides were washed again with PBS for 5 min, hydrated back through the ethanols, and air-dried for 10 min.
In situ hybridization DIG dUTP-labeled sense and antisense riboprobes were synthesized by in vitro transcription of the 650-bp cDNA fragment of the FUT1 cDNA in pGEM4Z (Promega). DIG-labeled mouse ß-actin riboprobes were used as a positive control. Sections were hybridized overnight at 65°C with the appropriate DIG-labeled riboprobes at a concentration of 1 µg/ml in hybridization buffer containing 50% formamide, 5-strength SSC (pH 4.5), 1% SDS, 50 µg/ml tRNA, and 50 µg/ml heparin. Immunological detection was carried out with the use of an anti-DIG antibody (Boehringer) conjugated to alkaline phosphatase, which was visualized by a color reaction using nitro blue tetrazolium chloride (NBT; Boehringer) and 5-bromo-4-chloro-3-indolyl-phosphate (BCIP; Boehringer) as substrates to give a purple precipitate.
(1–2)Fucosyltransferase Assay
Two acceptor substrates (phenyl-ß-D-galactoside and Galß(1–3)GalNAc; Dextra Labs, Reading, UK) were used to carry out a kinetic analysis of (1–2)FUT enzyme activity in mouse uterine epithelium on Day 1 of pregnancy. These acceptors were chosen because they function as specific acceptors for (1–2)FUT activity (other FUTs are unable to use them as substrates) and because of their ability to discriminate between FUT1 and secretor type (1–2)FUT (FUT2) in other species [43–47].
Assay conditions FUT assay conditions were based on previous work [35, 43, 45]. Conditions used for standard incubations were as follows: 0.425 nM GDP-[14C]fucose (Amersham; specific activity 10.8 Gbq/mmol), 30 µM GDP-fucose, 0.2 mM ATP, 2 mM MnCl2, 0.2 µl/ml Triton X-100 (Sigma), and 10 mM sodium cacodylate (pH 6.0). To determine apparent Michaelis-Menten constant (Km) values, the acceptor substrates were examined at a range of concentrations from 0.1 mM to 30 mM. Sonicated uterine epithelial samples were measured for protein using the BCA Reagent assay system (Pierce&Warriner, Cheshire, UK), and protein samples of 150 µg/ml were used in each reaction.
Assays were carried out in an Eppendorf tube (Hamburg, Germany) in a total volume of 80 µl and were incubated at 25°C for 4 h, then frozen at -80°C. All assays were performed in triplicate along with parallel control incubations that were identical to the standard incubations except that they lacked any acceptor substrates. This gave an indication of fucose incorporation into other available tissue substrates.
Paper chromatography Ascending paper chromatography was used to separate out the products of the FUT assays. Samples were freeze-dried and then resuspended into a smaller volume of 30 µl of 50% (v:v) ethanol. Samples were then dotted onto Whatman No. 3MM chromatography paper (Whatman, Kent, UK), 4 cm from the bottom on 3-cm-wide sample lanes.
Samples were chromatographed either in solvent A—ethylacetate:pyridine:H2O (10:4:3 by volume) [36, 46]—for phenyl-ß-D-galactoside, or in solvent B—ethanol:pyridine:butan-1-ol:H2O:glacial acetic acid (100:10:10:30:3 by volume) [46]—for the acceptor substrate Galß(1–3)GalNAc. The solvent front was allowed to run to approximately 20 cm (5 h at 20°C) before chromatograms were removed from the chromatography tank and dried.
Each chromatography lane (3 cm wide) was cut into 1- or 2-cm horizontal strips, and these were eluted with 500 µl H2O. Four and one half milliliters of scintillation fluid was added to each vial containing the strip and H2O, and the samples were then briefly vortexed. Radioactivity was then measured using an LKB (Rockville, MD) 1214 Rackbeta scintillation counter and converted to cpm. Sample lanes were cut into strips so as to separate peaks of endogenous activity from the acceptor substrate peaks. Apparent Michaelis-Menten constants (Km) were derived from Lineweaver-Burk plots of substrate concentration versus radioactive incorporation using the SIMFIT statistical analysis program [48].
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Northern analysis of 20 µg RNA extracted from uterine luminal epithelial cells from mice on Day 1 of pregnancy with the 650-bp DIG-labeled FUT1 antisense riboprobes identified a single transcript for FUT1 mRNA of approximately 6.2 kb based on migration of the RNA marker fragments and the 28S ribosomal RNA (Fig. 1). The sense probe did not hybridize to the blot.
|
Semiquantitative RT-PCR
Changes in FUT1 mRNA in luminal epithelium were examined from Days 1 to 5 of pregnancy by producing PCR products and comparing them to an endogenous standard, GAPDH mRNA, expression of which is known to remain fairly constant during pregnancy [38, 41]. Two separate pools of RNA were examined for each day of pregnancy, and each experiment was carried out in duplicate. Comparisons between GAPDH and FUT1 PCR products were made at 30 cycles, before the respective amplification reactions reached the plateau phase. Identity of the product was determined by sequencing.
From this set of experiments, FUT1 cDNA was consistently maximal on Day 1 of pregnancy, and only a very faint band could be detected on Day 5 of pregnancy. Through Days 2, 3, and 4 of pregnancy, FUT1 cDNA was also detected, but it was not possible to discern any consistent, clear differences in the amount of product obtained. However, the PCR product derived from RNA taken from Days 2 to 4 of pregnancy was always less than that on Day 1 of pregnancy but higher than that on Day 5 of pregnancy (Fig. 2a). GAPDH PCR product remained constant through Days 1–5 of pregnancy (Fig. 2b), and no genomic DNA contamination was seen with the control RT-PCR reactions (Fig. 2c).
|
In Situ Hybridization
Pre- and periimplantation period of pregnancy In situ hybridization analysis of FUT1 mRNA levels was carried out on sections of mouse uterus from Days 1 to 6 of pregnancy (Fig. 3, Table 2). The strongest signal for FUT1 mRNA was identified on Day 1 of pregnancy (Fig. 3, a and b). This was specific to the luminal and glandular epithelium; both the epithelia stained uniformly and with equal intensity. No signal was observed using the sense riboprobe (Fig. 3c). On Days 2 and 3 of pregnancy, signal was again localized to the luminal and glandular epithelium but was lower than on Day 1 of pregnancy (not shown). Little or no difference could be detected in FUT1 hybridization signal between Day 2 and Day 3. By Day 4 of pregnancy, however, hardly any signal could be detected in the luminal epithelium. In the glandular epithelium, only a modest signal was identified (Fig. 3e). There appeared to be a subtle difference in the distribution of the FUT1 mRNA among different regions of the uterus. The hybridization signal appeared to be more intense in the glandular epithelium in the anti-mesometrial region of the uterus. By Day 5 of pregnancy, no hybridization signal for FUT1 mRNA could be detected (Fig. 3f), nor was there any detectable signal on Day 6 (not shown). Although an occasional slight color reaction was observed in the stroma, this appeared to be nonspecific precipitation of the color substrate. The ß-actin antisense probe produced a strong hybridization signal in the epithelium and muscle, and a weaker signal in the stroma (Fig. 3d).
|
|
Estrous cycle FUT1 mRNA expression was examined on sections of mouse uterus taken from the 4 different stages of the estrous cycle (Fig. 4, Table 2). Sections of mouse uterus examined from the proestrus stage exhibited slight differences between animals in the signal for FUT1 mRNA. No signal could be detected from sections taken from two of the mice examined, but sections taken from two other mice at this stage showed faint signals in the luminal and glandular epithelium (Fig. 4a). No difference in signal strength was seen between the lumen and the glands.
|
All sections examined from the estrus stage of the estrous cycle exhibited strong hybridization signals for FUT1 mRNA in the luminal and glandular epithelium (Fig. 4b). The signal were uniformly distributed across both glandular and luminal epithelia.
Sections of mouse uterus examined at the metestrus stage of the estrous cycle also showed subtle variations in the signal detected. A signal was always detected in the luminal and glandular epithelium (Fig. 4c), but the intensity varied from mouse to mouse. This could be due to the fact that FUT1 mRNA is decreasing at this stage and subtle differences in the staging of the mice, i.e., early or late metestrus, could account for the changes seen.
No signal for FUT1 mRNA was detected in any of the sections examined from the diestrus stage of the estrous cycle (Fig. 4d).
Hormonal control A series of in situ hybridization studies were carried out on sections of mouse uterus taken from ovariectomized mice treated with E2 only, P4 only, or a combination of P4 and E2. (Fig. 5, Table 2). Under the influence of E2 alone, a strong and even distribution of the signal for FUT1 mRNA could be detected in both the luminal and glandular epithelium (Fig. 5a). However, under the influence of P4 alone, no signal could be detected (Fig. 5b). No signal was seen in sections of uterus taken from control animals (corn oil only: Fig. 5c) or in ovariectomized mice treated with P4 followed by E2 (results not shown).
|
Kinetic Analysis of (1–2)FUT Enzyme Activity in the Murine Endometrial Epithelium
Uterine epithelial extracts from Day 1 pregnant mice were used in these studies because this was the day of maximal (1–2)FUT enzyme activity [35]. Extracts from Day 5 pregnant mice were also examined, but no significant levels of enzyme activity could be measured using this methodology (results not shown).
Determining Km values for phenyl-ß-D-galactoside and Galß(1–3)GalNAc
FUT assays were carried out with varying concentrations of phenyl-ß-D-galactoside or Galß(1–3)GalNAc as acceptor molecules, and results were compared to those of control enzyme assays that lacked the acceptor substrate. From this study, using the SIMFIT statistical analysis package [48], Lineweaver-Burk plots were constructed, and apparent Km values were determined for the acceptor substrates (Fig. 6 and Table 3). With respect to phenyl-ß-D-galactoside and Galß(1–3)GalNAc, apparent Km values of 0.29 mM and 1.75 mM were obtained for (1–2)FUTs from uterine epithelial cell extracts from Day 1 pregnant mice. Statistical analysis of the data (using SIMFIT) also predicted that the kinetic profiles of the enzyme assays best fit the Michaelis-Menten model for a 1:1 reaction, i.e., the presence of only one type of (1–2)FUT enzyme activity in this tissue.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In situ hybridization determined the location of the FUT1 mRNA to be in the endometrial epithelial cells, and FUT1 mRNA expression was observed between Days 1 and 4 of pregnancy, confirming the RT-PCR results. Highest levels were detected on Day 1 of pregnancy, and signal levels fell gradually through Days 2–4 of pregnancy. On Day 4 of pregnancy, a signal was observed in glandular epithelium, but little or no signal could be detected in the luminal epithelium, suggesting slight differences in control of transcription or degradation between these epithelia. By Day 5, no FUT1 mRNA could be detected (Fig. 3).
The changes in early pregnancy suggested regulation by ovarian steroids; therefore, we first examined tissue from different stages of the estrous cycle. In situ hybridization on sections of mouse uterus taken from the different stages of the estrous cycle (Fig. 4) demonstrated that mRNA levels for FUT1 change quite markedly, with a strong correlation with the cycling levels of estrogen. A weak signal for FUT1 mRNA was detected in the luminal and glandular epithelium of uterine sections from mice in proestrus, when estrogen levels start to rise. The highest levels of signal were detected at estrus, when estrogen levels are at their peak. Estrogen levels wane at metestrus, and a variable but reduced level of signal may reflect variation between estrogen levels in different mice: no signal could be identified in tissue taken subsequently at diestrus.
To determine whether relative FUT1 mRNA levels were indeed under the control of E, in situ hybridization studies were carried out on sections of mouse uterus taken from ovariectomized mice treated with ovarian steroid hormones (Fig. 5). This confirmed the estrogenic stimulation of FUT1 mRNA expression: a strong signal was detected in the luminal and glandular epithelium of ovariectomized mice that had been supplemented with E2 only. After treatment with P4 or a combination of P4 and a single injection of estradiol (to mimic nidatory estrogen), or in corn oil-treated controls, no signal was detected.
The results from the in situ hybridization and RT-PCR studies on FUT1 mRNA levels during early pregnancy are in agreement. Both suggest a general trend of falling message levels through early pregnancy. There is a slight discrepancy on Day 5 of pregnancy insofar as RT-PCR analysis identified a very weak band for FUT1, but mRNA was not picked up by in situ hybridization. However, this could be explained by the greater sensitivity of RT-PCR to extremely low levels of transcribed mRNA or perhaps stable mRNA [49]. The data obtained in the present study also agree with previous work that examined changes in (1–2)FUT enzyme activity [35] and H epitope expression [19]. FUT1 mRNA, enzyme activity, and H epitope expression are all maximal on Day 1 of pregnancy and then follow a gradual decline so that by Day 5 of pregnancy, the day after implantation, they can hardly be detected. All three factors—FUT1 mRNA, enzyme activity, and H epitope expression—have also been shown to increase markedly under the influence of estrogen in ovariectomized mice, and during the estrous cycle have maximal expression or activity at estrus, when estrogen levels are at a peak [31, 35]. H-type-1 oligosaccharide has been shown to inhibit attachment of blastocysts to luminal uterine epithelial cells in vitro [30]. However, the expression patterns for, and control of, the H-carbohydrate epitope, the 1–2FUT gene, and the 1–2FUT enzyme activity, indicate that the appearance of this epitope is not responsible for uterine receptivity to blastocyst implantation. However, cessation of (1–2)FUT transcription or mRNA degradation and disappearance of the epitope may mark the onset of the refractory period on Day 5. Other factor(s) such as P4 and nidatory estrogen-dependent unmasking of this epitope may initiate uterine receptivity at the level of the luminal epithelium.
At the time this work was carried out, only one (1–2)FUT had been identified in the mouse, although a second (1–2)FUT gene known as the Secretor (Se) type (FUT2) had been identified in other species [46, 47, 50]. In order to determine whether FUT2, or other as yet unidentified (1–2)FUTs, played a role in the generation of the H epitope during pregnancy, a kinetic analysis of (1–2)FUT enzyme activity in the uterus was carried out with respect to two previously defined specific acceptors for (1–2)FUT activity. Although no kinetic analysis of mouse (1–2)FUTs have been carried out in the past, isoforms of these enzymes are well defined and display distinct differences in their affinity for specific acceptor substrates across species [43, 46, 47, 51, 52].
Kinetic analysis of (1–2)FUT enzyme activity with the acceptor phenyl-ß-D-galactoside showed that it maintains an apparent Km of 0.29 mM. This compares favorably to the Km values obtained for FUT1 rather than FUT2 using phenyl-ß-D-galactoside in other species, especially in similar studies carried out on crude human cell extracts that identified a Km value of 1.4 mM for FUT1, while FUT2 had a Km of 10.0 mM [43]. The Km value of 1.75 mM obtained with the acceptor sugar, Galß(1–3)GalNAc, although not a conclusive result by itself, was closer to the Km value obtained for FUT1 (3.5 mM) than for the FUT2 enzyme (4.6 mM) in rabbits [46].
Both sets of experiments, however, also indicated that only one type of enzyme (FUT1) was acting on the acceptor substrate in the enzyme assays because the data best fit the Michaelis-Menten profiles for a 1:1 reaction. This means that only one factor was affecting the kinetic profile of the enzyme assays, and that was the changing concentrations of the acceptor substrate. If two or more factors, i.e., more than one enzyme, were affecting the assays, a typical log profile would not be obtained (SIMFIT statistical analysis package) [48]. It is important to note that this study does not completely rule out the possibility of some FUT2 activity in the uterus. Some basal activity may exist that might not be detected with the current methodology, but the results from this study indicate that FUT1 mRNA level and enzyme activity is the major regulator of the H epitope.
Northern analysis of RNA extracted from uterine epithelial cells from Day 1 pregnant female mice identified a single transcript at approximately 6.2 kb for the FUT1 mRNA. The size of the transcript was surprisingly large, as the actual coding region is approximately 1.27 kb long. Recently, another study has determined the FUT1 transcript size to be generally 2.8 kb in a variety of mouse tissues by Northern blotting, although they did identify larger but weaker transcripts in the testis, thymus, and pancreas [53]. We propose that what we are seeing is a stage- and tissue-specific transcription of an alternative form of FUT1 in the murine endometrial epithelium that is linked to the hormonal control of FUT1 in this tissue. On the basis of the size of the coding region, the FUT1 transcript we have identified has an unusually large 5' and/or 3' untranslated region, which may play a role in mRNA stability and regulate the rate of translation initiation [49, 54]. The human FUT1 gene also has a large transcript, approximately 4.0 kb (as identified in human cancer cell lines) [55], and RACE (rapid amplification of cDNA ends) analysis of the 5' end of the human FUT1 transcript has indicated that it has two distinct transcription start sites and several forms produced by alternative splicing. Two distinct promoters are linked to both transcription start sites. One of these is thought to allow a basal level of transcription, while the other may allow developmental or tissue-specific (temporal and/or spatial) expression. This form of regulated expression (by distinct promoters and alternative splicing of the 5' ends) has also been reported for other glycosyltransferase genes, such as (2–6)sialyltransferase and (1–3)FUT [55–59]. Since Northern blotting revealed a transcript that was larger than expected from the size of the coding region, it is possible that similar complex regulation also occurs for the murine gene with a distinct hormonally regulated promoter. Previous work has not identified steroid hormone response elements directly associated with glycosyltransferase genes, but it is well known that estrogen has a stimulatory effect on N-linked glycosylation in the uterus by activation of the glycosylation apparatus [29]. However, it is also possible that the increased relative FUT1 mRNA levels we observed after estrogen stimulation may arise indirectly: via stimulation of gene expression by other regulatory molecules such as growth factors and cytokines [10, 11, 14], or through increased mRNA stability.
Changes in FUT1 gene expression in response to hormonal factors may also occur in rat and human endometrium. Immunostaining of rat endometrium indicates that the H-type structures are expressed under the influence of P4 [60]. However, in pregnancy, maximal expression of the H epitope is also seen on the rat luminal epithelium just before the time of implantation. Immunostaining in humans also showed that expression patterns of H-type structures fluctuate in relation to the cycling endometrium [60–63].
Recently, the full cDNA sequence for the mouse FUT1 gene was submitted to GenBank (accession no. U90553) [53], including 5.0 kb of sequence from the 5' end of the noncoding region. By use of the GCG program SIGNAL SCAN (Daresbury, UK), the sequence was analyzed for known transcriptional elements. Hundreds of potential signal sites were identified, but the most significant in relation to this study was the identification of two potential P4 receptor binding sites at 810 and 3767 bases upstream of the coding region of the FUT1 gene, and several potential glucocorticoid receptor sites (see Table 4). This appears to be the first evidence of steroid hormone response elements directly associated with a glycosyltransferase gene. Although it must still be determined whether these sites play a functional role, previous studies that reported (1–2)FUT enzyme activity in relation to ovarian steroids [35] and the in situ hybridization results from this paper suggest that P4 may have an inhibitory effect on FUT1 gene expression. It is therefore possible that liganded P4 receptor binds at the P4 receptor sites and has a direct effect in inhibiting FUT1 gene expression.
|
In summary, FUT1 mRNA levels are strictly regulated in the mouse uterus during early pregnancy and in the estrous cycle. Levels of FUT1 mRNA are increased by estrogen and may be inhibited by P4. Furthermore, down-regulation of the FUT1 gene or degradation of the mRNA appears to be a major factor in regulating the H epitope, which has been identified as an adhesion ligand for the embryo at implantation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Susan J. Kimber, School of Biological Sciences, University of Manchester, Stopford Building, Oxford Road, Manchester, M13 9PT, UK.
3 Current address: INSERM Unit 257, I.C.G.M., 24 rue du Faubourg Saint-Jacques, Paris 75014, France.
Accepted: September 3, 1998.
Received: May 28, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(1–2)-fucosyltransferase activity in the murine endometrial epithelium during the estrous cycle, early pregnancy, and after ovariectomy and hormone replacement. Biol Reprod 1994; 50:73–81.[Abstract]- and ß-subunits in the mouse uterus and vagina: potential mediators of estrogen action. Endocrinology 1995; 136:2325–2340.[Abstract]-2-L-fucosyltransferases in human serum and the genetics of blood group H substances. Am J Hum Genet 1985; 37:749–760.[Medline]-L-fucosyltransferase in transfected cells. J Biol Chem 1989; 264:11158–11167.1,2-fucosyltransferase. J Biol Chem 1995; 270:8844–8850.(1,2)fucosyltransferase gene (FUT2). J Biol Chem 1995; 270:4640–4649.(1,2)fucosyltransferase genes differentially expressed throughout the rat colon. Biochem J 1994; 300:623–626.-2-L-fucosyltransferase from human gastric mucosa. J Biochem 1995; 118:541–545.1,2-L-fucosyltransferase towards sulfated, sialylated and fucosylated acceptors: evidence for an inverse relationship between 1,2-L-fucosylation and 1,6-L-fucosylation of asparagine linked GlcNAc. Biochemistry 1996; 35:8914–8924.[CrossRef][Medline](1–2) fucosyltransferase expressed in thymic and epididymal epithelial cells. Biochem J 1997; 327:105–115.-L-fucosyltransferase gene (FUT1). J Biol Chem 1997; 272:7501–7505.-L-fucosyltransferase (FUT2). Eur J Biochem 1997; 246:750–755.[Medline](2–6)sialyltransferase gene. J Biol Chem 1992; 267:2512–2518.(1–3)-fucosyltransferase genes in normal tissues. J Biol Chem 1995; 270:20112–20122.This article has been cited by other articles:
|
|
 |
|
  A. P Ponnampalam and P. A W Rogers Expression and regulation of fucosyltransferase 4 in human endometrium Reproduction, July 1, 2008; 136(1): 117 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. White, J.-G. Zhang, L. A. Salamonsen, M. Baca, W. D. Fairlie, D. Metcalf, N. A. Nicola, L. Robb, and E. Dimitriadis From the Cover: Blocking LIF action in the uterus by using a PEGylated antagonist prevents implantation: A nonhormonal contraceptive strategy PNAS, December 4, 2007; 104(49): 19357 - 19362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
L.-J. Xiao, J.-X. Yuan, X.-X. Song, Y.-C. Li, Z.-Y. Hu, and Y.-X. Liu Expression and regulation of stanniocalcin 1 and 2 in rat uterus during embryo implantation and decidualization. Reproduction, June 1, 2006; 131(6): 1137 - 1149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X.-H. Ma, S.-J. Hu, H. Ni, Y.-C. Zhao, Z. Tian, J.-L. Liu, G. Ren, X.-H. Liang, H. Yu, P. Wan, et al. Serial Analysis of Gene Expression in Mouse Uterus at the Implantation Site J. Biol. Chem., April 7, 2006; 281(14): 9351 - 9360. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. D. Catalano, M. H. Johnson, E. A. Campbell, D. S. Charnock-Jones, S. K. Smith, and A. M. Sharkey Inhibition of Stat3 activation in the endometrium prevents implantation: A nonsteroidal approach to contraception PNAS, June 14, 2005; 102(24): 8585 - 8590. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. von Wolff, X. Wang, H.-J. Gabius, and T. Strowitzki Galectin fingerprinting in human endometrium and decidua during the menstrual cycle and in early gestation Mol. Hum. Reprod., March 1, 2005; 11(3): 189 - 194. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A H K El-Hashash and S J Kimber Trophoblast differentiation in vitro: establishment and characterisation of a serum-free culture model for murine secondary trophoblast giant cells Reproduction, July 1, 2004; 128(1): 53 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Domino and E. A. Hurd LacZ expression in Fut2-LacZ reporter mice reveals estrogen-regulated endocervical glandular expression during estrous cycle, hormone replacement, and pregnancy Glycobiology, February 1, 2004; 14(2): 169 - 175. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Ni, T. Sun, N.-Z. Ding, X.-H. Ma, and Z.-M. Yang Differential Expression of Microsomal Prostaglandin E Synthase at Implantation Sites and in Decidual Cells of Mouse Uterus Biol Reprod, July 1, 2002; 67(1): 351 - 358. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. E. Domino, L. Zhang, P. J. Gillespie, T. L. Saunders, and J. B. Lowe Deficiency of Reproductive Tract alpha (1,2)Fucosylated Glycans and Normal Fertility in Mice with Targeted Deletions of the FUT1 or FUT2 alpha (1,2)Fucosyltransferase Locus Mol. Cell. Biol., December 15, 2001; 21(24): 8336 - 8345. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. M. Quirk, R. M. Harman, and R. G. Cowan Regulation of Fas Antigen (Fas, CD95)-Mediated Apoptosis of Bovine Granulosa Cells by Serum and Growth Factors Biol Reprod, November 1, 2000; 63(5): 1278 - 1284. [Abstract] [Full Text]
|
|
|
|
|
|
J.K. Powell, S.R. Glasser, S. Woldesenbet, R.C. Burghardt, and G.R. Newton Expression of Carbohydrate Antigens in the Goat Uterus During Early Pregnancy and on Steroid-Treated Polarized Uterine Epithelial Cells In Vitro Biol Reprod, February 1, 2000; 62(2): 277 - 284. [Abstract] [Full Text]
|
|
|
|
|
|
S. E. Domino, L. Zhang, and J. B. Lowe Molecular Cloning, Genomic Mapping, and Expression of Two Secretor Blood Group alpha (1,2)Fucosyltransferase Genes Differentially Regulated in Mouse Uterine Epithelium and Gastrointestinal Tract J. Biol. Chem., June 22, 2001; 276(26): 23748 - 23756. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
