| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Biological Sciences, WSU/UI Center for Reproductive Biology, University of Idaho, Moscow, Idaho 83844-3051
b Departments of Molecular and Integrative Physiology and
c Obstetrics and Gynecology, Ralph L. Smith Research Center, University of Kansas Medical Center, Kansas City, Kansas 66160-7338
ABSTRACT
The multifunctional cytokine leukemia inhibitory factor (LIF) is presumed to participate in preparing the uterus for blastocyst implantation. Increased production of LIF is positively correlated with termination of embryonic diapause and preparation for implantation in the spotted skunk. This study examined changes in the expression, localization, and hormonal regulation of LIF receptor (LIFRß) gene expression in the uterus of the skunk. Changes in the uterine concentration of LIFRß mRNA during pregnancy or in response to hormones after ovariectomy were determined by Northern hybridization analysis and reverse-transcriptase polymerase chain reaction. The skunk uterus produces two LIFRß transcripts, the levels of which increase in concentration when the blastocysts resume their development but then decline somewhat during the latter stage of blastocyst activation. Ovariectomy significantly reduced uterine LIFRß expression. Estradiol and/or progesterone failed to significantly elevate LIFRß mRNA levels in ovariectomized animals. Prolactin significantly increased uterine concentrations of LIFRß mRNA to greater than those of ovariectomized controls, but these levels were not comparable to those observed during preimplantation. The LIFRß mRNA was predominately localized to stromal cells surrounding the uterine glands and in yolk sac endoderm, syncytiotrophoblast, and cytotrophoblast of postimplantation embryos.
cytokines, estradiol, implantation/early development, pregnancy, progesterone, prolactin, syncytiotrophoblast, uterus
INTRODUCTION
The western spotted skunk breeds during the fall months. Early development is continuous as the embryos traverse the oviducts, but development becomes arrested at the blastocyst stage for 200 ± 20 days on entering the uterus. The uterus is small, very similar in appearance to that of an anestrous animal during this period of arrested development, and secretes very little protein [1]. Several days before implantation, the uterus undergoes a series of changes, including increased weight [2], increased luminal concentrations of an inhibitor to plasminogen activator [3], loss of estrogen and progesterone receptors from the luminal epithelium [4], increased expression of epidermal growth factor-receptor mRNA [5], increased vascular permeability (as indicated by an influx of plasma proteins into the uterine lumen) [3], and up-regulation of COX-2 gene expression [2]. In addition, some changes in uterine physiology are temporally correlated with the resumption of embryonic development (i.e., blastocyst activation). These changes include increased synthesis and secretion of uterine-specific proteins [1] one of which is leukemia inhibitory factor (LIF) [6].
A multifunctional cytokine, LIF exhibits proliferative, differentiation, survival, and cell death-initiating functions depending on the cell type involved. It has been detected at relatively high concentrations in uteri of several species at the expected time of implantation [7]. In the western spotted skunk, LIF protein first becomes detectable and LIF mRNA concentration in the uterus increases significantly at the time that blastocysts resume their development [6]. Moreover, mutation of the LIF gene results in failure of implantation in the mouse, which can be overcome by a constant infusion of LIF or by transplantation of the blastocysts to pseudopregnant, wild-type recipients [8]. The precise action of LIF in the implantation process is unknown. However, LIF is not required for embryonic development in the mouse, because mutation of the LIF receptor (LIFRß) results in implantation of LIFR-/- blastocysts and their development to term [9]. Even so, the neonates do not survive, and they exhibit reduced bone mass and reduced numbers of astrocytes in the central nervous system [9]. Deficiencies in placental function, resulting from reduction in the fetal vascular component and accumulation of fibrin in the LIFRß placentas, are presumed to be partially responsible for the early death of these neonates, as reported by Ware et al. [9]. Based on current evidence, it appears that LIF plays an essential role in preparing the uterus for successful implantation and placentation.
Currently, LIF is believed to initiate signal transduction by binding solely to the LIFR complex, which consists of two subunits. The LIFRß binds LIF, oncostatin M, or cardiotrophin-1 and then forms a heterodimer with gp130 [10]. In the absence of gp130, LIF is incapable of initiating an intracellular signal. Activation of this heterodimer complex results in activation of cytoplasmic tyrosine kinases and subsequent modification of transcription factors [11]. The LIFRß exists as a transmembrane and soluble receptor [12, 13], and the liver is the primary source of soluble LIFR, which has a molecular mass of 90 kDa and binds LIF with approximately the same affinity as the transmembrane receptor [12, 14]. The soluble receptor, which circulates in the blood, is believed to inhibit the actions of LIF [12, 14]. Placental and uterine concentrations of LIFRß mRNA increase during pregnancy in the mouse [13]. To better understand the possible role of LIF in blastocyst activation and subsequent implantation in a species that exhibits an obligate delay of implantation, we examined changes in uterine expression, cell localization, and hormonal regulation of LIFRß in the western spotted skunk.
MATERIALS AND METHODS
Reagents
Guanidine thiocyanate, guanidine hydrochloride, buffer-saturated phenol, and cesium chloride were obtained from Ameresco (Dallas, TX). Tris-HCl, Tris base, sodium citrate, and sodium hydroxide were purchased from Fisher Scientific (Pittsburgh, PA). Specific DNA primers, Taq polymerase, Superscript II, and restriction enzymes were obtained from Gibco BRL (Gaithersburg, MD). RQ1 DNase, recombinant RNasin, and random hexamers were obtained from Promega (Madison, WI). AmpliTaq Gold polymerase was obtained from Perkin-Elmer (Branchbur, NJ). MAXIscript was obtained from Ambion (Austin, TX). All radioactive isotopes were obtained from NEN-DuPont (Boston, MA), and Nytran Plus membranes were purchased from Schleicher & Schuell (Keene, NH). All other chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).
Animals and Experimental Treatments
Seventy-nine pregnant western spotted skunks (Spilogale putorius latifrons) that had bred in the wild were obtained from a U.S. Department of Agriculture-licensed dealer (Jerry Eckstine) in Cottage Grove, Oregon. All skunks were maintained and samples collected according to the methods described by Hirzel et al. [6]. Fifty intact females were injected with an overdose of sodium pentobarbital at various times during the preimplantation and early postimplantation periods. Blood samples were collected in heparinized vacuum tubes by cardiac puncture and then stored at -20°C until the time of hormone assay. The uterus of each animal was rapidly removed, trimmed of mesentery, and flushed with 0.05 M PBS. The blastocysts were measured with an ocular micrometer, and the diameters of the embryos or implantation sites were used to determine the stage of pregnancy [15, 16]. Flushed blastocysts with diameters of 1.1 mm or smaller were classified as delayed implanting, those measuring 1.2–1.6 mm as early activated, and those measuring 1.7 mm or greater as late activated. The uteri were rapidly frozen in liquid propane at -196°C and stored at -80oC.
Twenty-nine skunks whose uteri contained delayed-implanting blastocysts were assigned to one of five treatment groups in early January 1995. All animals were anesthetized with sodium pentobarbital, bilaterally ovariectomized, and allowed to recover. Six skunks received six daily injections containing 0.5 mg of prolactin (PRL) dispersed in 5% beeswax and sesame oil beginning 3 days after ovariectomy. The remaining animals were anesthetized with ketamine 6 days after ovariectomy, and Silastic capsules, whose sizes and numbers were previously described by Hirzel et al. [6], were inserted into the interscapular region. Six females received capsules containing estradiol-17ß, five received capsules containing progesterone, six received capsules containing estradiol-17ß and progesterone, and six received empty capsules and daily injections of 5% beeswax and oil. Steroid treatments continued for 3 days. At the end of the treatments, blood and tissue samples were obtained as described earlier. Blood plasma was assayed for both steroids and PRL according to the methods described by Hirzel et al. [6].
RNA Isolation and cDNA Preparation
Total RNA was extracted from the uterine horns of individual animals by homogenization in guanidine thiocyanate and then separated by cesium chloride gradient centrifugation. Poly (A)+ RNA for use in Northern hybridization analysis was prepared by passing total RNA from individual animals through oligo (dT) columns. Aliquots of cDNA used in the reverse-transcriptase polymerase chain reaction (RT-PCR) assays reported here were taken from the same cDNA samples used in the report of uterine LIF gene expression by Hirzel et al. [6]. A 1-µl aliquot of each cDNA was mixed with specific primers and subjected to PCR (0.2 mM deoxyribonucleotides, 2 mM MgCl2, 0.3 µM of each primer, 1.25 U AmpliTaq Gold) to amplify skunk LIFRß (sLIFRß) sequences. Primers for synthesizing and amplifying the sLIFRß cDNA were designed to nucleotides 950 to 961 and 1703 to 1720 of the human LIFRß sequence [17]. These primers amplified a 781-base pair (bp) fragment, which was then separated by electrophoresis, cloned, and identified by sequencing in both directions. Amplification primers specific for the skunk LIFRß sequences were used in the quantitative RT-PCR assay and yielded a 405-bp PCR product from the cDNA templates that coded for the extracellular domain of the LIFRß. This portion of the mRNA had 88% sequence similarity to human LIFRß and 81% and 80% similarity to mouse and rat LIFRß protein. Three uterine samples from each stage of pregnancy were amplified for LIFRß, and the identity of the PCR products were verified by automated sequencing. The construction and verification of the identity of skunk cyclophilin, which was used as an internal control, were previously reported elsewhere [6].
RT-PCR
The appropriate cycle number to ensure exponential amplification of the sLIFRß cDNA (36 cycles) was determined in preliminary PCR experiments as described previously for cyclophilin [6]. Expression of both gene products was quantified in a separate set of PCR reactions consisting of cDNA samples (n = 5–8 per group) assayed in duplicate. Hot-start PCR was performed using 0.3 M of each primer, 0.2 mM deoxyribonucleotides, 11 µCi [32P]deoxyadenosine triphosphate (3000 Ci/mmol) and 11 µCi [32P]deoxycytidine triphosphate (3000 Ci/mmol), 2.5 U AmpliTaq Gold, 4% dimethylsulfoxide, 2.0 mM MgCl2, 5 µl cDNA, and 2.5 µl of vendor-supplied PCR buffer. The final volume was brought to 25 µl with the addition of distilled H2O. The PCR conditions consisted of an initial incubation (11 min at 95°C) followed by 36 (sLIFR) or 40 (cyclophilin) cycles (20 sec each at 94°C, 52°C, and 72°C. The PCR products were separated on 1% agarose gels, dried, and quantified using a GS-525 Molecular Imager (BioRad Laboratories, Hercules, CA). The LIFRß expression was normalized to that of cyclophilin.
Northern Hybridization Analysis
Poly(A)+ RNA was prepared from individual uteri of an additional 26 animals that contained delayed-implanting blastocysts (n = 8), early activated blastocysts (n = 6), activated blastocysts (n = 5), or implantation sites ranging from 3.8–16 mm in diameter (1 to 12 days postimplantation, n = 7). Only the uterine segments between implantation sites of the latter animals were extracted. Approximately 1.5 µg of RNA from each animal was denatured and separated on formaldehyde agarose gels. After transfer to nylon membranes, the RNA was cross-linked by ultraviolet (UV) radiation at 120 mJ in a Stragene 2400 UV Stratalinker. The blots were prehybridized for 4 h at 68°C in 5x SSPE (1x SSPE: 10 mM sodium phosphate [pH 7.4], 1 mM EDTA, 180 mM NaCl), 4% SDS, and 100 µg/ml sheared, denatured salmon-sperm DNA. Radiolabeled ([32P]UTP, 800 Ci/mmole) sense and antisense cRNA probes of skunk LIFRß and mouse rpL7 were obtained by T3- or T7-directed RNA synthesis from plasmids containing the cloned cDNAs. The construct for the rpL7 probe has been described by Das et al. [18]. The prehybridization solutions were discarded, and fresh hybridization solution containing 1 x 106 cpm/ml of the sLIFR-antisense probe was added and then incubated (18 h at 68°C). The blots were washed twice in 2x SSPE and 0.1% SDS for 30 min at room temperature, followed by two washes in 0.5 x SSPE and 0.1% SDS for 30 min at 55°C, and hybrids were then detected by autoradiography. To demonstrate specificity of the hybridization, the same blots were stripped and probed with the sLIFRß sense probe. To quantify RNA loading, the blots were stripped and probed with a mouse cRNA probe to rpL7. The major sLIFRß transcript and mouse rpL7 were quantified with the GS-525 Molecular Imager, and the data were expressed as the ratio of the two gene products.
In Situ Hybridization
Frozen sections (10 µm) of uteri were prepared from skunks collected during delayed implantation (n = 3), early activation (n = 3), late activation (n = 3), and early (4.0–6.8 mm or about 24–48 h) postimplantation (n = 3). Each poly-L-lysine-coated slide contained two cross-sections of uterus from the same animal. Each slide also contained sections of uterus from animals at each stage of pregnancy. The sections were hybridized with 35S-labeled sense or antisense skunk cRNA LIFRß probes. The sections were then washed as previously described and incubated with RNase A (20 µl/m) at 37°C for 20 min [19]. The RNase A-resistant hybrids were detected after 30 days of exposure to Kodak NTB-2 emulsion (Eastman Kodak, Rochester, NY). Adjacent sections were hybridized with the sense probe and served as negative controls. All slides were then stained with hematoxylin-and-eosin.
Statistical Analysis
The data are expressed as the mean ± SEM. The data sets were first examined by one-way ANOVA. Differences in the mean values among treatment groups were determined by pairwise multiple comparisons using the Student-Newman-Keuls or Mann-Whitney rank sum tests. A P value of less than 0.05 was considered to be statistically significant.
RESULTS
LIFR Expression During Pregnancy
Steady-state levels of LIFR mRNA in the uteri of spotted skunks during three stages of the preimplantation and the early (1–12 days) postimplantation period were determined by Northern hybridization analysis (Fig. 1). Two major transcripts of approximately 2.0 and 2.9 kilobases (kb) were observed during all stages of pregnancy that were examined, but these transcripts were most abundant among uteri in which the blastocysts were undergoing the initial stage of activation. The major transcript (2.9 kb) was significantly less abundant (P > 0.05) during delayed implantation and in segments of uterus located between the implantation sites of early postimplantation specimens. The LIFRß mRNA was still detectable in uterine segments located between implantation sites collected on Days 10 and 12 after implantation.
|
Changes in uterine expression of the LIFR gene during pregnancy, as determined by the RT-PCR method, are shown in Figure 2. The message was detectable but present in relatively low copy numbers during the winter months, when embryonic development was arrested. Uterine concentrations of LIFRß mRNA significantly increased (P < 0.05) at early blastocyst activation but declined during the later portion of the activation process.
|
Cell-specific autoradiographic signals for LIFRß message were faint and restricted to the deeper portions of the stroma in uteri containing delayed-implanting blastocysts (Fig. 3, A and B). In uteri containing early activating blastocysts, moderately intense hybridization signals for LIFRß mRNA were localized predominately in the stroma immediately adjacent to the uterine glands (Fig. 3, C and D). Localization of the message was the same but less intense in uteri containing late-activated blastocysts (Fig. 3, E and F). Two 24-h and one 48-h postattachment specimens were examined. Intense hybridization signals to the LIFRß message were localized in the syncytiotrophoblast, cytotrophoblast, and yolk sac endoderm of the embryos and in the endometrial stroma surrounding the basal regions of the uterine glands (Fig. 3, G and H). Very little, or perhaps no, hybridization signals to the LIFRß were observed in the inner cell mass (ICM) of one early activated blastocyst (data not shown). Positive hybridization signals were absent in all uterine sections hybridized with sense probe (Fig. 4).
|
|
Hormonal Regulation of LIFR Expression
Ovariectomy significantly reduced the ratio of LIFRß:cyclophilin mRNA (Fig. 5) compared with that in uteri containing delayed-implanting blastocysts as depicted in Figure 2 (0.114 ± 0.021 vs. 2.129 ± 0.275, P < 0.05). Although all steroid treatments both significantly elevated plasma concentrations of estrogen and progesterone to within the physiological range observed during the early postimplantation period and significantly increased uterine weight to greater than that of ovariectomized controls [6], none significantly enhanced uterine expression of LIFR (Fig. 5). Injection of PRL for 6 days elevated the plasma PRL concentrations to supraphysiological levels and significantly up-regulated uterine LIFRß mRNA concentrations beyond those of ovariectomized controls (Fig. 5). However, PRL had no effect on uterine weight [6], and it failed to generate uterine concentrations of LIFRß mRNA equivalent to those observed during any phase of the preimplantation period.
|
DISCUSSION
Our data indicate that at least two transcripts of LIFRß gene of approximately 2.0 and 2.9 kb are expressed in the uterus of the spotted skunk during all stages of pregnancy that we examined. The uterus and placenta of the mouse express three transcripts of approximately 11, 4.4, and 3 kb [14]. The latter transcript codes for the soluble form of the receptor and is only detectable in the uterus during pregnancy. Two major transcripts of approximately 6 and 4.5 kb and a minor transcript of 5 kb of LIFRß have been detected in human placenta [17]. One of the transcripts observed in the skunk uterus might code for a soluble form of the skunk LIFRß.
The LIFRß mRNA is present in low copy numbers in the uterine stroma of the spotted skunk during the prolonged period of delayed implantation, but it undergoes significant up-regulation precisely when the blastocysts cease diapause and resume their development. The highest ratios of LIFRß:rpL7 mRNA (2.2 and 2.3) were in animals with the earliest-activated blastocysts. The LIFRß expression remains elevated but appears to decline somewhat during the latter stages of blastocyst activation. This pattern of expression is consistent with a previous report that uterine production of LIF increases at precisely the same time-point in pregnancy [6].
Our data indicate that LIFRß mRNA is localized in the stroma of the skunk uterus rather than in the luminal or the glandular epithelium. In this respect, our data differ from those of other studies, in which LIFRß mRNA and/or protein have been detected in the luminal and glandular epithelium of the mouse [20, 21], rabbit [22], human [23], and rhesus monkey [24]. To our knowledge, only one other study has reported the appearance of LIFRß (protein) in the uterine stroma of the mouse on Day 5 of pregnancy [21]. Localization and up-regulation of the LIFRß message in the stroma at the immediate vicinity of the skunk uterine glands during early blastocyst activation suggests that LIF, which is predominately produced by the luminal epithelium and uterine glands of the spotted skunk [6], binds to LIFRß-bearing stromal cells, which, in turn, presumably release products that directly or indirectly promote blastocyst development and/or stimulate preparation of the uterus for implantation. The LIFRß hybridization signals in the ICM of an early activated blastocyst were perhaps slightly greater than background levels. Whether LIFRß is present in the ICM of delayed-implanting or fully activated blastocysts could not be determined from our specimens, however, because the tissue sections of these blastocysts did not contain any ICM. The LIFRß mRNA was consistently absent from trophoblasts of all preimplantation skunk blastocysts. The LIFRß mRNA is present in the ICM but not in the trophoblasts of mouse blastocysts [20], and both LIFRß and gp130 mRNA are detectable by RT-PCR in human morulae and blastocysts [25, 26]. Addition of LIF to defined culture medium reportedly has a beneficial effect on the growth of human preimplantation embryos [27]. If both receptor types are present in the ICM of delayed-implanting skunk blastocysts, one of the most plausible functions of LIF would be to promote the survival of ICM cells during diapause.
We observed intense autoradiographic LIFRß hybridization signals in yolk sac endoderm, syncytiotrophoblast, cytotrophoblast, and in stromal cells surrounding the basal regions of the glands of all three postimplantation skunk embryos. However, our Northern hybridization analysis of pieces from uteri devoid of embryos (interimplantation uterine segments) from all seven postimplantation specimens revealed markedly reduced LIFRß mRNA concentrations, which were comparable to those from delayed-implantation specimens. This suggests that LIFRß mRNA concentrations are higher at implantation sites than at the rest of the uterus. Our finding of LIFRß mRNA in fetal components of the placenta, coupled with the report that LIFRß-/- neonatal mice die presumably because of the disruption of normal placentation [9], suggests that LIF may enhance placental formation or function in the spotted skunk. Expression of the LIFRß gene has also been detected in chorionic villi of first- and second-trimester as well as term placentae and, at lower levels of expression, in the decidual tissue of humans [28]. Owczarek et al. [13] reported increased expression of LIFRß in the liver, placenta, and uterus of mice during pregnancy. However, Nichols et al. [20] reported that LIFRß transcripts were down-regulated in embryos after implantation but were abundant in decidual tissue of mice. Consequently, the specific role of LIF in placentation or placental function may differ between species, because the cells possessing LIFRß vary between species.
Ovariectomy significantly decreased uterine LIFRß mRNA concentrations, as determined by RT-PCR, compared with those in uteri of intact preimplantation animals (P < 0.05). This indicates that ovarian hormones play some role in up-regulating LIFRß mRNA. However, physiological levels of estradiol, administered either alone or in combination with progesterone, that induced a significant increase in uterine weight had no effect on uterine LIFRß mRNA concentrations among ovariectomized skunks. A tendency for progesterone to up-regulate uterine LIFRß mRNA was observed, but this effect was not statistically significant (P > 0.05). Prolactin significantly (P < 0.05) increased uterine LIFRß mRNA concentrations beyond those of the ovariectomized controls, but it failed to restore concentrations to those observed in intact pregnant females. Steroids induce rapid changes in specific genes [29]. Therefore, we assumed that constant administration of steroids for 3 days would be sufficient to elicit a positive response. Eighteen or more days of daily administration of PRL are required to induce early implantation in the skunk; consequently, we administered PRL for a longer duration. We cannot rule out the possibility that failure of the various hormone treatments to elevate LIFRß mRNA to concentrations similar to that in the uteri of intact animals may relate to the duration of hormone treatment.
Our data suggest that a combination of progesterone, or some other ovarian hormone, and PRL may be required to achieve maximal up-regulation of uterine LIFRß mRNA. By sequencing an 800-bp RT-PCR product, we have verified that the uterus of the spotted skunk contains mRNA coding for PRL receptors localized in the luminal and glandular epithelium as well as in the myometrium (unpublished data). In this regard, our data appear to be similar to those regarding the situation in the rabbit, in which PRL enhances uterine sensitivity to progesterone by increasing progesterone receptors and the progesterone-dependent binding of nuclear proteins to the uteroglobin gene, which further stimulates transcription of the uteroglobin gene [30, 31].
In summary, our results indicate that the uterus of the spotted skunk contains mRNA coding for LIFR, and that expression of this gene is up-regulated in the stromal cells beneath the glands by ovarian hormones and PRL. Increased transcription of the LIFRß gene appears to coincide with termination of embryonic diapause and resumption of embryonic development in the blastocysts.
ACKNOWLEDGMENTS
We gratefully acknowledge the technical assistance of Todd Garwood in the collection and partial analysis of the RT-PCR data.
FOOTNOTES
First decision: 3 February 2000.
1 Supported in part by NICHD grants HD34247 and DHHS/NIH RR11833 to R.A.M. and HD 12304 and HD29968 to S.K. Dey 
2 Correspondence. FAX: 208 885 7905; rmead{at}uidaho.edu
Accepted: March 6, 2000.
Received: December 29, 1999.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  K. Wanggren, P.G. Lalitkumar, F. Hambiliki, B. Stabi, K. Gemzell-Danielsson, and A. Stavreus-Evers Leukaemia inhibitory factor receptor and gp130 in the human Fallopian tube and endometrium before and after mifepristone treatment and in the human preimplantation embryo Mol. Hum. Reprod., June 1, 2007; 13(6): 391 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Sato, T. Tsubota, T. Komatsu, G. Watanabe, K. Taya, T. Murase, I. Kita, and T. Kudo Changes in Sex Steroids, Gonadotropins, Prolactin, and Inhibin in Pregnant and Nonpregnant Japanese Black Bears (Ursus thibetanus japonicus) Biol Reprod, October 1, 2001; 65(4): 1006 - 1013. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
