|
|
||||||||
Article |
a Department of Veterinary Basic Sciences, Royal Veterinary College, Potters Bar, Hertfordshire, EN6 1NB, United Kingdom
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
Furthermore, several IGFBP-5 specific proteases have been isolated [8, 9]. These proteases, thought to be matrix metalloproteinases, cleave the IGF-IGFBP-5 complex, lowering IGFBP's affinity for the IGF, resulting in the dissociation of the IGF from the complex. Interestingly, a 23-kDa proteolytically cleaved fragment of IGFBP-5 has been shown to stimulate mitogenesis in an IGF-independent manner [10]. Although there have been studies investigating the local production of IGFBP-5 in the reproductive tract of a number of different species (cow [11], rodent [1214], nonhuman primate [15], human [1618]), the ewe has, to our knowledge, been largely ignored. Therefore, in an attempt to further our understanding of the function and regulation of IGFBP-5 in the reproductive tract, we have mapped its spatial and temporal expression throughout the estrous cycle and during early gestation in the ovine uterus and placenta.
| MATERIAL AND METHODS |
|---|
|
|
|---|
Reproductive tracts were taken from 46 nonpregnant ewes throughout the estrous cycle and from 33 pregnant ewes ranging in gestational age from 13 to 55 days. In nonpregnant ewes, the stage of the estrous cycle was regulated by an i.m. injection with Estrumate (Coopers Animal Health Ltd., Crewe, UK), a prostaglandin (PG) analogue given in the midluteal phase to induce estrus. After treatment, estrus can be expected at 48 h, with ovulation occurring at approximately 65 h. The number of samples collected at each time point were as follows: nonpregnant24 h post-PG (pPG), n = 3; 36 h pPG, n = 6; 48 h pPG, n = 6; 60 h pPG, n = 5; 65 h pPG, n = 3; Days 23, n = 6; Day 5, n = 3; Days 79, n = 5; Day 13, n = 6; Day 15, n = 3; pregnantDays 1322, n = 9; Days 2530, n = 8; Days 3141, n = 9; Days 4555, n = 7. Uterine samples were also collected on Days 16 and 17 from 4 ewes with unilateral pregnancies that followed uterine transection before mating. The horn ipsilateral to an active corpus luteum was sectioned, the endometrium cauterized, and the myometrium of the two ends sutured [19]. Because both horns (the pregnant and nonpregnant) are subject to the same endocrine environment, but are surgically isolated from each other, the technique provides an opportunity to study the effect of local regulatory factors produced by the embryo on IGFBP-5 expression. All animals were kept and treated under the Home Office Animals (Scientific Procedures) Act 1986.
Uteri were dissected transversely into segments of approximately 23 cm in length, wrapped in aluminum foil, and frozen in liquid nitrogen-tempered isopentane. Samples were stored at -80°C until sectioning.
Oligonucleotide Probe
The 45-mer oligonucleotide antisense probe corresponded to nucleotides 291335 of the bovine IGFBP-5 gene [20] and had the following sequence; 5'-TCG-GAG-ATG-CGG-GTG-TGC-TTG-GGC-CGG-AAG-ATC-TTG-GGC-GAG-TAG-3'. A sense probe identical in sequence to the respective mRNA target was included as a negative control. To date, the ovine sequence has not been determined, so the region selected was chosen for 1) its highly homology with other species (pig 97%, human 95%, mouse 91%, rat 88%) and 2) its low homology with other IGFBPs.
In Situ Hybridization
The in situ hybridization procedure was performed as described previously [2]. All chemicals were purchased from Sigma Chemical Co. (Poole, Dorset, UK) or BDH (Poole, Dorset, UK) unless otherwise stated. Briefly, 10-µm cryostat sections were thaw-mounted onto poly-L-lysine-coated slides, fixed in 4% (w:v) paraformaldehyde in 0.01 M PBS, washed in 0.01 M PBS, and dehydrated sequentially in 70% and 95% ethanol. The oligonucleotide probe for IGFBP-5 was [35S]-labeled at the 3'-end using terminal deoxynucleotidyl transferase (Pharmacia Biotechnology, St Albans, Herts, UK). Sections were subsequently hybridized with a 110 000-cpm probe in 200 µl hybridization buffer and incubated overnight at 42°C. After incubation, slides were washed in single-strength SSC (0.15 M sodium chloride, 0.015 M sodium citrate), 0.2% (w:v) sodium thiosulfate pentahydrate solution for 30 min at room temperature, then at higher stringency in single-strength SSC, 0.2% (w:v) sodium thiosulfate pentahydrate solution for 60 min at 55°C. Finally the slides were dehydrated in a gradient of ethanol, air-dried, and exposed to x-ray film (Hyperfilm-ßmax, Amersham International, Aylesbury, UK), for 21 days. Samples of uterus previously shown to be positive for the IGFBP-5 probe [21] were included as a positive control. All samples were processed together in one batch.
Photographic Emulsions
Slides previously exposed to x-ray film were coated with a photographic emulsion, LM1 (Amersham International), according to the manufacturer's instructions, and stored at 4°C for 28 days. Slides were then developed and counterstained with hematoxylin and eosin in order to confirm microscopically the cellular localization of the radioactive signal.
Optical Density (OD) Quantification
The quantification method was performed as described previously [2]. An image analysis system (Seescan PLC, Cambridge, UK) was used to convert the radioactive signals to OD units for quantification using a linear gray scale of 02.1. All readings were on this part of the scale. The antisense autoradiograph was placed under the image analyzer lens, and a blank section of the film was measured. The region to be quantified was outlined, and the background reading was automatically subtracted from this. The corresponding section on the sense autoradiograph was measured in a similar fashion, and the reading obtained was subtracted from the antisense value to give the final OD value for the specific hybridization. A minimum of two antisense slides, each containing at least two sections, were measured for each animal, and the average OD value was calculated. The detection limit was taken as an OD > 0.01. The coefficient of variation for duplicate absorbance measurements between two slides was 5%.
Statistical Analysis
Values are given as mean OD ± SEM. Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS, Chicago, IL), version 6.1. Differences between time points were analyzed by unbalanced one-way ANOVA or by an ANOVA of regression. Fisher's tests were used to determine which time points differed. If the data did not exhibit homogeneity of variance, then the Kruskal-Wallis nonparametric one-way ANOVA was used instead. If no time-related changes were detected, all data from a particular region were pooled between time intervals. Results were considered statistically significant when P < 0.05.
| RESULTS |
|---|
|
|
|---|
|
|
During pregnancy, IGFBP-5 was produced in the same regions with additional localization to the endometrial glands (Figs. 1E and 3A). Highest levels of IGFBP-5 mRNA expression were found in the luminal epithelium and endometrial glands (Fig. 1E and Fig. 3, AC). In these regions, levels of expression were initially modest, but showed a significant increase as gestation progressed (Fig. 4, A and B, P < 0.01). In the inner circular layer of myometrium, concentrations of IGFBP-5 also increased with time (P < 0.05), with concentrations initially low and rising to moderate levels at Day 55 (Fig. 4C). Concentrations in the outer myometrium and endometrial stroma were below the limit of detection, OD < 0.01.
|
|
In the caruncles, there was an intense band of IGFBP-5 mRNA localized just below the tips of invading fetal villi (Figs. 1C and 3D). As the villi penetrated deeper into the caruncle, this band was observed to retreat, such that by Day 40 it occupied only the region of the placentome capsule (Fig. 1G). Photographic emulsions revealed that below this band, the IGFBP-5 mRNA hybridized to form a series of circles (Fig. 3, D and E, Fig. 5). Careful analysis confirmed that this was not IGFBP-5 mRNA hybridizing to blood vessel walls as was expected. Instead, the IGFBP-5 appeared to hybridize to specific areas of caruncular stroma, which had seemingly become organized into circular structures (Fig. 3F). The identity of these circular structures and their significance remains to be determined. There was no hybridization to any fetal tissue.
|
In the transected uteri on Days 1617 of gestation, concentrations in the luminal epithelium of the pregnant horn were significantly higher than levels in the nonpregnant horn (Fig. 6, P < 0.01). In all other regions there was no significant difference in IGFBP-5 concentration between horns.
|
| DISCUSSION |
|---|
|
|
|---|
During the estrous cycle, IGFBP-5 concentrations were highest in the luminal epithelium and caruncular stroma, with moderate expression in the inner myometrium and low expression in the endometrial stroma. Similar findings have been shown in the rat, with IGFBP-5 protein localized to the luminal epithelium and outer myometrium [18] and IGFBP-5 mRNA localized to both muscle layers of the myometrium [13]. In the cow, IGFBP-5 has been found in the caruncles, endometrial stroma, and myometrium [11], with stronger expression recorded in the inner than the outer muscle layers of the myometrium (unpublished observations). In all regions, with the exception of the endometrial stroma, IGFBP-5 showed cyclic variations in expression, with levels particularly high around the time of ovulation. Concentrations declined significantly in the luteal phase, suggestive perhaps of regulation by ovarian steroid hormones. In human endometrium, IGFBP-5 is more dominant in the follicular phase [16], and in ovariectomized nonhuman primates, estradiol positively regulates IGFBP-5 expression in the myometrium [15]. By comparison, in the rat, IGFBP-5 expression predominates in the luteal phase [13], with estradiol negatively regulating IGFBP-5 in uteri of ovariectomized rats [18]. Collectively, these discrepancies must represent a species-specific difference in the steroidal regulation of IGFBP-5, with the sheep more closely resembling the regulatory system of the nonhuman primate than that of the rat. In the 5' flanking region of the rat IGFBP-5 gene, there are multiple cis-regulatory elements, including progesterone and estrogen-receptor binding sites [22]. Whether such elements exist in the ovine promoter region remains to be determined. IGFBP-5 can both potentiate and inhibit the mitogenic actions of the IGFs. We speculate that its abundant, almost ubiquitous, expression in the cyclic uterus suggests that it is an important physiological modulator of IGF-I and IGF-II. We and others have previously reported that mRNAs for both these peptides and the type 1 IGF receptor are present in the ovine uterus during the cycle, with a significant up-regulation of endometrial IGF-I mRNA expression at estrus [2325].
Studies from our laboratory have also shown that IGFBP-5 is the most abundantly expressed of the IGF binding proteins in the pregnant ovine uterus (unpublished observations). Its expression appeared to be developmentally regulated, increasing in the luminal epithelium, endometrial glands, and myometrium as gestation progressed. In the rat, IGFBP-5 expression increases in the myometrium during pregnancy, peaks mid-gestation, and thereafter becomes limited [12]. In primates, local IGF-I mRNA levels were significantly correlated with local myometrial cell proliferation [15], and high levels of IGFBP-5 have been observed in developing embryonic muscles [26, 27]. These data taken together with the current data would support a role for IGFBP-5 in modulating uterine muscle growth.
There was also intense and increasing expression of IGFBP-5 in the endometrial glands and luminal epithelium during pregnancy. In the glands, IGFBP-5 in conjunction with the type 1 receptor could modulate glandular secretions contributing to the uterine milieu. The fact that IGFBP-5 is colocalized with four other IGF binding proteins (IGFBP-2, -3, -4, and -6) and IGF-2R at the fetal-maternal interface at this time [2, 25] implies the importance of tight regulation of IGF activity occurring in and around this region. It is possible that this multitude of IGFBPs functions to create a large reservoir of IGFs, which could then be targeted to certain tissues, fetal or maternal.
Interestingly, the up-regulation of uterine IGFBP-5 is coincidental with the down-regulation of uterine IGFBP-3 [2], a finding that is reciprocated in the nonhuman primate myometrium [15] and in human intestinal muscle [28]. However, why IGFBP-3 and IGFBP-5 should be regulated in reciprocal fashion in these tissues remains to be determined.
The most intriguing finding was the pattern of IGFBP-5 expression in the caruncles, the sites of trophoblast invasion. In this region, the IGFBP-5 formed a band followed by a series of rings just below the tips of the invading fetal villi. Recent evidence has implicated IGFBP-5 in the induction of apoptosis as it is up-regulated in atretic ovarian follicles [29], involuting thyroid, prostate, and mammary glands [30, 31], as well as in fetal interdigital webs just before their deletion (D.J. Flint, Hannah Research Institute, Ayr, UK, personal communication). In view of this, one could speculate that the IGFBP-5 localized in this region functions to create a route or a series of channels to allow the fetal villi access into the caruncular stroma. The formation of a network of such channels has been reported in ruminant caruncles just before villus penetration [32]. We are currently investigating this phenomenon.
In the transected uteri, IGFBP-5 was up-regulated in the luminal epithelium of the pregnant horn. This finding is in agreement with Keller et al. [11], who found that binding of [125I]-IGF-I by proteins in the 29- to 31-kDa range was greater in pregnant than nonpregnant cow endometrium on Days 13 and 15. Since both horns in our studies were subject to the same endocrine environment, the results suggest that an embryonic factor, acting via a paracrine mechanism, functions to up-regulate IGFBP-5 at the level of transcription. One possible candidate for this is IGF-II, which in rat osteoblast and chondrocyte cultures causes dose-dependent increases in the level of both IGFBP-5 mRNA and peptide [33, 34]. In sheep, IGF-II is secreted from the adjacent fetal mesoderm in significantly increasing amounts towards term [2]. The antiluteolytic hormone, ovine interferon-tau, is unlikely to up-regulate IGFBP-5 transcription since it is down-regulated by Day 21, when IGFBP-5 levels are still continuing to rise [35].
IGFBP-5 is abundantly expressed in rodent and human placenta and fetal membranes [14, 17], as well as in bovine blastocysts [36]. In this study, however, there was no evidence for hybridization of IGFBP-5 to any fetal tissue. Its presence in the preimplantation ovine blastocyst cannot, however, be excluded, as no attempt was made to preserve or include blastocysts in the tissue collected for the purposes of this study.
In summary, our data have shown that IGFBP-5 mRNA is abundantly expressed in ovine uterus. IGFBP-5 shows cyclic variations in expression, with levels peaking around ovulation, perhaps suggestive of regulation by estradiol. During pregnancy, IGFBP-5 appears to be developmentally regulated, with levels increasing with gestational age. Increased concentrations of IGFBP-5 in the pregnant horn of transected uteri indicate that an embryonic factor functions to stimulate IGFBP-5 via a paracrine mechanism. Unique rings of IGFBP-5 mRNA localized in the caruncles suggest that this binding protein may be involved in creating a route to allow the fetal villi access into the caruncular stroma. This hypothesis is supported by recent evidence for a role for IGFBP-5 in apoptosis. Intense glandular and luminal epithelial expression could enable IGFBP-5 to facilitate the transfer of IGFs across the maternal-fetal interface, while localization to the myometrium could suggest a role for IGFBP-5 in modulating uterine muscle growth.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
1 Financial support provided by the Wellcome Trust; J.C.O. was supported by a studentship from the Biotechnology and Biological Sciences Research Council. ![]()
2 Correspondence: T.S. Gadd, Department of Veterinary Basic Sciences, The Royal Veterinary College, Boltons Park, Hawkshead Road, Potters Bar, Herts, EN6 1NB, United Kingdom. FAX: 01707 647085; treynold{at}rvc.ac.uk ![]()
Accepted: December 31, 1999.
Received: July 27, 1999.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. C. Satterfield, K. Hayashi, G. Song, S. G. Black, F. W. Bazer, and T. E. Spencer Progesterone Regulates FGF10, MET, IGFBP1, and IGFBP3 in the Endometrium of the Ovine Uterus Biol Reprod, December 1, 2008; 79(6): 1226 - 1236. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Kim, G. Song, H. Gao, J. L. Farmer, M. C. Satterfield, R. C. Burghardt, G. Wu, G. A. Johnson, T. E. Spencer, and F. W. Bazer Insulin-Like Growth Factor II Activates Phosphatidylinositol 3-Kinase-Protooncogenic Protein Kinase 1 and Mitogen-Activated Protein Kinase Cell Signaling Pathways, and Stimulates Migration of Ovine Trophectoderm Cells Endocrinology, June 1, 2008; 149(6): 3085 - 3094. [Abstract] [Full Text] [PDF] |
||||
![]() |
O. Shynlova, P. Tsui, A. Dorogin, B. L. Langille, and S. J. Lye Insulin-like Growth Factors and Their Binding Proteins Define Specific Phases of Myometrial Differentiation During Pregnancy in the Rat Biol Reprod, April 1, 2007; 76(4): 571 - 578. [Abstract] [Full Text] [PDF] |
||||
![]() |
C K Hilleson-Gayne and J A Clapper Effects of decreased estradiol-17{beta} on the serum and anterior pituitary IGF-I system in pigs J. Endocrinol., December 1, 2005; 187(3): 369 - 378. [Abstract] [Full Text] [PDF] |
||||
![]() |
M Munoz-Gutierrez, P A Findlay, C L Adam, G Wax, B K Campbell, N R Kendall, M Khalid, M Forsberg, and R J Scaramuzzi The ovarian expression of mRNAs for aromatase, IGF-I receptor, IGF-binding protein-2, -4 and -5, leptin and leptin receptor in cycling ewes after three days of leptin infusion Reproduction, December 1, 2005; 130(6): 869 - 881. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Hayashi, K. D Carpenter, T. H Welsh Jr, R. C Burghardt, L. J Spicer, and T. E Spencer The IGF system in the neonatal ovine uterus Reproduction, March 1, 2005; 129(3): 337 - 347. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |