Paracrine Inducers of Uterine Endometrial Spermidine/Spermine N¹-Acetyltransferase Gene Expression during Early Pregnancy in the Pig¹

^a Departments of Animal Science, ^b Physiological Sciences, and ^c Dairy&Poultry Sciences, and the Interdisciplinary Concentration in Animal Molecular and Cell Biology, University of Florida, Gainesville, Florida 32611

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The endogenous factors that underlie the transient induction of the gene encoding spermidine/spermine N¹-acetyltransferase (SSAT), the rate-limiting enzyme in cellular polyamine catabolism, in pig uterine endometrium during periimplantation are not known. The present study examined a number of peptide growth factors and regulatory molecules that are present within the uterine environment at early pregnancy, coincident with maximal SSAT gene expression, for their ability to manifest endogenous SSAT gene-inducing activity. Basal SSAT expression in luminal epithelial cells was higher (p < 0.01) than that for glandular epithelial (GE) or stromal (ST) cells. Recombinant human insulin-like growth factor-I (IGF-I; 50 ng/ml) had no effect on steady-state SSAT mRNA levels, but it increased mitogenesis in all three cell types. In contrast, IGF-I caused a marked induction (p < 0.01) of SSAT mRNA levels in the human endometrial carcinoma cell line Hec-1-A. Uterine explants incubated with interleukin-6, transforming growth factor , epidermal growth factor (each at 1, 10, and 100 ng/ml), retinoic acid and retinol (each at 0.01, 0.1, and 1 µM), and estradiol-17ß (10 nM) had SSAT mRNA levels similar to controls. By contrast, leukemia inhibitory factor (LIF; at 10 and 100 ng/ml) caused a modest, but significant (p < 0.05), increase in SSAT mRNA levels over those of untreated explants. This effect of LIF, however, did not approach the level of induction observed in GE or ST cells after addition of medium conditioned by Day 12 or 17 porcine conceptuses and in endometrial explants supplemented with medium conditioned by Day 21 porcine conceptuses or a continuous cell line (Jag-1) derived from Day 14 porcine trophoblast. We suggest that transient induction of endometrial SSAT gene expression at implantation is mediated by the functional interactions of specific conceptus-derived regulatory factors, distinct from estrogen, with endometrial-derived factor(s) such as LIF. These complex interactions are probably requisite for the transient, yet dramatic, induction of SSAT gene expression and may be critical for successful implantation.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The complex mechanisms underlying uterine and embryonic growth and differentiation involve a diverse array of uterine endometrial- and conceptus-secreted proteins, the interactions of which are presumed to provide an optimal environment requisite for successful implantation [1]. Thus, the identification of uterine and conceptus genes and corresponding products that are highly regulated by changes in the uterine microenvironment has become an important initial approach to assessing the biological functions and modes of action of these molecules in periimplantation events. One unbiased technique that has proved successful in this regard is the use of mRNA differential display [2], which enables the isolation of genes that are expressed differentially in different physiological states. The subsequent analyses of these genes to confirm their preferential sites of synthesis, extent of differential expression, and specificity in cell-type regulation have to date provided important insights into biological functions not otherwise predicted by the candidate gene approach. Following this strategy, our laboratory has identified spermidine/spermine N¹-acetyltransferase (SSAT) as a porcine endometrial gene whose expression is preferential to gravid endometrium and maximal at early pregnancy, temporally coincident with conceptus elongation [3].

Unlike other identified porcine uterine endometrial proteins, which are largely secreted [1, 4–6], SSAT is intracellular in location and mediates the acetylation of polyamines, thus promoting their excretion from cells [7]. The maintenance of appropriate levels of intracellular polyamines is crucial to the balance between cell growth and differentiation, since polyamines mediate DNA synthesis, cell cycle events, and apoptosis via a variety of mechanisms that include interactions with key regulatory factors including the estrogen receptor [8], control of growth factor secretion [9] and action [10], and induced expression of immediate early and cyclin genes [11, 12]. Indeed, overexpression of human SSAT in Escherichia coli retarded cellular growth, while addition of polyamines to these cells abrogated the effect of increased SSAT [13]. At early pregnancy, when tissue differentiation is likely to be favored over growth, a conducive environment for successful implantation may require decreased intracellular polyamine concentrations. Thus, a rationale exists for the temporal and rapid increase in SSAT gene expression observed within this period. The more recent finding that female mice that overexpress SSAT are infertile and exhibit a hypoplastic uterus [14] is consistent with the importance of highly regulated SSAT gene expression for successful pregnancy.

The transient, albeit dramatic, increase in porcine uterine endometrial SSAT gene expression during the dynamic period of conceptus elongation suggests that a conceptus-derived product(s) may be responsible in part for inducing SSAT gene expression at this time. This is underscored by the finding that conditioned medium (CM) from Day 12 filamentous, but not spherical, conceptuses increased SSAT mRNA levels in Day 12 pregnant endometrial explants [3]. However, a number of growth and regulatory factors previously identified to be synthesized by the endometrium may also mediate the induction of endometrial SSAT. In the present study, we examined the SSAT gene-inducing activity of a number of peptide growth factors and regulatory molecules in vitro using endometrial explants and primary cultures of isolated endometrial cells. The level of induction observed with these factors was compared to that of CM from porcine conceptuses at different developmental stages or from a porcine trophoblast-derived cell line, Jag-1 [15].

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Materials

Recombinant human insulin-like growth factor-I (IGF-I), recombinant human interleukin-6 (IL-6), and recombinant human transforming growth factor (TGF) were purchased from Upstate Biotechnology, Inc. (Lake Placid, NY). Recombinant human epidermal growth factor (EGF) was from R&; Systems (Minneapolis, MN). The nick-translation kit was from Amersham Corp. (Arlington Heights, IL). [-³²P]Deoxycytidine triphosphate (SA 3000 Ci/mmol), [³H&; (SA 6.7 Ci/mmol), and BioTrans nylon membranes (0.2 µm) were purchased from ICN Radiochemicals (Irvine, CA). All molecular biology-grade chemicals were purchased from Fisher Scientific (Pittsburgh, PA). TRIzol and culture media were purchased from Gibco BRL (Gaithersburg, MD). Tissue culture dishes (30-mm 6-well plates) were from Costar Corp. (Cambridge MA). Recombinant human leukemia inhibitory factor (LIF), all trans-retinol, retinoic acid, antibiotic-antimycotic (ABAM), and fetal bovine serum (FBS) were from Sigma Chemical Co. (St. Louis, MO).

Animals

Animal-use protocols were approved by the University of Florida Institutional Animal Care and Use Committee. Prepubertal crossbred gilts were monitored daily with mature boars for the onset of pubertal estrus. The day of onset of estrus was defined as Day 0. Gilts exhibiting estrous cycles of normal duration (18–22 days) were mated with boars at first observed estrus and again 12 and 24 h later to ensure a fertile mating. Gilts were killed at the University of Florida Meats Processing Facility on Days 12, 17, and 21 of pregnancy or on Days 0 and 12 of the estrous cycle. Reproductive tracts were excised immediately after exsanguination, immersed in ice, and transferred to a laminar flow hood, where uterine horns were trimmed free of mesometrium and the conceptuses were flushed from each uterine horn. Conceptuses were cultured, and endometrium was used for explant culture and/or primary cell culture as described below.

Preparation of Conditioned Media

To demonstrate direct conceptus contribution to SSAT gene expression, medium was conditioned by conceptuses at pre- (Day 12), late- (Day 17)- and post- (Day 21) implantation stages and tested in primary cultures of GE and ST cells or explants from Day 12 pregnant endometrium. Conceptuses (embryo and associated membranes) from Day 12, 17, and 21 pregnant gilts were aseptically flushed from each uterine horn with 20 ml of sterile PBS (137 mM NaCl, 2.7 mM KCl, 10.1 mM Na₂HPO₄, 1.8 mM KH₂PO₄; pH 7.4) containing 2% ABAM. Conceptuses were gently washed in PBS-2% ABAM before culture. For Day 12 conceptuses, the entire litter of conceptuses was cultured for 24 h in 10 ml of minimal essential medium (MEM) supplemented with nonessential amino acids, glutamine (292 mg/L), MEM vitamin solution (10 ml/L), glucose (3.0 g/L), and 1% ABAM. For Days 17 and 21 of pregnancy, 3–5 conceptuses (including the embryo proper) were pooled and cultured in 15 ml of the same medium. Culture conditions were exactly as described for uterine explant cultures (see below). The possibility that trophoblast cells were a major source of SSAT gene-inducing activity was examined using CM from a continuous cell line (Jag-1; courtesy of J. Ramsoondar, Texas A&; University) derived from Day 14 porcine trophoblast. For preparation of Jag-1 CM, confluent monolayers of Jag-1 cells were washed with several volumes of Hank's balanced salt solution (HBSS) and incubated at 37°C in an atmosphere of 95% air:5% CO₂ for 48 h in serum-free Medium 199 containing phenol red dye [15]. After culture, CM was clarified by centrifugation at 2000 x g for 5 min and frozen at -20°C until analyzed.

Cell Cultures

Porcine endometrial luminal epithelial (LE), glandular epithelial (GE), and stromal (ST) cells were isolated from Day 12 pregnant uterine endometrium by enzymatic dispersion and sieve-filtration [16]. Primary cells were grown in RPMI 1640 supplemented with 10% FBS, 0.25 U/ml insulin, and 1% ABAM. Human endometrial adenocarcinoma cells (Hec-1-A; American Type Culture Collection, ATCC, Rockville, MD) were included in the study to determine whether SSAT induction by conceptus-derived CM was specific to pig uterine cells and to verify the activity of the molecules tested. Phorbol 12-myristate 13-acetate (PMA) and spermidine, two known inducers of SSAT [7], were included as positive controls. Hec-1-A cells were grown in McCoy's 5A medium supplemented with 10% FBS [17]. Medium was changed every other day. At 80–85% confluence, cells were washed with HBSS and cultured in serum-free, insulin-free RPMI 1640 (primary cells) or serum-free McCoy's 5A (Hec-1-A). After 24 h of incubation, fresh medium supplemented with the indicated treatment(s) was added, and cells were cultured for the indicated times.

Uterine Explant Cultures

Uterine horns from Day 0 and 12 pregnant pigs were opened longitudinally, and endometrial tissue was excised, washed in several volumes of PBS-2% ABAM, and minced into 1- to 3-mm³ sections. Explants (250–300 mg) were cultured in 3 ml of modified MEM (same as described for conceptus cultures) in 6-well tissue culture plates on a rocking platform at 37°C under an atmosphere of 50% N₂:47.5% O₂:2.5% CO₂. After a 2-h preincubation, medium was aspirated, and fresh medium supplemented with the indicated treatment(s) was added. The explants were further incubated in treatment media for 3–24 h under the same conditions. Treatments included uterine flushings, conceptus CM and Jag-1 CM (described above) at 50% v:v; IGF-I at 50 ng/ml, a dose representative of luminal IGF-I concentrations on Day 12 of pregnancy [18–20]; retinol and its derivative, retinoic acid, at 0.01, 0.1 and 1 µM; TGF, EGF, and IL-6 at 1, 10, and 100 ng/ml; and LIF at 10 and 100 ng/ml. The putative cooperative interactions among estradiol-17ß, IGF-I, and IL-6 were tested by co-addition of these molecules in a 2³ factorial arrangement of treatments. The choice of regulatory molecules tested was based on two criteria: 1) these molecules exhibit coincident expression with SSAT during early pregnancy, and 2) these molecules have been previously demonstrated to alter polyamine biosynthesis in other mammalian systems or have the potential to interact with putative cis-acting elements identified in the 5'-regulatory region of the mammalian SSAT gene [7].

RNA Isolation and Analyses

Total cellular RNA was isolated from uterine explants and primary culture cells with TRIzol according to the manufacturer's specifications. Northern blot analysis was performed as previously described [3]. For dot blot analyses, 5 or 10 µg of total RNA was blotted onto nylon membranes using a vacuum manifold apparatus (Schleicher and Schuell, Keene, NH). All membranes were hybridized to a ³²P-labeled gel-purified porcine SSAT cDNA insert as previously described [3]. Hybridization signals were quantified by phosphorimage analysis (Molecular Dynamics, Sunnyvale, CA). All membranes were stripped of radiolabeled SSAT probe [3] and rehybridized to either a nick-translated human ß-actin cDNA probe (ATCC, Rockville, MD) or a random-primed cDNA corresponding to the coding region of porcine 18S rRNA (courtesy of O.P. Perera, University of Florida) to confirm equal loading of RNA.

Thymidine Incorporation Assays

Cells were grown to confluence in 6-well culture dishes and then incubated in the appropriate serum-free medium for 24 h. Monolayers were washed with PBS, and medium containing the indicated treatments (50 ng/ml IGF-I or Jag-1 CM) was added. After 20 h, 2 µCi of [³H&; was then added. After labeling for 4 h, cells were washed with PBS, and trichloroacetic acid (TCA) precipitations were performed. Radioactivity incorporated into the TCA-precipitable material was determined by liquid scintillation counting.

Statistical Analyses

All hybridization volumes obtained from phosphorimage analysis were subjected to ANOVA using the General Linear Models procedures of the Statistical Analysis System [21]. For significant effects of treatment, dose, time, and all appropriate interactions, orthogonal contrasts or the Student-Neuman-Keuls sequential range test were used for specific mean comparisons. All results presented are the least-squares means ± SEM and have been adjusted for differences in RNA loading by a covariate analysis with either ß-actin or 18S rRNA hybridization volumes.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Source of SSAT Gene-Inducing Activity in the Uterine Microenvironment

To confirm the observed association of induced endometrial SSAT mRNA levels with developmental status of periimplantation conceptuses [3], flushings from uteri bearing either filamentous or spherical conceptuses at Day 12 of pregnancy or from pigs at Day 12 of the estrous cycle were added (50% v:v) to uterine endometrial explants. Endometrial explants cultured for 24 h with flushings from uteri containing filamentous conceptuses had increased (p < 0.01) steady-state levels of SSAT mRNA, relative to control explants (Fig. 1). In contrast, flushings from uteri with spherical conceptuses or from cyclic endometrium only tended (p < 0.08) to induce SSAT mRNA levels.

View larger version (23K): [in this window] [in a new window]   FIG. 1. Presence of endometrial SSAT gene-inducing activity in uterine luminal flushings. Endometrium from Day 12 pregnant pigs was cultured for 24 h in PBS or with uterine flushings collected from a Day 12 nonpregnant uterus (NP) and Day 12 pregnant uteri bearing spherical (SPH; n = 2) or filamentous (FIL; n = 2) conceptuses. Results represent the adjusted least-squares means ± SEM for 3 explant cultures per uterine flushing. Asterisks denote differences between explants treated with PBS and uterine flushings; p < 0.08 (*), p < 0.01 (**).

The presence of SSAT gene-inducing activity in uterine flushings is consistent with an inducing factor(s) that is a secretory product of the endometrium and/or conceptus. Media conditioned by Day 12 and 17 conceptuses increased (p < 0.05) steady-state levels of SSAT mRNA in primary cultures of endometrial GE and ST cells (Fig. 2A), demonstrating a direct conceptus contribution to SSAT gene expression. Similarly, media conditioned by Day 21 conceptuses increased (p < 0.01) SSAT gene expression in endometrial explants (Fig. 2B), at levels similar to that previously reported for CM from Day 12 filamentous conceptuses [3]. Furthermore, endometrial explants and isolated GE, but not ST, cells exhibited increased SSAT gene expression (p < 0.05) over corresponding controls, upon 24-h incubation in medium containing Jag-1 CM (Fig. 3, A and B).

View larger version (34K): [in this window] [in a new window]   FIG. 2. Effect of conceptus-conditioned media on SSAT gene expression in primary cultures of isolated pig endometrial cells or endometrial explants. A) GE and ST cells isolated from Day 12 pregnant pig uterine endometrium were cultured for 6 h in MEM conditioned by Day 12 porcine conceptuses (d12 CM; n = 2), Day 17 porcine conceptuses (d17 CM; n = 2), or MEM alone (MEM). Results represent the adjusted least-squares means ± SEM for cells isolated from 2 different pigs. Bars labeled with different superscripts differ (p < 0.05). B) Endometrium from Day 12 pregnant pigs was cultured for 6 h in the presence of medium conditioned by Day 21 porcine conceptuses (d21 CM). Results are from 3 independent experiments (n = 3 pigs), and CM obtained from 3 different litters of conceptuses. * Difference (p < 0.01) between control and d21 CM.

View larger version (35K): [in this window] [in a new window]   FIG. 3. Effect of media conditioned by a porcine trophoblast cell line (Jag-1) on endometrial SSAT gene expression. Primary cultures of isolated endometrial GE and ST cells (A) or endometrium from Day 12 pregnant pigs (B) were incubated for 6 h in media conditioned by Jag-1 cells (Jag-1 CM) or control medium (Medium 199). Results (least-squares means ± SEM) were obtained for cells isolated from 2 different pigs (A) or for endometrial explants isolated from 2 different pigs, and for each pig, n = 3 explant cultures per treatment (B). * Differences (p <0.05 for A; p < 0.01 for B) between treated and control cells or explants.

SSAT Gene Expression in Human Endometrial Hec-1-A Cells

To determine whether the inductive effect of conceptus-derived CM on SSAT mRNA levels was specific to pig uterine cells, the human endometrial carcinoma cell line Hec-1-A was treated with increasing doses (v:v) of Jag-1 CM. In addition, because SSAT expression is negatively correlated with cell growth [3, 7, 13], the postulated inhibitory effect of increasing concentrations of Jag-1 CM on Hec-1-A mitogenesis was also examined. Increasing concentrations of Jag-1 CM increased steady-state levels of SSAT mRNA in a dose-dependent fashion (Fig. 4A), while decreasing cell proliferation (Fig. 4B) of Hec-1-A cells. Moreover, spermidine, retinol, IGF-I, and PMA, but not estradiol-17ß, increased (p < 0.05) steady-state levels of SSAT mRNA in Hec-1-A cells after 12 h of culture. This induction was maintained (p < 0.05) for 24 h, but only with spermidine and IGF-I treatments (Fig. 5).

View larger version (33K): [in this window] [in a new window]   FIG. 4. Dose-dependent increase in steady-state levels of SSAT mRNA and concomitant inhibition of DNA synthesis in human endometrial carcinoma cells (Hec-1-A) treated with Jag-1 CM. A) Confluent monolayers were cultured for 24 h with increasing concentrations of Jag-1 CM. B) Duplicate plates were treated exactly as described above, except that cells were labeled with 2 µCi of [3H&; per well during the final 4 h of culture. Results represent the least-squares means ± SEM of 3 experiments, in which each experiment represents a different passage of Hec-1-A cells, and for each experiment, n = 3 wells per dose of CM. Means without a common superscript differ, p < 0.05.

View larger version (40K): [in this window] [in a new window]   FIG. 5. Inducers of SSAT gene expression in Hec-1-A cells. Subconfluent cultures of Hec-1-A cells were incubated in treatment media, and SSAT mRNA levels were examined by dot blot hybridization. Results are the adjusted least-squares means ± SEM for 3 experiments, in which each experiment represents a different passage of cells, and within each experiment, n = 3 wells of cells per treatment. Means without a common superscript differ, p < 0.05.

Effect of IGF-I on In Vitro Expression of SSAT Gene

To examine the question whether IGF-I contributes to the SSAT gene-inducing activity present in uterine flushings containing filamentous conceptuses, IGF-I was added to primary cultures of endometrial GE, LE, and ST cells isolated from Day 12 pregnant endometrium. Incubation with IGF-I for 24 h did not affect (p > 0.10) SSAT mRNA levels in any of the primary cell types examined (Fig. 6A). Consistent with previous findings [3], differences in basal steady-state levels of SSAT mRNA with cell types were observed; LE cells exhibited higher (p < 0.01) SSAT mRNA abundance than did GE or ST cells. The lack of an IGF-I inductive effect was not due to absence of cell responsiveness to IGF-I, since each cell type had 2- to 3-fold increased DNA synthesis with added IGF-I (Fig. 6B). IGF-I also did not affect SSAT mRNA levels in 12- to 24-h explant cultures of endometrium obtained at Day 12 of pregnancy and Day 0 of the estrous cycle, time points characterized by high and low intrauterine content of IGF-I, respectively (data not shown). By contrast, Hec-1-A cells had increased (p < 0.05) SSAT mRNA levels within 12 h of IGF-I addition, and this increase was maintained for 24 h (Fig. 5).

View larger version (24K): [in this window] [in a new window]   FIG. 6. Effect of IGF-I on steady-state levels of SSAT mRNA in primary cultures of porcine endometrial cells. A) Subconfluent primary cultures of GE, LE, and ST cells isolated from Day 12 pregnant uterine endometrium were incubated for 24 h in medium alone (Control) or medium supplemented with IGF-I. Steady-state SSAT mRNA levels were examined by dot blot hybridization. Bars represent the adjusted least-squares means ± SEM of cells isolated from 3 individual gilts. B) Effect of IGF-I on [3H&; incorporation in porcine uterine endometrial cells. Confluent GE, LE, and ST cells were treated with IGF-I for 20 h, and labeled with [3H&; for an additional 4 h. Bars represent least-squares means ± SEM of cells isolated from 3 gilts for GE and LE cells and 2 gilts for ST cells.

Effect of Retinoids on In Vitro Expression of SSAT Gene

Retinol and its derivative, retinoic acid, are found at significant levels in uterine flushings containing periimplantation conceptuses [22, 23]. To test whether these regulatory molecules are responsible in part for the SSAT gene-inducing activity of uterine flushings at Day 12 of pregnancy, explants of Day 12 pregnant endometrium were incubated with retinol or retinoic acid (0.01, 0.1, 1 µM) for 3, 6, 12, and 24 h. Neither factor at the concentrations tested had any effect on basal levels of SSAT mRNA levels (Fig. 7). SSAT mRNA abundance for primary cultures of GE cells isolated from Day 12 pregnant endometrium and Day 0 explant cultures was similarly unaffected by retinoic acid (0.1 µM) treatment for 1, 3, 6, and 24 h (data not shown). However, retinol increased (p < 0.05) steady-state levels of SSAT mRNA in Hec-1-A cells after 12 h (Fig. 5).

View larger version (53K): [in this window] [in a new window]   FIG. 7. Steady-state levels of SSAT mRNA in endometrial explants cultured in the presence of retinoids. Endometrial explants were isolated from Day 12 pregnant endometrium and cultured in the presence of increasing concentrations of all trans-retinol (left panel) or retinoic acid (right panel) for the indicated times. Results represent the adjusted least-squares means ± SEM of 2 experiments.

Effects of IL-6, TGF, EGF, and LIF on SSAT Gene Expression In Vitro

Other growth factors and cytokines known to be present in porcine uterine luminal fluids and synthesized by the endometrium, conceptus, or both [24–27], were examined for their effects on SSAT gene expression in vitro. Explants from Day 12 pregnant endometrium, incubated with IL-6 (1, 10, and 100 ng/ml) for 3, 6, and 15 h, had SSAT mRNA levels similar (p > 0.10) to those of untreated explants (Fig. 8A). Incubation of explants for 6 and 24 h with different combinations of IGF-I, IL-6, and estrogen also had no effect on SSAT mRNA relative to that of controls (Fig. 8B). TGF and EGF, alone or in combination, similarly had no effect (p > 0.10) on SSAT mRNA levels in endometrial explants (Fig. 8C). In contrast, LIF caused a modest increase (p < 0.05) in SSAT mRNA abundance after 6 and 24 h of culture (Fig. 8D). However, the effects of LIF did not approximate the levels of induction observed with flushings from uterine endometrium with filamentous conceptuses (Fig. 1) or medium conditioned by conceptuses or conceptus trophoblastic cells (Fig. 2).

View larger version (29K): [in this window] [in a new window]   FIG. 8. Effect of growth factors, cytokines, and steroids on endometrial SSAT gene expression. Uterine endometrial explants from Day 12 pregnant pigs were cultured in the presence or absence of increasing concentrations of human recombinant IL-6 (A); IGF-I, estradiol-17ß (E2) and IL-6 in different combinations (B); human recombinant TGF, EGF, and the combination thereof (C); or human recombinant LIF (D). Results represent the adjusted least-squares means ± SEM for 2 experiments, in which each experiment represents endometrium isolated from a different pig, and for each pig, n = 3 explant cultures per treatment.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The regulation of intracellular polyamine content is a complex process that is tightly linked to the expression of several anabolic and catabolic enzymes, the loss of function of which results in phenotypes of altered growth and differentiation [14, 28]. One such enzyme is SSAT, a protein that acetylates spermine and spermidine, resulting in their cellular excretion. Our previous finding that endometrial levels of SSAT are dramatically increased during early pregnancy, coincident with the rapid period of conceptus development, and subsequently become undetectable by later pregnancy stages, suggests that control of SSAT gene expression is physiologically important to periimplantation events. In the present study, we addressed the identity of molecules potentially responsible for the endogenous SSAT gene-inducing activity at early pregnancy, by measuring expression levels of the SSAT gene in vitro using primary cell culture systems and tissue explants from Day 12 pregnant endometrium, upon addition of a number of candidate regulatory factors known to be present in the porcine uterine microenvironment. Our results indicate that the endogenous endometrial SSAT gene-inducing activity at early pregnancy may comprise multiple factors that represent, in part, molecules that 1) are secreted by conceptuses at different developmental stages, one or more of which may be synthesized by the trophoblast; 2) are distinct from IGF-I, IL-6, TGF, EGF, estrogens, and the retinoids; 3) are of endocrine origin; and 4) are endometrium-derived; one of these is LIF. Taken together, these data suggest the potential autocrine, endocrine, and paracrine sources of endometrial SSAT gene inducers and underscores the complex regulation of this gene at periimplantation.

An interesting finding from these studies was the observed ability of medium conditioned by conceptuses of distinct periimplantation stages to induce endometrial SSAT gene expression. Thus, CM from Days 12, 17, and 21 conceptuses, as well as that from a cell line derived from Day 14 conceptus trophoblasts, exhibited inducer activity for primary cell cultures and explants derived from Day 12 pregnant endometrium. Although it is not known if the conceptus-secreted factors with SSAT gene-inducing activity are distinct for each stage of conceptus development, these results suggest that the inductive effect of the conceptus is maintained throughout the period of conceptus attachment and may reflect the importance of highly regulated polyamine levels during early pregnancy.

Surprisingly, despite their documented expression by the endometrium and/or the conceptus [6, 23–27, 29], temporally coincident with endometrial SSAT gene expression at early pregnancy, regulatory factors that include retinoids, estrogen, IGF-I, TGF, EGF, and IL-6 did not mimic conceptus induction of endometrial SSAT gene expression. Indeed, of those tested, only LIF caused an increase in SSAT mRNA levels, albeit its activity did not approximate that for CM from conceptuses. This effect of LIF was obtained at doses similar to those found to effect biological responses in some other systems [30]. Anegon et al. [26] reported the presence of LIF in uterine flushings from endometrium of pigs at Day 12 of pregnancy and the estrous cycle, with levels higher at pregnancy than during the estrous cycle, a finding consistent with the SSAT gene-inducing activity of these uterine flushings. The source of LIF in Day 12 flushings appears to be the endometrium, since periimplantation pig conceptuses do not express this gene, and LIF transcripts from corresponding endometrium were detected by reverse transcription-polymerase chain reaction [26, 27]. These results suggest an autocrine/paracrine role for LIF and its potential interactions with conceptus-derived factors, in modulating SSAT gene expression.

Previously, we suggested that IGF-I may induce uterine SSAT gene expression, on the basis of the following observations from this and other laboratories [3]. First, IGF-I concentrations in luminal flushings are higher in uteri bearing filamentous conceptuses than in uteri bearing spherical conceptuses [20], a pattern similar to that observed for endometrial steady-state SSAT mRNA levels. Second, IGF-I concentrations are higher in uterine flushings from the gravid than the nongravid horn of unilaterally pregnant pigs [19], the same experimental model previously used to identify SSAT as a conceptus-induced gene [3]. Third, progesterone increased uterine endometrial expression of both SSAT [3] and IGF-I [19] genes in ovariectomized gilts. Further, IGF-I has been shown to increase ornithine decarboxylase mRNA levels and enzyme activity in human breast cancer cells [31], resulting in subsequent increases in intracellular levels of spermine and spermidine, which are known stimulators of SSAT expression. This postulated linkage of IGF-I and endometrial SSAT gene expression, however, was not supported by the results presented here. Expression of the SSAT gene by endometrial explants recovered on estrus (Day 0) or Day 12 of pregnancy (representing low and high intrauterine content of IGF-I, respectively), was similarly unaffected by exogenous IGF-I. In contrast, IGF-I caused a marked increase in SSAT mRNA levels in the human endometrial cancer cell line, Hec-1-A, which was also responsive to induction of SSAT gene expression by the Jag-1 CM. Taken together, these results suggest that IGF-I is not responsible for the dramatic induction of SSAT in the early pregnancy endometrium and demonstrate distinct as well as common regulatory mechanisms underlying SSAT gene regulation in normal and cancer cells.

The temporal expression of the SSAT gene in pig uterus during pregnancy [3], the positive correlation of uterine SSAT with peripheral estrogens in the rat [32], and the resultant infertility associated with overexpression of SSAT in mice [14] point to pivotal role(s) for this enzyme in reproductive tract biology of multiple species. SSAT most likely has a permissive role in reproduction through its control of intracellular polyamine bioavailability. In turn, polyamines may participate in estrogen-mediated gene expression [8] and regulation of uterine cell growth and differentiation [33]. Paradoxically, overexpression of SSAT in transgenic mice is not accompanied by a concomitant decrease in the polyamine pools in reproductive tissues such as ovary, testes and uterus, although this may reflect long-term homeostatic changes [14]. Nevertheless, SSAT overexpression caused profound changes in the reproductive tract, such as a hypoplastic uterus and reduced number of tertiary follicles in the ovaries, indicating that growth of these tissues is tightly regulated by polyamine-mediated mechanisms.

The porcine periimplantation conceptus and corresponding uterine endometrium secrete a myriad of molecules, many of which are yet to be characterized and identified. Any one or a number of such molecules not tested in the present study, such as interferons, catecholamines, and catechol estrogens [34, 35], could potentially regulate endometrial SSAT gene expression during early pregnancy. The relative lack of success in assigning endogenous SSAT gene-inducing activity to a single or small group of factors, despite the systematic analysis of a number of these regulatory molecules based on their co-expression with endometrial SSAT gene, suggests a complex interaction among endometrium and conceptus-derived factors underlying the induction of SSAT gene expression. However, our observation of a strong inducing activity in media conditioned by conceptuses and by trophoblast cells reflects the active role of periimplantation conceptuses in establishing a uterine microenvironment conducive to their subsequent development. This model is further supported by the observed induction at sites of implantation of the porcine gene for secretory leukocyte protease inhibitor [36].

In conclusion, results of this study demonstrate that the major known secretory products of the endometrium and/or the developing conceptuses, which include estrogens, retinoids, and the growth factors IGF-I, EGF, TGF, and IL-6, are not responsible for the transient induction of endometrial SSAT gene expression during early pregnancy. Rather, this activity appears to reside in part with multiple factors that include conceptus-secreted molecules whose expression is maintained during the period of conceptus attachment, and endometrium-derived products, one of which may be LIF. The further identification of these individual factors and their modes of interaction is anticipated to contribute to an understanding of the control of normal and abnormal growth and development of reproductive, as well as nonreproductive, tissues by polyamines.

	ACKNOWLEDGMENTS

 
The authors gratefully acknowledge J. Ramsoondar (Texas A&M University) for the gift of the Jag-1 cell line and other members of our laboratories for assistance with animal breeding and management.

	FOOTNOTES

 
¹ This work was supported in part by USDA Postdoctoral Fellowship grant 95–37205–2315 (to M.L.G.), USDA grant 95–37206–2317 (to F.A.S. and R.C.M.S.), NIH grant HD21961 (to R.C.M.S), and a University of Florida Interdisciplinary Research Initiative grant. This is Journal Series No. R-06336 from the Florida Agricultural Experiment Station.

² Correspondence: R.C.M. Simmen, Department of Animal Science, University of Florida, P.O. Box 110910, Gainesville, FL 32611–0910. FAX: 352 392 7652; simmen{at}animal.ufl.edu

Accepted: July 8, 1998.

Received: April 1, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Simmen RCM, Simmen FA. Regulation of uterine and conceptus secretory activity in the pig. J Reprod Fertil Suppl 1990; 40:279–292.[Medline]
2. Liang P, Pardee AB. Differential display of eukaryotic messenger RNA by means of the polymerase chain reaction. Science 1992; 257:967–971.[Abstract/Free Full Text]
3. Green ML, Blaeser LL, Simmen FA, Simmen RCM. Molecular cloning of spermidine/spermine N¹-acetyltransferase from the periimplantation porcine uterus by messenger ribonucleic acid differential display: temporal and conceptus-modulated gene expression. Endocrinology 1996; 137:5447–5455.[Abstract]
4. Letcher R, Simmen RCM, Bazer FW, Simmen FA. Insulin-like growth factor-I expression during early conceptus development in the pig. Biol Reprod 1989; 41:1143–1151.[Abstract]
5. Simmen RCM, Michel FJ, Fliss AE, Smith LC, Fliss MFV. Ontogeny, immunocytochemical localization, and biochemical properties of the pregnancy-associated uterine elastase/cathepsin G protease inhibitor, antileukoproteinase (ALP): monospecific antibodies to a synthetic peptide recognize native ALP. Endocrinology 1992; 130:1957–1965.[Abstract/Free Full Text]
6. Simmen RCM, Green ML, Simmen FA. IGF system in periimplantation uterus and embryonic development. In: Molecular and Cellular Aspects of Periimplantation Processes. New York: Springer-Verlag, New York, Inc.; 1995: 185–204.
7. Casero RA Jr, Pegg AE. Spermidine/spermine N¹-acetyltransferase-the turning point in polyamine metabolism. FASEB J 1993; 7:653–661.[Abstract]
8. Thomas T, Gallo MA, Klinge CM, Thomas TJ. Polyamine-mediated conformational perturbations in DNA alter the binding of the estrogen receptor to poly (dG-m⁵dC)·poly(dG-m⁵dC) and a plasmid containing the estrogen response element. J Steroid Biochem Mol Biol 1995; 54:89–99.[CrossRef][Medline]
9. Cohen FJ, Manni A, Glikman P, Bartholomew M, Demers L. Interactions between growth factor secretion and polyamines in MCF-7 breast cancer cells. Eur J Cancer 1990; 26:603–608.
10. Blachowski S, Motyl T, Grzelkowska K, Kasterka M, Orzechowski A, Interewicz B. Involvement of polyamines in epidermal growth factor (EGF), transforming growth factor (TGF)-alpha and -beta I action on culture of L6 and fetal bovine myoblasts. Int J Biochem 1994; 26:891–897.[CrossRef][Medline]
11. Wang JY, McCormack SA, Viar MJ, Wang H, Tzen CY, Scott RE, Johnson LR. Decreased expression of protooncogenes c-fos, c-myc and c-jun following polyamine depletion in IEC-6 cells. Am J Physiol 1993; 265:G331–338.
12. Thomas T, Thomas TJ. Regulation of cyclin B1 by estradiol and polyamines in MCF-7 breast cancer cells. Cancer Res 1994; 54:1077–1084.[Abstract/Free Full Text]
13. Parry L, Lopez-Ballester J, Weist L, Pegg AE. Effect of expression of human spermidine/spermine N¹-acetyltransferase in Escherichia coli. Biochemistry 1995; 34:2701–2709.[CrossRef][Medline]
14. Pietila M, Alhonen L, Halmekyto M, Kanter P, Janne J, Porter CW. Activation of polyamine catabolism profoundly alters tissue polyamine pools and affects hair growth and female fertility in transgenic mice overexpressing spermidine/spermine N¹-acetyltransferase. J Biol Chem 1997; 272:18746–18751.[Abstract/Free Full Text]
15. Ramsoondar J, Christopherson RJ, Guilbert LJ, Wegmann TG. A porcine trophoblast cell line that secretes growth factors which stimulate porcine macrophages. Biol Reprod 1993; 49:681–694.[Abstract]
16. Reed KL, Badinga L, Davis DL, Chung TE, Simmen RCM. Porcine endometrial glandular epithelial cells in vitro: transcriptional activities of the pregnancy-associated genes encoding antileukoproteinase and uteroferrin. Biol Reprod 1996; 55:469–477.[Abstract]
17. Tabibzadeh S, Kaffka KL, Killian PL, Satyaswaroop PG. Human endometrial cell lines for studying steroid and cytokine actions. In Vitro Cell Dev Biol 1990; 26:1173–1179.[Medline]
18. Tavakkol A, Simmen FA, Simmen RCM. Porcine insulin-like growth factor-I (pIGF-I): complementary deoxyribonucleic acid cloning and uterine expression of messenger ribonucleic acid encoding evolutionarily conserved IGF-I peptides. Mol Endocrinol 1988; 2:674–681.[Abstract/Free Full Text]
19. Simmen RCM, Simmen FA, Hofig A, Farmer SJ, Bazer FW. Hormonal regulation of insulin-like growth factor gene expression in pig uterus. Endocrinology 1990; 127:2166–2174.[Abstract/Free Full Text]
20. Green ML, Simmen RCM, Simmen FA. Developmental regulation of steroidogenic enzyme gene expression in the periimplantation porcine conceptus: a paracrine role for insulin-like growth factor-I. Endocrinology 1995; 136:3961–3970.[Abstract]
21. Barr AJ, Goodnight JH, Sall JP, Blair WH, Chilko DM. SAS User's Guide. General Linear models procedure of the statistical analysis system. Cary, NC: Statistical Analysis System Institute, Inc.; 1979.
22. Bazer FW, Simmen RCM, Simmen FA. Comparative aspects of conceptus signals for maternal recognition of pregnancy. The primate endometrium. Ann NY Acad Sci 1991; 622:202–211.[Medline]
23. Trout WE, McDonnell JJ, Kramer KK, Baumbach GA, Roberts RM. The retinol-binding protein of the expanding pig blastocyst: molecular cloning and expression in trophectoderm and embryonic disc. Mol Endocrinol 1991; 5:1533–1540.[Abstract/Free Full Text]
24. Mathialagan N, Bixby JA, Roberts RM. Expression of interleukin-6 in porcine, ovine, and bovine preimplantation conceptuses. Mol Reprod Dev 1992; 32:324–330.[CrossRef][Medline]
25. Vaughan TJ, James PS, Pascall JC, Brown KD. Expression of the genes for TGF, EGF and the EGF receptor during early pig development. Development 1992; 116:663–669.[Abstract]
26. Anegon I, Cuturi MC, Godard A, Moreau M, Terqui M, Martinat-Botte F, Soulillou JP. Presence of leukaemia inhibitory factor and interleukin 6 in porcine uterine secretions prior to conceptus attachment. Cytokine 1994; 6:493–499.[CrossRef][Medline]
27. Modric T, Green ML, Simmen RCM, Simmen FA. Expression and function of leukemia inhibitory factor and interleukin-6 genes during porcine pregnancy. Biol Reprod 1997; 56(suppl. 1):202.
28. Hakovirta H, Keiski A, Toppari J, Halmekyto M, Alhonen L, Janne J, Parvinen M. Polyamines and regulation of spermatogenesis: selective stimulation of late spermatogonia in transgenic mice overexpressing the human ornithine decarboxylase gene. Mol Endocrinol 1993; 7:1430–1436.[Abstract/Free Full Text]
29. Perry JS, Heap RB, Amaroso EC. Steroid hormone production by pig blastocysts. Nature 1973; 245:45–47.[CrossRef][Medline]
30. Koshimizu U, Taga T, Watanabe M, Saito M, Shirayoshi Y, Kishimoto T, Nakatsuji N. Functional requirement of gp130-mediated signaling for growth and survival of mouse primordial germ cells in vitro and derivation of embryonic germ (EG) cells. Development 1996; 122:1235–1242.[Abstract]
31. Huber M, Poulin R. Permissive role of polyamines in the cooperative action of estrogens and insulin or insulin-like growth factor-I on human breast cancer cell growth. J Clin Endocrinol Metab 1996; 81:113–123.[Abstract]
32. Sessa A, Perin A. Spermidine N¹-acetyltransferase activity in rat uterus after 17ß-estradiol and during the estrous cycle. Biochim Biophys Acta 1991; 1074:31–35.[Medline]
33. Pegg AE. Recent advances in the biochemistry of polyamines in eukaryotes. Biochem J 1986; 234:249–262.[Medline]
34. La Bonnardiere C. Nature and possible functions of interferons secreted by the preimplantation pig blastocysts. J Reprod Fertil Suppl 1993; 48:157–170.[Medline]
35. Mondschein JS, Hersey RM, Dey SK, Davis DL, Weisz J. Catechol estrogen formation by pig blastocysts during the preimplantation period: biochemical characterization of estrogen-2/4-hydroxylase and correlation with aromatase activity. Endocrinology 1985; 117:2339–2346.[Abstract/Free Full Text]
36. Reed KL, Blaeser LL, Dantzer V, Green ML, Simmen RCM. Control of secretory leukocyte protease inhibitor gene expression in the porcine periimplantation endometrium: a case of maternal-embryo communication. Biol Reprod 1998; 5:448–457.