| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Animal Sciences and Interdisciplinary Concentration in Animal Molecular & Cell Biology, University of Florida, Gainesville, Florida 32611-0910
ABSTRACT
Insulin-like growth factor-I (IGF-I) and the polyamine catabolic enzyme spermidine/spermine N1-acetyltransferase (SSAT) are progesterone-regulated genes with maximal expression at peri-implantation in the porcine uterine endometrium. However, while IGF-I stimulates cell proliferation, SSAT, by acetylating the naturally occurring polyamines (PA) spermine (SPM) and spermidine (SPD), typically functions as a cell growth inhibitor. The present study examined the functional relationships of IGF-I, SSAT, and PA in the control of endometrial cell proliferation. Northern blot analysis indicated that SSAT mRNA levels change with distinct pregnancy stages, in contrast to those for the PA biosynthetic enzyme ornithine decarboxylase (ODC). Primary cultures of luminal and glandular epithelial (LE, GE) and stromal (ST) cells isolated from Day 12 pregnant pig endometrium had IGF-I mRNA levels for ST > LE > GE cells. The mRNA levels for SSAT and ODC were transiently diminished by IGF-I treatment, but only in GE cells. By contrast, SPM and SPD increased SSAT mRNA levels in GE and ST cells, but increased ODC mRNA levels only in GE cells. IGF-I, putrescine (PUT), and SPM individually increased cellular DNA synthesis as measured by tritiated thymidine incorporation in GE and ST cells, while SPD had an effect only in ST cells. IGF-I enhanced the proliferative effect of each PA in GE cells, but only of SPD in ST cells. The mitogen-activated protein kinase inhibitor, PD98059, inhibited the induction by SPM of GE cell DNA synthesis but not that of IGF-I. Wortmannin, a phosphatidylinositol-3-kinase inhibitor had no effect on either IGF-I or SPM induction of GE cell DNA synthesis. The relative concentrations of SPM, SPD, and PUT in uterine luminal fluids differed, with the levels for each PA higher at pregnancy Day 12 than at 11.5. These results suggest that IGF-I and PA act through distinct signaling pathways to mediate cell-type-specific growth of early pregnancy pig uterine endometrium. Further, SSAT, through its control of intracellular PA levels, likely plays a modulatory role in the establishment of an optimal uterine environment for successful embryo attachment.
embryo, growth factors, implantation, signal transduction, uterus
INTRODUCTION
Peri-implantation represents a critical period during early pregnancy when the developing embryo and maternal uterus must interact in a highly synchronous manner to ensure the proper initiation of embryo attachment/implantation leading to pregnancy success [1]. These orchestrated events are initiated, in part, by signal(s) from the blastocyst that, although distinct among different mammalian species [2–5], results in the inhibition of corpus luteum regression and hence, maintenance of critical levels of serum progesterone. Progesterone-regulated processes in reproduction are diverse [6] and in the uterine endometrium, these, in conjunction with the distinct effects of embryonic- or ovarian-derived estrogens, result in the synthesis and/or secretion of a variety of signaling molecules that include growth factors, cytokines, proteases, and protease inhibitors [7, 8]. These regulatory molecules likely mediate the integration of multiple signaling pathways that underlie the complex processes involved in embryo development, although the mechanisms by which these occur remain poorly understood.
One progesterone-regulated gene is that of insulin-like growth factor-I (IGF-I), a peptide regulatory molecule that is highly expressed in the uterine endometrium during early pregnancy in the pig [9, 10]. The endometrial expression of IGF-I is temporally correlated with the onset of rapid morphological changes in the porcine embryo and with peak embryonic estrogen production [11]. The presence of IGF-I receptors in porcine endometrial cells as well as in the embryo suggests paracrine and autocrine modes of action of IGF-I in the uterine microenvironment [8]. A role for IGF-I in embryonic growth and development has been gleaned from studies of IGF-I null mice [12, 13]. Moreover, female IGF-I (-/-) mice are distinguished by their characteristic hypoplastic uteri [14]. Taken together, these findings are consistent with IGF-I modulation of signaling pathways requisite for proper uterine and embryonic growth and development.
Using the mRNA differential display methodology [15], our laboratory previously identified spermidine/spermine N1-acetyltransferase (SSAT) as a porcine endometrial gene whose expression is maximal at peri-implantation and which is induced by conceptus-derived factors and by progesterone [16, 17]. SSAT is an intracellular catabolic enzyme that acetylates the naturally occurring polyamines (PA) spermidine (SPD) and spermine (SPM), thereby facilitating their back-conversion to putrescine (PUT) and/or promoting their excretion from cells [18]. The maintenance of appropriate intracellular concentrations of PA as being crucial for cell growth and proliferation has been demonstrated in vitro using SSAT overexpressing cells [19–21]; this role is likely due to the participation of these polycations in diverse cellular pathways that control DNA synthesis, cell cycle progression, DNA methylation, transcription factor activation, and apoptosis [22–28].
The recent demonstration that transgenic female mice overexpressing SSAT manifest hypoplastic/underdeveloped uteri and are consequently, infertile suggests an important role for PA in the control of endometrial tissue growth [29]. Moreover, the similarities in the uterine and reproductive phenotypes of IGF-1 null and SSAT transgenic mice imply a functional linkage between IGF-I and PA metabolism in pregnancy events. In the peri-implantation pig endometrium, however, the apparent coincidence in the maximal expression levels of SSAT and IGF-I mRNAs at Day 12 of pregnancy is not entirely consistent with the inverse correlation suggested between the functions of these molecules in the aforementioned mice models. To clarify further the relationship between IGF-I, SSAT, and PA in the control of endometrial proliferation, primary cultures of glandular and luminal epithelial (GE, LE) and stromal (ST) cells isolated from early pregnancy (Day 12) pig endometrium were examined for the expression of SSAT gene and, in parallel, that of the gene for the polyamine biosynthetic enzyme, ornithine decarboxylase (ODC), in response to exogenous IGF-I treatments. The effects of SPM and SPD on the expression of SSAT and ODC genes and in combination with IGF-I, on the proliferation of GE and ST cells, as measured by induction of DNA synthesis, were also evaluated. Further, using specific inhibitors of the mitogen-activated protein kinase (MAPK) and phosphatidylinositol-3-kinase (PI-3-K) pathways, we examined if the discrete effects of IGF-I and PA on cell proliferation are mediated by related or distinct signaling systems.
MATERIALS AND METHODS
Animals and Tissue Collection
Prepuberal cross-bred gilts were monitored daily with mature boars for the onset of estrus. Gilts exhibiting estrous cycles of normal duration (18–22 days) were mated with boars at estrus (defined as Day 0) and killed at the University of Florida Meat Processing Facility on Day 12 of pregnancy. Reproductive tracts were excised immediately, immersed in ice, and transferred to a sterile laminar flow hood where uterine horns were trimmed free of mesometrium, and the uterine endometrium separated from myometrium. Freshly isolated endometrial tissues were used for isolation of LE, GE, and ST cells (see below) or stored immediately at -80°C until used.
Primary Cell Culture
Endometrial LE, GE, and ST cells were isolated by enzymatic dispersion and sieve filtration, and cultured as previously described [30] at 37°C in an atmosphere of 5% CO2-95% air. Cells were grown in RPMI-1640 (Gibco-BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS), 0.25 U/ml insulin, and 1% antibiotic/antimycotic (ABAM) solution. Cells were left undisturbed for 3 days to allow attachment to the plates and then replenished with fresh medium every other day until they reached 95–100% confluence.
Mitogen Assay
Dispersed GE and ST cells were seeded into multiwell (six-well) plates and incubated until confluent monolayers were formed. The medium was aspirated, cells were washed with Hank balanced salt solution (HBSS) and then incubated with serum-free RPMI-1640 for 24 h. The medium was aspirated and the cells were incubated with the corresponding treatments for 20 h, followed by labeling with [3H]thymidine (1 µCi/well; ICN, Costa Mesa, CA) for 4 h. Cell-associated [3H]thymidine was determined following previously described protocols [17].
Treatments
Primary cultures of GE, LE, and ST cells, upon reaching confluence, were immediately transferred to serum-free, insulin-free RPMI-1640 containing 1% ABAM solution and incubated for an additional 24 h. Cells then received recombinant human IGF-I (rhIGF-I; 50 ng/ml) (Upstate Biotechnology, Lake Placid, NY), SPD, or SPM (50 µM each; Sigma Biochemicals, St. Louis, MO) for designated times in each study. The doses of PAs used in the study were within the range of concentrations found to stimulate SSAT gene expression [17] as well as other growth responses [24, 25] in other cell lines. Controls received no added IGF-I or PAs. Cells were recovered and immediately processed for RNA isolation. For mitogen assays, cells were treated with SPD, SPM, and PUT (50 µM each), alone or in combination with IGF-I (50 ng/ml). For studies using specific kinase inhibitors in the presence or absence of IGF-I and/or SPM, growth and treatment conditions were the same as described above, except that the inhibitors were added 1 h prior to IGF-I and/or SPM. The specific inhibitors used for MAPK and PI-3K pathways, respectively, were PD98059, a MEK-1 inhibitor (2.5 µM; Cell Signaling Technology, Beverly, MA), and wortmannin (2 nM; Upstate Biotechnology).
RNA Isolation and Northern Blot Analysis
Total RNA was isolated from cells using the TriZol reagent (Gibco-BRL), following the manufacturer's instructions. RNA (30 µg) was separated in a 1.5% agarose-formaldehyde gel in morpholinepropanesulfonic acid buffer [31], and sample integrity was evaluated by monitoring intensities of 18S and 28S ribosomal bands in ethidium bromide-stained gels. RNA was transferred to a Biotrans nylon membrane that was then exposed to a UV light source for 30 sec and baked for 20 min. Membranes were incubated for 2 h at 42°C in ultrasensitive hybridization (ULTRAhyb) buffer (Ambion Inc., Austin, TX) followed by an overnight incubation at 42°C in the same buffer containing the corresponding radiolabeled cDNA insert. In the case of hybridization with 18S rRNA DNA, the standard prehybridization (5x Denhardt, 4x SSC [1x = 0.15 M sodium chloride, 0.015 M sodium citrate], 50% formamide, 0.1% SDS, 250 µg/ml yeast RNA, 50 mM sodium phosphate) and hybridization (same as prehybridization buffer except for 1x Denhardt) buffers were used. Filters were sequentially washed in 2x SSC-0.1% SDS and in 0.1x SSC-0.1% SDS two times each at 42°C for 15 min and then exposed to x-ray film. The hybridization probes used included porcine ODC (263 base pairs [bp]), porcine SSAT (500 bp), porcine glyceraldehyde phosphate dehydrogenase (GAPDH; 593 bp), 18S rRNA DNA (100 bp), and porcine IGF-I (251 bp) cDNAs that were labeled with [32P]dCTP by nick translation.
Determination of Uterine Intraluminal PA Content
Polyamine levels in uterine flushes were analyzed by a combination of HPLC and positive ion electrospray ionization mass spectrometry of dansylated PA derivatives. Uterine luminal contents were collected by flushing each uterine horn from pigs at Days 11.5 and 12 of pregnancy with 20 ml of PBS. The flushes from each uterine horn of an animal were combined, and the total protein content was determined by the Bradford assay. Samples corresponding to 350 µg of total protein were precipitated in 1.2 N perchloric acid overnight at 4°C. The procedures for preparation of dansylated derivatives from the acidified samples (100 µl) followed those described by Bergeron et al. [32]. The samples were analyzed at the Spectroscopic Services Group (Department of Chemistry) of the University of Florida. Liquid chromatographic separation was achieved on a Waters Symmetry Shield RP18 column (2.1 x 150 mm; Milford, MA) with a mobile phase flow of 0.2 ml/min provided by a Beckman Instruments System Gold model 126 pump (Fullerton, CA). The mobile phases were: A = 1% acetic acid in water, and B = 1% acetic acid in methanol. The gradient of A:B = 80:20 (0–1 min), linear gradient of A:B = 5:95 over 30 min, and hold at A:B = 5:95 for 14 min, provided good separation of the PA derivatives. Mass spectrometry was performed on a Finnigan LCQ (San Jose, CA) operated in the electrospray ionization mode with a 3.5-kV spray voltage at a capillary temperature of 235°C. Polyamine dansylates in the uterine flushes were confirmed by comparison of their spectra to those obtained for dansylated SPM, SPD, and PUT standards, and concentrations were estimated from integrated peak areas of each.
Statistical Analyses
Numerical data obtained from densitometric analysis of hybridization bands, using the Alpha Imager 2000 Documentation & Analysis System (Alpha Innotech Corp, San Leandro, CA), were subjected to ANOVA following the general linear models procedures of the Statistical Analysis System (SAS) [33]. Results are presented as least-squares means ± SEM and were obtained from three independent experiments, where each experiment represented cells or tissues from a different gilt at the indicated day of pregnancy. Differences in RNA loading were adjusted by a covariate analysis with GAPDH or 18S rRNA hybridization values. Differences are considered significant at P 
0.05.
RESULTS
Uterine SSAT and ODC mRNA Levels Across Pregnancy
We previously demonstrated that steady-state levels of endometrial SSAT mRNA were high at Day 12 of pregnancy, and that these diminished to very low or undetectable levels at mid- to late pregnancy [16]. However, the pattern of SSAT gene expression specifically at peri-implantation has not been determined. To establish a physiological role for PA metabolic pathways mediated by SSAT and/or ODC during early pregnancy, Northern blot analysis was performed on total cellular RNA isolated from endometrium corresponding to pre (Days 10 and 12)-, peri (Days 14–15)-, and post (Days 18–19)-implantation and later pregnancy (Days 30, 60, 90) stages, using cloned porcine SSAT and ODC cDNA inserts as probes. Results indicated that for SSAT, the expression of both the mature (1.3-kilobase [kb]) and the preprocessed (3.5-kb) mRNA transcripts [18] were readily detectable from Day 10 to Day 19, after which time, a dramatic loss in expression was observed (Fig. 1A). By contrast, the transcript doublet of ca. 1.2–1.3 kb, characteristic of ODC mRNAs, was expressed relatively uniformly, when normalized to that of GAPDH, across pregnancy (Fig. 1B).
|
Endometrial Cell-Specific Expression of IGF-I mRNAs
Previously, we have shown that endometrial expression of SSAT is cell-type specific, with expression in LE > GE > ST cells of Day 12 pregnancy endometrium [16]. To correlate cell context-dependent expression of the SSAT gene with that of IGF-I, primary cultures of endometrial GE, LE, and ST cells isolated from pregnant gilts at Day 12 were analyzed for IGF-I gene expression. Northern blot analysis indicated that the levels of IGF-I mRNA were highest in ST > LE > GE cells, when normalized to the levels for GAPDH transcripts to account for sample loading differences (Fig. 2). The ST cells expressed all three well-characterized IGF-I transcripts of 8, 2.3, and 1.2 kb [8, 9]. The LE cells had detectable expression of the 8- and 1.2-kb transcripts, while only the smallest transcript (1.2 kb) was detected in GE cells (Fig. 2).
|
Insulin-Like Growth Factor-I Regulation of Endometrial SSAT and ODC mRNA Abundance
The above data suggested a correlation between SSAT and IGF-I gene expression as a function of cell type, with ST cells having the highest and the lowest IGF-I and SSAT gene expression, respectively, and GE cells coincidentally exhibiting low IGF-I and relatively high SSAT mRNA levels. To evaluate a direct relationship between local IGF-I concentration and SSAT gene expression, primary cultures of GE, LE, and ST cells were incubated with IGF-I (50 ng/ml) for 3, 6, and 12 h. We had previously demonstrated no effect of IGF-I on SSAT gene expression in these cells after 24 h of exogenous treatment [17]. Northern blot analysis for SSAT mRNA followed by densitometric analysis of resultant hybridization bands, indicated a temporal response to IGF-I by GE (Fig. 3), but not by LE and ST (data not shown), cells. Steady-state SSAT mRNA levels in GE cells were modestly reduced (P 0.05), relative to untreated controls, within 3 h of IGF-I addition. However, no significant change in SSAT mRNA levels was observed, relative to untreated cells, after either 6 or 12 h incubation with IGF-I (Fig. 3, left).
|
The effect of IGF-I on ODC gene expression was similarly evaluated in all three cell types. The same Northern blots used in the above experiments were stripped and subsequently hybridized with labeled porcine ODC cDNA. Within 3 h of IGF-I addition, a significant decrease (P 0.05) in steady-state ODC mRNA levels was observed in GE (Fig. 3, right), but not in LE or ST (data not shown) cells, relative to their respective control (untreated) cells. No significant change in ODC mRNA levels was observed with IGF-I treatment at any of the other time points examined.
Polyamine Regulation of Endometrial SSAT and ODC mRNA Levels
The ability of IGF-I to rapidly perturb the PA metabolic pathway genes in a cell-type-specific manner, through its regulation of the steady-state ODC and SSAT mRNA levels (above), and the well-documented findings that SPD and SPM are involved in growth-regulatory pathways [18] prompted an evaluation of SPD and SPM effects on SSAT and ODC gene expression in endometrial cells. Northern blot analysis for SSAT mRNAs (Fig. 4A) demonstrated a temporal response of GE cells to exogenously added SPD and SPM, with a significant increase in steady-state mRNA levels, relative to controls, observed within 3 h, but not after 24 h of PA addition. Similarly, ODC mRNA levels were increased but only after 3 h of incubation with either polycation (Fig. 4B).
|
Parallel analysis of ST cells indicated that after 3 h of incubation, only SPD significantly increased (P 0.05) SSAT mRNA levels (Fig. 5A). The effect of SPM on these cells was observed only after 24 h of incubation (P 0.05). By contrast, no significant changes in ODC mRNA levels were observed with any of the treatments at all the time points examined (Fig. 5B).
|
Gene expression of IGF-I was also evaluated in ST cells, as a function of SPD and SPM treatments, because these cells, among the three major endometrial cell types, exhibited the highest IGF-I gene expression. Results demonstrated that only SPM at 24 h of treatment significantly increased (P 0.05) IGF-I mRNA levels (data not shown).
Effects of IGF-I ± PAs and Specific Inhibitors of MAPK and PI-3K Pathways on Endometrial DNA Synthesis
The role of PAs and their specific functional interactions with IGF-I, in mediating endometrial cell proliferation, was evaluated by determining the levels of DNA synthesis in cells treated with PUT, SPD, and SPM alone or in combination with IGF-I. At 50 ng/ml concentration, IGF-I significantly increased DNA synthesis in GE and ST cells, consistent with previously published results from this laboratory [17] (Fig. 6, A and B). Putrescine and SPM similarly increased DNA synthesis in both cell types, although SPD was only effective on ST cells. In GE cells, cotreatment with IGF-I significantly increased (P 0.05) the individual effects of each PA (Fig. 6A). In ST cells, however, only the effect of SPD was increased (P 0.05) by IGF-I, with no effect observed above that elicited by SPM or PUT, with IGF-I coaddition (Fig. 6B).
|
To determine whether well-characterized inhibitors of the MAPK or PI-3K signaling pathways could block the individual or combined effects of PA and IGF-I, GE cells were pretreated with PD98059 (MEK inhibitor) or wortmannin (PI-3K inhibitor), 1 h prior to treatment with SPM, IGF-I, or their combination. Neither inhibitor had any effect on the extent of DNA synthesis in untreated GE cells (serum-free medium alone) (Fig. 7). Surprisingly, these inhibitors also did not affect the IGF-I-mediated induction of DNA synthesis in these cells. PD98059 diminished the inductive effect of SPM on DNA synthesis, which was not observed upon cotreatment of cells with IGF-I + SPM. Wortmannin had no significant effect (P > 0.05) on the induction of GE DNA synthesis by SPM, alone or in combination with IGF-I (Fig. 7).
|
Polyamine Content in Uterine Luminal Fluids
To begin to evaluate the potential physiological significance of the observed changes in endometrial SSAT mRNA levels during early pregnancy, the levels of PAs in the uterine lumen of gilts at pregnancy Days 11.5 and 12 were quantified by HPLC/electrospray ionization-mass spectroscopy. The results presented here are to be interpreted with the caveat that this analysis does not measure absolute concentrations but rather provides estimated levels based on the differences in integrated peak areas between the unspiked extracts and extracts spiked with known amounts of each PA. Results indicated that the levels (nM/µg total protein) of these polycations were higher at Day 12 than at Day 11.5 of pregnancy, with estimated 3 (0.07 versus 0.023)-, 10 (1.37 versus 0.14)-, and 2 (1.31 versus 0.67)-fold increases, respectively, for SPM, SPD, and PUT amounts between these days.
DISCUSSION
Polyamines and IGF-I are potent regulatory molecules implicated in the control of cell growth, differentiation, and apoptosis [34–37]. The functional linkage between these molecules in the regulation of uterine tissue growth that underlies, in part, establishment of pregnancy in mammals is suggested by the demonstrated similarities in the uterine and overall reproductive phenotypes manifested by SSAT overexpressing transgenic and IGF-I null, female mice, respectively [13, 14, 29]. In these mouse models, the consequence of abnormally high uterine SSAT expression, resulting in depletion of intracellular PA pools [29] mimicked those lacking uterine IGF-I expression; these observations suggest a potential causal relationship between IGF-I and SSAT, although this has not been previously examined. Our findings using primary cultures of distinct endometrial cell types isolated from early pregnancy porcine endometrium provide evidence for IGF-I regulation of SSAT as well as ODC gene expression, but only in GE cells. These data suggest an important role for IGF-I in endometrial PA metabolism, raising the possibility that its effects on cellular growth, differentiation, and apoptosis may be mediated in part by intracellular PA bioavailability for a given cell type.
Our previous identification of the PA catabolic enzyme SSAT as a peri-implantation conceptus-induced gene in the porcine endometrium [16] and the demonstration that its maximal expression is temporally coincident with that of IGF-I ([11, 17] this study) initially appeared paradoxical, because IGF-I is a growth promotant, in contrast to SSAT that generally inhibits cell growth. To clarify this, we evaluated the expression levels of the IGF-I gene in the three major cell types of the Day 12 pregnancy endometrium that differed in SSAT gene expression (LE > GE > ST) [16]. Results presented here indicate the apparent inverse correlation between SSAT and IGF-I, as ST cells with the highest IGF-I mRNA levels exhibited the lowest levels of SSAT mRNA, while GE cells have correspondingly lower IGF-I and relatively higher SSAT mRNAs, respectively, than ST cells. These findings are consistent with the observed inhibition, albeit transient, by IGF-I of SSAT mRNA levels in GE cells in vitro and provide support for a paracrine/autocrine mode of IGF-I action in the control of SSAT gene expression in utero. The lack of a comparable inhibition of SSAT gene expression in LE and ST cells by exogenous IGF-I is not simply due to an absence of functional IGF-I receptors in these cells, nor to their higher endogenous levels of IGF-I production, because both are known to respond to the mitogenic effects of IGF-I [17]. However, numerous studies have demonstrated the regulation of SSAT gene expression by hormones, growth factors, and PAs to be highly cell-type specific and controlled at the level of transcription [38, 39]; this may underlie the observations reported here for ST and LE cells. Consistent with this, we have previously demonstrated the distinct inducibility of SSAT gene expression in primary cultures of GE compared to ST, cells by conditioned medium derived from the porcine trophoblast cell line Jag-1 or from Day 12 filamentous embryos [17]. Thus, the differential regulation of PA metabolism by IGF-I in distinct cell types of the peri-implantation endometrium might be necessary for these cells' separate and nonoverlapping functions during this period. Although SSAT enzymatic activity and intracellular PA levels in response to IGF-I were not determined here, the direct correlation between SSAT activity and PA pool dynamics with SSAT mRNA accumulation has been extensively evaluated in numerous cell types [19, 21, 40] and is anticipated to be operative as well in the endometrial cells examined in the present study.
The transient inhibition by IGF-I of SSAT gene expression in GE cells in vitro, which is consistent with the observed high SSAT and low IGF-I gene expression, respectively in these cells likely leads to rapid changes in intracellular levels of PAs that could mediate the well-known IGF-I mitogenic effects in endometrial cells. In the present study we provide evidence that PAs affect cellular DNA synthesis, independent of IGF-I. First, we show that PUT, SPD, and SPM have distinct DNA synthesis-stimulating activities; albeit this is a function of cell type. Second, the combined effects of IGF-I and of each PA on DNA synthesis in GE cells were additive. In contrast, ST cells exhibited additive effects only for IGF-I and SPD, with no such effects observed between IGF-I and PUT or SPM. Third, the use of specific inhibitors to the MAPK and PI-3K signaling systems indicated their distinct consequences on IGF-I- and PA-mediated mitogenesis. Indeed, several interesting observations from the latter study are worth noting. One is the inability of specific inhibitors to either signaling system to inhibit IGF-I effects in GE cells. This is surprising as IGF-I in a number of target cells has been shown to utilize both of these signaling systems [41] and implies the use of yet another signaling system for IGF-I action in endometrial epithelial cells. Additional studies will be needed to clarify this point further. However, the MAPK inhibitor PD98059 specifically attenuated the activity of SPM, suggesting that this PA likely employs the MAPK pathway in inducing DNA synthesis in GE cells. Consistent with this, an interaction between PA and the MAPK signaling pathway has been previously reported in human MCF-10A breast epithelial cells genetically engineered to overexpress ODC [42]. The lack of effect of this inhibitor on the other hand, on the combined actions of IGF-I and SPM, may reflect the predominant mitogenic effects of IGF-I over that of SPM on GE cells. Taken together, these findings provide support to the distinct actions of PA (e.g., SPM) and IGF-I in mediating the proliferation of endometrial cells in vitro, and most likely in vivo.
An important point of regulation in PA biosynthesis is the feedback mechanism employed by these polycations on the cellular expression of SSAT and ODC. We examined this possibility using isolated endometrial GE and ST cells for two reasons. First, previous studies demonstrating this regulation were conducted mostly in immortal and malignant cells [39, 43] that respond to stimuli or regulatory factors in ways that are at times distinct from normal cells [16]. Second, we wanted to determine if PAs can mimic the effects of IGF-I on SSAT and ODC gene expression in uterine endometrial cells, as both are growth-regulatory molecules for many other cell types. The results presented here demonstrate that the effects elicited by SPD and SPM on GE cell expression of SSAT and ODC genes, albeit similarly occurring in a transient manner, are opposite to those observed with IGF-I. However, the induction by SPM of SSAT gene expression in both GE and ST primary cell cultures is consistent with the positive regulation by this polycation of SSAT activity previously reported in other cell types [39, 40, 43, 44]. Collectively, these results provide support for nonsequential (independent) pathways by which IGF-I and PAs mediate their proliferative effects, establish the distinct actions of SPM versus SPD on specific expression of ODC and SSAT genes, and demonstrate a positive feedback loop exerted by PAs on their biosynthesis in primary endometrial cell cultures.
It is tempting to speculate on the biological rationale for the enhanced endometrial SSAT gene expression during a restricted period of peri-implantation in the pig. On the one hand, the increase in the relative levels of PAs (3- and 10-fold, respectively, for SPM and SPD) found in uterine luminal fluids between Days 11.5 and 12 of pregnancy is suggestive of a large pool of PAs that is available for rapid cellular uptake by a highly efficient transport system and that likely contributes to uterine and/or rapid embryonic growth occurring during this time. On the other hand, the increase in uterine SSAT mRNA levels coincident with this implies the presence of a similarly effective system to regulate any excessive increase in intracellular PA pools, as acetylated PAs do not stimulate cell growth [45]. These results suggest that during the dynamic period of embryo development when transformation from large tubular to attachment-competent filamentous forms occur, the maintenance of a threshold level of endometrial intracellular PAs may be needed to allow the endometrium and the embryo to coordinate their development and hence initiate proper conceptus attachment. The dramatic loss of SSAT gene expression after Day 19 of pregnancy, when conceptus attachment is completed, lends support to this possibility.
A role for PA in control of apoptosis has been invoked for many cell types. Indeed, high PA pools have been postulated to trigger apoptosis [46, 47], possibly via their induction of caspase-3 activity [48]. On the other hand, progesterone has been correlated with inhibition of apoptosis in uterine-derived cell lines, including endometrial epithelial cells [49, 50], suggesting that progesterone induction of endometrial SSAT that we had previously demonstrated [16] may represent a physiologically relevant mechanism for control of apoptosis in these cells. Thus, the induction by progesterone of both IGF-I and SSAT synthesis [10, 16] and the transient decrease in SSAT gene expression mediated by IGF-I in endometrial GE cells (this study) may affect the relative progression of specific endometrial cell types to mitogenesis versus apoptosis, resulting in the establishment of an optimal uterine environment for embryo attachment.
In summary, the present data suggest that SSAT gene expression during a limited period of peri-implantation likely has functional relevance for the successful initiation of embryo attachment. Because SSAT is a conceptus-induced gene [16, 17], this suggested role for SSAT represents yet another form of embryo-maternal communication during the pregnancy recognition period that is likely important for embryo development [51]. These studies also raise the possibility of novel mechanisms by which endometrial IGF-I and progesterone, via their regulation of SSAT gene expression, mediate a delicate balance between cell growth and apoptotic events in support of a successful pregnancy.
ACKNOWLEDGMENTS
The authors thank Frank Michel and Ge Zhao for expert technical assistance and other members of R.C.M. Simmen's and F.A. Simmen's laboratories for help with tissue collection and animal management.
FOOTNOTES
First decision: 21 February 2001.
1 This research was supported by National Institutes of Health grant HD-21961 and by the Florida Agricultural Experiment Station and approved for publication as Journal Series R-08053. 
2 Correspondence: Rosalia C.M. Simmen, Department of Animal Sciences,
Building 459, Shealy Drive, University of Florida, Gainesville, FL 32611-
0910. FAX: 352 392 7652; simmen{at}animal.ufl.edu
Accepted: April 4, 2001.
Received: January 23, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. E. Pegg Spermidine/spermine-N1-acetyltransferase: a key metabolic regulator Am J Physiol Endocrinol Metab, June 1, 2008; 294(6): E995 - E1010. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
