| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Molecular Biology Laboratory, Istituto Auxologico Italiano (A.M., M.R., S.M., E.P., L.A., P.V., A.M.D.B.), 20135 Milan, Italy; Second Department of Obstetrics and Gynecology, University of Milan School of Medicine (M.V.), 20122 Milan, Italy
Address all correspondence and requests for reprints to: Dr. A. M. Di Blasio, Molecular Biology Laboratory, Istituto Auxologico Italiano, Viale Montenero 32, 20135 Milan, Italy. E-mail: a.diblasio{at}auxologico.it.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The female reproduction system undergoes profound cyclic changes involving both cell proliferation and angiogenesis. The molecular and cellular mechanisms that mediate these processes in the endometrium are still poorly understood, but these modifications are related to the cyclical change in circulatory levels of ovarian steroids, growth factors (6), and cytokines, such as IL-6 (7). Basic fibroblast growth factor (bFGF) is a heparin-binding cationic protein with mitogenic and angiogenic properties (8). The presence of bFGF has been demonstrated in endometriotic tissue by immunohistochemistry (9), and we have previously demonstrated the presence of bFGF and its receptor mRNAs in purified primary cultures of stromal cells isolated from eutopic and ectopic endometrial samples (10). Dysregulation of bFGF expression has been implicated in a variety of pathological conditions involving angiogenesis, vascular smooth muscle cell proliferation, and solid tumor growth (11, 12, 13, 14, 15, 16, 17). However, the transcriptional and posttranscriptional mechanisms regulating bFGF expression are complex and still largely unknown. The presence of an endogenous antisense mRNA of bFGF (FGF-AS) has been demonstrated in different mammalian cells (18, 19, 20), including human, and indirect evidence supports the hypothesis that this antisense RNA may regulate bFGF expression. Indeed, steady state levels of the antisense RNA are inversely related to the level of bFGF mRNA during rat embryonic development and in a variety of tumor cell lines (20, 21, 22, 23). In Jurkatt cells, it has been demonstrated that as a result of alternative splicing, two forms of human FGF-AS mRNA are present, each consisting of five exons. Both share exons 2–5 and differ only in the first exon, either 1A or 1B (1A FGF-AS and 1B FGF-AS) (23). The sense and the two antisense transcripts share a 641-bp region of sequence homology at their 3' ends (Fig 1
). The FGF-AS gene that codifies for the two transcripts has been localized on the same chromosomal site as the bFGF gene, so it could be defined as a human endogenous antisense gene (24). Only the 1B FGF-AS contains an open reading frame of 316 amino acids, predicting a translation product of 35 kDa (GFG) which is 75% and 83% homologous with Xenopus and rat clones, respectively. GFG contains a conserved region, the MutT domain, that is present in a family of proteins involved in preventing mutations and in cleaning the cells of potentially deleterious endogenous metabolites (22, 25).
|
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Endometrium was collected from 53 women scheduled for laparoscopy for ovarian cysts or infertility or pelvic pain. Laparoscopic examination demonstrated the presence of endometriosis in 24 patients. The extent of endometriosis was staged, according to the Revised American Society for Reproductive Medicine system (26), as stage I in 3 patients, stage II in 2, stage III in 15, and stage IV in 4 cases. Twenty-nine women without laparoscopic evidence of endometriosis served as a control group. Intraoperative diagnosis demonstrated benign ovarian cysts in 9 cases, adhesions in 3, myomas in 8, hydrosalpingitis in 2, and normal pelvis in 7 cases. Based on the data of the last and following menstrual periods and of the histological examination of the samples, specimens of endometrium were assigned to the early proliferative (EP) phase in 10 controls and 6 patients, to the LP phase in 12 controls and 10 patients, and to the luteal (L) phase in 7 controls and 8 patients. This research project was granted by the local human institutional investigation committee. All patients were informed in detail about the aims and procedures of the study and subsequently gave their written consent to sample collection. No participant received any hormonal medications in the 3 months before surgery.
Cell preparation and culture
We established stromal cell monolayer from eutopic endometrial samples as previously described (10). Briefly, the tissues were gently minced into small pieces (1–2 mm3) and washed in fresh medium to remove mucus or debris. Thereafter, they were incubated for 2 h at 37 C in a shaking water bath in 10 ml Ham’s F-10 containing 0.2% collagenase. At the end of the incubation, cell clumps were mechanically dispersed by aspiration through a Pasteur pipette. Single stromal cells were separated from large clumps of epithelium during a 10-min period of differential sedimentation at single gravity. The top 8 ml medium, containing predominantly stromal cells, were then slowly removed, and the cells were collected by centrifugation (200 x g). The stromal-enriched fraction was washed twice in Ham’s F-10 supplemented with 10% fetal calf serum and antibiotics and was allowed to adhere selectively to 25-cm2 tissue culture dishes for 15 min at 37 C in a 95% air and 5% CO2 incubator. Thereafter, nonattached epithelial cells still present were removed, and a purified stromal preparation was obtained. Total RNA was extracted when the cultures became subconfluent; this was generally achieved within 10–15 d of cell culture.
RNA extraction and RT
Total RNA was extracted from human endometrial stromal cells according to the method of Chomczynski and Sacchi (27). Three micrograms of total RNA were heated for 10 min at 70 C in presence of 1 µg random primers. Then the RNA and random primers were incubated for 60 min at 37 C in presence of RT buffer, 200 U Moloney murine leukemia virus reverse transcriptase, 0.5 mM of each deoxynucleotide triphosphate, and 25 U recombinant RNasin ribonuclease inhibitor (Promega Corp., Madison, WI) in a total volume of 25 µl.
cDNA normalization
Amplification of the hypoxanthine guanine phosphoribosyltransferase (HGPRT) first strand cDNA was performed using 200 ng total cDNA in a 50-µl volume containing 1x Taq polymerase buffer (1.5 mM magnesium chloride), 0.2 mM of each deoxynucleotide triphosphate, 12.5 µM of each HGPRT primer, and 2.5 U Taq polymerase (Promega Corp.). The HGPRT PCR conditions were: initial denaturation at 94 C for 5 min, followed by 28 cycles consisting of denaturation for 30 sec at 94 C, annealing for 30 sec at 52 C, and extension for 30 sec at 72 C. Final extension was performed for 5 min at 72 C. To exclude the possible genomic DNA contamination, HGPRT-specific primers were designed to amplify, after RT, a 97-bp product from mRNA and a 267-bp product from DNA. HGPRT primers are reported in Table 1. PCR products were electrophoresed through 3% agarose gels and visualized with ethidium bromide. In every HGPRT amplification, a constant amount of cDNA (resulting from RT of 120 ng total RNA) derived from K562 cells was used. The fluorescence of HGPRT amplification products was analyzed by a densitometer, and the DNA fragment derived from K562 cells was used as the cDNA external standard for cDNA normalization as previously described (28). We performed this analysis to assure that subsequent amplification reactions used the same amount of cDNA for each sample.
|
Competitor construction.
We constructed specific competitors for bFGF, 1A FGF-AS, and 1B FGF-AS cDNAs that were amplified under the same conditions as those used for the human correspondent cDNA, but could be distinguished from the targets. To construct the bFGF competitor, we amplified the rat bFGF cDNA fragment included between the specific primers indicated in Table 1
. The human and rat bFGF cDNA have a complete homology in these primers’ sequences, the length of the amplification product is the same, and the overall homology of the entire two fragments is 89%. In the rat fragment the 35 nucleotide is different from the human one, and thus an HincII restriction site is missing. We cloned the DNA fragment obtained from the rat cDNA amplification into the pTarget vector (Promega Corp.) and subsequently amplified it by PCR. The purified DNA was quantified by a spectrophotometer, and different competitor dilutions were prepared.
To construct the 1A FGF-AS and 1B FGF-AS competitors, we amplified both human cDNA sequences included between the couples of primers for 1A FGF-AS and 1B FGF-AS (1B FGF-AS forward a and reverse b; Table 1). We cloned the two products of amplification into the pTarget vector (Promega Corp.). There was an ApaI restriction site in the sequences of both 1A FGF-AS and 1B FGF-AS cloned fragments. At this site we inserted a small DNA fragment of 160 bp. Thereafter, the two cloned fragments were amplified by PCR and quantified using a spectrophotometer. Different competitor dilutions were then prepared.
Competitive PCR analysis.
A constant amount of cDNA was coamplified with known concentrations of the specific competitors in 50 µl PCR mixture containing 1x Taq polymerase buffer (1.5 mM magnesium chloride), 0.2 mM of each deoxynucleotide triphosphate, 25 µM of each specific primer, and 2.5 U Taq polymerase (Promega Corp.). Oligonucleotide primers specific for bFGF, 1A FGF-AS, and 1B FGF-AS are reported in Table 1. After an initial denaturation step at 95 C for 5 min, PCR conditions were: 1) for bFGF amplification, 27 cycles of 94 C for 30 sec, 60 C for 30 sec, and 72 C for 30 sec; 2) for 1A FGF-AS amplification, 33 cycles of 94 C for 45 sec, 64 C for 30 sec, and 72 C for 1 min; and 3) for 1B FGF-AS amplification, 30 cycles of 94 C for 30 sec, 65 C for 30 sec, and 72 C for 1 min. Ten microliters of bFGF PCR products were digested with 10 U of the restriction enzyme HincII at 37 C overnight and then resolved on 4% agarose gel stained with ethidium bromide. PCR products of 1A and 1B FGF-AS amplifications were directly resolved on 2% agarose stained with ethidium bromide. In each experiment the fluorescence of both target and competitor products was quantified by a densitometer as absorbance. The competitor/target absorbance ratio was plotted against known concentrations of the competitor. The amount of each mRNA was determined by the equation obtained from the regression curve of the plotted values. The results were expressed as femtograms of the specific mRNA per nanogram of total RNA.
Semiquantitative PCR for bFGF
A constant amount of cDNA was amplified in 50 µl PCR mixture containing 1x Taq polymerase buffer (1.5 mM magnesium chloride), 0.2 mM of each deoxynucleotide triphosphate, 25 µM of each specific primer, and 2.5 U Taq polymerase (Promega Corp.). Oligonucleotide primers specific for bFGF or HGPRT are reported in Table 1. PCR conditions were the same as those used in the competitive PCR and for the cDNA normalization. bFGF and HGPRT amplification products were resolved on a 3% agarose stained with ethidium bromide. The fluorescence of both fragments was quantified by a densitometer, and the bFGF/HGPRT absorbance ratio was used to estimate bFGF mRNA levels in each sample.
Sequence analysis
The identities of all PCR DNA fragments were confirmed by sequence analysis. Sequencing reactions were prepared employing ABI Big Dye terminator chemistry (PE Applied Biosystems, Foster City, CA) and were analyzed using the ABI PRISM 310 genetic analyzer (PE Applied Biosystems).
Protein extraction
Human endometrial stromal cells were resuspended in 5% sodium dodecyl sulfate and 0.125 M Tris (pH 6.8) and boiled for 5 min. The cell lysis products were then sonicated, and protein concentrations were determined by the bicinchoninic acid protein assay (Pierce Chemical Co., Rockford, IL).
Western blot analysis
Equal amounts of protein (40 µg) from cell lysates were separated on a sodium dodecyl sulfate-15% polyacrylamide gel, electrophoretically transferred to nitrocellulose membranes, and allowed to stick to the filters for 1 h at room temperature in 5% nonfat milk in Tris-buffered saline with 0.5% Tween 20. Membranes were incubated with dilutions of primary antibodies in 5% nonfat milk in Tris-buffered saline with 0.5% Tween 20. The antibodies used were a monoclonal antihuman bFGF antiserum (Sigma-Aldrich Corp., St. Louis, MO) and a polyclonal antiserum that recognizes the MutT domain of the rat FGF-AS protein (gift from Dr. P. R. Murphy). Bound antibody was detected using sheep antimouse or antirabbit IgG conjugated to horseradish peroxidase (Amersham Pharmacia Biotech, Arlington Heights, IL) and the enhanced chemiluminescence reagent (Amersham Pharmacia Biotech). The stained immunoblots were placed against x-ray film (Kodak, Rochester, NY) to generate chemiluminographs.
Transfection and expression of 1B FGF-AS cDNA
The full-length 1B FGF-AS cDNA was amplified using primers 1B FGF-AS forward (c) and reverse (d) and was subcloned into the eukaryotic expression vector pTarget vector (Promega Corp.). The clone was transfected into wild-type HEK 293T cells using the calcium phosphate method (29). The expression of the construct was confirmed by Western blot.
Statistical analysis
Data are expressed as the mean ± SEM. Differences between groups were determined, as appropriate, by unpaired t test; ANOVA, followed by Fisher’s least significant difference test, was used as the posttest. The correlation between bFGF sense and antisense mRNA levels was evaluated by linear regression analysis. Statistical significance was considered when P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Levels of bFGF mRNA were examined in eutopic endometrial stromal cells derived from 28 controls and 22 patients affected by endometriosis using competitive PCR.
Figure 2 shows the results of bFGF mRNA quantification in the different phases of the menstrual cycle. In both groups of women, levels of bFGF mRNA were between 0.5–5.9 fg/ng total RNA. In controls, bFGF mRNA did not significantly change during the entire menstrual cycle. In contrast, in women with endometriosis, bFGF mRNA levels were higher in the LP phase (mean ± SEM, 3.8 ± 0.40 fg/ng total RNA) compared with both the EP and the L phases (mean ± SEM, 1.9 ± 0.38 and 2.3 ± 0.40 fg/ng total RNA, respectively). These differences were statistically significant (P = 0.006 for LP vs. EP; P = 0.01 for LP vs. L).
|
The results of competitive PCR quantification of 1B FGF-AS mRNA obtained from 27 controls and 23 patients are summarized in Fig. 3. 1B FGF-AS mRNA levels in endometrial cells derived from controls did not vary significantly as a function of the different phases of the menstrual cycle. The mean values were 2.9 ± 0.7 3.9 ± 0.51, and 2.9 ± 0.75 fg/ng total RNA in the EP, LP, and L phases, respectively. As observed for the sense transcript, 1B FGF-AS mRNA levels present in endometrial cell samples derived from women with endometriosis showed a different pattern of expression compared with those derived from the control group. Indeed, during the LP phase, 1B FGF-AS mRNA levels were significantly lower than those in the EP phase (P = 0.03) and were lower, although not significantly, compared with those in the L phase.
|
Expression of bFGF and 1B FGF-AS in the LP phase
Linear regression analysis indicated that bFGF and 1B FGF-AS mRNA levels were significantly correlated in controls (P = 0.04), whereas in women with endometriosis this correlation was not present (data not shown). Taken together, the results obtained suggested that transcription of bFGF and 1B FGF-AS in endometrial stromal cells could be differently regulated in the two groups of women studied, and the more relevant differences were present during the late proliferative phase. As shown in Fig. 4A, in this phase bFGF mRNA levels were significantly higher in cells derived from women with endometriosis than in those from controls (mean ± SEM, 3.8 ± 0.40 and 2.7 ± 0.33 fg/ng total RNA, respectively; P = 0.04). The opposite pattern was observed for 1B FGF-AS mRNA, as its levels were significantly lower in cells derived from patients compared with those from controls (mean ± SEM, 1.9 ± 0.24 and 3.9 ± 0.51 fg/ng total RNA, respectively; P = 0.004; Fig. 4B).
|
shows that the bFGF/1B FGF-AS mRNA ratios were 0.77 ± 0.14 and 2.31 ± 0.55 in controls and patients, respectively, and the difference was statistically significant (P = 0.009). These data indicate that in endometrial stromal cells, bFGF mRNA levels were more than 2-fold higher than 1B FGF-AS mRNA levels in women with endometriosis, but were slightly lower in controls.
|
To investigate whether the differences found between women with and without endometriosis at the mRNA level were also present at the protein level, we determined, by semiquantitative PCR and Western blot, bFGF mRNA and protein levels present in the same sample derived from patients and controls. In each experiment we evaluated one control and one patient with endometriosis, and a total of eight pairs of women were examined. In every experiment the differences between patients with endometriosis and controls at the mRNA level were also maintained at the protein level (Fig. 6, A and B).
|
1B FGF-AS promoter sequences
We hypothesized that the different 1B FGF-AS expression observed in the two groups of women could be caused by mutations in the antisense promoter sequence. For this reason we sequenced the two regions upstream of the 1A (409 bp upstream) and 1B (187 bp upstream) exons (23) in eight patients and five controls. In these regions different potential binding sites for transcription factors, two half estrogen-responsive element (GGTCA) (30), and one dioxin-responsive enhancer (GCGTG) (31) were present at nucleotides –261, –114, and –33 from the transcription start site, respectively. We did not find any mutations.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
bFGF is constitutively present in the human endometrium and, due to its mitogenic and angiogenic activities, is most likely involved in determining endometrial tissue modifications during the menstrual cycle (9, 10, 32, 33). We as well as other investigators have also demonstrated the presence of bFGF and its receptor in eutopic and ectopic human endometrium (9, 10). Recent evidence suggests that the expression of bFGF could be regulated by the presence of a naturally occurring antisense transcript, FGF-AS. Indeed, several examples of natural endogenous antisense transcripts have been reported in eukaryotic systems (34, 35, 36, 37), but in the majority of the cases the functions of these molecules are still largely unknown. To date, the expression of the FGF-AS transcript in the human species has been investigated in few tissues, including normal and neoplastic pituitaries (19, 24). Asa and co-workers (24) demonstrated that in normal pituitary, FGF-AS is expressed, whereas bFGF is undetectable. Instead, pituitary adenomas expressed bFGF, but had reduced levels of FGF-AS. These data support the hypothesis that the antisense transcript could regulate the expression of bFGF (24). Based on these findings it is possible to postulate that different expressions of bFGF and FGF-AS could also modulate the proliferative capacity of human endometrial cells.
The results of the present study are in line with this hypothesis. In eutopic endometrial cells derived from control women there is a significant correlation between bFGF sense and antisense mRNA levels, whereas this is not present in cells derived from women with endometriosis. Indeed, during the LP phase of the menstrual cycle, patients had bFGF mRNA levels significantly higher and 1B FGF-AS mRNA levels significantly lower than those detected in controls. As a consequence, the mean bFGF/1B FGF-AS ratio was higher in patients than in controls, and the difference was statistically significant. These data strongly suggest that the antisense transcript could negatively regulate bFGF expression by inducing degradation of its mRNA. Similar to what observed in the Xenopus oocyte (18), in human endometrial stromal cells higher levels of 1B FGF-AS mRNA corresponded to lower levels of both bFGF mRNA and protein. In contrast, in rat glioma cells the presence of the antisense transcript influenced bFGF protein levels, but not mRNA levels (38).
It is worth noting that the different expressions of the sense and antisense bFGF transcripts in endometrial cells of women with and without endometriosis are evident during the LP phase of the menstrual cycle, when estrogen levels reach their peak. The effect of estrogen on bFGF expression is still a matter of controversy. Some studies indicate that bFGF expression correlates with the expected estrogen levels throughout the menstrual cycle (33, 39), whereas others, performed in both humans and rats, have shown no correlation (32, 40). In contrast to these previous reports, in the present study we evaluated bFGF expression in highly purified primary cultures of endometrial stromal cells. In cells derived from controls, we did not find relevant differences in bFGF mRNA levels as a function of the menstrual cycle, thus confirming an absence of the estrogen effects. In contrast, in cells derived from patients, bFGF mRNA levels were significantly higher in the LP phase, and this is most likely a consequence of the concomitantly lower expression of the 1B FGF-AS transcript. These results lead us to speculate that in women with endometriosis, estrogens could indirectly influence bFGF expression by modulating the transcription of its antisense RNA. It should be noted that in the promoter region of the 1B FGF-AS gene there are two half estrogen-responsive elements (30) and two dioxin-responsive elements (31). However, no mutations in the nucleotide sequences of these regions were observed in any of the patients examined. Thus, the possible role of estrogens in the different expressions of bFGF and 1B FGF-AS needs to be further clarified.
Besides the effects on the regulation of bFGF expression, the 1B FGF-AS transcript could be directly involved in the pathogenesis of endometriosis acting through its protein product GFG. This protein contains a conserved region, the MutT domain, that is present in a family of proteins that are involved in preventing mutations and in cleaning the cells of potentially deleterious endogenous metabolites caused by environmental contaminants and oxidative damage (25, 41). Several studies have highlighted that oxidative stress and, in particular, exposure to dioxin might induce the development of endometriosis (41, 42, 43, 44). Based on these two observations, it is tempting to speculate that a decreased synthesis of GFG protein could be among the causes of increased endometrial cell sensibility to the effects of oxidative stress. At present this remains a speculative, as in this study we could not detect GFG protein in endometrial stromal cells. We cannot rule out that this is due to the antibody used, which may not have been sensitive enough to detect low levels of the protein. Few data are currently available on the activity of the GFG protein in humans; thus, we are undertaking further studies to shed new light on its presence and function in endometrial cells.
In conclusion, the results presented herein clearly indicate that in women with endometriosis, eutopic endometrial stromal cells are characterized by an increased expression of bFGF, which physiologically is a mitogenic and angiogenic peptide. Recently, Nisolle et al. (45) reported that in nude mice during the initial phases of development of endometriosis, endometrial stromal cells are the cells that attach to the mesothelium, proliferate, and then induce the early angiogenesis. The present findings in association with these experimental data, allow us to postulate that an altered expression of bFGF could be one of the genetic determinants that contributes to the development of endometriosis lesions.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received September 11, 2002.
Accepted March 4, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
in endometriotic tissue and in endometrium. Fertil Steril 75:489–495[CrossRef][Medline]
transcriptional unit. Mol Cell Biol 9:1128–1136This article has been cited by other articles:
 |
|
 |
|
  J. D. Wren, Y. Wu, and S.-W. Guo A system-wide analysis of differentially expressed genes in ectopic and eutopic endometrium Hum. Reprod., August 1, 2007; 22(8): 2093 - 2102. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Gentilini, M. Busacca, S. Di Francesco, M. Vignali, P. Vigano, and A.M. Di Blasio PI3K/Akt And ERK1/2 signalling pathways are involved in endometrial cell migration induced by 17{beta}-estradiol and growth factors Mol. Hum. Reprod., May 1, 2007; 13(5): 317 - 322. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. P. Milling Smith, S. Battersby, K. J. Sales, H. O. D. Critchley, and H. N. Jabbour Prostacyclin Receptor Up-Regulates the Expression of Angiogenic Genes in Human Endometrium via Cross Talk with Epidermal Growth Factor Receptor and the Extracellular Signaling Receptor Kinase 1/2 Pathway Endocrinology, April 1, 2006; 147(4): 1697 - 1705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Barclay, A. W. Li, L. Geldenhuys, M. Baguma-Nibasheka, G. A. Porter, P. J. Veugelers, P. R. Murphy, and A. G. Casson Basic Fibroblast Growth Factor (FGF-2) Overexpression Is a Risk Factor for Esophageal Cancer Recurrence and Reduced Survival, which Is Ameliorated by Coexpression of the FGF-2 Antisense Gene Clin. Cancer Res., November 1, 2005; 11(21): 7683 - 7691. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
