Background
The growth rate of corpora lutea (CL) immediately after ovulation is 4- to 20-fold more intense than that of some of the most malignant human tumours and is sustained by the formation of new blood vessels from capillaries of the theca interna compartment
After the stimulation of endothelial cells mitosis for neovascular tubes development, the remodelling of extracellular matrix is necessary and different proteolytic enzymes are involved
To allow repeated opportunities of fertilization, the duration of CL life is rigorously programmed, in fact, after a first period of rapid growth the tissue becomes stably organized and prepares to switch to the phenotype required for its next apoptotic regression
The switch between growth and regression is mainly regulated by LH (luteinising hormone – luteotrophyn) and PGF2α (prostaglandin F2α – luteolysin) and their receptors balance as well as by cytokines, growth factors, apoptosis/oncogenes related factors and plasminogen activator/matrix metalloproteinase activators and inhibitors
This study was aimed at characterizing the temporal expression pattern of VEGF, MMP-2, MMP-9 and Ca++/Mg++-dependent endonuclease throughout the lifespan of swine CL and investigating whether the expression levels of those molecules are related to CL functional stage.
Results
CL functional stage assessment
Both luteal progesterone (P4) concentration and prostaglandin F2α receptor (FPr) mRNA expression followed typical swine luteal phase patterns. P4 concentrations increased gradually during the formation of CL, reaching the highest level during the mid-late phase (days 10 and 14), and then declined significantly on day 17 (Fig
P4 and FPr mRNA levels in swine CL during different luteal phases and pregnancy
P4 and FPr mRNA levels in swine CL during different luteal phases and pregnancy. A: Changes in P4 levels (mean ± SEM). B: Changes in FPr mRNA expression in relation to day 1; error bars represent the range of relative expression. The statistical analysis were based upon the mean of 5 CLs per animal (n = 4/time point). Different letters represent significant differences (p < 0.05).
DNase activity assay
A regulated Ca++/Mg++-dependent endonuclease expression and activity were detected in swine CL. Fig
A: Representative electrophoresis profile of low molecular weight DNA from CL at day 1, 14 and 17
A: Representative electrophoresis profile of low molecular weight DNA from CL at day 1, 14 and 17. Each lane contains 10ug of DNA extracted from luteal nuclei after treatment with different cations. B: Nuclease activity in swine CL nuclei during different luteal phases and pregnancy. For each time point only No Salt (open bars) and Ca+++Mg++ (closed bars) samples were presented. Data represent means ± SEM of percentage of low molecular weight DNA (≤ 2000 bp). The statistical analysis were based upon the mean of 5 CLs per animal (n = 4/time point). Different capital and small letters represent significant differences (p < 0.05) for No Salt and Ca+++Mg++ groups, respectively.
The amount of activable DNase gradually increased during the oestrous cycle, maintaining high at pregnancy. Moreover, a high level of spontaneous apoptosis in the control group (No Salt) at day 17 was also observed (Fig
VEGF content
Both VEGF mRNA and protein presented similar patterns of expression throughout the oestrous cycle. The highest values of VEGF were observed immediately after ovulation, decreasing significantly at day 3, and remaining unchanged during the mid-luteal phase. At day 17, a second significant drop was observed. At pregnancy VEGF mRNA and protein levels were intermediate between those of day 1 and 3 (Fig
VEGF mRNA and protein levels in swine CL during different luteal phases and pregnancy
VEGF mRNA and protein levels in swine CL during different luteal phases and pregnancy. A: changes in VEGF mRNA expression in relation to day 1; error bars represent the range of relative expression. B: changes in VEGF content (mean ± SEM). The statistical analysis were based upon the mean of 5 CLs per animal (n = 4/time point). Different letters represent significant differences (p < 0.05).
MMPs activity assay
Three distinct bands of gelatinase activity corresponding to latent MMP-9 (proMMP9), latent MMP-2 (proMMP2) and active MMP-2 (actMMP2) were evidenced in the swine CL (Fig
A: Representative gelatin substrate zymography gel showing gelatinase activity in swine CL during different luteal phases and pregnancy
A: Representative gelatin substrate zymography gel showing gelatinase activity in swine CL during different luteal phases and pregnancy. B: Relative abundance of MMP-2 as determined by densitometric analysis expressed as means ± SEM. C: Relative abundance of MMP-9 as determined by densitometric analysis expressed as means ± SEM. The statistical analysis were based upon the mean of 5 CLs per animal (n = 4/time point). Different letters represent significant differences (p < 0.05).
Discussion
The CL is a transient endocrine gland, which is formed from the remnants of an ovulated follicle
In this study, proteases with gelatinolytic activity that is consistent with MMP-9 and MMP-2 family members
Members of MMP-system may be involved in several of the proteolytic events that take place in the ovary during the reproductive cycle
Our findings demonstrated the presence of MMP-2 and MMP-9 in the corpus luteum during different ovarian phases. MMP-2 activity was basal during early-mid luteal phase and at pregnancy while was maximal during luteolysis. MMP-2 plays a role in various tissue remodelling processes, including trophoblast invasion
Concerning MMP-9 activity, high levels were observed during early luteal phase, luteolysis and pregnancy. MMP-9 is probably involved in the extensive tissue remodelling that occurs during CL formation, when extracellular material, composed primarily of follicular elements that represent a basement membrane-type ECM, is removed
A different activity profiles for both MMP-2 and MMP-9 were also observed in human
In the early luteal phase, we also detected the highest levels of VEGF. As expected, VEGF mRNA and protein were detected during all days of the oestrous cycle and showed similar pattern of expression. High levels were verified immediately after ovulation, concomitantly with the intense luteal vascular growth and CL formation. Thus supporting the role of VEGF in the angiogenesis of the newly formed CL. Changes of VEGF levels in CL during the oestrous cycle have been reported also in other species
Taken together, our results provide that in addition to intense angiogenesis (characterized by maximal VEGF concentration), elevated luteal gelatinases may contribute to the extensive luteal ECM and tissue remodelling that occurs as the postovulatory follicle is transformed into the CL
After the CL is fully formed, steroidogenesis is maximal during the midluteal period, and MMP activity and VEGF concentration are at basal levels. Interestingly, we previously reported an increased expression of the less abundant and matrix-attached VEGF splice variants (VEGF188 and VEGF182) as well as of the two VEGF receptors in this moment
The elevated expression of VEGF associated to the presence of MMP-9 at pregnancy suggests that initial angiogenic process during the early luteal phase may be renewed in swine pregnant CL. Increased angiogenesis at pregnancy in luteal tissue is controversial. Wulff et al.
Conclusion
We have demonstrated that CL development and regression is a very useful model for studying VEGF/MMPs relationships. During the very early luteal phase, high MMPs activities coupled with high VEGF levels drive the tissue to an "angiogenic phenotype", allowing CL growth under LH stimulus. In the late luteal phase, low VEGF and elevated MMPs levels may play role in the apoptotic tissue and ECM remodelling during structural luteolysis.
Methods
Animal model
Synchronized folliculogenesis was obtained in 28 prepubertal Large White gilts (96 ± 2.13 Kg, mean ± SEM) by administration of 1250 IU equine chorionic gonadotropin (eCG; Folligon, Intervet, Holland) and 750 IU hCG (Corulon, Intervet) 60 h later. In this model, ovulation occurs around 42–44 h after hCG administration (day 0). Four animals were artificially inseminated 40 h after hCG administration and pregnancies were determined by ultrasonography 35 days after. Ovaries were recovered by surgical laparotomy on days 1, 3, 5, 10, 14 and 17 after ovulation (n = 4 animals/time point) and at day 60 of pregnancy. Animals were pre-anesthetized by using azaperone (240 mg/gilt; Stresnil, Janssen, Belgium) and atropine sodium salt (2 mg/gilt; Industria Galenica Senese, Italy), and maintained under thiopental sodium (1.5 g/gilt; Pentothal Sodium; Gellini, Latina, Italy) anaesthesia. Five CLs from each gilt, chosen totally random from both ovaries, were isolated, cut in three parts with a razor blade and employed to perform all the analysis.
One third was weighed and homogenized in PBS (0.1 g/ml) on ice bath by an Ultra Turrax. The homogenate obtained was processed as follows: 500 μl were centrifuged at 2000 × g for 10 min at 4°C and supernatant was stored at -20°C until VEGF determination and the remainder of the homogenate was kept frozen until progesterone (P4) measurement and MMPs activity evaluation.
Total RNA was isolated from the second third of the CLs with the Tri-Pure reagent (Roche Diagnostic GmBH, Mannheim, Germany) and stored at -80°C until VEGF and Prostaglandin F2α receptor (FPr) mRNA levels quantification. The last part of samples was immediately processed for nuclei extraction and Ca++/Mg++-dependent endonuclease activity evaluation. All the sampling procedures were executed within 2 hours from the surgical removal of the ovaries.
All animals were housed and used according to EEC animal care guidelines. The experimental procedures had been previously approved by the Ethical Committee of Bologna University.
P4 assay
Aliquots of 20 μl from each homogenate CL were extracted with 5 ml petroleum ether. After centrifugation, ether was collected and dried under a N2 stream. Dried ether extracts were resuspended in 1 ml phosphate buffer, diluted 1:50 and aliquots of 50 μl were then assayed by a validated RIA as previously described
The sensitivity of the assay was 3.7 pg/tube. The intra- and interassay coefficients of variation were 6.3 and 19.6%, respectively. The results are expressed in ng/mg tissue.
RNA extraction and Real-time quantitative RT-PCR
Total RNA from CLs, homogenized in Tri-Pure reagent (50 mg/ml), was extracted according to manufacturer's instructions (Roche Diagnostic GmBH, Mannheim, Germany). Purified RNA was resuspended in RNase-free water and quantified (A260 nm). One microgram of total RNA was reverse-transcribed to cDNA using iScript cDNA Synthesis Kit (Bio-RAD Laboratories Inc., CA, USA) in a final volume of 20 μl, according to the manufacturer's instruction. Transcription reactions without reverse transcriptase were performed to control for an eventual DNA contamination.
Swine primers were designed for VEGF, FPr and HPRT (Hypoxanthine Guanine Phosphorybosyl Transferase), using the Beacon Designer 3.0 Software (Premier Biosoft International, Palo Alto, Ca, USA). Their sequences, expected PCR product length and accession number are shown in Table
Sense and antisense primers sequences used for real time RT-PCR.
Primer
Sequence (5'-3')
Product size (bp)
HPRT sense
GGACAGGACTGAACGGCTTG
HPRT antisense
GTAATCCAGCAGGTCAGCAAAG
115
VEGF sense*
CCTTGCCTTGCTGCTCTACC
VEGF antisense*
CGTCCATGAACTTCACCACTTC
101
FPr sense
TCAGCAGCACAGACAAGG
FPr antisense
TTCACAGGCATCCAGATAATC
151
*VEGF primers were located on a common region for all VEGF isoforms.
Real-time efficiencies were acquired by amplification of a standardised dilution series and corresponding slopes and PCR efficiencies were calculated using iCycler iQ Real Time PCR Detection System (Bio-Rad Laboratories Inc.).
Ca++/Mg++-dependent endonuclease activity assay
Tissues were immediately treated to isolate nuclei and to determine endonuclease activity in the samples
In brief, minced tissue was homogenized (1/10, w/v) with a Dounce homogenizer in a homogenization solution containing 10 mM Tris-Cl (pH 7.4), 3 mM MgCl2, 3 mM EGTA, and 250 mM sucrose. The homogenate was filtered and then centrifuged at 800 g for 15 min at 4°C. The resulting pellet was resuspended in the homogenization solution supplemented with 0.5% (v/v) nonidet P40, incubated for 15 min at 4°C and then centrifuged at 800 g for 15 min at 4°C. The resulting pellet was resuspended in a solution containing 10 mM Tris-Cl (pH 7.4), 25 mM NaCl, and 340 mM sucrose. The reaction mixture for the DNA fragmentation assay was performed with 30 mg of luteal tissue and 1 mM Ca++ and 5 mM Mg++, with or without the addition of 2 mM Zn++. A control reaction, without salts, was also carried out. The reactions were performed at 37°C for 10 min, after which low and high molecular weight DNA were extracted; residual RNA was removed by addition of RNase A. DNA content was evaluated by densitometric scanning under a UV transilluminator after 2% agarose gel electrophoresis run. Low molecular weight DNA (≤ 2000 bp) was expressed as percent of total DNA in the sample.
VEGF assay
Luteal VEGF concentrations were measured in 100 μl samples of homogenate supernatants by a specific enzyme linked immune-adsorbent assay (ELISA, Quantikine, R&D Systems, Minneapolis, MN, USA) previously validated for the measurement of porcine VEGF
MMPs activity assay
MMP-2 and -9 activities were analyzed by use of gelatin zymography on 10% Tris-Glycine poliacrylamide pre-cast gels with 0.1% gelatin (10% Novex Zymogram Gels, Invitrogen U.K.). Aliquots containing 30 μg of total proteins, mixed with an equal volume of sample buffer (Novex Tris-Glycine SDS sample Buffer 2X, Invitrogen U.K.) were loaded into the gel. Electrophoresis was performed under non-reducing conditions at a constant voltage (125 V for 120 minutes). Following electrophoresis, gels were washed for 30 minutes in Novex Zymogram Buffer (Invitrogen U.K.), equilibrated at room temperature for 30 minutes in developing buffer (Novexα Zymogram Developing Buffer, Invitrogen U.K.) and then incubated at 37°C for 22–24 hours in fresh developing buffer. Band of gelatinolytic activity were developed after staining gels for 6–8 hours with Simply Blue Safe stain (Invitrogen U.K.) by comparison with a MMP-2 and -9 human standard (Chemicon International, CA, USA).
Gel images were captured with a computerized system (Geldoc 1000, Bio-Rad), and gelatinolytic bands were measured with densitometric analysis software (Quantity One, Bio-Rad). The resulting data are expressed as arbitrary units (AU).
Statistical analysis
The statistical analysis were based upon the mean of 5 CLs per animal (n = 4/time point), since no significant differences among CLs within the same animal were observed. Differences in relative mRNA expression of VEGF and FPr (using the ΔCt values), VEGF protein levels, progesterone contents and MMPs and Ca++/Mg++-dependent endonuclease activities were determined using one-way ANOVA (SPSS Version 13.0, Inc, Chicago, IL, USA), followed by the Duncan's post-hoc test. Data are presented as mean ± SEM. A value of p < 0.05 was considered significant.
Authors' contributions
All authors participated in experimental design and collected biological material. LAR carried out RNA extraction and real-time RT-PCR. MET performed MMPs activity and VEGF assays. AZ carried out endonuclease activity and P4 assays. MLB was responsible for animal care and surgical procedures. MF conceived and supervised the study. LAR and MF wrote the manuscript. All authors read and approved the final manuscript.