Impact of uteroplacental insufficiency on ovarian follicular pool in the rat

All protocols for the study were approved by the Committee for Animal Research of Tor Vergata University, Rome, Italy. Animal experiments were performed according to the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (NIH Publication No. 85–23, revised 1996). All procedures complied with Italian regulations for laboratory animal care, according to the guidelines and under supervision of the Animal Technology Station, Interdepartmental Service Center, Tor Vergata University, Rome, Italy.

The animal model was developed as previously described [24]. Briefly, time-dated Sprague-Dawley pregnant rats (Charles River Laboratories Inc., Italy) were delivered to the animal facilities at least 3 days prior to the first surgery. IUGR was induced by bilateral ligation of the uterine arteries on day 19 of gestation, according to the method described from Wigglesworth in 1964 [26]. Control animals (shams) were born from mothers that underwent the same surgical procedure, with the exception of uterine artery ligation. Dams delivered spontaneously during the night between day 22 and 23. Animals were selected as significant IUGR if their birth weight (BW) was more than 2 standard deviations (SDS) below the mean BW of the control litter. Newborn pups were assigned to feeding dams with a maximum number of six offspring per litter. Pup weight was recorded on day 0 and thereafter once a week until the end of experiments. However, it was not possible to follow individual animals until sacrifice or until the age of 3 weeks when the ear tag was placed. At 21 days post-partum (dpp), pups were weaned onto standard rat chow diet (Mucedola S.R.L., Milano, Italy), separated from dams and placed in groups of three to five, with the males and females separated. For the purpose of this study only female pups were evaluated. The results of the analyses on male pups have been reported earlier [24]. Animals were sacrificed through cervical dislocation at postnatal days 5, 20 and 40 dpp. These age points represent in the rat the neonatal (end of follicular assembly), juvenile (appearance of the first large antral follicles and atresia of medullary follicles) and peripubertal (first ovulation) periods [27]. In order to ensure biological variability of the observed phenotype, each group of IUGR and shams at 20 and 40 dpp included six animals from at least three different litters. For the 5 dpp old animals, starting material included eight IUGR and four sham animals, also derived from three different litters. For all of them body and ovary weight were measured. Gene expression analysis was carried in 6 IUGR and 3 sham animals. Histological evaluation was performed in 4 IUGR and 4 sham animals.

At sacrifice, ovaries were immediately excised from euthanized animals and processed as previously described [24]. In brief, for histological purpose, one gonad was fixed in 4% paraformaldehyde (PFA; P/N15812–7, Sigma-Aldrich, MO, USA) overnight at 4 °C, serially dehydrated in increasing concentration of aqueous ethanol followed by 100% butyl acetate (P/N 45860, Sigma-Aldrich, MO, USA) and finally embedded in paraffin (Paraplast X-TRA®; P/N P3808, Sigma-Aldrich) at 61 °C overnight. Ovaries were cut serially to a thickness of five μm, using a Biocut sectioning machine (Reichert-Jung, NY, USA), mounted on microscope slides (P/N10143352, Superfrost Plus, Thermo Scientific, MA, USA) and placed at 37 °C overnight.

For immunohistochemical (IHC) analysis, three sections per ovary corresponding to ~ 25, 50 and 75% points of the entire tissue block (i.e. 25th, 50th and 75th sections when the total number was 100) were stained and evaluated [28]. Tissue sections were dewaxed with xylene (P/N 02080, HistoLab, Gothenburg, Sweden) for 10 min and then serially rehydrated with decreasing concentrations of aqueous ethanol. Antigen retrieval was performed using citrate buffer (pH 6.0) at 95 °C for 10 min. Slides were then incubated with 3% H₂O₂ in methanol for 10 min at RT for endogenous peroxidase blocking followed by incubation with 2% horse serum in 1× tris-buffered saline (TBS) for 30 min at RT to prevent nonspecific antibody binding. Afterwards, samples were incubated with primary antibody against AMH (P/N MCA2246, Biorad, CA, USA) or unspecific IgGs (for negative control) dissolved in 2% horse serum in TBS overnight at 4 °C. After washing with TBS plus 0.01% Tween20 (P/N P1379, Sigma Aldrich,), slides were incubated with biotinylated secondary antibody (P/N BA1400, Vector laboratories, CA, USA), and then with avidin-biotin-peroxidase complex prepared using Vectastain ABC kit (P/N PK-6100, Vector Laboratories, CA, USA) for 30 min each at 37 °C. After washing again twice, slides were stained with DAB (P/N SK-4105, Vector Laboratories,) at RT, counter-stained with hematoxylin solution modified according to Gill III, rinsed for 5 min with running tap water, dehydrated with gradually increasing concentrations of ethanol, cleared with xylene and finally mounted using Pertex (Cell Path, Hemel Hempstead, UK) and cover slips.

For histological evaluations, ovarian sections were stained with hematoxylin and eosin. In brief, following dewaxing in xylene and serial rehydration in ethanol, samples were washed twice with distilled water, incubated for 3 min with hematoxylin solution modified according to Gill III and then rinsed with tap water for 20 min, followed by 70 and 95% ethanol baths for 5 min each. Slides were finally incubated with eosin for 30 s, dehydrated with 100% ethanol for 5 min and 100% xylene twice for 5 min, and finally mounted with Pertex (Cell Path, Hemel Hempstead, UK) and cover slips. Three sections per ovary, corresponding to ~ 25, 50 and 75% points of the entire tissue block [28], were selected for follicle quantification. For each section, only follicles with a visible nucleus were counted and classified as follows: (1) primordial follicles (PF), with one layer of flattened pre-granulosa cells (GCs); (2) primary, with two or more cuboidal GC up to one complete layer of cuboidal GCs; (3) secondary, with a least two layers of GCs but no antrum; and (4) tertiary, where the antrum was visible [29, 30]. Two blinded observers evaluated the same sections independently and the results were compared showing an inter-observer concordance above 90%.

The ovary was approximated to a prolate ellipsoid [31] and the ovarian volumes (mm³) extrapolated using the ellipsoid formula 4/3 π a b², where ‘a’ is the length of the entire ovary derived by multiplying the total number of sections (n) by 0.005 (representing the thickness of each section), and π b² is the Area of the middle section (Am) of the ovary. Therefore, ovarian volume (V) = 4/3 * Am * n * 0.005.

Total RNA was extracted from whole frozen ovaries and stored at − 80 °C using an RNeasy Mini Kit (P/N 74104, Qiagen, Veno, Netherlands). In brief, samples were dissolved in Qiagen lysis buffer and then homogenized twice for 30 s in an ULTRA-TURRAX T25 homogenizer (JANKE and KUNKEL, Staufen, Germany). Subsequently, the supernatant was mixed with half its volume of 70% ethanol. The mixture was then transferred to RNeasy spin columns and processed according to the manufacturer’s protocol.

RNA concentrations were quantified using a single-beam UV/vis spectrophotometer (P/N 6132000032, Eppendorf, Hamburg, Germany). To generate cDNA, one μg of total RNA was reverse-transcribed on a thermocycler (P/N 4359659, Applied Biosystems, MA, USA) using random hexamers in a total reaction volume of 20 μl. The IScript™ cDNA synthesis kit (P/N 170–8891, Bio-Rad, CA, USA) was used as instructed by the manufacturer.

TLDA cards (P/N 4342259, ABI, Hilden, Germany) were used for comparative analysis of gene expression of IUGR and control animals, according to the manufacturer’s protocol, as previously described [24]. Briefly, the TLDA cards are based on TaqMan chemistry where gene expression of a panel of 96 genes is analyzed in one run. The TLDA cards were pre-loaded with 96 TaqMan gene expression assays of importance for proliferation, apoptosis, cellular energy, germ cell and somatic cell function and differentiation, and six endogenous controls assigned for normalization. Gene expression was normalized to the mean of five out of six endogenous controls (Actin beta, Beta-2-microglobulin, Catenin beta 1, Eukaryotic translation elongation factor 1 alpha 1 and Glyceraldehyde-3-phosphate dehydrogenase) of the same sample (dCt), selected according to their stability. Data from one randomly chosen animal were used as a calibrator (ddCT) and data from all other animals were normalized to it. Gene expression was finally presented as relative expression [using the fold change (2 − ddCT) method for calculation].

TaqMan Gene expression Master Mix (P/N 4369510, Applied Biosystems, MA, USA) was used when running the TLDA assays.

Body weights (BW) and ovarian weights (OW) of IUGR and sham animals are shown in Table 1.

Mean BW in IUGR rats was still significantly reduced at 5 dpp compared to sham rats but caught up by 20 dpp. Ovarian weight (OW) and OW normalized per body weight (OW:BW) were not significantly different in the two groups (Table 1).

In both groups, PF density declined from 5 to 40 dpp, in line with the expected age-related decrease in the number of PF from birth to puberty, that has been well-described in other mammalian species, i.e. humans and primates [32‐34] (Additional file 1: Figure S1A). However, in IUGR rats at 5 dpp PF and total follicle showed a tendency to be lower compared to controls, although the comparison did not reach statistical significance. At 20 dpp, the number of PF and total follicles was significantly reduced in the ovaries of IUGR rats compared to sham (Fig. 1).

The ratio between growing follicles (primary, secondary and tertiary) and total follicles was calculated in order to evaluate a possible change in activation of PF and their subsequent transition into the pool of growing follicles. No difference was detected in the ratio between the groups at any age (Fig. 1).

Extrapolated ovarian volumes and CL number was also comparable between the groups (data not shown).

We analyzed the relative expression of a panel of 90 genes that were selected because of their known functions in the gonads (Table 2).

In IUGR ovaries, 24 genes were differentially regulated as compared to controls (Table 3).

Among those, six genes showed a significantly higher expression and 18 genes showed a significantly lower expression in IUGR versus sham animals (Fig. 2).

Six genes involved in cell proliferation, survival and cycle regulation were significantly dysregulated in IUGR rats at 40 dpp compared to controls. In addition, five genes were modified concordantly towards a reduction in the proliferation rate of ovarian cells, namely Proliferation-related Ki67 antigen (MKi67), Proliferating cell nuclear antigen (Pcna), Cyclin-dependent kinase inhibitor 1B (Cdkn1b), DNA topoisomerase II (Top2a) and Thymidine kinase 1 (Tk1). The gene encoding for Caspase 9 (Casp9) was also downregulated at this age.

Three genes involved in ovarian angiogenesis were dysregulated in 20 dpp IUGR ovaries compared to sham ones, Angiopoietin 1 (Angpt1) and Transforming growth factor beta 2 (Tgfb2) expression was reduced, whereas Platelet derived growth factor receptor alpha polypeptide (Pdgfra) was upregulated.

Furthermore, IUGR animals showed a lower expression of Insulin-like growth factor (IGF) 1 receptor (Igf1r) both at 20 and 40 dpp, that together with the increased expression of IGF binding protein (Igfbp3) at 40 dpp pointed towards a downregulation of the IGF signaling within the ovary of IUGR rats.

The expression of four genes involved in the regulation of cellular metabolism was altered in IUGR animals at different time points, as compared to sham; in particular Lactate dehydrogenase-c (Ldh-c) and Citrate synthase (Cs) were downregulated at 5 dpp and 40 dpp, respectively. In IUGR rats at 20 dpp Ldh-b and Ldh-c were both upregulated and Solute carrier family 2 member 1 (Slc2a1) was upregulated at 40 dpp.

Gene encoding for AMH (Amh) showed a reduced expression in IUGR ovaries at 20 dpp compared to controls. Inhibin B (Inhbb) expression was also lower at 40 dpp in IUGR rats versus sham ones. In Fig. 3 the expression of 8 representative genes from IUGR and sham ovaries at the three age points is reported.

Given the role of AMH as an indirect marker of ovarian reserve [35], we evaluated serum AMH levels in the two groups of animals. No significant difference was detected although a tendency to lower AMH concentrations in IUGR rats at all ages was observed (Additional file 1: Figure S1B).

In order to analyze whether a tendency to lower serum AMH and a lower expression of the gene encoding for AMH was related to a lower protein synthesis from the follicles, expression of ovarian AMH was evaluated by IHC; the analysis did not show any clear difference in the intensity of protein expression in granulosa cells between the groups (Additional file 1: Figure S1C).