Abstract
The objective of our study was to investigate glycemic, oxidative/antioxidative and inflammatory status in letrozole and estradiol valerate induced polycystic ovarian syndrome (PCOS) models. Sixty adult female Wistar rats were divided into four groups: L (0.2 mg letrozole/0.5 ml carboxymethyl cellulose (CMC), daily for 30 days), the control group CL, EV (one i.m. injection of 5 mg EV/0.5 ml sesame oil) and its corresponding control group CEV. After 30 days, ovarian morphology was assessed through ultrasound, serum free testosterone was determined, and an oral glucose tolerance test was performed. Blood, muscle, liver and periovarian adipose tissue (POAT) were collected for oxidative/antioxidative and inflammatory status evaluation. Free testosterone was increased only in the L group, while fasting glycemia was higher in the EV group. Both L and EV led to a significantly decreased level of muscle malondialehyde (MDA) and liver glutathione peroxidase (GPx) activity, while in POAT, MDA level diminished and GPx activity increased. The only difference between the two protocols was in muscle, where after L administration, GPx activity was significantly lower. Implementation of both protocols resulted in an increased expression of pNFKB in muscle, liver and POAT. The expression of monocyte chemoattractant protein 1 (MCP1) increased in liver and POAT after L administration, while in the EV group, MCP1 and STAT3 decreased in POAT. Our study shows that both protocols are characterized by an inflammatory environment in the usually insulin resistant tissues of human PCOS, without generating oxidative stress. In addition, EV has mild metabolic effects and unexpected interference with MCP1 expression in POAT, which require further investigation.
Introduction
Polycystic ovarian syndrome (PCOS) is the most common cause of anovulatory infertility (Broekmans et al. 2006), affecting 6–20% of reproductive age women (Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group 2004). For its diagnosis, two out of the following criteria must be met: oligoovulation/anovulation, clinical/biochemical signs of hyperandrogenism and polycystic ovaries (Azziz et al. 2009). The definition is far from being exhaustive, the clinical presentation of women with PCOS being extremely variable, involving frequently abnormalities of lipid and glucose metabolism. The alteration of glucose metabolism results in hyperglycemic states such as impaired fasting glycemia, impaired glucose tolerance or both (Karakas et al. 2010). Moreover, oxidative stress (Sabuncu et al. 2001) and low-grade inflammation (Duleba & Dokras 2012) are emerging as factors in the pathogenesis of the syndrome. Women with PCOS have an increased oxidative status which is related to central obesity, age, blood pressure, serum glucose, insulin and triglyceride levels, and insulin resistance, correlated with alteration of the antioxidant status (Sabuncu et al. 2001). In addition, markers of oxidative stress and inflammation are correlated with androgen level (González 2012).
Since the etiology of PCOS remains unclear, the use of animal models has offered valuable insights into the pathophysiology of the syndrome. However, as the array of pathological modifications observed in human PCOS widens, it becomes clear that the existing animal models should be characterized also from an inflammatory and oxidative/antioxidative point of view. This may lead to a better understanding of the underlying mechanisms of PCOS, which is essential for shifting the therapy from an incomplete, symptomatic approach to a curative one.
PCOS experimental models have been developed on various animals, such as rodents (reviewed by Shi & Vine 2012, Walters et al. 2012, McNeilly & Colin Duncan 2013, Maliqueo et al. 2014), sheep (reviewed by Padmanabhan & Veiga-Lopez 2013) or rhesus monkeys (reviewed by Abbott et al. 2013). Rhesus monkeys and sheep models provide promising perspectives and obvious advantages related to their intrinsic physiological characteristics which resemble those of humans. However, the rodents are the most commonly used because they offer a practical way to study PCOS and they are also cost-effective. Two widely used models for PCOS study, the ones induced by letrozole (L) and estradiol valerate (EV) respectively were concerned especially with morphological, endocrinological and/or metabolic aspects, while the study of oxidative stress and inflammation was not enough investigated. Letrozole is a third generation aromatase inhibitor that blocks the conversion of testosterone to estrogen, thus inducing hyperandrogenism. Estradiol valerate is a long-acting estrogen which acts by disturbing hypothalamic gonadotrophin-releasing hormone secretion, which in turns interferes with the secretion and storage of LH (Simard et al. 1987).
The aim of our study was to investigate glycemic, oxidative/antioxidative inflammatory status in the above described murine experimental models. These investigations included tissues involved in maintaining glucose homeostasis such as liver and muscle, with a special focus on the periovarian adipose tissue (POAT).
Materials and methods
Reactives
The Bradford protein assay kit, SDS–PAGE gels, PVDF membranes, SDS Laemli sample buffer, Tris–glycine buffers, nonfat dry milk, Tween 20 were purchased from Bio-Rad and kit for free testosterone from MyBioSource (San Diego, CA, USA). Antibodies against phosphorylated Ser 311 of nuclear factor kappa-light-chain-enhancer 88 of activated B cells p65 subunit (pNFKB), NFKB p65, secondary corresponding HRP linked antibodies, anti-human TRP1 antibody-conjugated with Alexa Fluor 488, were from Santa Cruz Biotechnology. Monocyte chemoattractant protein 1 (MCP1) antibody was purchased from Linaris Biologische Produkte (Mannheim, Germany), and anti-GAPDH from Trevigen-Biotechnology (Gaithersburg, MD, USA). Letrozole and estradiol valerate were purchased from Sigma–Aldrich Chemicals GmbH.
Animals
Sixty female Wistar rats, 16–18 weeks of age, weighing 170–200 g, were obtained from the Animal Department of ‘Iuliu Haţieganu’ University of Medicine and Pharmacy, Cluj-Napoca, Romania. For acclimatization, the female rats were housed in the Physiology Department's Biobase a week before and throughout the experiment at a controlled temperature of 22–24 °C, having a 12 h light:12 h darkness cycle. The rats had access to pellet food and tap water ad libitum. All experiments were performed according to the approved animal care protocols of the Ethics Committee on Animal Welfare of the ‘Iuliu Haţieganu’ University in accordance with European Convention for the Protection of Vertebrate Animals used for Experimental and other Scientific Purposes, Council of Europe no. 123, Strasbourg 1985.
Experimental design
Letrozole protocol
Thirty female Wistar rats were divided in two groups: the rats in the control group CL (n=8) received 0.5 ml carboxymethyl cellulose (CMC) 0.5%, while the rats in the L group (n=22) received 0.2 mg letrozole dissolved in 0.5 ml CMC 0.5%. Administration of the substances was done daily, for 30 days through gavage.
Estradiol valerate protocol
Thirty female Wistar rats were divided into two groups: the rats in the control group CEV (n=8) received a single i.m. injection of 0.5 ml sesame oil, while the rats in the EV group (n=22) received a single i.m. injection of 5 mg EV dissolved in 0.5 ml sesame oil.
The doses were chosen based on previous studies in which letrozole or estradiol valerate was used to induce PCOS (Stener-Victorin et al. 2000, Kafali et al. 2004). At the end of the experiment (days 30–32), under anesthesia with 90 mg/kg ketamine and10 mg/kg xylazine injected intraperitoneally, the ovarian morphology was assessed through ultrasound; an oral glucose tolerance test (OGTT) was performed; the rats were sacrificed and blood, liver, muscle, and POAT were taken for analysis.
Ultrasound evaluation
At the end of the experiment, an ultrasound examination was performed in the L and EV groups in order to locate and evaluate the morphologic aspect of the ovaries. We used the Ultrasonix Sono Touch Series ultrasound diagnostic system with a linear transducer, 1.5 cm aperture, wide frequency selection range (8–40 MHz), 40 MHz operating frequency, depth of focus <1 cm, spatial resolution <0.1 mm. Transversal sections were performed in order to localize and measure the width and length of every ovary and also to measure the maximum size of every endoovarian fluid formation observed.
Free testosterone assay
Serum free testosterone level was determined by using an ELISA kit purchased from MyBioSource. The minimum rat detectable free testosterone was 0.05 ng/ml. The assay was performed according to the manufacturer's instructions.
OGTT protocol
OGTT test was performed on all rats using an Accu Chek Active glucometer (Roche Diagnostics Ltd). The rats were fasted overnight for 12 h before the fasting glycemia was measured by using blood samples from the tail vein. Each rat received per os (orally or through gavage) 0.5 ml solution containing 400 mg glucose. After 2 h, the glycemia was again assessed.
Oxidative/antioxidative status assessment
Muscle, liver and POAT preparation for oxidative stress parameters determination
Liver and muscle tissue samples were homogenized with a Polytron homogenizer (Brinkman Kinematica, Lucern, Switzerland) for 3 min on ice in phosphate buffered saline (pH 7.4) added at a ratio of 1:4 (w/v). The suspension was centrifuged for 5 min at 3000 g and 4 °C in order to prepare the cytosolic fraction.
POAT was homogenized using a Polytron homogenizer (Brinkman Kinematica, Switzerland). Minced tissue was mixed with a HNTG lysis buffer solution which contained 5.96 g HEPES (50 mM); 4.38 g NaCl (150 mM); 50 ml glycerol (10%); 5 ml triton-X-100 (1%). The homogenate was transferred in 2 ml Eppendorf tubes and was rotated at 4 °C for 30 min. The obtained lysate was centrifuged at 18 000 g at 4 °C for 30 min, resulting three layers. For analysis, only the middle layer was used. Protein concentration was determined using the Bradford assay (Noble & Bailey 2009).
Oxidative stress assay
Oxidative stress analysis included the determination of malondialehyde (MDA), reduced/oxidized glutathione ratio (GSH/GSSG) and glutathione peroxidase (GPx) activity in muscle, liver and POAT. In serum, MDA and GSH/GSSG were also determined.
MDA, the marker most frequently used for lipid peroxidation, was determined by the fluorimetric method with 2-thiobarbituric acid (TBA) as previously described (Conti et al. 1991). Tissue homogenates were heated in a boiling water bath for 1 h in 75 mM K2HPO4, pH=3, containing 10 mM TBA. After cooling, the solution was extracted with 3 ml n-butanol in 0.6 ml TBA. MDA was determined spectrofluorimetrically in the organic phase using a synchronous technique with excitation at 534 nm and emission at 548 nm. MDA is reported as nmol/mg protein.
GSH and GSSG were measured fluorimetrically using O-phtalaldehyde (Hu 1994). Samples were treated with trichloroacetic acid (10%) and centrifuged. A solution of O-phtalaldehyde (1 mg/ml in methanol) was added to supernatants diluted with sodium phosphate buffer 0.1 M/EDTA 5 mM, pH 8.0. After 15 min, the fluorescence was recorded (350 nm excitation and 420 nm emission). GSH and GSSG concentrations were determined using a standard curve and expressed as nmoles/mg protein in tissue homogenate, and as nmoles/ml in serum.
GPx activity was determined with Flohe & Gunzler (1984) method, with some modifications. The reaction mixture consisting 1 mM GSH, 0.24 U/ml glutathione reductase and 0.15 mM NADPH (final concentrations) in 50 mM phosphate buffer (pH 7.0) was incubated at 37 °C for 5 min with the appropriate amounts of tissue homogenates. The assay was initiated with a 12 mM t-butyl hydroperoxide solution. The decrease in absorbance at 340 nm was recorded for 3 min. GPx activity was expressed as U/g protein in muscle and liver, and as U/mg protein in POAT.
Inflammatory markers assessment
Muscle, liver and POAT preparation for WB
Samples of snap-frozen tissue samples were homogenized in lysis buffer containing Igepal-nonidet 1% (Sigma–Aldrich Chemicals GmbH), 1% protease inhibitor complex (Sigma–Aldrich Chemicals GmbH) in PBS for 1 h, on ice. Cell extracts were spun at 14 000 g for 30 min at 4 °C. Supernatant was collected and 50 μl were used to determine the protein content by the Bradford method. Lysates were mixed 1:2 (v/v) with Laemli sample buffer (Bio-Rad) containing 2-mercaptoethanol and the proteins were denaturized at 95 °C for 10 min. The protein concentration in each sample was 4 mg/ml.
Western blot analysis of NFKB, pNFKB and MCP1
Lysates (20 μg protein/lane) were separated by electrophoresis. Samples were separated by 12% SDS–PAGE gels and transferred to PVDF membranes using Bio-Rad Miniprotean system (Bio-Rad). Blots were blocked for 1 h at room temperature and then were incubated with the primary antibodies against NFKB, pNFKB, MCP1 and GAPDH 1:200, overnight at 4 °C. After washing with PBS-T, the blots were incubated with specific secondary antibodies for 90 min at room temperature. The proteins were visualized and detected using Supersignal West Femto Chemiluminiscent substrate (Thermo Fisher Scientific, Rockford, IL, USA), and were quantified using Image Lab analysis software (Bio-Rad); GAPDH was used as protein loading control.
ELISA analysis
TNFα and STAT3 were determined in the POAT, which was prepared as was described for the WB analysis. The TNFα and STAT3 immunoassay kits were purchased from R&D Systems, Inc. (Minneapolis, MN, USA) and from Life Technologies respectively and were used according to manufacturer's instructions.
For the ELISA analysis (free testosterone, TNFα and STAT3) and for the investigation of OS parameters the number of analyzed samples was determined by using the ‘resource equation’ (Mead 1990), while the WB analysis was performed in triplicate.
Statistical analysis
Obtained data were analyzed with GraphPad Prism Software, version 5.0 (San Diego, CA, USA) by using unpaired t-tests. D'Agostino & Pearson omnibus test was used to check the normality of data. All values in text and figures were expressed as mean±s.d., with a limit of statistical significance of P<0.05.
Results
Ultrasound evaluation
At the end of the experiment, ultrasound assessment of the ovaries was performed in the L and EV groups. In the L group, mean ovarian dimensions (length/width) were 4.68±0.39/2.85±0.56 mm and no cystic modifications were observed (Fig. 1A). In the EV group, the mean ovarian dimensions were 2.63±0.34/1.68±0.36 mm with transonic images of 0.73/1.51 mm in average (Fig. 1B). Both length and width of the ovaries were higher in the L group as compared with EV group (P<0.001 and P<0.01 respectively).
Ultrasound assessment of the ovaries at the end of the experiment. In the L group increased dimensions of the ovaries were observed (A), while in the EV group transonic images were encountered (B).
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
Free testosterone assay
Serum free testosterone was significantly higher in the L group as compared with CL (1.48±0.38 ng/ml vs 0.58±0.03 ng/ml, P<0.05), while after EV administration, free testosterone level was not significantly modified in the EV group as compared with CEV (0.85±0.15 ng/ml vs 0.98±0.25 ng/ml, P>0.05) (Fig. 2A).
Serum free testosterone level and OGTT results at the end of the experiment. (A) The serum free testosterone level was increased in the L group as compared with CL (*P<0.05), while no significant changes were noted in the EV experiment. (B) In the L experiment there were no significant differences regarding neither fasting glycemia (G0h) nor 2-h glycemia (G2h). (C) In the EV experiment G0h was significantly increased in the EV group as compared with the CEV (***P<0.001).
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
Oral glucose tolerance test
In the L experiment, there were no significant changes between groups, neither regarding fasting glycemia (65.75±11.76 mg/dl in the L group compared with 71.50±10.20 mg/dl in the CL group; P>0.05) nor 2-h glycemia (97.57±6.53 mg/dl after L administration compared with 102.6±15.6 mg/dl in the corresponding control group; P>0.05) (Fig. 2B). Fasting glycemia was significantly increased in the EV group as compared with CEV (105.2±9.13 mg/dl vs 75.75±8.73 mg/dl, P<0.001), but no significant difference was observed regarding 2-h glycemia (99.73±18.32 mg/dl compared with 110.4±22.09 mg/dl in the CEV group; P>0.05, Fig. 2C).
Oxidative/antioxidative status assessment
For the oxidative/antioxidative status assessment MDA level, GSH/GSSG ratio and GPx activity were evaluated in muscle, liver and in POAT. In addition, MDA level and GSH/GSSG ratio were assessed in serum.
In muscle, MDA level was 0.20±0.03 nmol/mg protein in the L group and 0.26±0.03 nmol/mg protein in the CL group (P<0.05, Fig. 3A) and 0.17±0.02 nmol/mg protein in the EV group and 0.30±0.05 nmol/mg protein in the CEV group (P<0.001, Fig. 3A). The GSH/GSSG ratio was not significantly modified by either L or EV administration (Fig. 3B). GPx activity was significantly decreased in the L group as compared with CL (28.08±3.12 U/g protein vs 44.8±9.83 U/g protein, P<0.05, Fig. 3C).
Oxidative stress parameters determined in muscle. (A) The level of MDA decreased in the L group as compared with CL (*P<0.05) and in the EV group as compared with CEV (***P<0.001). (B) The GSH/GSSG ratio was not significantly modified. (C) The GPx activity decreased in the L group as compared with CL (*P<0.05) and was not significantly modified in the EV group as compared with CEV.
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
In liver, MDA level and GSH/GSSG ratio were not significantly modified by either L or EV administration (Fig. 4A and B). The activity of GPx was significantly diminished in the L group as compared with CL (161.4±16.46 U/g protein vs 188.3±3.87 U/g protein, P<0.05, Fig. 4C) and in the EV group as compared with CEV (190.4±11.33 U/g protein vs 309.2±22.38 U/g protein, P<0.001, Fig. 4C).
Oxidative stress parameters determined in liver. The level of MDA (A) and the GSH/GSSG ratio (B) were not significantly modified. (C) The GPx activity diminished significantly in the L group as compared with CL (*P<0.05) and in the EV group as compared with CEV (***P<0.001).
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
In POAT, MDA level was significantly decreased in the L group as compared with CL (1.05±0.16 nmol/mg protein vs 2.04±0.59 nmol/mg protein, P<0.01, Fig. 5A) and in the EV group as compared with CEV (0.77±0.3 nmol/mg protein vs 1.46±0.56 nmol/mg protein, P<0.05, Fig. 5A). The GSH/GSSG ratio was not significantly modified in either experiments (Fig. 5B), while the activity of GPx enhanced in the L group as compared with CL (28.94±3.3 U/mg protein vs 21.68±2.05 U/mg protein, P<0.01, Fig. 5C) and in the EV group as compared with CEV (31.16±6.33 U/mg protein vs 23.5±3.95 U/mg protein, P<0.05, Fig. 5C).
Oxidative stress parameters determined in POAT. (A) The level of MDA was significantly decreased in the L group as compared with CL (**P<0.01) and in the EV group as compared with CEV (*P<0.05). (B) The GSH/GSSG ratio was not significantly modified. (C) GPx activity was significantly increased in the L group as compared with CL (**P<0.01) and in the EV group as compared with CEV (*P<0.05).
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
In serum, neither MDA level nor GSH/GSSG ratio was significantly modified (Fig. 6A and B). In the L group was observed a decreasing tendency of GSH/GSSG ratio as compared with CL (P=0.0509, Fig. 6B).
Oxidative stress parameters determined in serum. The MDA level (A) and the GSH/GSSG ratio (B) were not significantly modified.
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
Inflammatory markers assessment
This evaluation included the determination of NFKB, pNFKB and MCP1 expression in muscle, liver and POAT. Moreover, in POAT, the levels of TNFα and STAT3 were quantified.
In muscle, the expression of pNFKB increased significantly in both L (0.52±0.008 pNFKB/GAPDH) and EV (0.39±0.01 pNFKB/GAPDH) groups as compared with their corresponding controls: 0.33±0.02 pNFKB/GAPDH in CL, and 0.23±0.01 pNFKB/GAPDH in CEV respectively (both P<0.001, Fig. 7C). NFKB and MCP1 expression was not significantly modified (Fig. 7B and D).
Inflammatory parameters determined in muscle. (A) Western blot analysis of NFKB, pNFKB and MCP1, with GAPDH as loading control protein. The expression of NFKB (B) was not significantly modified while the active form pNFKB (C) increased significantly in L group as compared with CL (***P<0.001) and in the EV group as compared with CEV (***P<0.001). The expression of MCP1 (D) was not significantly changed.
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
In liver, pNFKB expression was significantly higher in the L group (0.68±0.02 pNFKB/GAPDH) as compared with CL (0.21±0.01 pNFKB/GAPDH, P<0.001, Fig. 8C). Likewise, the expression of MCP1 increased in the L group (1.23±0.11 MCP1/GAPDH) as compared with control CL (0.8±0.04 MCP1/GAPDH, P<0.05, Fig. 8D). In the EV group, the expression of NFKB (0.46±0.02 NFKB/GAPDH) diminished significantly as compared with CEV (0.60±0.01 NFKB/GAPDH, P<0.05, Fig. 8B), while pNFKB increased at statistically significant levels (0.36±0.01 pNFKB/GAPDH vs 0.1±0.01 pNFKB/GAPDH in CEV group, P<0.001, Fig. 8C).
Inflammatory parameters determined in liver. (A) Western blot analysis of NFKB, pNFKB and MCP1, with GAPDH as loading control protein. The expression of NFKB (B) was not significantly modified in the L group as compared with CL and was diminished in the EV group as compared with CEV (*P<0.05). The expression of pNFKB (C) was significantly increased in L group as compared with CL (***P<0.001) and in the EV group as compared with CEV (***P<0.001). MCP1 (D) expression was augmented in the L group as compared with CL (*P<0.05) and was not significantly changed in the EV group as compared with CEV.
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
In the POAT, the expression of NFKB was higher in the L group (1.22±0.06 NFKB/GAPDH) as compared with CL (0.57±0.12 NFKB/GAPDH, P<0.01, Fig. 9B). A significant increase was also observed regarding pNFKB expression: 2.83±0.11 pNFKB/GAPDH in the L group vs 1.99±0.09 pNFKB/GAPDH in the CL group (P<0.001, Fig. 9C) and MCP1 expression: 0.89±0.01 MCP1/GAPDH in the L group as compared with 0.62±0.01 MCP1/GAPDH in the CL group (P<0.001, Fig. 9D). In the EV group, the levels of NFKB (1.1±0.05 NFKB/GAPDH) was higher as compared with CEV (0.6±0.03 NFKB/GAPDH, P<0.001, Fig. 9B). A similar increase was observed regarding pNFKB level: 1.18±0.1 pNFKB/GAPDH in the EV group vs 0.68±0.04 pNFKB/GAPDH in the CEV group (P<0.01, Fig. 9C). The level of MCP1 was significantly lower in the EV group (2.16±0.16 MCP1/GAPDH) as compared with CEV (2.82±0.22 MCP1/GAPDH, P<0.05, Fig. 9D). TNFα secretion was not significantly modified in either L or EV group (Fig. 10A), while STAT3 level was lower in the EV group (4.82±1.02 U/mg protein) as compared with CEV (11.03±1.92 U/mg protein, P<0.01, Fig. 10B).
Inflammatory parameters determined in POAT. (A) Western blot analysis of NFKB, pNFKB and MCP1, with GAPDH as loading control protein. The expression of NFKB (B) increased significantly in L group as compared with CL (**P<0.01) and in the EV group as compared with CEV (***P<0.001). The expression of pNFKB (C) increased significantly in L group as compared with CL (***P<0.001) and in the EV group as compared with CEV (**P<0.01). The expression of MCP1 (D) enhanced in the L group as compared with CL (***P<0.001) and significantly decreased in the EV group as compared with CEV (*P<0.05).
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
TNFα and STAT3 levels in POAT determined through ELISA. (A) The level of TNFα was not significantly modified. (B) STAT3 level was not significantly modified in the L group as compared with CL and decreased significantly in the EV group as compared with CEV (**P<0.01).
Citation: REPRODUCTION 151, 4; 10.1530/REP-15-0352
Discussion
Since PCOS is associated with glucose metabolism abnormalities and the role of inflammation and oxidative stress emerge as factors in the pathogenesis of the syndrome, our study adds new information to the characteristics of the letrozole and estradiol valerate induced PCOS in rats.
As usually observed in these murine models (reviewed by Maliqueo et al. 2014b), the free testosterone level was increased after letrozole administration, but was not significantly modified in the EV group at the end of the experiment. Our results show that administration of L or EV has different effects on the glucose metabolism. Both protocols caused inflammation in the examined tissues, but influenced differently the expression of the pro-inflammatory cytokine MCP1. However, neither L nor EV has generated oxidative stress. In addition, the ultrasound assessment of ovaries revealed increased dimensions of the ovaries in the L group as compared with the EV group, result which is comparable with other histopathological findings (Lim et al. 2011). Also, transonic images were visualized after EV administration, in agreement with our previous ultrasonographic and histopathological results (Dăneasă et al. 2014).
Glucose metabolism parameters should be systematically evaluated in PCOS models since they are involved in the long-term complications of the syndrome. While L had no significant effect from this perspective, in the EV group we observed an increased level of fasting glycemia. This result was consistent with our previous study (Dăneasă et al. 2014), but contrary to the results obtained in literature (Stener-Victorin et al. 2005, Dikmen et al. 2012). Slightly elevated glycemic levels are able to promote mitochondrial reactive oxygen species (ROS) production (Nishikawa et al. 2000). A drawback of our study was that the blood and the tissues were not collected in a fasting state, thus the full extent of oxidative/antioxidative modifications could not be quantified. This would have been particularly important in the EV group, which presented increased levels of fasting glycemia.
However, an unexpected finding was that neither protocol generated oxidative stress. On the contrary, the levels of MDA, marker of ROS attack on membrane lipids, were lower in muscle and POAT or were not significantly modified in liver and serum. Dikmen et al. investigated glycemic and oxidative/antioxidative status in an EV-induced PCOS rat model and found similar results in serum. The authors hypothesize that this was probably due either to compensatory mechanisms activated by the oxidative stimuli or to the antioxidative role of sesame oil used as vehicle (Dikmen et al. 2012). However, in our study, in the L group in which CMC was used as vehicle, the MDA level was also lower, so the first hypothesis is more plausible.
Based only on the oxidative stress parameters evaluated in this study, we cannot establish a link with the inflammatory environment observed in muscle, liver and POAT. Both L and EV led to an increase expression of pNFKB at these sites, which are usually involved in insulin resistance in human PCOS. NFKB controls the synthesis of cytokines involved in inflammation and can be activated through phosphorylation by various stimuli, among which ROS, hyperglycemia and hyperandrogenism. This results in a subsequent increase in pro-inflammatory cytokine production, such as TNFα, which mediates insulin-resistance (González et al. 2006, 2012). Indeed, in our study the level of TNFα was increased in the POAT after administration of both L and EV, but not sufficiently to be considered statistically significant. Another study confirms the increased mRNA expression of TNFα in the parametrial adipose tissue of a transgenic PCOS mouse model induced by letrozole (Kauffman et al. 2015). TNFα stimulates the hyperplasia of androgen producing theca cells which promotes hyperandrogenism (Spaczynski et al. 1999). Also, in the long run, inflammation can interfere with insulin signaling, which is an important step in the development of metabolic disturbances. Thus, the pro-inflammatory molecules became targets for various therapeutical interventions. Concomitant administration of carvedilol or semelil (Angipars) with letrozole decreased the level of inflammatory biomarkers, among which TNFα; OS markers were diminished, while the antioxidant capacity was preserved. Moreover, the range of beneficial effects extended to ovarian morphology and physiology (Rezvanfar et al. 2015).
The transcription of pro-inflammatory cytokine MCP1 is also regulated through NFKB pathway. MCP1 is involved in atherogenesis, thus in the long-term complications of PCOS (Niu & Kolattukudy 2009). In our study, the subsequent step following higher levels of pNFKB would have been a parallel increase of MCP1. However, this was not observed in all the cases, suggesting that MCP1 expression is not exclusively dependent on NFKB pathway and that it is tissue specific. The increased expression of MCP1 in POAT after letrozole administration is consistent with the results obtained in the parametrial adipose tissue in another study (Kauffman et al. 2015), but our findings in the POAT after EV administration raised some questions: Why MCP1 was significantly decreased, even though the expressions of NFKB and pNFKB were elevated? Was it possible that EV administration as part of the induction protocol to interfere with MCP1 release in POAT? Some studies revealed that estradiol administration slowed down wound healing process in parallel with the decrease of local MCP1 (Plackett et al. 2015). By contrast, its administration amplified the inflammatory cytokine production in temporomandibular joint, but in this second study, the expression of MCP1 was not determined (Kou et al. 2011).
If indeed EV administration had anti-inflammatory effects, why NFKB and pNFKB were not subsequently decreased? Which other factors are involved in the expression of MCP1 in the POAT following estradiol administration? It is known that STAT3 is both an important transcription factor that controls adipose tissue (Richard & Stephens 2014) and reproductive functions (Robker et al. 2014). Also, its interaction with the NFKB inflammatory pathway was documented (Sarközi et al. 2015). Therefore STAT3 was analyzed in POAT and the same descending trend was observed, which could indicate a possible role in MCP1 release at this site. Because adipose tissue is a highly active metabolic organ, the study should be extended to other possible candidates known for their interaction with NFKB such as Sp1 (Ping et al. 2000) or AP1 (Fujioka et al. 2004). Jun N-terminal kinase (JNK), which is known as ‘Stress Activated Protein Kinase’, is an important component of AP1 transcription factor. The ovarian alterations observed in the EV-induced PCOS rat model were reduced after the inhibition of JNK. The obtained results were assumed to be a consequence of the diminished inflammatory response and fibrosis observed in the treated group (Bulut et al. 2015).
As a conclusion, we observed that both protocols induced an inflammatory environment in the tissues involved in insulin resistance, without generating OS. However, we believe that the EV-induced PCOS, which is used primarily to investigate ovarian morphology, can be appropriate for the study of early glucose metabolism modifications encountered in PCOS. Moreover, the intriguing results found in the POAT raised new questions about the intricate pathways involved in inflammatory cytokines expression, which need further exploration. Last but not least, EV-induced PCOS protocol is easier and more accessible to implement.
Critical analysis of various animal models available has had an indisputable value in the current understanding of PCOS. Deciphering the mechanisms involved in its pathogenesis is important since it is the most frequent cause of anovulatory infertility. Moreover, PCOS is more than a reproductive disorder and the long-term cardiovascular and metabolic risks should be seriously taken into consideration. Inflammation and oxidative stress are relatively new aspects observed in PCOS, thus is important to investigate these models from this perspective, with the purpose of highlighting pathogenetic mechanisms, which ultimately could lead to the development of curative treatments.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This study was funded by the University of Medicine and Pharmacy ‘Iuliu Haţieganu’ Cluj-Napoca, Romania, internal grant no. 1493/3/28.01.2014. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.
Acknowledgements
We would like to express our gratitude to Ms Nicoleta Decea for technical contribution and to Mr Remus Moldovan for animal handling.
References
Abbott DH, Nicol LE, Levine JE, Xu N, Goodarzi MO & Dumesic DA 2013 Nonhuman primate models of polycystic ovary syndrome. Molecular and Cellular Endocrinology 373 21–28. (doi:10.1016/j.mce.2013.01.013)
Azziz R, Carmina E, Dewailly D, Diamanti-Kandarakis E, Escobar-Morreale HF, Futterweit W, Janssen OE, Legro RS, Norman RJ & Taylor AE et al. 2009 The Androgen Excess and PCOS Society criteria for the polycystic ovary syndrome: the complete task force report. Fertility and Sterility 91 456–488. (doi:10.1016/j.fertnstert.2008.06.035)
Broekmans FJ, Knauff EAH, Valkenburg O, Laven JS, Eijkemans MJ & Fauser BC 2006 PCOS according to the Rotterdam consensus criteria: change in prevalence among WHO-II anovulation and association with metabolic factors. BJOG 113 1210–1217. (doi:10.1111/j.1471-0528.2006.01008.x)
Bulut G, Kurdoglu Z, Dönmez YB, Kurdoglu M & Erten R 2015 Effects of jnk inhibitor on inflammation and fibrosis in the ovary tissue of a rat model of polycystic ovary syndrome. International Journal of Clinical and Experimental Pathology 8 8774–8785.
Conti M, Morand PC, Levillain P & Lemonnier A 1991 Improved fluorometric determination of malonaldehyde. Clinical Chemistry 37 1273–1275.
Dăneasă A, Cucolaş C, Furcea M, Bolfa P, Dudea S, Olteanu D, Alupei MC, Mureşan A & Filip GA 2014 Spironolactone and dimethylsulfoxide effect on glucose metabolism and oxidative stress markers in polycystic ovarian syndrome rat model. Experimental and Clinical Endocrinology & Diabetes 122 154–162. (doi:10.1055/s-0033-1363685)
Dikmen A, Ergenoglu AM, Yeniel AO, Dilsiz OY, Ercan G & Yilmaz H 2012 Evaluation of glycemic and oxidative/antioxidative status in the estradiol valerate-induced PCOS model of rats. European Journal of Obstetrics, Gynecology, and Reproductive Biology 160 55–59. (doi:10.1016/j.ejogrb.2011.09.042)
Duleba AJ & Dokras A 2012 Is PCOS an inflammatory process? Fertility and Sterility 97 7–12. (doi:10.1016/j.fertnstert.2011.11.023)
Flohé L & Günzler WA 1984 Assays of glutathione peroxidase. Methods in Enzymology 105 114–121. (doi:10.1016/S0076-6879(84)05015-1)
Fujioka S, Niu J, Schmidt C, Sclabas GM, Peng B, Uwagawa T, Li Z, Evans DB, Abbruzzese JL & Chiao PJ et al. 2004 NF-κB and AP-1 connection: mechanism of NF-κB-dependent regulation of AP-1 activity. Molecular and Cellular Biology 24 7806–7819. (doi:10.1128/MCB.24.17.7806-7819.2004)
González F 2012 Inflammation in polycystic ovary syndrome: underpinning of insulin resistance and ovarian dysfunction. Steroids 77 300–305. (doi:10.1016/j.steroids.2011.12.003)
González F, Rote NS, Minium J & Kirwan JP 2006 Reactive oxygen species-induced oxidative stress in the development of insulin resistance and hyperandrogenism in polycystic ovary syndrome. Journal of Clinical Endocrinology and Metabolism 91 336–340. (doi:10.1210/jc.2005-1696)
González F, Nair KS, Daniels JK, Basal E & Schimke JM 2012 Hyperandrogenism sensitizes mononuclear cells to promote glucose-induced inflammation in lean reproductive-age women. American Journal of Physiology. Endocrinology and Metabolism 302 E297–E306. (doi:10.1152/ajpendo.00416.2011)
Hu ML 1994 Measurement of protein thiol groups and glutathione in plasma. Methods in Enzymology 233 380–385. (doi:10.1016/S0076-6879(94)33044-1)
Kafali H, Iriadam M, Ozardali I & Demir N 2004 Letrozole-induced polycystic ovaries in the rat: a new model for cystic ovarian disease. Archives of Medical Research 35 103–108. (doi:10.1016/j.arcmed.2003.10.005)
Karakas SE, Kim K & Duleba AJ 2010 Determinants of impaired fasting glucose versus glucose intolerance in polycystic ovary syndrome. Diabetes Care 33 887–893. (doi:10.2337/dc09-1525)
Kauffman AS, Thackray VG, Ryan GE, Tolson KP, Glidewell-Kenney CA, Semaan SJ, Poling MC, Iwata N, Breen KM & Duleba AJ et al. 2015 A novel letrozole model recapitulates both the reproductive and metabolic phenotypes of polycystic ovary syndrome in female mice. Biology of Reproduction 93 69. (doi:10.1095/biolreprod.115.131631)
Kou XX, Wu YW, Ding Y, Hao T, Bi RY, Gan YH & Ma X 2011 17β-estradiol aggravates temporomandibular joint inflammation through the NF-kB pathway in ovariectomized rats. Arthritis and Rheumatism 63 1888–1897. (doi:10.1002/art.30334)
Lim SC, Jeong MJ, Kim SE, Kim SH, Kim SC, Seo S-Y, Kim T, Kang SS & Bae CS 2011 Histologic comparison of polycystic ovary syndrome induced by estradiol valerate and letrozole. Korean Journal of Obstetrics and Gynecology 54 294–299. (doi:10.5468/KJOG.2011.54.6.294)
Maliqueo M, Benrick A & Stener-Victorin E 2014 Rodent models of polycystic ovary syndrome: phenotypic presentation, pathophysiology, and the effects of different interventions. Seminars in Reproductive Medicine 32 183–193. (doi:10.1055/s-0034-1371090)
McNeilly AS & Colin Duncan W 2013 Rodent models of polycystic ovary syndrome. Molecular and Cellular Endocrinology 373 2–7. (doi:10.1016/j.mce.2012.10.007)
Mead R 1990. The Design of Experiments: Statistical Principles for Practical Applications. p587. Cambridge: Cambridge University Press
Nishikawa T, Edelstein D, Du XL, Yamagishi S, Matsumura T, Kaneda Y, Yorek MA, Beebe D, Oates PJ & Hammes HP et al. 2000 Normalizing mitochondrial superoxide production blocks three pathways of hyperglycaemic damage. Nature 404 787–790. (doi:10.1038/35008121)
Niu J & Kolattukudy PE 2009 Role of MCP-1 in cardiovascular disease: molecular mechanisms and clinical implications. Clinical Science 117 95–109. (doi:10.1042/CS20080581)
Noble JE & Bailey MJA 2009 Quantitation of protein. Methods in Enzymology 463 73–95. (doi:10.1016/S0076-6879(09)63008-1)
Padmanabhan V & Veiga-Lopez A 2013 Sheep models of polycystic ovary syndrome phenotype. Molecular and Cellular Endocrinology 373 8–20. (doi:10.1016/j.mce.2012.10.005)
Ping D, Boekhoudt G, Zhang F, Morris A, Philipsen S, Warren ST & Boss JM 2000 Sp1 binding is critical for promoter assembly and activation of the MCP-1 gene by tumor necrosis factor. Journal of Biological Chemistry 275 1708–1714. (doi:10.1074/jbc.275.3.1708)
Plackett TP, Gregory MS & Kovacs EJ 2015 Effects of high estrogen levels on monocyte chemoattractant protein-1 and wound healing. Advances in Wound Care 4 92–99. (doi:10.1089/wound.2014.0597)
Rotterdam ESHRE/ASRM-Sponsored PCOS Consensus Workshop Group Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome Fertility and Sterility 81 2004 19–25. (doi:10.1016/j.fertnstert.2003.10.004)
Rezvanfar MA, Saeedi S, Mansoori P, Saadat S, Goosheh M, Shojaei Saadi HA, Baeeri M & Abdollahi M 2015 Dual targeting of TNF-α and free radical toxic stress as a promising strategy to manage experimental polycystic ovary. Pharmaceutical Biology 54 80–90. (doi:10.3109/13880209.2015.1014922)
Richard AJ & Stephens JM 2014 The role of JAK-STAT signaling in adipose tissue function. Biochimica et Biophysica Acta 1842 431–439. (doi:10.1016/j.bbadis.2013.05.030)
Robker RL, Watson LN, Robertson SA, Dunning KR, McLaughlin EA & Russell DL 2014 Identification of sites of STAT3 action in the female reproductive tract through conditional gene deletion. PLoS ONE 9 1–9. (doi:10.1371/journal.pone.0101182)
Sabuncu T, Vural H, Harma M & Harma M 2001 Oxidative stress in polycystic ovary syndrome and its contribution to the risk of cardiovascular disease. Clinical Biochemistry 34 407–413. (doi:10.1016/S0009-9120(01)00245-4)
Sarközi R, Corazza U, Osterkamp J-P, Pirklbauer M, Mayer G & Schramek H 2015 Synergistic induction of CCL2/MCP-1 expression driven by oncostatin M and IL-1 in human proximal tubular cells depends on STAT3 and p65 NF B/RelA. Physiological Reports 3 e12298. (doi:10.14814/phy2.12298)
Shi D & Vine DF 2012 Animal models of polycystic ovary syndrome: a focused review of rodent models in relationship to clinical phenotypes and cardiometabolic risk. Fertility and Sterility 98 185–193. (doi:10.1016/j.fertnstert.2012.04.006)
Simard M, Brawer JR & Farookhi R 1987 An intractable, ovary-independent impairment in hypothalamo-pituitary function in the estradiol-valerate-induced polycystic ovarian condition in the rat. Biology of Reproduction 36 1229–1237. (doi:10.1095/biolreprod36.5.1229)
Spaczynski RZ, Arici A & Duleba AJ 1999 Tumor necrosis factor-α stimulates proliferation of rat ovarian theca-interstitial cells. Biology of Reproduction 61 993–998. (doi:10.1095/biolreprod61.4.993)
Stener-Victorin E, Lundeberg T, Waldenström U, Manni L, Aloe L, Gunnarsson S & Janson PO 2000 Effects of electro-acupuncture on nerve growth factor and ovarian morphology in rats with experimentally induced polycystic ovaries 1. Biology of Reproduction 63 1497–1503. (doi:10.1095/biolreprod63.5.1497)
Stener-Victorin E, Ploj K, Larsson B-M & Holmäng A 2005 Rats with steroid-induced polycystic ovaries develop hypertension and increased sympathetic nervous system activity. Reproductive Biology and Endocrinology 3 44. (doi:10.1186/1477-7827-3-44)
Walters KA, Allan CM & Handelsman DJ 2012 Rodent models for human polycystic ovary syndrome. Biology of Reproduction 86 1–12. (doi:10.1095/biolreprod.111.097808)
