Theriogenology 84 (2015) 818–826

Contents lists available at ScienceDirect

Theriogenology journal homepage: www.theriojournal.com

Collagen and matrix metalloproteinase-2 and -9 in the ewe cervix during the estrous cycle M. Rodríguez-Piñón a, *, C. Tasende a, D. Casuriaga a, A. Bielli b, P. Genovese b, E.G. Garófalo a a

Biochemistry Area, Department of Molecular and Cellular Biology, Universidad de la República, Montevideo, Uruguay Histology and Embryology Area, Department of Morphology and Development, Veterinary Faculty, Universidad de la República, Montevideo, Uruguay b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 December 2014 Received in revised form 21 May 2015 Accepted 21 May 2015

The cervical collagen remodeling during the estrous cycle of the ewe was examined. The collagen concentration determined by a hydroxyproline assay and the area occupied by collagen fibers (%C), determined by van Gieson staining, were assessed in the cranial and caudal cervix of Corriedale ewes on Days 1 (n ¼ 6), 6 (n ¼ 5), or 13 (n ¼ 6) after estrous detection (defined as Day 0). In addition, the gelatinase activity by in situ and SDS-PAGE gelatin zymographies and matrix metalloproteinase-2 and -9 (MMP-2 and MMP-9, respectively) expression by immunohistochemistry were determined. The collagen concentration and %C were lowest on Day 1 of the estrous cycle (P < 0.04), when MMP-2 activity was highest (P < 0.006) and the ratio of activated to latent MMP-2 trend to be highest (P ¼ 0.0819). The MMP-2 activity was detected in 73% of the homogenized cervical samples, and its expression was mainly detected in active fibroblasts. By contrast, the MMP-9 activity was detected in 9% of the samples, and its scarce expression was associated with plasmocytes, macrophages, and lymphocytes. Matrix metalloproteinase-2 expression was maximal on Day 1 in the cranial cervix and on Day 13 in the caudal cervix and was lower in the cranial than in the caudal cervix (P < 0.0001). This time-dependent increase in MMP-2 expression that differed between the cranial and caudal cervix may reflect their different physiological roles. The decrease in the collagen content and increase in fibroblast MMP-2 activity in sheep cervix on Day 1 of the estrous cycle suggests that cervical dilation at estrus is due to the occurrence of collagen fiber degradation modulated by changes in periovulatory hormone levels. Ó 2015 Elsevier Inc. All rights reserved.

Keywords: Metalloproteinase Collagen Cervical remodeling Estrous cycle Ewe

1. Introduction The tortuous nature of the ovine cervix restricts transcervical artificial insemination and embryo transfer procedures [1–3]. However, natural cervical dilatation occurs at estrus [4], and many studies have examined the physiological mechanism of cervical dilatation for transcervical cannulation improvement [5,6]. Fibrillar collagen and

* Corresponding author. Tel.: þ598 2622 1195; fax: þ598 2628 0130. E-mail address: [email protected] (M. Rodríguez-Piñón). 0093-691X/$ – see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.theriogenology.2015.05.017

high-molecular-weight proteoglycan complexes are the main components of the extracellular matrix (ECM) of the cervical connective tissue [7–9]. The biochemical interactions between these structural elements are critical to the cervical remodeling process that results in cervical dilation [10,11]. In the ewe, the proposed model for cervical dilatation at estrus involves a central role of periestrous endocrine changes that drive ECM remodeling processes and, consequently, cervical dilatation [4–6]. These periestrous endocrine changes include the preovulatory increase of estradiol and gonadotropins [4–6,12] and the activation of the

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

prostaglandin E2/oxytocin (PGE2/Ox) system [13–15]. Cervical production of PGE2 stimulates smooth muscle relaxation and hyaluronan-like glycosaminoglycan (GAG) synthesis via an autocrine or paracrine mechanism, causing disaggregation of collagen fibers and cervical dilatation [16,17]. In other species, some evidence indicates that degradation of collagen fibers may also be involved in cervical dilatation. For example, a decrease in collagen content was measured chemically [18] and histologically in pregnant women at term [19]. The degradation of collagen in the ECM depends on the activity of matrix metalloproteinases (MMPs), which are the only enzymes capable of degrading denatured fibrillar collagen [20]. In particular, the expression of MMP-2 and -9 (also called gelatinases A and B, respectively) increases in the human cervix at the end of pregnancy [21], indicating a likely role of MMP-2 and -9 in the cervical dilatation process. We hypothesized the coexistence of increased collagen fiber disaggregation and increased enzymatic collagen degradation in the sheep cervix around the estrus. The changes in the collagen content and distribution and in the MMP-2 and -9 abundance and activity along the cervix of the ewe during the estrous cycle were examined, particularly at the expected time of artificial insemination and embryo transfer (Days 1 and 6 after estrus, respectively).

819

longitudinally cut into four equal segments (500–600 mg/ segment). One longitudinal segment for each cervical zone was used to determine the water content by drying until a constant weight was reached at 80  C for 3 hours; this water content was expressed as percentage humidity (%). Another longitudinal segment was weighed, sliced, and homogenized in PBS buffer (1/10, wt/vol) with a Polytron Homogenizer (Polytron Homogenizer PT-10; Kinematica AG, Littau Luzern, Switzerland). Aliquots of homogenates were stored at 80  C until the spectrophotometric and SDS-PAGE zymography assays were performed. The third longitudinal segment was immediately fixed by immersion in buffered 4% formaldehyde (pH 7.4) for 24 hours and then stored in 70% ethanol for 10 days. Fixed cervices were then dehydrated and embedded in paraffin until histochemistry and immunohistochemistry assays were performed. The fourth longitudinal segment was embedded in tissue-freezing medium without fixation and stored at 80  C until the in situ zymography was performed. 2.3. Collagen and total soluble protein content determined by spectrophotometry

The experiment was carried out at the experimental field of the Veterinary Faculty of the University of Uruguay, Canelones, Uruguay (35 S), during the breeding season of Corriedale ewes (February through March). Animal experimentation was performed in compliance with regulations set by the Veterinary Faculty of the University of Uruguay. The adult Corriedale ewes were kept under natural daylight conditions. They grazed on native pastures and were given water ad libitum. Vasectomized rams fitted with marking crayons were kept with the ewes for 2 months before the start of the study to confirm the normal cyclic conditions of the ewes. The estrus was synchronized with two doses of a PGF2a analogue (intramuscularly, 150 mg, Glandinex; Laboratorio Universal, Montevideo, Uruguay) administered 6 days apart. From Day 10 after the second PGF2a treatment, ewes were checked twice daily (at 6 and 18 hours) for service marks of two vasectomized rams carrying marking crayons (day of estrus ¼ Day 0). Seventeen ewes (bodyweight, mean  pooled standard error of the mean, 39.0  1.1 kg) showing spontaneous estrus were slaughtered on Days 1 (n ¼ 6), 6 (n ¼ 5), or 13 (n ¼ 6) after the estrus detection. The day of the estrous cycle for each animal was confirmed by concentrations of circulating estradiol-17b (E2) and progesterone [12].

The collagen content was measured indirectly by a hydroxyproline assay adapted from Bannister and Burns [22]. Aliquots of frozen homogenates were hydrolyzed in constantly boiling hydrochloric acid (3 N) at 90  C for 24 hours. After a partial neutralization (pH ¼ 2.3) with 3-N NaOH, the hydrolyzed samples were exposed to an oxidizing agent (chloramine-T 7%: water: PBS buffer, 7:100:500, v:v:v) for 15 minutes at room temperature (RT). Under these conditions, hydroxyproline was liberated by acid hydrolysis and oxidized to a pyrrole, which then reacted with the color reagent (30 g of 4-dimethylaminobenzaldehyde, 45 mL of 60% perchloric acid, and 250 mL of propan-2-ol) at 70  C for 15 minutes. The absorbance of the resulting colored product was read at 550 nm. Readings were calibrated against standards prepared from L-4-hydroxyproline (Fluka, 56250) dissolved in 0.01-N HCl (0.5–30 mg/mL, r ¼ 0.9976, P < 0.0001). All samples were analyzed in a single assay, with a sensitivity of 0.5 mg/mL and intra-assay coefficient of variation of 7%. The collagen concentration was calculated assuming that the hydroxyproline/collagen ratio is 14% [23] and was expressed relative to dry tissue mass (mg/g of dry tissue). The total soluble protein concentration in the aliquots of frozen homogenates was determined by the method of Lowry et al. [24], using BSA (Fraction V, Sigma Chemical, St. Louis, MO, USA) as the standard (0.05–0.8 mg/mL, r ¼ 0.9960, P < 0.0001). All samples were analyzed in a single assay, with a sensitivity of 0.05 mg/mL and intraassay coefficient of variation of 4%. The total soluble protein concentrations (mg/g of dry tissue) were positively correlated with the amount of tissue used (r ¼ 0.7123, n ¼ 86, P < 0.0003), showing that the total proteins extracted were similar among cervical samples.

2.2. Cervical samples, wet weight, and water content

2.4. Collagen distribution determined by van Gieson staining

The cervices were weighed and dissected at a temperature of 0  C to 4  C into three transversal segments of equal length labeled cranial, middle, and caudal cervical zones (2–2.5 g per cervical zone). The cranial and caudal cervical zones were

Van Gieson’s picrofuchsin stain was used to observe the connective tissue fibers in deparaffinized and rehydrated cervical sections (5 mm). Sections were stained with iron hematoxylin for 10 minutes, washed in running water for

2. Materials and methods 2.1. Animals and treatments

820

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

15 minutes, and rinsed with distilled water. After differentiation with 0.5% acid alcohol for 2 minutes, the slides were stained with van Gieson stain (alcoholic picric acid and acid fuchsine) for 5 minutes, dehydrated, and mounted in Entellan (Merck, Darmstadt, Germany). All cervical samples were analyzed in a single assay. After a general inspection of each slide, the stroma was divided into four histologic compartments (SHC) depending on cell density and localization according to Rodríguez-Piñón et al. [15]: superficial fold stroma (SFS), deep fold stroma (DFS), superficial wall stroma (SWS), and deep wall stroma (DWS). To estimate the percentage of collagen fibers (red stain), a quantitative image analysis was performed on selected digitized images captured by an Olympus microscope and infinity camera connected to a computer. Ten digitized images ( 400) of systematic randomly selected fields of each SHC for two slides per cervical zone per ewe were analyzed separately by removing the other SHCs (Photoshop 6.0). Color-discrimination software (Image-Pro Express 6.0) was used to apply a threshold for red staining by a color detection system; it creates a binary image from which the percentage of the total area that contains collagen fibers (%C) is automatically estimated. 2.5. Gelatinase and collagenase activity detected by in situ zymography Frozen tissue sections were examined for gelatinase and collagenase activity using in situ zymography. Fluorescein conjugates DQ gelatin from pig skin and DQ collagen type 1 from bovine skin (Molecular Probes, Inc., Eugene, OR, USA) were used as substrates. Seven-micron sections were cut from each frozen cervical segment, placed on poly-Llysine–coated slides, fixed in 10% buffered formalin for 5 minutes at 48  C, and washed three times with cold Trisbuffered saline (TBS). Nuclear counterstaining was performed with propidium iodide (Molecular Probes) diluted 1:50 (wt/vol) in TBS for 8 minutes at RT. After washing, the slides were maintained in a darkened TBS bath at 48  C until use. The desired substrate (DQ gelatin or DQ collagen) was dissolved to a final concentration of 25 mg/mL in a mixture of 2% gelatin, 2% sucrose, and 0.02% sodium azide in TBS. The substrate solution was layered over the tissue section, covered with a coverslip, and incubated in a darkened humidity chamber at 37  C for 16 hours. For control sections, the broad-spectrum MMP inhibitor 1,10phenanthroline (Sigma) was added at a final concentration of 10 mM and incubated at 37  C for 1 hour before counterstaining. Each section was viewed using an Olympus Corp. (Birkerød, Denmark) fluorescent microscope with a fluorescein isothiocyanate filter. 2.6. MMP-2 and -9 activity detected by gel electrophoresis (SDS-PAGE zymography) Aliquots of frozen homogenates were diluted 1:1 in a 50mM Tris/HCl sample buffer containing 10% glycerol, 2% SDS, and 0.0025% bromophenol blue at pH 7.6 and incubated at 37  C for 1 hour before electrophoresis. The samples were loaded (10 mL) onto 1-mm-thick polyacrylamide gels (10%) copolymerized with gelatin (2.5 mg/mL, G9391; Sigma) and

electrophoresed at 100 V for 2 hours. After washing, the gels were incubated in 50-mM Tris/HCl buffer (5 mM of CaCl2, 200 mM of NaCl, and 0.005% Brij 35, Sigma, pH 7.8) for 24 hours at 37  C. The gels were then stained with 1% Coomassie blue R-250 (Sigma) and destained. Gelatin degradation was manifest as clear zones on a blue-stained gel. The gels were scanned, and the images were analyzed using the menu option Analyze->Measure in ImageJ 1.46r to measure the area and number of pixels for each clear band, then converted to standard units using the Set Scale function. The concentration of both latent (L) and activated (A) forms of MMP-2 and -9 in the samples were calculated from calibration curves prepared from human recombinant standards (M9445, Sigma; 5–0.08 ng/10 mL, r ¼ 0.9654 and r ¼ 0.9794 for L and A MMP-2 forms, respectively, and M8945, Sigma; 0.5–0.008 ng/10 mL, r ¼ 0.9629 and r ¼ 0.9756 for L and A MMP-9 forms, respectively, P < 0.0001). Low, medium, and high MMP-2 and -9 standard concentrations were loaded in each gel, with interassay coefficients of variation of 11%, 8%, and 26%, and 6%, 11%, and 23%, respectively. Two samples were treated with 20-mM EDTA (Sigma) as negative controls, and two were treated with 1-mM p-aminophenylmercuric acetate (Sigma), for a metalloproteinase-specific activation test. The activities of both L and A forms of MMP-2 and -9 in the homogenates were expressed in ng/mg protein using the total soluble protein concentration (see Section 2.4), and the A/L ratios were calculated for both MMP-2 and -9 isoenzymes. 2.7. MMP-2 and -9 histology distribution assessed by immunohistochemistry Sections (5 mm) of cervical samples were deparaffinized and rehydrated before antigen retrieval treatment, consisting of steam heating in 10-mM sodium citrate buffer, pH 6.0 for 30 minutes at 100  C. After rinsing, endogenous peroxidase activity was blocked by 30% hydrogen peroxide in methanol for 10 minutes at RT in a humidified chamber. To prevent nonspecific reactions, samples were incubated with normal horse serum (Vectastain Elite ABC Kit; Vector Laboratories, Burlingame, CA, USA) for 30 minutes at RT and then treated with a primary antibody. Goat polyclonal MMP-2 and -9 primary antibodies against the C-terminus amino acidic sequence of human MMPs (hMMPs) were used (K-20, sc-8835 and C-20, sc-6840, respectively, Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA); they recognize both L and A forms. Primary antibodies (dilution 1/100) were incubated at RT for 1 hour. Replacement of the primary antibody with an equivalent concentration of normal horse serum (sc-2025; Santa Cruz Biotechnology) was used as a negative control. After primary antibody binding, sections were rinsed and incubated with a biotinylated horse antimouse IgG secondary antibody (Vectastain Elite ABC Kit, Mouse IgG, Cat # PK-6102) at a dilution of 1:200 in normal horse serum for 60 minutes. After rinsing, the sections were incubated in a horseradish avidin–biotin peroxidase complex (Vectastain Elite ABC Kit) for 60 minutes and then in 3,30 -diaminobenzidine (DAB Kit, sk-4100; Vector Laboratories) for 60 seconds. All sections were counterstained with Mayer’s hematoxylin and mounted with Entellan. Due to scarce and sporadic positive

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

immunostaining obtained with the MMP-9 antibody, a quantitative image analysis was only performed on the selected digitized images for MMP-2 immunostaining. The number of positive (brown stained) and negative (blue stained) cells of each SHC (defined in Section 2.4) was counted for a sample of 1000 cells at a magnification of  400. Brown-stained cells were considered to be positive, irrespective of the intensity of the color. The proportion of positive cells relative to the total number of cells was calculated (%MMP-2), and the MMP-2–positive cell density was estimated (MMP-2 cells/mm2) using the area measurement tool of the software (Image-Pro Express 6.0). 2.8. Statistical analysis Data were analyzed by ANOVA using the MIXED procedure implemented in Statistical Analysis Systems (SAS Institute, Cary, NC, USA). The model included the fixed effects of day of estrous cycle (Days 1, 6, or 13), cervical zone (cranial or caudal), and their interactions. For immunohistochemistry and histochemistry determinations, the ANOVA model also included the fixed effect of SHC (SFS, DFS, SWS, or DWS) and their interactions. For cervical weight, only the effect of day of estrous cycle was considered. The %MMP-2 and the MMP-2–positive cell density showed skewed and nonnormal distributions (Kolmogorov–Smirnov Test, Statgraphics Centurion XV, 2011). Before the analysis, these variables were log transformed and the variance homogeneity between groups was confirmed (Statgraphics Centurion XV, 2011). The results are expressed as the least-square mean  pooled standard deviation, and P < 0.05 was considered statistically significant, unless otherwise specified. 3. Results 3.1. Cervical weight and water content The cervical wet weight (g) was greater on Day 1 than on Days 6 and 13 (10.4  0.4, 7.4  0.5, and 7.9  0.9, respectively, P < 0.005). No effect of day of estrous cycle or cervical zone on water content was found, which ranged from 66% to 74%. 3.2. Collagen and total soluble protein content A significant effect of day of estrous cycle on collagen concentration (P < 0.04) was found. The collagen concentration was lower on Day 1 than on Days 6 and 13 (Table 1).

821

There was an effect of day of estrous cycle on total soluble protein concentration (P < 0.0002). The total soluble protein concentration was lower on Days 1 and 6 than on Day 13 (Table 1). 3.3. Collagen distribution There was an effect of day of estrous cycle (P < 0.002), cervical zone (P < 0.0001), and SHC (P < 0.0001) on %C, as well as an interactive effect among stroma types (P < 0.005). The %C was lower on Days 1 and 6 than on Day 13 (37.8  1.7%, 37.9  1.5%, and 43.0  1.8%, respectively) and was greater in the cranial than in the caudal cervix (47.1  1.4% and 32.7  2.7%, respectively). The %C differed between all SHCs evaluated (25.0  2.1%, 34.4  3.7%, 44.5  2.8%, and 53.0  1.8% for SFS, DFS, SWS, and DWS, respectively). In SFS, the %C was greater on Days 1 and 6 than on Day 13 in the cranial cervix, whereas it was lower on Day 1 than on Day 13 in the caudal cervix (Table 2). In DFS, the %C was greater on Day 1 than on Days 6 and 13 in the cranial cervix and was greater on Day 1 than on Day 13 in the caudal cervix. In SWS and DWS, no significant difference between days in the %C was found in the cranial cervix. However, it increased from Day 1 to Day 13 in SWS and from Days 1 and 6 to Day 13 in DWS in the caudal cervix (Fig. 1). 3.4. Gelatinase and collagenase activity estimated by in situ zymography Using in situ analysis, we found that gelatinase and collagenase activity were localized primarily in the extracellular space, although pericellular localization was also visible (not shown). Negative controls performed with 1,10phenanthroline (a zinc ion chelator) completely inhibited the fluorescence, showing that both gelatinase and collagenase activities were due to MMP activity. The gelatinase activity was localized in the epithelium and in both fold and wall stromata (Fig. 2A), whereas the collagenase activity was restricted to the epithelium (Fig. 2B). 3.5. MMP-2 and -9 activity detected by SDS-PAGE zymography Using SDS-PAGE, two intense bands that migrated at 72 and 62 kDa and two weaker ones that migrated at 92 and 86 kDa were found, corresponding to the clear bands obtained with the L and A forms of hMMP-2 and -9 standards, respectively (Fig. 3A). A clear band of w130 to 140 kDa of unknown origin was found in four cervical

Table 1 Collagen and total soluble protein concentration (mg/g of dry tissue), gelatinase activity (ng/mg of protein) of latent (L) and activated (A) forms of matrix metalloproteinase-2 (MMP-2), and A/L ratio in cervices of ewes on Days 1, 6, and 13 after estrus (Day ¼ 0). Days after the estrous detection (Day 0)

Collagen concentration (mg/g dry tissue)

Total soluble proteins concentration (mg/g dry tissue)

L MMP-2 (ng/mg of protein)

A MMP-2 (ng/mg of protein)

A/L ratio

Day 1 (n ¼ 6) Day 6 (n ¼ 5) Day 13 (n ¼ 6)

151  15a 200  18b 201  12b

270  18a 294  25a 411  20b

8.0  2.8a 2.4  1.0b d

22.3  6.3a 2.4  1.1b d

3.4  1.1 1.0  0.4 d

a,b Values (mean  standard error of the mean) within each column that are marked with different letters differ significantly (P < 0.04). On day 13, the L and A forms of MMP-2 were detected only in one (1.0 ng/mg of protein) and two samples (1.1 and 0.2 ng/mg of protein), respectively.

822

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

Table 2 Area occupied by collagen fibers expressed as a percentage (%C) in different stromal histologic compartments of the cervices of ewes on Days 1, 6, and 13 after estrus (Day ¼ 0). Days after the estrous detection (Day 0)

SFS

DFS

SWS

DWS

Cranial

Caudal

Cranial

Caudal

Cranial

Caudal

Cranial

Caudal

Day 1 (n ¼ 6) Day 6 (n ¼ 5) Day 13 (n ¼ 6)

36.0  1.1a 37.4  1.1a 26.5  2.2b

13.6  4.3a 17.5  3.7ab 22.4  5.0b

58.5  4.1a 48.4  3.3b 47.2  3.2b

24.1  2.6a 13.4  2.2b 18.4  2.7ab

51.2  0.8 51.0  1.2 55.3  2.3

18.4  2.7a 40.1  2.3b 51.9  5.6c

48.8  1.5 52.1  4.3 51.4  1.2

52.3  1.9a 45.7  1.4a 69.5  2.3b

a,b,c Percentages (mean  standard error of the mean) within a column that are marked with different letters differ significantly (P < 0.005). SFS, DFS, SWS, and DWS are histologic compartments. Abbreviations: DFS, deep fold stroma; DWS, deep wall stroma; SFS, superficial fold stroma; SWS, superficial wall stroma.

samples (Fig. 3A, line 5). A clear band of gelatinase activity at w200 kDa in the hMMP-9 standard and in two of the 34 cervical samples was found (Fig. 3B, line 1); these bands were artifacts of MMP-9 dimerization. Treatment of samples with EDTA removed both MMP-2 and -9 gelatinase activity bands, whereas p-aminophenylmercuric acetate treatment reduced the bands corresponding to the L-form and increased the bands corresponding to the A-form (Fig. 3B, line 1), indicating that gelatinase activity is specifically exerted by MMPs. The gelatinase activity corresponding to the L and A forms of MMP-2 was detected in 25 of the 34 cervical samples, whereas gelatinase activity corresponding to the L and A forms of MMP-9 was only detected in three of the 34 cervical samples (Fig. 3A). Therefore, only bands corresponding to the L and A MMP-2 forms were quantified (Table 1). There was an effect of day of estrous cycle (P < 0.006) on the activity of both L and A MMP-2 forms and a tendency on the effect of day of estrous cycle on A/L ratio (P ¼ 0.0819). The L and A MMP-2 activities were higher on Day 1 than on Day 6, and they were only detected in one and two samples on Day 13, respectively. The ratio of A/L MMP-2 trends to be higher on Day 1 than on Day 6 (P ¼ 0.0819).

occasionally leukocytes (Fig. 4A, B). Weak and sporadic MMP-9 immunostaining was detected, which was limited to the cytosol of some stromal cells. Positive cells were mainly plasmocytes, some macrophages, a few lymphocytes, and very few fibroblasts (Fig. 4A, C). Therefore, only the immunohistochemical signal corresponding to MMP-2 was quantified. There was an effect of cervical zone (P < 0.0001), SHC (P < 0.007), and an interaction between day of estrous cycle and cervical zone (P < 0.0001) on %MMP-2. The %MMP-2 (%) was lower in the cranial (0.32  0.01%) than in the caudal (1.96  0.06%) cervix. It was lower in DWS than in DFS and SWS and trends to be lower in SFS than in SWS (0.49  0.03%, 1.14  0.06%, 1.34  0.06%, and 0.24  0.02% in SFS, DFS, SWS, and DWS, respectively). The %MMP-2 decreased throughout the estrous cycle in the cranial cervix but was higher on Day 13 than on Days 1 and 6 in the caudal cervix (Table 3). There was an effect of cervical zone (P < 0.0001) and SHC (P < 0.0004) on MMP-2–positive cell density. The MMP-2 cells density was lower in the cranial (569  106) than in the caudal (2917  401) cervix. It was higher in SFS, DFS, and SWS than in DWS (1644  265, 1180  213, 2070  367, and 409  89, respectively).

3.6. Characterization of MMP-2 and -9 by immunohistochemistry

4. Discussion

The MMP-2 immunostaining was localized in the extracellular space for all SHCs and was primarily detected in active fibroblasts but also in inactive fibroblasts and

We reported for the first time in the ovine cervix that the collagen content is lower and MMP-2 activity is higher in estrus than in the luteal phase of the natural estrous cycle.

Fig. 1. Images of collagen fibers with van Gieson staining from superficial wall stroma of the caudal cervix from the same ewe on Days 1 (A) and 13 (B) after estrus detection (Day 0). Note the lower density of fibers on Day 1 than on Day 13.

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

823

Fig. 2. Gelatinase (A) and collagenase (B) activities by in situ zymography ( 400) in the cervix of cycling ewes. Note the presence of gelatinase activity in both the epithelium and stroma, whereas the collagenase activity was restricted to the epithelium.

Both the cervical collagen concentration and distribution were lower 1 day after estrus detection compared with during the luteal phase, suggesting that either the preovulatory estrogen levels increase collagen degradation or the progesterone luteal levels inhibit it. Consistent with these interpretations, the cervical collagen concentration and distribution remain unchanged throughout most of ovine pregnancy but decrease in the final month [25], when circulating progesterone begins to decrease and estrogen increases [26]. Because the collagen concentration was expressed with respect to the dry tissue and the water content did not change during the estrous cycle, the decrease in collagen 1 day after estrus detection is due to a genuine collagen degradation and not a dilution effect. In the cervix of cycling cows, both water and collagen content (based on dry tissue) were not associated with the progesterone status [27], suggesting interspecific differences in the magnitude of collagen degradation during cervical softening at estrus. The wet weight of the cervix was higher 1 day after estrus detection than in the luteal phase, despite similar water contents; this is in agreement with the increase in cervical wet weight without changes in water content reported by Regassa and Noakes [25] during ewe pregnancy. Interestingly, there was maximal cervical weight around the estrus, when the collagen content, total soluble proteins, cell proliferation [28], and nuclear density [15] were minimal. These data suggest that the increase in the cervical wet weight around estrus is due to an increase in the nonprotein component of the cervical ECM. Hyaluronanlike GAGs are 80% to 90% of all GAGs in the sheep cervix and increase before the LH preovulatory surge in the cervix of estrus-synchronized ewes [16,17]. Overall data suggest that the ovine cervical dilatation around the estrus is a consequence of a tissue remodeling process involving both an increase in collagen degradation and an increase in hyaluronan-like GAG synthesis, which have been suggested in the human cervix during pregnancy [7,29]. In the present work, the observed gelatinase activity in the cervix of cycling ewes was found by in situ and SDS-PGE gelatin zymographies. Using in situ zymography, we observed both collagenase and gelatinase activities in the luminal epithelium but only gelatinase activity in the stroma. The stromal gelatinase activity may be due to

the presence of MMP-2 and -9 (gelatinases) because the immunostaining of both MMP-2 and -9 was restricted to stroma. The gelatinase activity bands corresponding to the L and A forms of MMP-2 were detected in all cervical samples, but slight bands of the L and A MMP-9 forms were detected in very few samples. Immunohistochemical findings confirmed these observations because MMP-2–positive cells were detected in all samples, but MMP-9–positive cells were scarce detected or undetected. These results suggest that the gelatinase activity in the cervix of cycling ewes is predominantly caused by the MMP-2 isoenzyme. In addition, MMP-2 immunostaining was mainly associated with active fibroblasts, whereas MMP-9 immunostaining was associated with cells related to immune and inflammatory processes. These data are in agreement with those of Stygar et al. [21], who reported that stromal fibroblasts of the human cervix are the main source of MMP-2, whereas MMP-9 is restricted to leukocytes. The scarce MMP-9 activity levels and its association with cells related to immune and inflammatory processes suggest that this type of processes are not predominant in the cervical ECM modifications during the estrous cycle. No reports describing the occurrence of inflammatory and immune-mediated–like processes were found in ruminant cervices during the estrous cycle. In the ovine cervix, although an increase in cervical interleukin 8 (a proinflammatory cytokine) was detected at estrus [30], administration of interleukin 8 had no effect on cervical penetrability [31]. Interestingly, the levels of the activated form of cervical MMP-2 were approximately 10 to 20 times higher around the estrus, when cervical penetrability is maximal [4], than during the luteal phase. Concomitant with this increase in MMP-2 activity 1 day after estrus detection, a decrease in the cervical collagen concentration and the percentage of collagen fibers was found. Raynes et al. [32] failed to detect changes in cervical collagenase activity in ewes during gestation using a semisynthetic collagen-like substrate. The higher levels of cervical MMP-2 activity found after estrous detection could result from the induction of MMP-2 expression by estrogen during the preovulatory peak and/or the lack of luteal progesterone. The MMP hormonal regulation could occur at the transcriptional and posttranscriptional levels via changes in the rate of messenger RNA synthesis and/or stability (half-life) [33].

824

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

Fig. 3. Gelatin zymography of cervical sample homogenates of ewes during the estrous cycle. (A) Lines 1, 2, and 3: 10 mL of mixed human recombinant standards of latent and activated matrix metalloproteinase (MMP)-2 (72 and 62 kDa, respectively; Sigma–Aldrich, M9445; 5.0, 0.62, and 0.08 ng in each line, respectively) and latent and activated MMP-9 (92 and 86 kDa, respectively; Sigma–Aldrich, M8945; 0.5, 0.062, and 0.008 ng in each line, respectively). Lines 4, 5, and 6: 10 mL of cervical sample homogenates of different ewes on Day 1 after estrus (Day 0). (B) Line 1: 10 mL of cervical sample homogenates. Line 2: 10 mL of cervical sample homogenates preincubated with 1-mM p-aminophenylmercuric acetate (APMA). Note the clear bands of approximately 130 to 140 kDa of unknown origin in panel A, line 5 and 200 kDa at the top of panel B, line 1.

Anuradha and Thampan [34] reported that E2-mediated enhancement of collagenase activity in the rat uterus is inhibited by actinomycin D and cycloheximide, indicating a transcriptional inductive effect of estrogens on collagenase expression. In agreement with this, an increase in uterine MMP-2 expression at 48 hours after E2 treatment has been observed in rats [35]. The relatively prolonged latency between the E2 stimulus and the increase in uterine MMP-2 expression could be explained, at least in part, by the 12to 36-hour half-life of MMP transcripts [36]. Data suggest that the increase in cervical MMP-2 activity 1 day after estrus detection could reflect a previous stimulatory effect of preovulatory estrogens on MMP-2 expression, occurring before the onset of estrus, under maximal concentrations of circulating E2 [37]. This early preovulatory estrogenstimulatory effect on MMP-2 expression could be maintained during the estrus via high levels of cervical estrogen receptors [12,15]. In addition, the MMP-2 A/L ratio trends to be higher on Day 1 after estrus detection than during the luteal phase, indicating that the estrogen-induced increase in MMP-2 activity may be due to an increase in both protein expression and enzyme activation. The increased activity of MMP-2 1 day after the onset of estrus may be due to an inductive effect of preovulatory estrogens, but other stimulatory factors cannot be ruled out. Cervical PGE2 production increases in response to LH [38]

and Ox [39,40] in the bovine cervix, although both PGE2 receptor types 4 and 2 have been reported in the cervix of the ewe [16,41,42] which can mediate MMP secretion [43,44]. In an elegant in vitro experiment in the human cervix, PGE2 treatment decreased the incorporation of [3H] glycine but increased [3H] glucosamine, precursors of collagen, and GAGs in samples obtained during the follicular phase, and the opposite was found in samples obtained during the luteal phase [45]. Overall, the data report an association between various hormones at the control of cervical collagen remodeling. Moreover, low-molecular-weight hyaluronic acid (HA) can induce collagenase and gelatinase activity in the rabbit cervix [46]. The rearrangement and dissociation of collagen fibers and bundles is thought to contribute to cervical relaxation in the ewe at estrus [47,48]. Kershaw et al. [48] found a higher proportion of collagen relative to smooth muscle before the LH surge than during the medium luteal phase (on Day 9), and this result was attributed to estrogeninduced separation of collagen bundles and fibers via HA accumulation [16,17]. In the present work, a pre-LH stage was not assessed; however, 1 day after estrous detection (probably after the LH preovulatory surge), the area occupied by collagen fibers and the collagen concentration were lower, and the cervical MMP-2 activity and A/L ratio were higher than they were in the late luteal phase (on Day 13).

Fig. 4. Immunohistochemical detection of matrix metalloproteinase-2 and -9 (MMP-2 and MMP-9, respectively). Negative controls omitted primary antibodies (bar ¼ 50 mm, A), MMP-2 (bar ¼ 20 mm, B), and MMP-9 (bar ¼ 10 mm, C). Note that the MMP-2–positive immunostaining localized pericellularly to active fibroblasts (arrows, B), whereas the MMP-9–positive immunostaining localized in the cytoplasm of plasmocytes (arrows, C).

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826 Table 3 Matrix metalloproteinase-2–immunopositive cells (%) in the cranial and caudal cervices of ewes on Days 1, 6, and 13 after estrus (Day ¼ 0). Days after the estrous detection (Day 0)

Cranial

Caudal

Day 1 (n ¼ 6) Day 6 (n ¼ 5) Day 13 (n ¼ 6)

0.77  0.03a,* 0.34  0.03a,* 0.10  0.01c

0.94  0.09a 1.17  0.07a,b 3.10  0.07b

a,b,c Values (mean  standard error of the mean) marked with different letters differ significantly (P < 0.0001) and those marked with an asterisk differ but not significantly (P ¼ 0.0892).

Both Kershaw et al. [48] and the present report indicate that collagen fiber disaggregation occurs from the late luteal phase until ovulation, followed by enzymatic collagen degradation. These results strongly suggest that the dissociated and partially denatured collagen fibers that result from HA accumulation during the early stage of the follicular phase are degraded by MMP-2 during the late follicular phase. Thus, around the estrus, both disaggregation and degradation of cervical collagen fibers coexist and cooperate for the remodeling of the ECM. The temporal relationship between the cervical collagen degradation and the periovulatory hormonal environment should be taken into account in the design of treatments for induction of cervical dilation in artificial insemination and embryo transfer techniques. Interestingly, although the percentage of MMP-2– positive cells did not change during the estrous cycle, its pattern of variation differed between cranial and caudal cervical regions, suggesting that there is a differential timedependent stimulation of MMP2 protein synthesis along the longitudinal axis of the cervix. The increase in MMP-2 expression before ovulation (on Day 13 after estrus detection) in the caudal cervix could play a permissive role for upward progress of sperm at copulation [49] by inducing cervical relaxation. The increase of MMP-2 expression around the time of ovulation (on Day 1 after estrus detection) in the cranial cervix could be related to its role as a spermatic reservoir [49] by softening the cervical folds. This differential MMP-2 protein expression between the cranial and caudal cervix could be regulated directly or indirectly through local tissue- or cell-dependent factors that interact with regulation via ovarian steroid hormones [20]. Differential expression of the receptors of estrogen, progesterone, LH, FSH, Ox, and PGE2, as well as of COX-2 and hyaluronan synthase enzymes between the cranial and caudal cervix has also been reported in the ewe [12,15,16,48,50,51]. Interestingly, the percentage and density of MMP-2 protein-positive cells were greater in the caudal than in the cranial cervix, and the percentage of collagen fibers was lower in the caudal than in the cranial cervix, suggesting a high potential rate of collagen degradation in the caudal cervix. The MMP-2–positive cell density showed a similar pattern of variation to the total nuclear density previously reported [15]. Both the MMP-2–positive cells and the total nuclear density were lower in the cranial than in the caudal cervix, and maximum in SWS and minimum in DWS, suggesting that the MMP-2–positive cell density is strongly influenced by the total cell density. Overall, these data indicate molecular and functional differences between the

825

cranial and caudal ovine cervix; these differences are important considerations in the design of local treatment protocols to induce cervical dilatation. In this study, a decrease in the cervical collagen levels around the ovine estrus was reported, which occurs concomitant with an increase in the activity of MMP-2, which is produced by stromal fibroblasts and activated by changes in periovulatory hormone levels. Additionally, a time-dependent differential increase in MMP-2 expression along the longitudinal axis of the cervix was detected, occurring before ovulation in the caudal cervix and around the time of ovulation in the cranial cervix, reflecting their different physiological roles. Acknowledgments The authors would like to thank P. Rubianes for technical assistance, M. Marco and G. Lin for technical assistance in gelatin zymography, and C. Maeda Takiya, V. Samoto, and A. Dantas Medeiros for the generous and selfless technical assistance in in situ gelatin zymography. The authors received financial support from Comisión Sectorial de Investigación Científica (CSIC) and Programa de Desarrollo de Ciencias Básicas (PEDECIBA), Universidad de la República; Comisión de Investigación y Desarrollo Científico (CIDEC) and Programa de Posgrados, Facultad de Veterinaria; Fondo Clemente Estable, Agencia Nacional de Investigación e Innovación (ANII); and Dirección Nacional de Ciencia y Tecnología (DINACYT), Ministerio de Educación y Cultura, Uruguay. References [1] Kraemer DC. Embryo collection and transfer in small ruminants. Theriogenology 1989;31:141–8. [2] Halbert GW, Dobson H, Walton JS, Buckrell BC. The structure of the cervical canal of the ewe. Theriogenology 1990;33:977–92. [3] Salamon S, Maxwell WMC. Frozen storage of ram semen II. Causes of low fertility after cervical insemination and methods of improvement. Anim Reprod Sci 1995;8:1–36. [4] Kershaw CM, Khalid M, McGowan MR, Ingram K, Leethongdee S, Wax G, et al. The anatomy of the sheep cervix and its influence on the transcervical passage of an inseminating pipette into the uterine lumen. Theriogenology 2005;64:1225–35. [5] Candappa IBR, Bartlewski PM. A review of advances in artificial insemination (AI) and embryo transfer (ET) in sheep, with the special reference to hormonal induction of cervical dilation and its implications for controlled animal reproduction and surgical techniques. Open Reprod Sci J 2011;3:162–75. [6] Robinson JJ, McKelvey WA, King ME, Mitchell SE, Mylne MJ, McEvoy TG, et al. Traversing the ovine cervixda challenge for cryopreserved semen and creative science. Animal 2011;5:1791–804. [7] Golichowski AM, King SR, Mascaro K. Pregnancy-related changes in rat cervical glycosaminoglycans. Biochem J 1980;192:1–8. [8] Fosang AJ, Handley CJ. Connective tissue remodelling in the ovine cervix during pregnancy and at term. Connect Tissue Res 1988;17:277–85. [9] Uldbjerg N. Cervical connective tissue in relation to pregnancy, labour, and treatment with prostaglandin E2. Acta Obstet Gynecol Scand Suppl 1989;148:1–40. [10] Word RA, Li XH, Hnat M, Carrick K. Dynamics of cervical remodeling during pregnancy and parturition: mechanisms and current concepts. Semin Reprod Med 2007;25:69–79. [11] Myers KM, Paskaleva AP, House M, Socrate S. Mechanical and biochemical properties of human cervical tissue. Acta Biomater 2008;4:104–16. [12] Rodríguez-Piñón M, Tasende C, Puime P, Garófalo EG. Oestrogens and progesterone receptor binding proteins and oestrogens receptor alpha expression (ERa mRNA) along the cervix in cycling ewes. Reprod Fertil Dev 2008;20:350–6.

826

M. Rodríguez-Piñón et al. / Theriogenology 84 (2015) 818–826

[13] Matthews EL, Ayad VJ. Characterization and localization of a putative oxytocin receptor in the cervix of the oestrous ewe. J Endocrinol 1994;142:397–405. [14] Falchi L, Scaramuzzi RJ. The expression of ERa, OTR, cPLA2, COX-2, and PPARg in the cervix of the ewe during the estrous cycle. Theriogenology 2013;79:40–7. [15] Rodríguez-Piñón M, Gonzalez R, Tasende C, Bielli A, Genovese P, Garófalo EG. Cervical changes in estrogen receptor alpha, oxytocin receptor, LH receptor, and cyclooxygenase-2 depending on the histologic compartment, longitudinal axis, and day of the ovine estrous cycle. Theriogenology 2014;81:813–24. [16] Kershaw-Young CM, Khalid M, McGowan MR, Pitsillides AA, Scaramuzzi RJ. The mRNA expression of prostaglandin E receptors EP2 and EP4 and the changes in glycosaminoglycans in the sheep cervix during the estrous cycle. Theriogenology 2009;72:251–61. [17] Perry K, Haresign W, Wathes DC, Khalid M. Intracervical application of hyaluronan improves cervical relaxation in the ewe. Theriogenology 2010;74:1685–90. [18] Uldbjerg N, Ulmsten U. The physiology of cervical ripening and cervical dilatation and the effect of abortifacient drugs. Baillieres Clin Obstet Gynaecol 1990;4:263–82. [19] Junqueira LC, Zugaib M, Montes GS, Toledo OM, Krisztán RM, Shigihara KM. Morphologic and histochemical evidence for the occurrence of collagenolysis and for the role of neutrophilic polymorphonuclear leukocytes during cervical dilation. Am J Obstet Gynecol 1980;38:273–81. [20] Hulboy DL, Rudolph LA, Matrisian LM. Matrix metalloproteinases as mediators of reproductive function. Mol Hum Reprod 1997;3: 27–45. [21] Stygar D, Wang H, Vladic YS, Ekman G, Eriksson H, Sahlin L. Increased level of matrix metalloproteinases 2 and 9 in the ripening process of the human cervix. Biol Reprod 2002;67:889–94. [22] Bannister DW, Burns AB. Adaptation of the Bergman and Loxley technique for hydroxyproline determination to the autoanalyzer and its use in determining plasma hydroxyproline in the domestic fowl. Analyst 1970;95:596–600. [23] Etherington DJ, Sims TJ. Detection and estimation of collagen. J Sci Food Agric 1981;32:539–46. [24] Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951;193:265–75. [25] Regassa F, Noakes DE. Changes in the weight, collagen concentration and content of the uterus and cervix of the ewe during pregnancy. Res Vet Sci 2001;70:61–6. [26] Challis J, Matthews S, Gibb W, Lye SJ. Endocrine and paracrine regulation of birth at term and preterm. Endocr Rev 2000;21: 514–50. [27] Breeveld-Dwarkasing VN, de Boer-Brouwer M, te Koppele JM, Bank RA, van der Weijden GC, Taverne MA, et al. Regional differences in water content, collagen content, and collagen degradation in the cervix of nonpregnant cows. Biol Reprod 2003;69:1600–7. [28] Ramos JG, Varayoud J, Bosquiazzo VL, Luque EH, Muñoz-de-Toro M. Cellular turnover in the rat uterine cervix and its relationship to estrogen and progesterone receptor dynamics. Biol Reprod 2002; 67:735–42. [29] Uldbjerg N, Ekman G, Malmström A, Olsson K, Ulmsten U. Ripening of the human uterine cervix related to changes in collagen, glycosaminoglycans, and collagenolytic activity. Am J Obstet Gynecol 1983;147:662–6. [30] Mitchell SE, Robinson JJ, King ME, McKelvey WA, Williams LM. Interleukin 8 in the cervix of non-pregnant ewes. Reproduction 2002;124:409–16. [31] Croy BA, Prudencio J, Minhas K, Ashkar AA, Galligan C, Foster RA, et al. A preliminary study on the usefulness of huIL-8 in cervical relaxation of the ewe for artificial insemination and for embryo transfer. Theriogenology 1999;52:271–87. [32] Raynes JG, Anderson JC, Fitzpatrick RJ, Dobson H. Increased collagenase activity is not detectable in cervical softening in the ewe. Coll Relat Res 1988;8:461–9.

[33] Crawford HC, Matrisian LM. Mechanisms controlling the transcription of matrix metalloproteinase genes in normal and neoplastic cells. Enzyme Protein 1996;49:20–37. [34] Anuradha P, Thampan RV. Hormonal regulation of rat uterine collagenase. Arch Biochem Biophys 1993;303:81–9. [35] Russo LA, Peano BJ, Trivedi SP, Cavalcanto TD, Olenchock BA, Caruso JA, et al. Regulated expression of matrix metalloproteinases, inflammatory mediators, and endometrial matrix remodeling by 17beta-estradiol in the immature rat uterus. Reprod Biol Endocrinol 2009;7:124. [36] Overall CM, Wrana JL, Sodek J. Transcriptional and posttranscriptional regulation of 72-kDa gelatinase/type IV collagenase by transforming growth factor-beta 1 in human fibroblasts. Comparisons with collagenase and tissue inhibitor of matrix metalloproteinase gene expression. J Biol Chem 1991;266:14064–71. [37] Goodman RL, Inskeep EK. Neuroendocrine control of the ovarian cycle of the sheep. In: Neill JD, editor. Knobil and Neill’s physiology of reproduction. St. Louis, USA: Elsevier Academic Press; 2006. p. 2389–448. [38] Mizrachi D, Shemesh M. Expression of functional luteinising hormone receptor and its messenger ribonucleic acid in bovine cervix: luteinising hormone augmentation of intracellular cyclic AMP, phosphate inositol and cyclooxygenase. Mol Cell Endocrinol 1999; 157:191–200. [39] Shemesh M, Dombrovski L, Gurevich M, Friedman S, Shore LS, Fuchs AR, et al. Regulation of bovine cervical secretion of prostaglandins and synthesis of cyclooxygenase by oxytocin. Reprod Fertil Dev 1997;9:525–30. [40] Fuchs AR, Graddy LG, Kowalski AA, Fields MJ. Oxytocin induces PGE2 release from bovine cervical mucosa in vivo. Prostaglandins Other Lipid Mediat 2002;70:119–29. [41] Audicana L, Aughey E, O’Shaughnessy PJ. Sensitivity of the early luteal phase ovine cervix to prostaglandin E2 (PGE2) and expression of EP3 receptor mRNA. Res Vet Sci 1998;64:177–9. [42] Schmitz T, Levine BA, Nathanielsz PW. Localization and steroid regulation of prostaglandin E2 receptor protein expression in ovine cervix. Reproduction 2006;131:743–50. [43] McCarthy JB, Wahl SM, Rees JC, Olsen CE, Sandberg L, Wahl LM. Mediation of macrophage collagenase production by 39-59 cyclic adenosine monophosphate. J Immunol 1980;124:2405–9. [44] Zeng L, An S, Goetzl EJ. Selective regulation of RNK-16 cell matrix metalloproteinases by the PTGER4 subtype of prostaglandin E2 receptor. Biochemistry 1996;35:7159–64. [45] Norström A. Acute effects of prostaglandins on the biosynthesis of connective tissue constituents in the non-pregnant human cervix uteri. Acta Obstet Gynecol Scand 1984;63:169–73. [46] El Maradny E, Kanayama N, Kobayashi H, Hossain B, Khatun S, Liping S, et al. The role of hyaluronic acid as a mediator and regulator of cervical ripening. Hum Reprod 1997;12:1080–8. [47] Owiny J, Fitzpatrick R, Spiller D, Appleton J. Scanning electron microscopy of the wall of the ovine cervix uteri in relation to tensile strength at parturition. Res Vet Sci 1987;43:36–43. [48] Kershaw CM, Scaramuzzi RJ, McGowan MR, Wheeler-Jones CP, Khalid M. The expression of prostaglandin endoperoxide synthase 2 messenger RNA and the proportion of smooth muscle and collagen in the sheep cervix during the estrous cycle. Biol Reprod 2007;76: 124–9. [49] Hawk HW, Conley HH, Cooper BS. Number of sperm in the oviducts, uterus, and cervix of mated ewe as affected by exogenous estradiol. J Anim Sci 1978;46:1300–8. [50] Leethongdee S, Kershaw-Young CM, Scaramuzzi RJ, Khalid M. Intracervical application of Misoprostol at estrus alters the content of cervical hyaluronan and the mRNA expression of follicle stimulating hormone receptor (FSHR), luteinizing hormone receptor (LHR) and cyclooxygenase-2 in the ewe. Theriogenology 2010;73:1257–66. [51] Perry K, Haresign W, Wathes DC, Pitsillides AA, Khalid M. Cervical expression of hyaluronan synthases varies with the stage of the estrous cycle in the ewe. Theriogenology 2012;77:1100–10.