Progesterone receptors, oestrogen receptor α and glucocorticoid receptors in the bovine intercaruncular uterine wall around parturition

Animal Reproduction Science 103 (2008) 215–227

Progesterone receptors, oestrogen receptor ␣ and glucocorticoid receptors in the bovine intercaruncular uterine wall around parturition M. Sch¨aubli a,1 , N. Ritter b , M. H¨assig c , H. Zerbe b , U. Bleul d , A. Boos a,∗ a

b

Institute of Veterinary Anatomy, Vetsuisse Faculty, University of Zurich, Winterthurerstrasse 260, CH-8057 Zurich, Switzerland Clinic for Ruminants, Veterinary Faculty, University of Munich, Sonnenstr. 16, D-85764 Oberschleißheim, Germany c Clinic of Reproductive Medicine, Section for Herd Health, Vetsuisse Faculty, University of Zurich, Winterthurerstrasse 260, CH-8057 Zurich, Switzerland d Clinic of Reproductive Medicine, Vetsuisse Faculty, Winterthurerstrasse 260, CH-8057 Zurich, Switzerland Received 24 March 2006; received in revised form 6 December 2006; accepted 6 December 2006 Available online 17 December 2006

Abstract The bovine intercaruncular uterine wall expresses steroid hormone receptors throughout pregnancy. Concentrations of specific hormones undergo massive changes during the peripartal period and modulate the synthesis of their own receptors. This is well documented for the placentome, but respective data concerning the intercaruncular uterine wall are completely lacking. Thus, intercaruncular uterine wall segments from cows (I) being 8 and 9 months pregnant (slaughtered cows) and (II) cows undergoing a premature caesarean section 269–282 days after artificial insemination (AI) with (IIa, b) or without (IIc) induction of birth with PGF2␣ agonist or (III) receiving a caesarean section during severe dystocia (n = 6, 5, 5, 5, 6 and 4 animals, respectively) were studied. In four naturally calving cows (IV) endometrial biopsies were obtained within 30 min after the expulsion of the calf. All tissue probes were fixed for 24 h in 4% formaldehyde, routinely embedded in paraffin, and cut at 4 ␮m. Progesterone receptors (PR), estrogen receptor ␣ (ER␣) and glucocorticoid receptors (GR) were assessed using specific antibodies and staining intensities were documented employing an immunoreactive score (IRS). PR, ER␣ and GR exhibited cell type- and location-specific distribution patterns. IRS for PR and

∗ 1

Corresponding author. Tel.: +41 44 635 87 96; fax: +41 44 635 89 43. E-mail address: [email protected] (A. Boos). Tel.: +41 44 635 87 96; fax: +41 44 635 89 43.

0378-4320/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.anireprosci.2006.12.015

216

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

ER␣ did not differ between groups. GR-IRS of endometrial stromal cells, however, were higher in animals undergoing premature caesarean section after induction of birth compared to animals slaughtered during month 8 or 9 of pregnancy or animals receiving caesarean section following dystocia. Results of the present study indicate that steroid hormone receptor amounts within the intercaruncular uterine wall do not (PR, ER␣) – or in a tissue-specific manner (GR) only – change during the peripartal period, although respective hormones undergo massive changes during this period. This is in strict contrast to the placentome. Comparatively lower local tissue estrogen concentrations around term may be one cause for this difference. © 2006 Elsevier B.V. All rights reserved. Keywords: Progesterone receptors; Oestrogen receptor ␣; Glucocorticoid receptors; Immunohistochemistry; Cattle; Uterus; Peripartal

1. Introduction In an immunohistochemical study based on tissue probes collected from slaughtered animals Boos et al. (2006) reported that the bovine intercaruncular uterine wall expresses progesterone receptors (PR), oestrogen receptor ␣ (ER␣) and glucocorticoid receptors (GR) in cell type-specific, location-specific and pregnancy stage-specific patterns. PR were present at high levels in myometrium and endometrial stromal cells and at low levels in endometrial glands. During ongoing pregnancy, PR did not exhibit any defined pattern—i.e. increased in superficial stromal cells, decreased in glandular tubules and remained almost unchanged in gland openings, intermediate and deep stromal cells and in myometrial smooth muscle cells. ER␣ were present at high levels in myometrial smooth muscle cells, intermediate and deep stromal cells and at low levels in surface epithelium, superficial stromal cells, glands and endometrial and myometrial blood vessels. During the progression of pregnancy ER␣ decreased significantly in surface epithelium, glandular tubules, superficial and intermediate stromal cells and myometrial blood vascular cells. GR immunoreactivity was recorded in surface epithelium, glandular openings, endometrial and myometrial blood vessels. GR increased significantly in endometrial surface epithelium and myometrial blood vessels towards the end of pregnancy. Thus, PR remained fairly unchanged during pregnancy while ER␣ and GR exhibited distinct changes paralleling findings obtained in placentomes (Boos et al., 2000, 2006). Induction of parturition with a prostaglandin F2␣ agonist resulted in significant and temporary increased of immunoreactivity for PR, ER␣, and GR in several cell types of placentomes which were, however, no longer recorded in animals after spontaneous calving (Boos et al., 2000). Since the peripartal period was not included in the study dealing with the intercaruncular uterine wall (Boos et al., 2006), it is not known if this compartment of the bovine uterus exhibits similar endocrinological changes during this period of reproduction. The latter is sensitive to disorders often leading to impaired calf viability or lowered fertility after more or less severe dystocia (Zhang et al., 1999a,b; Wischral et al., 2001; Kindahl et al., 2002, 2004; Taverne et al., 2002). Information concerning the intercaruncular uterine wall at term, however, is of great interest since it is known from literature that the uterus secretes prostaglandin F2␣ , which is synthesised mainly within placentomes and by the intercaruncular uterine surface epithelium and thus requires discrete modulation (see, e.g. Danet-Desnoyers et al., 1994; Asselin et al., 1996). From studies in cattle and sheep it is believed that this process is mainly oestrogen-dependent (Asselin et al., 1996; Challis et al., 2000; Whittle et al., 2001), but a direct effect of glucocorticoids cannot not be ruled out in the light of these recent findings (Boos et al., 2000, 2006).

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

217

Thus, intercaruncular tissue specimens of uteri of pregnant cows were obtained during two periods from animals at slaughter (I), during caesarean sections (II) and by using biopsy forceps after the expulsion of the calf (III). Tissue probes were handled differently during dehydration and embedding – i.e. manually in period 1 and automatically in period 2 – and analysed immunohistochemically using light microscopy for steroid hormone receptors to characterise peripartal changes. 2. Materials and methods 2.1. Animals The 35 animals (see Table 1) enrolled in the present study all belonged to the “Holstein Frisian” breed. During the first sampling period in 1997, uteri of 11 late pregnant cows were collected at the local abattoir. Six cows were in the 8th (Ia, crown rump length 60.0–67.0 cm, see Schnorr and Kressin, 2001) and five cows in the 9th month of gestation (Ib, crown rump length 70.0–80.0 cm). Immediately after being stunned by bolt pistol, the animals were eviscerated and the samples were taken. Following a short macroscopical assessment, a complete cross-section was done through the intercaruncular uterine wall. Additionally five pre-partum cows being 274–282 days post insemination were housed at the Clinic for Cattle of the University of Veterinary Medicine Hanover, Germany. Each cow received 1 mg Cloprostenol-Na (Estrumate® , Essex Tierarznei, Munich, Germany) per cow to induce calf lung maturation (Zaremba et al., 1997) and induce birth. Caesarean sections were performed 27 h later under local anaesthesia with cows standing when the dams did not show any clinical sign of labour. Four more cows were at term in labour, suffering from severe dystocia and received a caesarean section as therapy. Samples were taken from these cows within the intercaruncular segments of the uterine wall along the cut of the uterus after the delivery of the calf. During the second period – tissue samples collected in 2004 focussing on the very peripartal period – the effect of PGF2␣ agonist and of spontaneous birth was assessed. Six cows (272–276 days after insemination, housed at the Federal Agricultural Research Centre, D-31535 NeustadtMariensee, Germany) received a premature caesarean section without any pre-treatment (group IIb) or 27 h after the application of 1 mg Cloprostenol-Na per cow (n = 5, 269–273 days after insemination; cows housed at the Clinic for Cattle, University of Veterinary Medicine Hanover, Germany). Tissue samples were collected as mentioned above. Another 4 cows – also housed at the Federal Agricultural Research Centre – spontaneously delivered living calves without any problems. Following expulsion of the offspring, biopsy samples were taken from the Table 1 Origin of tissue probes used in this study Sampling period

Group

Number of animals (n)

Pregnancy stage

Treatment

1 1 1 1 2 2 2

Ia Ib IIa III IIb IIc IV

6 5 5 4 6 5 4

8th month 9th month d274–d282 aai At term d272–d276 aai d269–d273 aai <30 min after birth

Slaughter Slaughter 1 mg Clp, pcs cs after dystocia pcs 1 mg Clp, pcs Biopsy

aai: after artificial insemination; Clp: Cloprostenol-Na; cs: caesarean section; pcs: premature caesarean section.

218

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

intercaruncular endometrium per vaginam using biopsy forceps (Narco Pilling, Fort Washington, PA, USA). Animal experiments were approved by administration of the district Hanover, and performed according to the German Law for the Protection of Animals (TierschG) and the recommendations of the German Society of Laboratory Animal Science (GV-SOLAS). 2.2. Sample preparation Immediately after excision the samples were immersed in 4% neutral buffered formalin for 24 h. Thereafter they were rinsed in tap water for 20 h. Subsequently the samples (period 1, 1997) were dehydrated in 50%, 70%, 80% (24 h each) and 96% alcohol (4–5 h), transferred to acetic acid-n-butyl ester (3–4 h; Riedel-de-Ha¨en, Seelze, Germany) and immersed sequentially in hot liquid Paraplast Plus® (Sheerwood Medical, St. Louis, MO, USA) for 12, 2 and 2 h at 60 ◦ C and finally embedded in Paraplast Plus® . For period 2 (sampling of tissues in 2004), further processing of tissue probes was done using an automatic tissue processor (TP1020® , Leica, Wetzlar, Germany). The duration of each step was 2× 4 h each and the reagents were ethanol (70%, 96%, 100%), benzoic acid methyl ester, xylene and paraffin (Histowax® , Leica, Wetzlar, Germany). Following this procedure, tissue samples were embedded in paraffin. Four micrometer thick sections were cut using a rotation microtome (RM2155® , Leica, Wetzlar, Germany), placed upon adhesive glass slides (SuperFrost Plus® , Menzel-Gl¨aser, Braunschweig, Germany) and heat dried at 60 ◦ C for 30 min. 2.3. Immunohistochemistry Wax was removed using xylene (3× 5 min), sections were hydrated through serial dilutions of ethanol to water and antigen retrieval, i.e. unmasking of cross-linked epitopes (see, e.g. Shi et al., 2001) was carried out by incubating the sections in citrate buffer (0.01 mol l−1 , pH 6.0; ProTaqs® citrate buffer concentrate and diluted with H2 O to finally obtain 1000 ml solution; Medite, Nunningen, Switzerland) in a microwave oven (600 W, 1× 10 min and 3× 4 min) for PR and ER␣ (period 1, respectively). Antigen retrieval for ER␣ in period 2 probes was performed using TEC buffer (pH 7.8, 2.5 g trizma base, 5.0 g EDTA and 3.2 g trizma sodium citrate in 1000 ml H2 O) for 30 min at 95–97 ◦ C and GR epitopes were unmasked employing trizma-EDTA buffer (pH 9.0, 1.21 g trizma base, 0.37 g EDTA in 1000 ml H2 O and 0.5 ml Tween 20) for 15 min at 95–97 ◦ C. Thereafter sections were incubated in the above mentioned buffer solutions and cooled down to room temperature (RT). Sections were encircled with a Dako Pen® (Dako, Hamburg, Germany) and tissue-specific peroxidases were quenched using 3% (PR) or 2% H2 O2 (ER␣ and GR, respectively) for 10 min and sections were incubated in trizma-buffered saline (TBS, buffer stock solution: 6.1 g trizma base, 50 ml H2 O and 37 ml 1N HCl, diluted with H2 O to 1000 ml solution, adjust pH to 7.6; working solution: 100 ml buffer stock solution plus 900 ml saline, 0.85%). Subsequent immunohistochemistry was performed at RT according to the instructions of the supplier using Vectastain Elite ABC kit for mouse-IgG (Vector, Burlingame, CA, USA). 2.3.1. Progesterone receptors Blocking was performed by incubating sections in a humidified chamber for 20 min with normal horse serum and a serum-free protein block (Dakocytomation, Zug, Switzerland) for 10 min.

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

219

PR were detected with a mouse monoclonal antibody ␣PR6 (diluted 1:500, i.e. 1.22 ␮g ml−1 in TBS) (Sullivan et al., 1986; gift of D.O. Toft, Mayo Medical School, Rochester, MI, USA), directed against amino acids 113–125 of the chicken progesterone receptor (D.O. Toft, pers. commun.) and cross-reacting with the bovine antigen (see Hurd et al., 1991). This amino acid sequence is common to forms A and B of the receptor (see Gronemeyer et al., 1987). Incubation for 60 min was followed by rinsing with TBS for 5 min, incubation with a horse anti-mouse-IgG antibody for 30 min, rinsing with TBS (5 min) and incubation with ABC complex for 30 min. This step was followed by rinsing with TBS (5 min), incubation with diaminobenzidine tetrahydrochloride (DAB) chromogen and H2 O2 (DAB substrate kit, Vector, Burlingame, CA, USA) for 2 min and rinsing with TBS (5 min). Finally, nuclei were counterstained with haemalaun according to Mayer (B¨ock, 1989) for 20 s, rinsed with tap water for 10 min, dehydrated, cleared in xylene and automatically mounted (Medite, RCM 2000® , Nunningen, Switzerland) in Pertex® (Medite, Nunningen, Switzerland). Negative controls were performed employing non-immune mouse-IgG (Vector, Burlingame, CA, USA) in equal concentration instead of the primary antibody. Positive controls were performed on uterine sections of a cow in pro-oestrus. 2.3.2. Oestrogen receptor α Blocking was performed by incubating sections in a humidified chamber for 20 min with normal horse serum and a serum-free protein block (Dakocytomation, Zug, Switzerland) for 10 min. ER␣ was detected with a mouse monoclonal antibody Ab-8 (clone AER311, Lab Vision, Fremont, CA, USA) diluted 1:50 in TBS and directed against calf uterine ER␣ protein. Incubation for 20 h at 4 ◦ C was followed by rinsing with TBS for 5 min, incubation with a horse anti-mouse-IgG antibody for 30 min, rinsing with TBS (5 min) and incubation with ABC complex for 30 min. This step was followed by rinsing with TBS (5 min), incubation with diaminobenzidine tetrahydrochloride (DAB) chromogen and H2 O2 (DAB substrate kit, Vector, Burlingame, CA, USA) for 2.5 min (period 1 probes) or 3.5 min (period 2 probes) and rinsing with TBS (5 min). Counterstaining and mounting were performed as described for PR. Controls were performed employing non-immune mouse-IgG (Vector, Burlingame, CA, USA) in equal concentration instead of the primary antibody. Positive controls were performed on uterine sections of a cow in pro-oestrus. 2.3.3. Glucocorticoid receptors Blocking was performed employing an avidin–biotin blocking kit (Vektor, Burlingame, CA, USA) for 20 min each, normal goat serum (20 min) and a serum-free protein block (Dakocytomation, Zug, Switzerland) for 10 min. An affinity purified polyclonal antibody (rabbit, no. PA1-511, Dianova-Immunotech, Hamburg), directed against a 22 amino acid synthetic peptide (aa 346–367, common to receptor forms ␣ and ␤; see Encio and Detera-Wadleigh, 1997) from the amino terminus of the human glucocorticoid receptor (Cidlowski et al., 1990), diluted 1:300 in PBS, was used for glucocorticoid receptor detection. This antibody cross-reacts with ruminant GR (Saoud and Wood, 1996) and has already been used on cattle reproductive organs (Boos et al., 2000, 2006). Incubation for 20 h at 4 ◦ C was followed by rinsing with TBS (3× 3 min), incubation with a goat anti-rabbit-IgG antibody for 30 min, rinsing with TBS (3× 3 min) and incubation with ABC complex for 30 min. This step was followed by rinsing with TBS (3× 3 min), incubation with diaminobenzidine tetrahydrochloride (DAB) chromogen and H2 O2 (DAB substrate kit, Vector, Burlingame, CA, USA) for 5 min and rinsing with TBS (3× 3 min). Counterstaining and mounting were performed as described for PR.

220

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

Controls were performed employing non-immune rabbit IgG (Vector, Burlingame, CA, USA) in equal concentration instead of the primary antibody. Positive controls were performed on placental sections of a cow at term. 2.4. Semiquantitative evaluation of histochemical reactions Immunoreactivity was evaluated by a single observer at a magnification of 400× in randomly selected microscopic fields in different tissue locations: (I) superficial endometrium with surface epithelium (SE), glandular openings (GO) and subepithelial stroma (SS); (II) intermediate and deep endometrium with respective connective tissue (DS) and glandular tubules (GT); and (III) myometrium (MSMC). The intensity of staining (SI) of 500 nuclei of each cell type (see below) was scored as negative = 0, weak = 1, intermediate = 2, or strong = 3, correlating to absence of brown (that is, blue counterstain only), light brown, brown or dark brown staining, respectively (i.e. SI0–SI3). The frequency of the different staining intensities was assessed separately for each type of cell. Five hundred cells per type were evaluated and results were expressed as a percentage of positively stained cells of the respective type. Immunoreactivity of the different cell types in the three tissue locations described above was expressed using scores (progesterone receptor immunoreactive score, PR-IRS; oestrogen receptor ␣ immunoreactive score, ER␣-IRS; or glucocorticoid receptor immunoreactive score, GR-IRS), calculated according to the following procedure (see Boos et al., 2006): PR-IRS or ER␣-IRS or GR-IRS = 0 × n (SI0) + 1 × n (SI1) + 4 × n (SI2) + 9 × n (SI3), where n = percentage of cells of a specific type that demonstrated characteristic staining intensities, SI0 = staining intensity negative, SI1 = staining intensity weak, SI2 = staining intensity intermediate, and SI3 = staining intensity strong. IRS fluctuates between 0 (no specific stain) and 900 (all cells exhibit strong immunoreactivities). 2.5. Progesterone determination in blood plasma Vena jugularis externa blood samples were taken from cows at the time of uterine wall tissue collection during caesarean section during sampling period 2 (6 cows with versus 5 cows without induction of labour with 1 mg Cloprostenol-Na). Plasma was harvested, frozen and stored at −20 ◦ C until analysis at the endocrine laboratory of the Clinic for Cattle of the University of Veterinary Medicine Hannover. Progesterone was determined by radioimmunoassay procedure according to Behrens et al. (1993). Intra-assay variation coefficient was 4.5% and inter-assay variation coefficient was 9.7%. 2.6. Statistical methods Immunoreactive scores for receptors were analysed statistically using StatView® software (release 5.1, SAS Institute Inc., NC, USA) for each type of cell by comparing data from all groups of animals within one sampling period (period 1 or period 2). Factorial analysis of variance (ANOVA) was used to evaluate the influence of time (represented by groups of animals). Significant differences between groups were evaluated using Bonferroni–Dunn procedure as post hoc test. Data are presented as mean ± S.E. of the mean. Differences were considered as significant, if

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

221

p < 0.05. Differences between progesterone concentrations during sampling period 2 were tested using Student’s t-test. 3. Results 3.1. Progesterone receptors Only nuclei reacted specifically with the anti-progesterone receptor antibody ␣PR6, i.e. there was no cytoplasmic stain (Fig. 1a and b). The endometrial surface epithelium was devoid of progesterone receptor protein, while glands generally exhibited a very faint immunoreaction (Fig. 1a and b). Stronger staining was evident in stromal cells and the most prominent immunoreaction was noticed in myometrial smooth muscle cells (Fig. 1b and b ). These cell type-specific staining intensities result in many significant differences between the immunoreactive scores of the cell types assessed (see Table 2). However, no significant differences could be established between the groups of animals included in the present study (see Table 2). 3.2. Oestrogen receptor α Several cell types exhibited positive nuclear immunolabelling, when Ab-8 (clone AER311) was applied (Fig. 1c and d). Endometrial surface epithelial cells (Fig. 1d), glandular epithelial cells (Fig. 1d ) and myometrial smooth muscle cells (Fig. 1d ) exhibited stronger ER␣ staining as compared to endometrial stromal cells—reflected by several significant differences between the immunoreactive scores obtained from these groups of cell types (see Table 2). No significant differences were, however, detected between the groups of animals enrolled in this study (see Table 2). 3.3. Glucocorticoid receptors Nuclei reacted strongly with the polyclonal rabbit anti-glucocorticoid receptor antibody PA1511 and a very faint cytoplasmic stain also occurred (Fig. 1f and g). Surface epithelial cells (Fig. 1e and f), stromal cells within the superficial endometrial connective tissue (Fig. 1e and f) and myometrial smooth muscle cells (Fig. 1g) exhibited stronger immunoreactive labelling compared to the stromal cells within the deep endometrial connective tissue and the glandular epithelial cells (Fig. 1e ). This is reflected by corresponding significant differences established in sampling period 1 (see Table 2). In sampling period 2, differences between cell types were found to be non-significant (Table 2). GR-IRS of superficial (8th and 9th month of pregnancy) and deep stromal cells (9th month of pregnancy) were significantly lower in tissues collected from such animals at slaughter compared to animals undergoing premature caesarean section after induction of birth with a PGF2␣ agonist (Fig. 1e and e versus f and f ; sampling period 1, see also Table 2). 3.4. Progesterone concentration in blood plasma Animals without induction of labour and undergoing a premature caesarean section had significantly higher progesterone concentrations compared to animals without such pre-treatment (4.30 ± 1.07 ng/ml versus 1.58 ± 0.30 ng/ml; p < 0.05).

222

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

Fig. 1. Immunolocalization (brown staining) of progesterone receptors (PR), oestrogen receptor ␣ (ER␣) and glucocorticoid receptors (GR) in the bovine intercaruncular uterine wall during the peripartal period. PR (see a, b, and b ) were confined to nuclei and evident in glandular epithelial cells (GE), stromal cells within the superficial (SS) and deep stroma (DS), and in smooth muscle cells (SMC) of the myometrium (MM); G: uterine glands. ER␣ (see c, d, d , and d ) was also confined to nuclei and detected in surface epithelium (SE), glandular epithelial cells (GE), stromal cells of the superficial (SS) and deep stroma (DS), smooth muscle cells (SMC) of the myometrium (MM), and endothelial cells (E) of blood vessels (BV). Immunoreactions for GR (e, f, f , g, and g ) were evident in nuclei and cytoplasm and were visualised in many cell types of the uterine wall: surface epithelium (SE), glandular epithelial cells (GE), stromal cells of the superficial (SS) and deep stroma (DS), endothelial cells (E), and smooth muscle cells (SMC) of the myometrium (MM; see g). Stromal cells exhibited weaker immunoreactions for GR within the superficial (SS) and deep stroma (DS) of probes gained at slaughter during the 9th month of pregnancy as compared to probes taken from animals during premature caesarean 27 h after the application of a PGF2␣ agonist (e, e vs. f, f , respectively).

Table 2 Immunoreactive scores (IRS, mean ± S.E.) evaluated in sections after immunohistochemical detection of progesterone receptors (PR), oestrogen receptor ␣ (ER␣) and glucocorticoid receptors (GR) Uterine structure

Sampling period 1 Iaa 8th month of pregnancy

PR-IRS PR-IRS PR-IRS PR-IRS

GT SS DS MSMC

5 213 154 362

± ± ± ±

4 (2) 13 (1) 11 (1) 21 (3)

ER␣-IRS ER␣-IRS ER␣-IRS ER␣-IRS ER␣-IRS

SE GT SS DS MSMC

108 107 67 64 96

± ± ± ± ±

GR-IRS GR-IRS GR-IRS GR-IRS GR-IRS

SE GT SS DS MSMC

92 15 46 21 53

± ± ± ± ±

Sampling period 2 Iba 9th month of pregnancy

IIaa d274–d282 aai, Clp, pcs

IIIa cs after dystocia

IIba d272–d276 aai, pcs

IIca d269–d273 aai, Clp, pcs

IVa biopsy <30 min after birth

3 201 115 350

± ± ± ±

1 (3) 21 (1) 13 (2) 16 (4)

58 179 160 383

± ± ± ±

17 (2) 12 (1) 12 (1) 28 (3)

16 153 118 309

± ± ± ±

2 (2) 17 (1) 10 (1) 15 (3)

16 157 99 289

± ± ± ±

3 (3) 6.11 (1) 10 (2) 23 (4)

15 156 118 261

± ± ± ±

3 (2) 9 (1) 12 (1) 7 (3)

54 138 109 291

± ± ± ±

14 (2) 8 (1) 15 (1,2) 38 (3)

9 (11,13) 5 (11,13) 4 (12,13) 4 (12) 5 (13)

98 99 75 57 94

± ± ± ± ±

4 (11) 2 (11) 7 8 (12) 6 (11)

109 111 86 74 107

± ± ± ± ±

12 (11) 3 (11) 8 6 (12) 7

96 114 75 62 90

± ± ± ± ±

6 3 (11) 4 (12) 5 (12) 5

72 100 50 42 100

± ± ± ± ±

3 (11) 4 (12) 4 (11,13) 2 (13) 6 (12)

51 92 36 29 89

± ± ± ± ±

4 (11) 7 (12) 3 (11) 2 (11) 3 (12)

58 92 45 46 101

± ± ± ± ±

7 (11) 6 (12) 5 (11) 5 (11) 0 (12)

4 (21) 3 (22) 4 a (23,24) 4 (22,24) 3 (23)

89 11 46 18 51

± ± ± ± ±

4 (21) 4 (22) 8 a (23,24) 2 c (22,24) 3 (23)

102 38 89 47 73

± ± ± ± ±

8 (21) 9 (22) 7 b (21,23,24) 7 d (22,24) 4 (24)

95 29 74 28 55

± ± ± ± ±

9 (21) 7 (22) 3 (23) 3 (22) 2 (22,23)

104 87 86 50 90

± ± ± ± ±

3 (21) 13 (21) 4 3 (22) 5 (21)

80 62 70 48 66

± ± ± ± ±

5 13 5 9 6

92 91 83 57 88

± ± ± ± ±

4 12 5 6 2

aai: after artificial insemination; Clp: Cloprostenol-Na, cs: caesarean section; SE: surface epithelium; GT: glandular tubules; SS: stromal cells of the superficial stroma; DS: stromal cells of the deep stroma; MSMC: myometrial smooth muscle cells; pcs: premature caesarean section. Mean values with different letters differ significantly (p < 0.05), i.e. between groups indicated by letters along the lines (a–d, see GR-IRS, SS and DS) and between the various cell types/tissue layers indicated by numbers in parentheses within columns (1–4 for PR-IRS, 11–13 for ER␣-IRS and 21–24 for GR-IRS). a Group (see Table 1).

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

Parameter

223

224

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

4. Discussion The results of the present study indicate that progesterone receptors, oestrogen receptor ␣ and glucocorticoid receptors are expressed in the bovine intercaruncular uterine wall during late pregnancy and around parturition and exhibit cell type-specific distribution patterns. Locationdependent expression of progesterone receptors and glucocorticoid receptors was evident under specific circumstances in stromal cells within the bovine endometrium, i.e. superficial connective tissue versus deep connective tissue. This implies that at the end of pregnancy – as during early and mid-pregnancy (Boos et al., 2006) – various cell types of the intercaruncular uterine wall demonstrate specific sensitivities to progesterone, estrogens and glucocorticoids, which finally may result in cell type-specific reaction patterns. This confirms and extends results of a former study which dealt with intercaruncular uterine wall during pregnancy but did not include the period near parturition and after the expulsion of the foetus. Within the myometrium, almost all smooth muscle cells expressed progesterone receptor during late pregnancy and at parturition indicating a direct influence of progesterone on the contractile activity of this layer, finally resulting in myometrial quiescence until the beginning of the expulsion of the foetus (see, e.g. Challis et al., 2000; Taverne et al., 2001, 2002; Whittle et al., 2001). Additional effects of prostaglandin E2 and ocytocin via their receptors may also play an important role in this respect (Fuchs et al., 1999; Arosh et al., 2003, 2004a,b; Wehbrink et al., 2005). It is interesting to note that during late pregnancy and at term glucocorticoid receptors were immunohistochemically detectable in nuclei of endometrial surface epithelial cells. These findings are supported by results recently obtained in sheep (Gupta et al., 2003). Maternal surface epithelial cells express enzymes for endometrial prostaglandin synthesis during pregnancy (Fuchs et al., 1999; Arosh et al., 2004a,b; Wehbrink et al., 2005). The increase in luminal epithelial glucocorticoid receptor protein expression during ongoing pregnancy is in accordance with clinical observations that initiation of premature birth by glucocorticoids is only possible beyond the second trimester of gestation (Aurich and Aurich, 1994). It may be suggested in this connection that glucocorticoids directly stimulate endometrial and placental prostaglandin F2␣ synthesis and secretion leading to luteolysis and labour or premature birth or abortion if given during late pregnancy. The regulation of uterine prostaglandin F2␣ synthesis and secretion during spontaneous labour or initiated parturition by estrogens and/or glucocorticoids warrants further investigation since the presence of glucocorticoid receptors on endometrial surface epithelial cells and glandular openings was not known until recently (Boos et al., 2006). In this respect – i.e. prostaglandin F2␣ synthesis in the domestic cow – in vivo studies focused mainly on the whole uterus of pregnant animals or only placental tissues were analysed employing glucocorticoids (see, e.g. Gross and Williams, 1988; Hoedemaker et al., 1991; Izhar et al., 1992; Wood, 1999; K¨onigsson et al., 2001). Recent papers employing COX-2 immunohistochemistry reveal that the foetal aspect of the bovine placenta should contribute to uterine prostaglandin F2␣ synthesis at term since this key enzyme is localised mainly in uninucleated placental trophoblast cells (Arosh et al., 2004a; Schuler et al., 2006). The quantitative and possibly time-dependent contribution of the intercaruncular uterine wall (see Wehbrink et al., 2005) and the placentomes to uterine PGE2 and PGF2␣ production around term remains to be elucidated. It is interesting to note, that the two different protocols of tissue dehydration and embedding in wax – tissue fixation with 4% neutral buffered formaldehyde solution was identical – as performed during sampling period 1 in Hanover and sampling period 2 in Zurich had a great impact on the intensity of immunohistochemical staining depending on the antibodies used, i.e. antibodies

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

225

against PR, ER␣ or GR. While for PR no differences were visible, these were clearly evident in ER␣ and in GR—necessitating a completely different protocol of antigen retrieval. Thus, comparison of studies employing similar techniques must be regarded cautiously since both, (I) tissue processing – here: acetic acid-n-butyl ester versus benzoic acid methyl ester and xylene as solvents, both of which may differently interfere with the tertiary conformation of the receptor proteins assessed – and (II) differing antibody specificities, i.e. immunogenic epitopes (Schuler et al., 2002) can influence results of immunohistochemical procedures. Finally it should be mentioned, that in the intercaruncular uterine wall no changes of the overall progesterone and estrogen receptors are detected towards term. Any observed changes are cell type-specific (GR increase in endometrial stromal cells) and only evident during sampling period 1 and not 2 (expected between non-induced and induced premature caesarean sections) although changes in hormone concentrations are tremendous. This is in contrast to the placentome, where immunoreactive scores of PR, ER␣ and GR are significantly higher in most cell types of animals during induced parturition compared to cells in tissue specimens collected at slaughter during the last trimester of pregnancy (Boos et al., 2000). In the present study, changes in plasma progesterone concentrations were significantly lower 27 h after PGF2␣ agonist application, which is in accordance with other reports (Kindahl et al., 2002) that indicate a drop of progesterone concentrations below 1 ng/ml as early as 8 h after prostaglandin application (Taverne et al., 2002). The cause of absence or only slight reaction compared to the placentomes onto the massive endocrine changes taking place at parturition may be explained by the fact, that placentomal tissue is the source and target of estrogens (Hoedemaker et al., 1990; Hoffmann and Schuler, 2002) and prostaglandins (Gross and Williams, 1988; Hoedemaker et al., 1991). These are the most important molecules in this respect and should therefore be found in higher concentrations within this uterine compartment compared to the intercaruncular uterine wall. Hormones secreted within placentomes may reach the intercaruncular uterine wall via the blood stream or counter-current mechanisms within the broad ligament (see, e.g. Einer-Jensen, 1988; Kotwica and Krzymowski, 1989; Banu et al., 2005) resulting in strong dilution and time lag. It may be speculated from the present study that the expression level of the receptors detected here is sufficient to mediate the actions of these hormones at the cellular level. 5. Conclusion During late pregnancy and around term the intercaruncular uterine wall expresses receptors for progesterone, estrogens and glucocorticoids, which are distributed in cell type and locationspecific patterns (PR, GR) implying a highly specific hormonal regulation. Hormonal changes around parturition only slightly influence GR expression, while PR and ER␣ remain unchanged. The influence of progesterone, estrogens and glucocorticoids on their own receptors and on uterine functions, such as prostaglandin production, ocytocin and prostaglandin receptor expression, uterine contractility – placentomes and intercaruncular uterine wall separately assessed – remain to be elucidated to find out the contribution of both compartments to normal reproduction and undisturbed fertility. Acknowledgements The authors would also like to thank D.O. Toft, Mayo Medical School, Rochester for the provision of antibodies. This study was supported by a grant of Vetsuisse to A. Boos.

226

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

References Arosh, J.A., Banu, S.K., Chapdelaine, P., Emond, V., Kim, J.J., MacLaren, L.A., Fortier, M.A., 2003. Molecular cloning and characterization of bovine prostaglandin E2 receptors EP2 and EP4: expression and regulation in endometrium and myometrium during estrous cycle and early pregnancy. Endocrinology 144, 3076–3091. Arosh, J.A., Banu, S.K., Chapdelaine, P., Fortier, M.A., 2004a. Temporal and tissue-specific expression of prostaglandin receptors EP2, EP3, EP4, FP, and cyclooxygenases 1 and 2 in uterus and fetal membranes during bovine pregnancy. Endocrinology 145, 407–417. Arosh, J.A., Banu, S.K., Kimmins, S., Chapdelaine, P., MacLaren, L.A., Fortier, M.A., 2004b. Effect of interferon ␶ on prostaglandin biosynthesis, transport and signalling at the time of maternal recognition of pregnancy in cattle: evidence of polycrine actions of PGE2 . Endocrinology 145, 5280–5293. Asselin, E., Goff, A.K., Bergeron, H., Fortier, M.A., 1996. Influence of sex steroids on the production of prostaglandins F2 alpha and E2 and response to oxytocin in cultured epithelial and stromal cells of the bovine endometrium. Biol. Reprod. 54, 371–379. Aurich, J.E., Aurich, C., 1994. Geburtseinleitung bei Haustieren. Prakt. Tierarzt 75, 742–750. Banu, S.K., Arosh, J.A., Chapdelaine, P., Fortier, M.A., 2005. Expression of prostaglandin transporter in the bovine uterus and fetal membranes during pregnancy. Biol. Reprod. 73, 230–236. Behrens, C., Aurich, J.E., Klug, E., Naumann, H., Hoppen, H.O., 1993. Inhibition of gonadotrophin release in mares during the luteal phase of the oestrous cycle by endogenous opioids. J. Reprod. Fertil. 98, 509–514. B¨ock, P., 1989. Romeis Mikroskopische Technik, 17th revised ed. Urban and Schwarzenberg, Munich, Germany. Boos, A., Kohtes, J., Janssen, V., M¨ulling, C., Stelljes, A., Zerbe, H., H¨assig, M., Thole, H.H., 2006. Pregnancy effects on distribution of progesterone receptors, oestrogen receptor ␣, glucocorticoid receptors, Ki-67 antigen and apoptosis in the bovine interplacentomal uterine wall and foetal membranes. Anim. Reprod. Sci. 91, 55–76. Boos, A., Kohtes, J., Stelljes, A., Zerbe, H., Thole, H.H., 2000. Immunohistochemical assessment of progesterone, oestrogen and glucocorticoid receptors in bovine placentomes during pregnancy, induced parturition, and after birth with or without retention of foetal membranes. J. Reprod. Fertil. 120, 351–360. Challis, J.R.G., Matthews, S.G., Gibb, W., Lye, S.J., 2000. Endocrine and paracrine regulation of birth at term and preterm. Endocr. Rev. 21, 514–550. Cidlowski, J.A., Bellingham, D.L., Power-Oliver, F.E., Luan, D.B., Sar, M., 1990. Novel antipeptide antibodies to the human glucocorticoid receptor: recognition of multiple receptor forms in vitro and distinct localization of cytoplasmic and nuclear receptors. Mol. Endocrinol. 4, 1427–1437. Danet-Desnoyers, G., Wetzels, C., Thatcher, W.W., 1994. Natural and recombinant bovine interferon ␶ regulate basal and oxytocin-induced secretion of prostaglandins F2␣ and E2 by epithelial cells and stromal cells in the endometrium. Reprod. Fertil. Dev. 6, 193–202. Einer-Jensen, N., 1988. Countercurrent transfer in the ovarian pedicle and its physiological implications. Oxf. Rev. Reprod. Biol. 10, 348–381. Encio, I.J., Detera-Wadleigh, S.D., 1997. PubMed Protein Data Bank, Accession AAB64354. Fuchs, A.R., Rust, W., Fields, M.J., 1999. Accumulation of cyclooxygenase-2 gene transcripts in uterine tissues of pregnant and parturient cows: stimulation by oxytocin. Biol. Reprod. 60, 341–348. Gronemeyer, H., Turcotte, B., Quirin-Stricker, C., Bocquel, M.T., Meyer, M.E., Krozowski, Z., Jeltsch, J.M., Lerouge, T., Garnier, J.M., Chambon, P., 1987. The chicken progesterone receptor: sequence, expression and functional analysis. EMBO J. 20, 3985–3994. Gross, T.S., Williams, W.F., 1988. In-vitro steroid synthesis by the placenta of cows in late gestation and at parturition. J. Reprod. Fertil. 83, 565–573. Gupta, S., Gyomorey, S., Lye, S.J., Gibb, W., Challis, J.R.G., 2003. Effect of labour on glucocorticoid receptor (Grtotal , GR␣, and GR␤) proteins in ovine intrauterine tissues. J. Soc. Gynecol. Investig. 10, 136–144. Hoedemaker, M., Weston, P.G., Wagner, W.C., 1990. Influence of cortisol and different steroidogenic pathways on estrogen synthesis by the bovine placenta. Am. J. Vet. Res. 51, 1012–1015. Hoedemaker, M., Weston, P.G., Wagner, W.C., 1991. Arachidonic acid metabolism by bovine placental tissue during the last month of pregnancy. Prostaglandins 41, 75–84. Hoffmann, B., Schuler, G., 2002. The bovine placenta; a source and target of steroid hormones: observations during the second half of gestation. Domest. Anim. Endocrinol. 23, 309–320. Hurd, C., Nakao, M., Eliezer, N., Moudgil, V.K., 1991. Immunoanalysis of calf uterine progesterone receptor: modulation of receptor-associated 90 kDa heat-shock protein. Mol. Cell. Biochem. 105, 73–83. Izhar, M., Pasmanik, M., Marcus, S., Shemesh, M., 1992. Dexamethasone inhibition of cyclooxygenase expression in bovine term placenta. Prostaglandins 43, 239–354.

M. Sch¨aubli et al. / Animal Reproduction Science 103 (2008) 215–227

227

Kindahl, H., Kornmatitsuk, B., Gustafsson, H., 2004. The cow in endocrine focus before and after calving. Reprod. Domest. Anim. 39, 217–221. Kindahl, H., Kornmatitsuk, B., K¨onigsson, K., Gustafsson, H., 2002. Endocrine changes in late bovine pregnancy with special emphasis on fetal well-being. Domest. Anim. Endocrinol. 23, 321–328. K¨onigsson, K., Kask, K., Gustafsson, H., Kindahl, H., Parvizi, N., 2001. 15-Ketodihydro-PGF2␣ , progesterone and cortisol profiles in heifers after induction of parturition by injection of dexamethasone. Acta Vet. Scand. 42, 151–159. Kotwica, J., Krzymowski, T., 1989. Role of mesovarium and mesosalpinx in counter current exchange of hormones. Acta Physiol. Pol. 40, 12–22. Saoud, C.J., Wood, C.E., 1996. Developmental changes and molecular weight of immunoreactive glucocorticoid receptor protein in the ovine fetal hypothalamus and pituitary. Biochem. Biophys. Res. Commun. 229, 916–921. Schnorr, B., Kressin, M., 2001. Embryologie der Haustiere, 4th revised ed. Enke, Stuttgart, Germany. Schuler, G., Teichmann, U., Kowalewski, M.P., Hoffmann, B., Madore, E., Fortier, M.A., Klisch, K., 2006. Expression of cyclooxygenase-II (COX-II) and 20␣-hydroxysteroid dehydrogenase (20␣-HSD)/prostaglandin F-synthase (PGFS) in bovine placentomes: implications for the initiation of parturition in cattle. Placenta 27, 1022–1029. Schuler, G., Wirth, C., Teichmann, U., Failing, K., Leiser, R., Thole, H., Hoffmann, B., 2002. Occurrence of estrogen receptor ␣ in bovine placentomes throughout mid and late gestation and at parturition. Biol. Reprod. 66, 976–982. Shi, S.R., Cote, R.J., Taylor, C.R., 2001. Antigen retrieval techniques: current perspectives. J. Histochem. Cytochem. 49, 931–937. Sullivan, W.P., Beito, T.G., Proper, J., Krco, C.J., Toft, D.O., 1986. Preparation of monoclonal antibodies to the avian progesterone receptor. Endocrinology 119, 1549–1557. Taverne, M.A.M., Breeveld-Dwarkasing, V.N.A., Van Dissel-Emiliani, F.M.F., Bevers, M.M., De Jong, R., Van Weijden, G.C., 2002. Between prepartum luteolysis and onset of expulsion. Domest. Anim. Endocrinol. 23, 329–337. Taverne, M.A.M., De Schwartz, N.C.M., Kankofer, M., Bevers, M.M., Van Oord, H.A., Schams, D., Gutjahr, S., Van Weijden, G.C., 2001. Uterine responses to exogenous oxytocin before and after pre-partum luteolysis in the cow. Reprod. Domest. Anim. 36, 267–272. Wehbrink, D., Ritter, N., Zerbe, H., H¨assig, M., Boos, A., 2005. COX-2 and prostaglandin receptors EP2 and FP in the interplacentomal bovine uterine wall during the peripartal period. Schweiz. Arch. Tierheilkd. 147, 83. Whittle, W.L., Patel, F.A., Alfaidy, N., Holloway, A.C., Fraser, M., Gyomorey, S., Lye, S.J., Gibb, W., Challis, J.R.G., 2001. Glucocorticoid regulation of human and ovine parturition: the relationship between fetal hypothalamic-pituitaryadrenal axis activation and intrauterine prostaglandin production. Biol. Reprod. 64, 1019–1032. Wischral, A., Verreschi, I.T., Lima, S.B., Hayashi, L.F., Barnabe, R.C., 2001. Pre-parturition profile of steroids and prostaglandin in cows with or without foetal membrane retention. Anim. Reprod. Sci. 67, 181–188. Wood, C.E., 1999. Control of parturition in ruminants. J. Reprod. Fertil. Suppl. 54, 115–126. Zaremba, W., Grunert, E., Aurich, J.E., 1997. Prophylaxis of respiratory distress syndrome in premature calves by administration of dexamethasone or a prostaglandin F2␣ analogue to their dams before parturition. Am. J. Vet. Res. 58, 404–407. Zhang, W.C., Nakao, T., Moriyoshi, M., Nakada, K., Ohtaki, T., Ribadu, A.Y., Tanaka, Y., 1999a. The relationship between plasma oestrone sulphate concentrations in pregnant dairy cattle and calf birth weight, calf viability, placental weight and placental expulsion. Anim. Reprod. Sci. 54, 169–178. Zhang, W.C., Nakao, T., Moriyoshi, M., Nakada, K., Ribadu, A.Y., Ohtaki, T., Tanaka, Y., 1999b. Relationship of maternal plasma progesterone and estrone sulfate to dystocia in Holstein-Friesian heifers and cows. J. Vet. Med. Sci. 61, 909–913.