SEARCH

SEARCH BY CITATION

Contents

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgement
  7. Conflicts of interest
  8. Author contributions
  9. References

The luteal phase on pregnant and non-pregnant bitches is characteristic of this species and resembles significantly with respect to the growth pattern and luteal regression. Histological and immunostaining studies of the corpus luteum (CL) may help to elucidate differences between the CL of pregnant and non-pregnant bitches. The purpose of this study was to characterize histologically and localize by immunohistochemistry the cell proliferation (Ki-67) and vascular endothelial growth (VEGF) factors in the CL of pregnant and non-pregnant bitches. Eighteen bitches were analysed and distributed into three groups: In group I (gestational diestrous), seven bitches were subjected to two inseminations at 4 and 6 days after the pre-ovulatory LH surge and ovariohysterectomized (OSH) at 8–21 days after the first insemination. In group II (cyclic diestrous; control), 6 (Ki-67) or 8 (VEGF) bitches that were determined as non-pregnant were OSH at 12–25 days of the pre-ovulatory LH surge. In group III (late pregnancy), three bitches had their ovary removed during caesarean section at 62–64 days after the pre-ovulatory LH surge. Portions of the ovarian cortex containing CLs were cut and stored for histological and immunohistochemical analysis. Histological evaluation of the ovarian cortex showed a marked similarity in the morphological pattern among the CLs in all three groups. The morphology and expression pattern of VEGF and Ki-67 factors in CLs of cyclic and gestational diestrous bitches were similar but significantly lower than that of late pregnant bitches (p < 0.05).


Introduction

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgement
  7. Conflicts of interest
  8. Author contributions
  9. References

The reproductive cycle of female dogs differs from the cycle of other domestic mammalian species. Bitches are monoestrous, ovulate once or twice a year with interestrous intervals of 5–12 months. Puberty, in most of the breeds, occurs at 6–14 months of age and the luteal phase of non-pregnant females last for approximately 60 days, and it is similar in length to that of pregnant females. Canine estrous cycle has been classified in four phases (Evans and Cole 1931). As Phase 1: proestrous with duration of 5–20 days, phase 2: estrous with a duration of 5–15 days, phase 3: diestrous lasting from 50 to −80 days and phase 4: anestrous ranging from 80 to 240 days. These four phases reflect the follicular phase with oestrogenic domain and the luteal phase with increase in progesterone concentrations and oestrogen decline (Concannon 2011).

In bitches, corpus luteum (CL) growth pattern and luteal regression seem to be not different as in polyestrous species that have been hysterectomized In fact, luteolysis depends on uterus prostaglandin production, while the luteal growth phase involves an increase in prostaglandin sintetase (PG sintetase), cyclooxygenase-2 (COX-2) and prostaglandin E (PGE) production. The PGE is a mitogenic, angiogenic, antiapoptotic, vasodilator and luteotrophic factor. A histological study in bitches of periovulatory follicles and early CL formation suggests changes involving the loss of granulosa cells and an increase in theca interna cells and blood vessels during and after the LH surge (Kowalewski et al. 2007; Concannon 2011).

Development of a CL and its endocrine function are based on endothelial cell proliferation and vascular growth. Cellular and molecular regulation of angiogenesis involves a list of potential regulators; however, vascular endothelial growth factor (VEGF) is thought to be the principal angiogenic growth factor in CL formation (Mariani et al. 2006). The VEGF system has been described in the CL of other mammalian species (bovine, pig, human), in which time-dependent localization of VEGF receptors in luteal cells has suggested a role of VEGF in luteal cell function. Likewise, it has been reported for bitches also (Mariani et al. 2006).

Corpus luteum expresses VEGF and its receptors at 10 and 70 days after ovulation, with higher factor concentrations in luteal cells, and immunostaining intensity varies between moderate and intense, with a significant decrease in 70 days after ovulation (Mariani et al. 2006). The authors concluded that VEGF is present in the CL of bitches, and that VEGF and its receptors expression, even as its function, vary with the cycle stage and the cell type, and possible, act as a paracrine and/or autocrine factor. In contrast, the expression of cell proliferation (Ki-67) factor was not detected in CL cells but was present in all ovarian cells at proestrous and estrous phases (Srisuwatanasagul et al. 2009).

Accordingly, the reproductive cycle of the bitch differs from most of the domestic mammalian female species; mainly, when concern to the cycle of the luteal phase. In fact, the gestational and/or cyclic diestrous (progestational phase) is long in the bitch, the progesterone level is similar among these two phases (pregnancy and diestrous) and the CL is the only structure responsible for maintaining the pregnancy and/or the cyclic diestrous. Therefore, the lifespan, morphology and interaction among biochemical/molecular aspects of CL in this specie need to be clarified and is the purpose of this study.

The aim of this study was to histologically characterize and immunolocalize the cell proliferation (Ki-67) and the vascular endothelial growth (VEGF) factor in the CL of non-pregnant (cyclic diestrous) and pregnant bitches at 25 and 62 days after the LH surge, respectively.

Materials and Methods

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgement
  7. Conflicts of interest
  8. Author contributions
  9. References

This study was performed at the Laboratory of Small and Wild Animal Reproduction, Department of Animal Reproduction and Veterinary Radiology and at the Immunohistochemistry Laboratory, Department of Veterinary Clinics, both at the College of Veterinary Medicine and Animal Science (Sao Paulo State University, UNESP, Botucatu, Brazil).

Animals

Eighteen bitches not bred, with ages between 1 and 8 years old, equivalent in weight, in good health conditions and with a proven fertility history, examined in the Ambulatory of Small Animal Reproduction or belonging to APA (Animals Protection Association of Botucatu City) whose owners have an interest to castrate their animals, were used in this study. Bitches were submitted to a general clinical examination, gynecological examination, vaginal cytology and haemogram blood test. To confirm the phase of the estrous cycle, progesterone was measured in the serum.

Vaginal cytology

Vaginal cells were recovered with a vaginal brush, rolled onto microscope slides and then stained with Panoptic solution (Hematocor®, Biologica Comercial Ltda, Sao Paulo, Brazil). Cell type's identification was based in the criteria adopted in the study by Johnston et al. (2001). Vaginal cytology was performed in alternate days since the beginning of the proestrous until the determination of the first day of cytological diestrous.

Haemogram blood test

To determine the health status of bitches as a pre-operative action, blood samples were collected at the day of the clinical examination and at the day before of surgery. Blood samples (1–2 ml) were collected from jugular vein puncture in tubes containing EDTA, and the total of white and red blood cells and platelet cells were counted in an automatic haematology analyser (Hemascreen Ebram 18, Ebram, Brazil). The reference values were adopted from the Clinical Laboratory of Veterinary Hospital – UNESP/Botucatu.

Progesterone measurement

Blood samples (3.0 ml) were collected by jugular vein puncture in tubes without anticoagulant. To obtain serum, blood samples were centrifuged (Centrifugal LS-3; Celm®Cia. Equipadora de Laboratórios Modernos, Barueri, Brazil) for 10 min at 1500× g. Progesterone levels were detected by radioimmunoassay method (RIA) in solid phase by using a commercial kit (Diagnostics Procedure Corporation – progesterone Coat- a-Count, Los Angeles, CA, USA).

Ovariohysterectomy

Bitches were fasted for food and water from 12 to 3 h, respectively, before the surgery. Pre-anaesthesia was induced by administration of 0.05 mg/kg acepromazine and 0.5 mg/kg morphine, intramuscularly, and after 10 min, anaesthesia was induced with the application of 5 mg/kg ketamine and 0.25 mg/kg diazepam, intravenously, and anaesthesia was maintained by inhalation of halothane. As post-operative treatment, 0.1 mg/kg of meloxicam was administered for 2 days.

Collection and quantification of CL

Ovaries were isolated and a longitudinal section was made to count the number of CL's. Thereafter, ovaries were placed in plastic cassettes for inclusion and fixed in 10% buffered formalin for 24 h and stored in 70% ethanol until embedding in paraffin.

Immunohistochemical and histological evaluation

For histological analysis, samples of ovarian cortex were collected and embedded in paraffin, sectioned in 4 μm and stained with HE (haematoxylin and eosin). Samples were examined using the criterion previously described in the study by Bacha and Bacha (2003).

For immunohistochemistry, tissue sections of four micrometers were mounted onto glass slides that were previously treated with Organosilane (Sigma Chemical Co., St Louis, MO, USA). Mounted tissue sections were deparaffinized with xylene and rehydrated in graded alcohol and then, washed 10 times in distilled water. For antigen retrieval of the Ki-67 factor, sections were placed in a Pascal pressure chamber (Dako, Glostrup, Denmark) in 10 mm of sodium citrate solution (pH = 6.0) and for VEGF, slides were maintained in a pre-heated water bath at 96°C in Tris–EDTA solution (pH = 9.0) for 20 min. After heating, slides were allowed to cool for 20 min and then, washed ten times in tap water. Endogenous peroxidase activity was quenched with 8% of peroxidase solution for 20 min (two times of 10 min each), followed by 10 baths in tap water and blocked for 1 h at RT in 3% slim milk solution for protein block.

After blocking, slides were washed in 10 mm of Tris-buffered solution (pH = 7.4) and incubated with monoclonal mouse antihuman Ki-67 (Ki-67 Antigen, clone MIB-1, 1:50 dilution; Dako) or monoclonal mouse antihuman vascular endothelial growth factor antibody (VEGF Antigen, clone VG1, 1:50 dilution; Dako) as primary antibodies for 18 h at 4°C in a humidified chamber. Slides were then washed in Tris-buffered solution and incubated with the secondary antibody for 1 h (Advance; Dako) or with the secondary biotinilated antimouse antibody and streptavidin-biotin for 30 min (ABC Kit Vectastain; Vector, Burlingame, CA, USA) for detecting Ki-67 and VEGF immunostaining, respectively, in a humidified chamber at room temperature, according to manufacturer instructions.

Each tissue section was washed in Tris-buffered solution, and DAB chromogen (3,3′-diaminobenzidina; Dako) was added as a chromogen staining substrate for 5 min. Reaction was stopped by rinsing in Tris-buffered solution. Tissue sections were counterstaining with Mayers haematoxylin for 1 min, dehydrated and preserved in Permount mounting medium (Fisher Scientific, Fair Lawn, NJ, USA). For negative controls, a tissue section was incubated with mouse immunoglobulin ready to use (N-Universal Negative Control Mouse; Dako) and other tissues were incubated with a rabbit immunoglobulin at an equal protein concentration as the primary antibody (N-Universal Negative Rabbit Control; Dako), both replacing the primary antibody and the next steps of the reaction were maintained. For positive control, a section of lymph node was used and processed equally as the experimental group's slides.

Immunoreactivity for Ki-67 or VEGF was observed in a light microscope at 400× magnification. For Ki-67, the number of positive nuclei was counted in five randomly selected microscopic fields, and for VEGF, the intensity of ten randomly selected microscopic fields was classified in weak (i), moderate (ii) or strong (iii).

Experimental design – Bitches were distributed into three groups

Group I (early gestational diestrous)

Seven bitches were inseminated in the estrous phase of the cycle with fresh semen by two intravaginal artificial inseminations at 4 and 6 days after pre-ovulatory LH surge. The LH surge was estimated by progesterone concentration (P4 equal to 5 ng/ml was considered as the ovulation moment; Reynaud et al. 2005).

Semen was collected from a healthy adult male that was negative to brucellosis and leptospirosis. Sperm motility and morphology were evaluated immediately after semen collection, and total sperm concentration was determined after appropriate dilution (1:20). A minimum concentration of 2 × 109 total spermatozoa was used to inseminate each female. After artificial insemination, bitches were ovariohysterectomized (OSH) among days 8 and 25 after the pre-ovulatory LH surge.

Group II (cyclic diestrous; control)

For analysis of Ki-67 and VEGF, six and eight non-pregnant bitches, respectively, were OSH in the diestrous phase among 8 and 25 days after the pre-ovulatory LH surge.

Group III (late gestational diestrous)

Three bitches were naturally bred, and both ovaries were collected during Caesarean. Immunohistochemical and histological evaluation of Ki-67 and VEGF factors were evaluated in ovaries collected from each group.

Statistical analysis

Data were analysed by graphpad prism 5 (GraphPad Software Inc., La Jolla, CA, USA) and are expressed as mean ± SD for Ki-67 or percentage of occurrence of each score for VEGF. The percentage of positive nuclei for Ki-67 that failed for normality (Shapiro–Wilk test) was analysed by non-parametric Kruskal–Wallis anova followed by Dunn's test. The score representative of VEGF among groups was compared by Qui-square test.

Results and Discussion

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgement
  7. Conflicts of interest
  8. Author contributions
  9. References

After ovulation, the place previously occupied by the follicle is invaded with fibroblasts, smooth muscle cells, immune cells, endothelial cells, theca interna cells and granulose cells that suffer hyperplasia and/or hypertrophy (Senger 2003). Initially, this set of cells promotes the formation of a structure called corpus hemorrhagicum that reorganize to constitute the CL (reviewed by Salles and Araújo 2010). The luteal function regulation is controlled by a complex group of luteotrophic and angiogenic agents (like insulin, VEGF and their receptors, hypoxia inductive agent HIF1A and others) that interact to each other to ensure the gland homoeostasis (La Paz et al. 2007).

In this study, the histological evaluation of the ovarian cortex showed a marked similarity of the morphological pattern in CL's between cyclic and gestational diestrous bitches. CL's contained luteal cells with vesicular cytoplasm located at the place previously occupied by lipids in live tissue (Luz et al. 2006), which were removed by the buffered formalin during fixation. Luteal cells exhibited a spherical and central nucleus with a prominent nucleolus, whose presence seems to be related to the luteal cells ability to secret progesterone (Hansel et al. 1987). In addition, it was possible to observe more cells in apoptosis and few cells in mitosis (mitotic figure). Apoptosis was visualized by cromatin condensation, higher cytoplasmatic eosinophilia and the presence of karyorrhexis and karyolysis. Connective tissue was usually found within the CL, and there was a large quantity of capillaries, which is indicative of a highly vascularized structure.

The CLs samples analysed in the late pregnancy treatment group did not show marked differences in their morphology to that of the cyclic or gestational diestrous. Nonetheless, although the total number of lipids in CL's were not counted, there was an apparent increase in cytoplasmic vacuoles containing lipids in the late pregnancy CL. It has been reported that the presence of a high numbers of lipids is associated with a higher capacity of luteal cells to secret progesterone (Luz et al. 2006). Therefore, the higher numbers of lipids in CL's of late pregnant bitches observed in this study, indicate that the luteal cells are still functional by the end of the gestation, and possibly are required to be present until the pre-partum period. Yet, serum progesterone concentrations measured in blood were not different among pregnant and non-pregnant bitches.

The mean intensity of Ki-67 in CL's was significantly different between all treatment groups (gestational diestrous = 5.6 ± 2.7; cyclic diestrous = 10.0 ± 4.3; late pregnancy = 22.3 ± 8.8; p < 0.05; Fig. 1), while VEGF intensity was similar between cyclic and gestational diestrous (p = 0.053), but both were significantly different than that of late pregnancy, which have a higher intensity (p < 0.0001; Fig. 1). The mean intensity of VEGF at moderate staining was 36% of the cells for gestational diestrous; 25.63% for cyclic diestrous and 40% for late pregnancy.

image

Figure 1. Immunolocalization of Ki-67 (a, b, c) and vascular endothelial growth factor (d, e, f) in bitch corpus luteum during early pregnancy (a, d), early diestrous (b, e) and late pregnancy (c, f). Negative and positive controls, insert in c and f. Immunohistochemistry, DAB, counterstaining with Harry's haematoxylin, magnification of 200×

Download figure to PowerPoint

Recently, it was reported that bitches luteal cells at different stages of the estrous cycle do not express the Ki-67 factor (Srisuwatanasagul et al. 2009). Contrary, in this study, we observed positive expression of Ki-67 factor in the nuclei of luteal cells at all analysed periods. Moreover, proliferation index in luteal cells (reflected as mitotic cells) was increased at late pregnancy that could be related with a mechanism of selfrenewal and attempting to survive, which means that the gland stimulates a survive mechanism trying to maintain their lifespan. These results suggest that luteal cells at late gestation may have higher capacity of hormonal synthesis than at the other cycle stages. The Ki-67 is a cell proliferation indicator, so mitosis figures observed in the histological examination can be related to this marker.

Signs of apoptosis were also observed during the morphological examination and were defined by chromatin condensation and enhanced cytoplasmic eosinophilia, as well as karyolysis and karyokinesis. We did not measure the expression of any of the markers for apoptosis; therefore, we could not determine the occurrence of apotosis in each of the experimental groups. Yet, the morphological signs of apoptosis observed in luteal cells in each of the groups suggest that there is perhaps a dynamic mechanism of cellular proliferation and degeneration during CL's survival period.

In this study, we observed a higher intensity of VEGF at late gestational diestrous (day 60 after ovulation). In contrast, Mariani et al. (2006) reported that cyclic diestrous CL obtained 70 days after ovulation had a significant reduction in the levels of the VEGF factor. These observations could reinforce the idea that a different pattern of VEGF between the late cyclic and gestational diestrous exist and confirm that there are differences between CLs in early and late gestational diestrous.

The VEGF immunostaining was clearly observed during the evaluated period which detects a angiogenic stimuli which was confirmed by the histological analysis where a lot numbers of capillaries could be seen in all studied groups.

The analysis of late gestational diestrous was fundamental to indentify immunohistochemical differences between early and late gestational diestrous. Probably, there is a higher vascular and cellular proliferation associated with an increase in Ki-67 and VEGF factors at late pregnancy.

The endocrinology of cyclic and gestational diestrous has not been completely established in the canine species; providing that the effects of the prostaglandin (PGF2α) are just seen during the parturition and not in late cyclic diestrous. Perhaps, this hormonal condition has influence on CL's morphology and/or function. Additional research needs to be carried out to establish a consistent hormonal pattern at the estrous cycle in bitches, as well as in gestation and parturition.

Author contributions

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgement
  7. Conflicts of interest
  8. Author contributions
  9. References

In this study, all the authors contributed to the conception of the study; the authors Ribeiro AA, Martin I, Derussi AAP and Ferioli RB contributed to the acquisition of data; the authors Martin I, Derussi AAP, Laufer-Amorim R and Lopes MD contributed to analysis and interpretation of data and Ribeiro AA, Martin I, Derussi AAP drafted paper. Martin I and Lopes MD reviewed the final approval of the version to be published.

References

  1. Top of page
  2. Contents
  3. Introduction
  4. Materials and Methods
  5. Results and Discussion
  6. Acknowledgement
  7. Conflicts of interest
  8. Author contributions
  9. References
  • Bacha WJ Jr, Bacha LM, 2003: Atlas Colorido de Histologia Veterinária, 2ed. Editora Roca, Sao Paulo, pp. 110.
  • Concannon PW, 2011: Reproductive cycles of the domestic bitch. Anim Reprod Sci 124, 200210.
  • Evans HM, Cole HH, 1931: An introduction to the study of the cicle in the dog. Mem Univ Calif 9, 65118.
  • Hansel W, Alila HW, Dowd JP, 1987: Control of steroidogenesis in small and large bovine luteal cells. Aust J Biol Sci 40, 331347.
  • Johnston SD, Kustritz MVR, Olson PNS, 2001: Canine and Feline Theriogenology. WB Saunders Company, Philadelphia.
  • Kowalewski MP, Mutembei HM, Hoffmann B, 2007: Canine prostaglandin F2alpha receptor (FP) and prostaglandin F2alpha synthase (PGFS): molecular cloning and expression in the corpus luteum. Anim Reprod Sci 107, 161175.
  • La Paz MN, Fonseca VU, Campos DB, Artoni LP, Sousa LMMC, Papa PC, 2007: Produção de progesterona in vitro pelas células do corpo lúteo bovino ao longo da gestação. Pesq Vet Bras 27, 370376.
  • Luz MR, Cesário MD, Binelli M, Lopes MD, 2006: Canine corpus luteum regression: apoptosis and caspase-3 activity. Theriogenology 66, 14481453.
  • Mariani TC, Prado C, Silva LG, Paarmann FA, Lima MC, Carvalho I, Campos DB, Artoni LP, Hernandez-Blazquez FJ, Papa PC, 2006: Immunohistochemical localization of VEGF and its receptors in the corpus luteum of the bitch during diestrous and anestrous. Theriogenology 66, 17151720.
  • Reynaud K, Fontbonne A, Marseloo N, Thoumire S, Chebrout M, Viaris de Lesegno C, Chastant-Maillard S, 2005: In vivo meiotic resumption, fertilization and early embryonic development in the bitch. Reproduction 130, 193201.
  • Salles MGF, Araújo AA, 2010: Corpo lúteo cíclico e gestacional: revisão. Rev Bras Reprod Anim 34, 185194.
  • Senger PL, 2003: Pathways to Pregnancy and Parturition, 2nd ed. Current Conceptions Inc., Washington, 368p.
  • Srisuwatanasagul S, Manee-In R, Raksaserh P, Kongkiat S, 2009: Proliferation and apoptosis of the bitch ovary during the different stages of the oestrous cycle. Thai J Vet Med 39, 217224.