In Situ and in Vitro Immunolocalization of Oviductin, an Oviduc-Specific Glycoprotein, in the Hamster Uterine Epithelial Cells

IN SITU AND IN VITRO IMMUNOLOCALIZATION OF OVIDUCTIN BINDING SITES ON HAMSTER UTERINE EPITHELIAL CELLS AND DETECTION OF A HAMSTER OVIDUCTIN HOMOLOGUE IN THE FEMALE RAT REPRODUCTIVE TRACT

YING ZHENG

A thesis submitted to the Department of Anatomy and Cell Biology in conformity with the requirements for the degree of Master of Science

Queen’s University Kingston, Ontario, Canada February 2008

Oviductin is an oviduct-specific and high-molecular-weight glycoprotein that has been suggested to play important roles in the early events of reproduction. The present study was undertaken to localize the oviductin binding sites in the uterine epithelial cells of the golden hamster (Mesocricetus auratus) both in situ and in vitro, and to detect a hamster oviductin homologue in the female rat reproductive tract. Immunohistochemical localization of oviductin in the hamster uterus revealed certain uterine epithelial cells reactive to the monoclonal anti-hamster oviductin antibody. In order to study the interaction between hamster oviductin and the endometrium in vitro, a method for culturing primary hamster uterine epithelial cells has been established and optimized.

Study with confocal microscopy of the cell culture system showed a labeling pattern similar to what was observed using immunohistochemistry. Pre-embedding immunolabeling of cultured uterine epithelial cells also showed gold particles associated with the plasma membrane and microvilli. These results demonstrated that hamster oviductin can bind to the plasma membrane of certain hamster uterine epithelial cells, suggesting the presence of a putative oviductin receptor on the uterine epithelial cell surface. In the second part of the present study, using the monoclonal anti-hamster oviductin antibody that cross-reacts with the rat tissue, we have been able to detect an oviduct-specific glycoprotein, with a molecular weight of 180~300kDa, in the female rat reproductive tract. Immunohistochemical labeling of the female rat reproductive tract revealed a strong immunolabeling in the non-ciliated oviductal epithelial cells and a faint immunoreaction on the cell surface of some uterine epithelial cells. Ultrastructurally,

ii immunogold labeling was restricted to the secretory granules, Golgi apparatus, and microvilli of the non-ciliated secretory cells of the oviduct. In the uterus, immunogold labeling was observed on the cell surface of some uterine epithelial cells. Furthermore, electron micrographs of ovulated oocytes showed an intense immunolabeling for rat oviductin within the perivitelline space surrounding the ovulated oocytes. The findings of the present study demonstrated that oviductin is present in the rat oviduct and uterus, and it appears that, in the rat, oviductin is secreted by the non-ciliated secretory cells of the oviduct.

iii

Dedicated to my parents, parents in law, and husband

Thank you for your love and constant support

that led me all the way to here.

I love you.

iv ACKNOWLEDGEMENTS First and foremost, I would like to sincerely thank my graduate supervisor, Dr. Frederick.W.K. Kan for his support, patience, guidance and encouragement in the past years. Without his support and help, it would be impossible for me to finish this thesis. Studying at Queen’s and working in his laboratory were unforgettable and valuable experience in my life. I also want to express many thanks to my co-workers in the lab. Changnian Shi was our technician who patiently taught me useful techniques and gave me good suggestions throughout my project. Bo Kong and Laurelle Saccary are my labmates who have made my overseas studying so fulfilling. I would like to thank all of you for such enjoyable memories that we shared. I’d like to extend a special thank to Mr. John Dacosta and Dr. Xiaohu Yan for their wonderful technical assistance. Over the past several years, I received tremendous support and help from members within and outside the department. I want to say thank you to Dr. Kan, Dr. Reifel, Dr. Oko, and Dr.Mackenzie for their wonderful lectures and lessons. I want to show my appreciation to Dr. Oko, Dr. Graham and Dr. Leda Raptis for letting me use their facilities. Thanks to Dr. Virginia Walker and Dr. Inka Brockhausen for kindly providing me with cell lines. Wei Xu, Yang Yu, Judy and Zheng Qin from Dr. Oko’s lab were so friendly that I can easily talk to and ask questions. A special thank is given to my good friend, Jun Liu, Ph.D candidate in the Department of Biology. Not only did I learn many molecular biology techniques from him, but also I learned a great deal of scientific faith from him. In a word, without their help, it would have been impossible for me to finish my project. To the departmental technical and administrate staffs, including Mr. Rick Hunt, Mr. Donaldson Earl, Mrs. Anita Lister, Mrs. Marilyn McAuley, Ms. Ita A McConnell and others, thank you for your kind assistance. To all my fellow graduate students and my friends, because of you, I have a wonderful and unforgettable experience studying at Queen’s University, Canada. Thank you indeed! Finally, but not the least, I am extremely thankful to my beloved families, especially to my husband Zhuoran Cai, my parents and parents in law. Without their true love and unconditional support, I would not have had the current achievement.

v TABLE OF CONTENTS Page

ABSTRACT ………………..……………………………………….……...………... ii

DEDICATION ....……………….………………………………………..….………. iv

ACKNOWLEDGEMENTS ….…………………………………………..………… v

TABLE OF CONTENTS …..……………………………………………..………… vi

LIST OF FIGURES AND ILLUSTRATIONS ….……..…………………..……… x

LIST OF TABLES ….…………………………………………………..…………… xii

LIST OF ABBREVIATIONS …..………………………………………..…..……... xiii

CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW …….……... 1

1 OVERVIEW …....……………………………………………………..………. 2

2 LITERATURE REVIEW ……………………………………………..….….. 4 2.1 The Female Reproductive Tract ……………………………..…...... ……... 4 2.2 Uterus ……………………………….………..…………….…..….…….... 5 2.2.1 Anatomical Structure of the Uterus …………………………...….. 5 2.2.2 Histological Features of the Uterus ………….…………..…...…... 6 2.2.3 Hormonal Regulation in the Uterus …………….…………..……. 7 2.2.4 Cyclic Changes in the Endometrium of the Uterus …...……….…. 8 2.2.5 Implantation in the Uterus …...………..………………..……..….. 12 2.2.6 Uterine Epithelial Cell Culture System ……………..………..…... 13 2.3 Oviduct ……………………………………………...…………….……… 16 2.3.1 Anatomical Structure of the Oviduct ……………………....……... 16 2.3.2 Histological Features of the Oviduct ………………..……..……... 17 2.4 Oviductal Fluid …………………………………………………..…..…… 20 2.5 General Review of Oviductin ………………………….….………...……. 21 2.6 Hamster Oviductin ……………………………………..………..………... 26 2.6.1 Ontogeny of Hamster Oviductin ………………………..…...……. 26 2.6.2 Biochemical Characterization of Hamster Oviductin ……..….…... 26 2.6.3 Hormonal Regulation of Hamster Oviductin Synthesis and Glycosylation ……………………………..……………….…. 28

vi 2.6.4 Localization of Hamster Oviductin ………………………..……... 29 2.6.5 Possible Roles of Hamster Oviductin ………………………..….... 31 2.7 Rat Oviductin ………………………………………………………..……. 33

CHAPTER 2: IN SITU AND IN VITRO IMMUNOLOCALIZATION OF OVIDUCTIN BINDING SITES ON HAMSTER UTERINE EPITHELIAL CELLS ………………………………………………………………………..……... 34

1 ABSTRACT …...…………………………………………………….………… 35

2 INTRODUCTION …...……………….…………………………….………… 37

3 HYPOTHESIS …..…………………………………………………….……… 39

4 OBJECTIVES …...…………………………………………………….……… 39

5 MATERIALS AND METHODS …..……………...……..…………...…….... 40 5.1 Animals ………………………………………………….………………... 40 5.2 Monoclonal Anti-Hamster Oviductin Antibody …….……………….…… 40 5.3 Lectin-Affinity Purification of Oviductin from Hamster Oviducts ….….... 41 5.4 Immunodetection of Purified Hamster Oviductin ………………….…..… 42 5.5 Immunohistochemical Labeling of Hamster Oviductin in the Female Reproductive Tract ………………………………………………..….…… 42 5.6 Primary Hamster Uterine Epithelial Cell Culture ……………………..….. 43 5.7 Immunoperoxidase Labeling of Cytokeratin in Cultured Primary Uterine Epithelial Cells ……………………………………………………....……. 45 5.8 Double Immunofluorescence Labeling in Cultured Primary Uterine Epithelial Cells …………………………………………………..….…….. 46 5.9 Preparation of Cultured Cells for Morphological Analysis under the Electron Microscope ………………………….………….………….……. 47 5.10 Pre-embedding Immunocytochemical Labeling of Oviductin on Cultured Primary Hamster Uterine Epithelial Cells ……………………….….……. 49 6 RESULTS …….……………………………………………….……………..... 50 6.1 Lectin-Affinity Purification of Hamster Oviductin ….……..…..……..….. 50 6.2 Immunohistochemical Labeling of Hamster Oviductin in the Female Reproductive Tract ……………………………………………..…….….... 52 6.3 Primary Cell Culture of Hamster Uterine Epithelial Cells ...…...………… 55 6.4 Detection of Cytokeratin in the Cultured Hamster Uterine Epithelial Cells …………………………………………………..….…….. 57

vii 6.5 Double Immunofluorescence Labeling of Oviductin and Cytokeratin in Cultured Primary Uterine Epithelial Cells ………………..………….…… 58 6.6 Ultrastructural Morphology of Cultured Hamster Primary Uterine Epithelial cells and Stromal Cells ……………………………...……….… 62 6.7 Pre-embedding Immunocytochemical Labeling of Oviductin on Cultured Primary Uterine Epithelial Cells ………………………………..….……... 64 7 DISCUSSION ……..…………………………………….…………………….. 68

CHAPTER 3: DETECTION OF A HAMSTER OVIDUCTIN HOMOLOGUE IN THE FEMALE RAT REPRODUCTIVE TRACT ………….………………… 75

1 ABSTRACT ………..…………………………………………………………. 76

2 INTRODUCTION ……...……..………...……………………….…………… 77

3 HYPOTHESIS …..……………………………………………….…………… 79

4 OBJECTIVES ……..……………………………………………..…………… 79

5 MATERIALS AND METHODS ……...... ……………………………...……. 80 5.1 Animals ..………………………………………………………..…….…... 80 5.2 Monoclonal Antibody Anti-Hamster Oviductin Antibody...... 80 5.3 Western Blot Analysis of Rat Oviduct-Specific Glycoprotein in the Female Rat Reproductive Tract …………………………………...…….... 80 5.4 Immunohistochemical Labeling of Rat Oviduct-Specific Glycoprotein in the Female Reproductive Tract ………………………………………..….. 81 5.5 Immunocytochemical Labeling of Rat Oviduct-Specific Glycoprotein in the Female Reproductive Tract …………………………………….….….. 82 5.6 Superovulation of Female Rats and Collection of Rat Oviductal Oocytes………………………………………………………….…….…... 83 5.7 Immunocytochemical Labeling of Oviduct-Specific Glycoprotein in the Rat Oviductal Oocytes ………………..……………………………...…… 84 5.8 cDNA Cloning of Putative Rat Oviductin (NP_997469)………………..... 85 6 RESULTS ….…………………………...………………………………..……. 87 6.1 Western Blot Detection of Oviductin Expression in the Female Rat Reproductive Tract ……………………………………..……………..…... 87 6.2 Immunohistochemical Labeling of Oviductin in the Female Rat Reproductive Tract ………………………………………...………….…... 88

viii 6.3 Immunocytochemical Labeling of Oviductin in the Female Rat Reproductive Tract ………………………………………………………... 91 6.4 Immunodetection of Oviductin in Rat Oviductal Oocytes …..…...…....…. 97 6.5 Search for Candidate Protein of Rat Oviductin and cDNA Cloning of the Putative Rat Oviductin (NP_997469) …………………....………..….…... 101 7 DISCUSSION ………...……………………………………………………….. 106

SUMMARY AND CLOSING REMARKS …….……………………………..…… 113

REFERENCES …...……………………………………….………………………… 115

ix LIST OF FIGURES AND ILLUSTRATIONS Page Figure 1. Diagrammatic representation of the internal organs of the female reproductive system …………………………………………………… 4

Figure 2. Schematic representation of the mouse-like rodent female reproductive system …………………………………………………………………. 5

Figure 3. Human menstrual cycle with the illustration of the serum profiles of pituitary gonadotropins and ovarian hormones ...……………………... 9

Figure 4. Estrous cycle of the golden hamster showing the serum profiles of pituitary gonadotropins and ovarian steroid hormones ……………….. 11

Figure 5. Schematic representation of the hamster oviduct located between the ovary and uterus ………………………….…………………..……….. 16

Figure 6. Electron photomicrograph showing two types of oviductal epithelial cells………………………….…………………………..……………... 19

Figure 7. Western blot analysis examining the lectin-affinity purified oviductin from the proestrus stage ……………………….……….……………... 51

Figure 8. Immunohistochemical localization of hamster oviductin in the female hamster reproductive tract using monoclonal anti-hamster oviductin antibody …………………………………………………………….…. 53

Figure 9. Light micrographs of primary cultured hamster uterine epithelial cells ………..………………………………………………….……….. 56

Figure 10. Immunodetection of cytokeratin in the cultured hamster uterine epithelial cells …………………………………………………………. 57

Figure 11. Confocal microscopy revealing the association of oviductin with cultured hamster uterine epithelial cells ………………………………. 59

Figure 12. Transmission electron micrographs of cultured primary hamster uterine epithelial cells and stromal cells …………………….………… 63

x Figure 14. Western blot analysis examining the protein expression of a hamster oviductin homologue in the rat ……………………………....………... 87

Figure 15. Immunohistochemical localization of oviduct-specific glycoprotein in the rat reproductive tract using monoclonal anti-hamster oviductin antibody ……...………………………………………………………... 89

Figure 16. Electron micrographs of thin-sections of LR-White embedded rat oviduct labeled with monoclonal anti-hamster oviductin antibody….… 92

Figure 17. Electron micrographs of thin-sections of LR-white embedded rat uteri labeled with monoclonal anti-hamster oviductin antibody ………….... 95

Figure 18. Electron micrographs of ultra-thin LR-White sections of superovulated rat oviductal oocytes labeled with monoclonal anti-hamster oviductin antibody ……………………………………………...………………... 98

Figure 19. Domain of hamster oviductin homologue in the rat predicted by SMART …………………………………………………….………….. 103

Figure 20. Agarose gel electrophoresis analysis of cloned cDNA of putative rat oviductin (NP_997469) in pGEM-T vector ………….….…………….. 104

Figure 21. cDNA sequence of putative rat oviductin (NP_997469)…………….… 105

xi LIST OF TABLES Page Table 1. Summary of published methods used for culturing uterine epithelial cells prepared from various species ….….……..……………………... 15

Table 2. BLAST results using hamster oviductin amino acid sequence in search of a putative rat oviductin in the rat genome ………………………..… 102

xii LIST OF ABBREVIATIONS Ab Antibody BLAST Basic local alignment search tool Bp Base pairs BSA Bovine serum albumin °C Degrees Celsius cDNA Complimentary deoxyribonucleic acid C2GnT2 Type 2 core 2 β-1, 6-N-acetylglucosaminyltransferase C- terminal Carboxyl-terminal DAB 3,3’-diaminobenzidine tetrahydrochloride ddH2O Double distilled water Di1 Diestrus 1 stage of the estrous cycle Di2 Diestrus 2 stage of the estrous cycle DMEM Dulbecco’s modified eagle medium EDTA Ethylenediaminetetracetic acid EGF Epidermal growth factor E.M. Electron microscopy Epon Bisphenol A/epichlorohydrin based epoxy resin Est Estrus stage of the estrous cycle et al. et alia (and others) FBS Fetal bovine serum FSH Follicle-stimulating hormone GalNAc N-acetyl-galactosamine GlcNAc N- acetyl-glucosamine GnRH Gonadotropin-releasing hormone HBSS Hank’s buffered saline solution hCG Human chorionic gonadotropin HCl Hydrochloric acid HPA Helix pomatia agglutinin HRP Horseradish peroxidase

xiii ICC Immunocytochemistry IEF Isoelectric focusing IHC Immunohistochemistry IgG Immunoglobulin G i.p. Intraperitoneal IU International units kDa Kilodalton(s) L Liter(s) LH Luteinizing hormone L.M. Light microscopy M Molar mAb Monoclonal antibody Met Metestrus stage of the estrous cycle mRNA Messenger ribonucleic acid MUC Mucin NaCl Sodium chloride NGS Normal goat serum N- terminal Amino-terminal PBS-T Phosphate-buffered saline Tween 20 PCR Polymerase chain reaction pI Isoelectric point PMSF Phenylmethanesulphonyl fluoride PMSG Pregnant mare’s serum gonadotropin Pro Proestrus stage of the estrous cycle PVDF Polyvinyl difluoride PVS Perivitelline space rhaOvm Recombinant hamster oviductin RNA Ribonucleic acid RT-PCR Reverse transcription-polymerase chain reaction SDS-PAGE Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

xiv Ser Serine TBS-T Tris-buffered solution-Tween 20 Thr Threonine µ Micron WGA Wheat germ agglutinin ZP Zona pellucida

CHAPTER 1.

INTRODUCTION AND LITERATURE REVIEW

1 1. OVERVIEW

Infertility affects many couples in North America. In Canada (1995) there were approximately 330,000 couples experiencing infertility and the total annual cost of infertility management was estimated to be $415 million (Collins et al.1997). Infertility may result from problems with the processes of sperm capacitation, fertilization, embryo development and implantation. However, among these factors, defects leading to problems with implantation are probably the most common and constitute the major cause of unexplained infertility (Bulletti et al. 1996). Accumulating evidence has shown that cell surface glycoconjugates play an important role in uterine receptivity during implantation.

Research in our laboratory has focused on a high-molecular-weight glycoprotein, oviductin, secreted by the oviductal epithelial cells. Oviductin or oviduct-specific glycoprotein has been identified and characterized in several mammalian species including the baboon (Verhage et al. 1989; Verhage et al. 1997b; Jaffe et al. 1996), cow

(Boice et al. 1990; Sendai et al. 1994), goat (Abe et al. 1995), hamster (Araki et al. 1987;

Leveille et al. 1987; Kan et al. 1988; Oikawa et al. 1988; Kan et al. 1990; Suzuki et al.

1995), mouse (Kapur and Johnson 1985; Sendai et al. 1995), monkey (Verhage et al.

1997a), pig (Brown and Cheng 1986; Buhi et al. 1990; Buhi et al. 1996), rabbit (Oliphant and Ross 1982), sheep (Sutton et al. 1984; Gandolfi et al. 1991; DeSouza and Murray

1995) and human (Verhage et al. 1988; Arias et al. 1994; O'Day-Bowman et al. 1995).

We have previously shown that hamster oviductin can gain access to the lumen of the uterus and bind to the cell surface of a certain population of uterine epithelial cells

2 (Martoglio and Kan 1996). Metabolic and immunocytochemical experiments showed that epithelial cells lining the hamster endometrium do not synthesize oviductin. These results led us to hypothesize that there are potential receptor sites present on the surface of uterine cells for oviductin binding. This could be of importance in modulating uterine receptivity. In order to determine the presence of putative oviductin receptor sites and their function in uterine receptivity, we have established a method of culturing primary hamster uterine epithelial cells. In the present study, an effort has been made to develop and optimize the conditions for this primary uterine epithelial cell culture system.

Subsequent confocal microscopic and immunocytochemical studies were performed to determine the precise location of hamster oviductin on the cell surface of cultured uterine epithelial cells.

Although oviductin has been identified in many mammalian species, very little is known about oviductin in the rat species. Numerous studies have shown that oviductins from all species that have been examined become associated with the zona pellucida of postovulatory oocytes (Abe and Oikawa 1990b; Kan et al. 1993; Gandolfi et al. 1991;

Boice et al. 1992; Buhi et al. 1993; Boice et al. 1991; O’Day-Bowman et al. 1996) except the mouse (Kapur and Johnson 1988) in which species oviductin is localized in the perivitelline space surrounding the oviductal oocytes. Thus, the purpose of the second part of the present study was to determine whether oviductin is present in the female rat reproductive tract and, if so, where it is localized ultrastructurally in the ovulated oocytes.

A pilot experiment was also conducted to clone the cDNA of putative rat oviductin.

3 2. LITERATURE REVIEW

2.1 The Female Reproductive Tract

The female reproductive tract is made up of the ovaries, oviducts, uteri, the vagina and the external genitalia (Fig. 1) (Junqueira and Carneiro 2003). Except for the external genitalia, the rest of the organs are also called female internal genital organs. The present review focuses mainly on the studies involving the oviduct and uterus of the female reproductive tract.

Figure 1. Diagrammatic representation of the internal organs of female reproductive system (coronal view; slightly modified from Junqueira and Carneiro 2003).

4 2.2 Uterus

2.2.1 Anatomical Structure of the Uterus

The uterus is a thick-walled, pear-shaped, hollow muscular organ in the human. It is located anteriorly and centrally in the pelvis, supported by the round, broad, uterosacral, and cardinal ligaments (Baggish et al. 2007). The uterus is divided into three portions: the cervix, corpus, and fundus (Fig. 1) (Junqueira and Carneiro 2003). The corpus forms the main mass of the uterus and has a heavy muscular wall. However, the gross anatomy of the uterus differs slightly among mammalian species. In the mouse-like rodents including the rat, mouse and hamster, the uterus appears as two horns and a caudal section – undivided corpus uteri, like a “Y” with a very short stem (Fig. 2) (Rugh 1968). The uterus is the site where the embryo grows, the implantation occurs and the fetus develops until parturition.

Figure 2. Schematic representation of the mouse-like rodent female reproductive system (ventral view, modified from Rugh 1968).

5 2.2.2 Histological Features of the Uterus

The uterine wall consists of three layers: the perimetrium, the myometrium, and the endometrium (Junqueira and Carneiro 2003). The perimetrium is an outer lining of serosa and the myometrium is mainly made up of smooth muscle (Young & Heath 2000). The endometrium consists of a simple columnar epithelium with two cell types: ciliated cells and non-ciliated microvillous cells, and an underlying thick connective tissue stroma.

Many simple tubular uterine glands, also lined by a single layer of columnar epithelial cells, extend through the entire thickness of the stroma. The stromal cells are embedded in a network of fibroblasts, leukocytes, mast cells and plasma cells. Barberini et al. (1978) observed that extensive changes occurred on the surfaces of non-ciliated microvillous epithelial cells in the rabbit during the progestational stages, involving the synthesis and secretion of glycoproteins and displaying numerous apical protrusions that contain material of varying electron density. Thus, the epithelial lining, the stroma plus the glands of the endometrium play an important role in the uterine receptivity during implantation.

Functionally, the endometrium can be divided into two layers: the stratum basalis and the stratum functionalis. The basalis is adjacent to the myometrium and is not shed off during menstruation, but functions as a regenerative zone for the functionalis in the next menstrual cycle. The stratum functionalis is the luminal part of the endometrium. It is sloughed off during every menstruation and is the site of cyclic changes in the endometrium. This layer is nourished by the helical arteries that are from the arcuate arteries and are subjected to hormonal control by progesterone. The functionalis layer can

6 be further divided into stratum spongiosum and stratum compactum (Young & Heath

2000).

2.2.3 Hormonal Regulation in the Uterus

In the human, the hormonal regulation of the uterus is a very complicated process that involves the hypothalamic-pituitary-ovarian-uterus axis (Baggish et al. 2007). When the hypothalamus receives the neuronal signals from higher brain regions, it releases gonadotropin-releasing hormone (GnRH). GnRH is secreted in pulses with intervals from

60 to 90 minutes depending on the stage of the cycle. In response to GnRH, gonadotropes in the anterior pituitary region secrete follicle-stimulating hormone (FSH) and luteinizing hormone (LH) to the circulation system. Since FSH induces the expression of LH receptors present on the granulosa cells in the ovary, when FSH and LH act together, they enhance the granulosa cells to produce estrogen. Increasing levels of estrogen can cause the pituitary to lower its FSH secretion. At low concentrations, estrogen inhibits LH release, while at high concentrations, it stimulates LH production. Following ovulation with no occurrence of fertilization and under the influence of LH, the ruptured follicle becomes the corpus luteum and secretes progesterone. Progesterone inhibits the secretion of FSH in the pituitary. When the level of progesterone declines, the pituitary secretes FSH again.

However, if fertilization occurs, the trophoblasts will synthesize a new hormone, luteotropin, which keeps the progesterone level high, preparing the uterine tissue for implantation (Baggish et al. 2007). Thus, the endometrium of the uterus intimately acts as an end organ that changes in response to estrogen and progesterone stimulation.

7 2.2.4 Cyclic Changes in the Endometrium of the Uterus

A. The Human

The endometrium in the normal cycling female is an ever-changing structure; these cyclic changes are divided into several phases during a typical 28-day human menstrual cycle: the menstrual phase, the proliferative (or follicular) phase, the secretory (or luteal) phase, and the premenstrual (or ischemic) stage (Fig. 3) (Young and Heath 2002; Baggish et al. 2007).

The first day of vaginal bleeding is usually considered the first day of the menstrual cycle. This phase usually lasts 4~5 days in humans, during which the stratum functionalis is completely shed. The shedding at this phase is only restricted to primates.

Following the menstrual phase is the proliferative or follicular phase, which is from day 6 to day 13. In this phase, the basal endometrium, stroma and glands begin to grow under the influence of estrogen. The lengthened glands are mainly straight and tubular and becoming open to the luminal surface in the late proliferative phase. The endometrium thickens remarkably to 8mm from 1mm at the menstrual phase (Baggusg et al. 2007). The surface microvilli increase in length and remain straight. Mitotic features can be observed both in the epithelial and the stromal cells (Young and Heath 2002).

Ovulation marks the beginning of the secretory phase which lasts from day 15 to 28.

Remarkable endometrial changes occur during this phase. With secretion of progesterone from the corpus luteum, the endometrium continuously thickens and the glandular epithelium is stimulated to synthesize glycogen. The glycogen accumulates to form vacuoles in the basal portion of the epithelial cells on day 16, which is the characteristic of

8 early secretory phase (Young and Heath 2002). At the late phase, the vacuoles migrate to the supra-nuclear region and eventually are secreted into the lumen. Meanwhile, the glands become highly convoluted and significantly dilated. The lamina propia and stroma turn slightly edematous (Baggish et al.2007).

In the human, a short stage (around 2 days) prior to the menstrual phase is termed the premenstrual or ischemic stage. Triggered by the reduction of progesterone, the coiled arteries constrict causing hypoxia in the endometrium; the glands stop to secrete. This leads to a new menstrual phase (Young and Heath 2000).

Figure 3. Human menstrual cycle with the illustration of the serum profiles of pituitary gonadotropins and ovarian hormones. FSH: Follicle-stimulating hormone; LH: Luteinizing hormone (Young and Heath 2000).

9 B. The Hamster

Not only have the changes in the endometrium during the menstrual cycle been characterized in human as described above, they have also been investigated in many other species in their corresponding estrous cycles. Since rodents have a relatively short uterine cycle, more and more attention has been given to them in studies of reproduction. For instance, the golden hamster has a 4-day estrous cycle. On the morning after ovulation which is referred to as the metestrus stage (day 1 of the estrous cycle), the hamster produces a vaginal discharge. This vaginal discharge is recognizable by its copious nature, sticky consistency and tangy odor, so its existence provides us with an indication of the onset of the metestrus stage (Lisk 1985). Early in the diestrus 1 stage (in the evening on the 1st day of the estrous cycle), progesterone increases dramatically and drops later on diestrus 2 stage (the 3rd day of the estrous cycle). Thus, the metestrus and diestrus stages in the hamster correspond to the luteal or secretory phase in the human. At noon on day 4 of the estrous cycle (the proestrus stage), estrogen level remarkably surges, consequently stimulating LH, FSH and progesterone secretion to reach a peak. This period is equivalent to the follicular phase in the human (Lisk 1985). All these hormones go down to a normal level when ovulation initiates on the estrus stage (day 4 of the estrous cycle) (Fig. 4) (Lisk

1985).

Figure 4. Estrous cycle of the golden hamster showing the serum profiles of pituitary gonadotropins and ovarian steroid hormones. The five estrous stages are indicated as: M:

Metestrus; D1: Diestrus 1; D2: Diestrus 2; Pr: Proestrus; Es: Estrus. P: Progestrone; E: Estrogen; FSH: Follicle-stimulating hormone; LH: Luteinizing hormone (Slightly modified from Lisk 1985).

11 2.2.5 Implantation in the Uterus

Implantation is a critical and complex process by which the blastocyst comes into intimate physical and physiological contact with the uterine endometrium. Early in 1967,

Ender and Schlafke classified the process of embryo implantation into three stages: apposition, adhesion, and penetration. During these stages, the components involved in implantation are mainly the embryonic trophectoderm and the apical surface of the uterine epithelial cells. It is thought that under the effect of steroid hormones, the luminal epithelial surface of the uterus differentiates to a state, termed “uterine receptivity” that allows the blastocyst to attach and subsequent events of implantation to occur (Paria et al.

2000).

Although it has been suggested that ovarian steroid hormones are the main factors that regulate implantation (Bagchi et al. 2001), growing evidence indicates that growth factors, cytokines (Sharkey 1998), extracellular matrix-degrading enzymes (Brenner et al.

1989) and cell surface glycoproteins may also play important roles in this process (Carson

2002). The possible role of glycoproteins in uterine receptivity was first brought into light based on the observation of the reduction in the thickness of the glycocalyx on the uterine epithelial cells (Morris and Potter 1984; Anderson and Hoffman 1984). Under most circumstances, the uterine epithelium is in the non-receptive state that is characterized by the existence of mucin glycoproteins. For example, MUC1, one component of the apical glycocalyx, is a transmembrane mucin with a high molecular weight (>200kDa) (Carson et al.1998). Studies showed that MUC1 expression was markedly reduced in luminal epithelia during the receptive phase in the rat (DeSouza et al. 1998). Loss of MUC1 is

12 believed to be necessary for the initial attachment of embryos to the uterus and the conversion of the uterine luminal epithelium to a functionally receptive state (Aplin and

Kimber 2004). The second proposed process involved in peri-implantation is the functional expression of complementary adhesion-promoting molecules at both uterine epithelial and embryo cell surfaces. Studies have indicated that heparan sulfate proteoglycans and their corresponding binding sites play important roles in this process

(Carson 2002).

Due to research limitations and ethical considerations, understanding of the implantation and the molecules involved in humans has grown slowly in the last few decades. Studies have been extensively carried out in animal models such as the mouse and rat (Lee and DeMayo 2004). Therefore different animal models will continue to be used for the study of embryo-uterine interactions and efforts should be guided to set up some reliable in vitro system, facilitating researchers to gain a better understanding of implantation in the human.

2.2.6 Uterine Epithelial Cell Culture System

Several epithelial-stromal cell culture systems have been developed and established initially to study the functions of steroid hormones in regulating the growth and differentiation of these cells. To directly study the biological properties of uterine epithelial cells and the interaction between the blastocyst and uterine endometrium, several laboratories have reported methods to isolate and culture uterine epithelial cells from the human (Liszczak et al. 1977; Kirk et al. 1978; Varma et al. 1982; Classen-Linke et al.

1997), porcine (Bowen et al. 1996), bovine (Thibodeaux et al. 1991 and 1992), rat

13 (Richardson et al. 1995; Sharpe et al. 1992; Glasser et al. 1988), mouse (Tomooka et al.

1986), guinea-pig (Chaminadas et al. 1986), and rabbit (Mulholland et al. 1988).

Significant improvement has been made to isolate the uterine epithelial cells. For instance, enzymatic dissociation followed by filtration or gradient centrifugation after enzymatic digestion has been applied in human uterine epithelial cell culture (Classen-Linke 1997) and mouse uterine cells culture (Tomooka et al. 1986). On the other hand, chemically modified culture serum also has positive effects on the cells during the culture procedure.

In addition to the steroid hormones that regulate uterine cell differentiation, factors such as epidermal growth factor (EGF) and Ca2+ concentration are other important regulators during culture. In the study of Tomooka et al. (1986), the authors demonstrated that growth of mouse uterine epithelial cells can occur and can be enhanced in medium containing EGF

(10ng/mL) and calcium (5X10-5 M). Since it is not possible to describe in detail the procedures used for each species that has been studied, a summary of the published procedures regarding cell dissociation, isolation and medium used in culture is presented in

Table 1.

Table 1. Summary of published methods used for culturing uterine epithelial cells prepared from various species. (DMEM: Dulbecco’s Modified Eagle’s Medium; TCM: Tissue culture medium; CMRL: Connaught Medical Research Laboratories. EDTA: Ethylenediaminetetracetic acid)

Authors Year Species Stage of the Method of Method of Culture Passage of Cycle Dissociation Isolation Medium Cells

Classen-Linke et al. 1997 human proliferative 0.125% type IA filtration DMEM/ maintained for collagenase through seives Ham’s F12 10-14 days

Bowen et al. 1996 porcine day 15 of 4.8mg/mL centrifugation DME/F12 N/A pregnancy dispase/ pancreatin 15 Thibodeaux et al. 1991 bovine estrus 0.25% trypsin/ centrifugation TCM-199 3rd subpassage 0.2mg/mL EDTA frozen

Glasser et al. 1988 rat immature 0.5% trypsin/ centrifugation+ DMEM/ detach by 2.5% pancreatin sedimentation Ham’s F12 12-14 day

Richardson et al. 1995 rat estrus 0.5% trypsin/ centrifugation+ DMEM/ maintained for 2.5% pancreatin sedimentation Ham’s F12 8 days

Tomooka et al. 1986 mouse immature 0.5% bovine pancreatic centrifugation DMEM/ N/A trypsin Ham’s F12

Chaminadas et al. 1986 guinea-pig estrus 0.25% collagenase centrifugation CMRL 1066 N/A

2.3 Oviduct

2.3.1 Anatomical Structure of the Oviduct

The mammalian oviduct, also known as Fallopian tube or uterine tube in the human, provides a suitable environment for the transport of gametes, maturation of gametes, development of the early embryo (Leese 1988). In humans, the oviduct is divided in four segments: 1) the infundibulum, a funnel-shaped distal segment with finger-like projections called fimbriae; 2) the ampulla, a wide middle segment where fertilization occurs; 3) the isthmus, a narrow segment adjacent to the uterus; 4) the intramural portion that continues with the uterus (Fig. 1) (Young and Heath 2002; Junqueira and Carneiro 2003). In mouse-like rodents such as the hamster, the oviduct displays as a highly convoluted and coiled tube between the uterus and ovary (Figs. 2 and 5) (Lisk 1985).

Figure 5. Schematic representation of the hamster oviduct (grey) located between the ovary and uterus. The hamster oviduct is highly convoluted (Slightly modified from Boatman et al. 1994).

16 2.3.2 Histological Features of the Oviduct

The oviductal wall consists of three layers: an outer layer of serosa made of peritoneum; a poorly defined muscularis containing external longitudinal and internal circular smooth muscle fibers; and a folded mucosa consisting of loose connective tissue

(lamina propria) and a layer of luminal epithelium (Young and Heath 2002).

The luminal epithelium consists of two main cell types: ciliated and non-ciliated cells

(Fig. 6) (El-Mestrah and Kan 1999a; Abe and Oikawa 1991). The ciliated cells are short and characterized by their long cilia projecting into the lumen of the oviduct (Young and

Heath 2000). The nucleus of the ciliated cells is centrally located and palely stained as seen under the light microscope. Electron microscopy reveals clusters of mitochondria and membrane-bound vesicles are present in the apical region of the ciliated cells

(El-Mestrah and Kan 1999a; Junguiera and Carneiro 2003). Ciliated cells are responsible for transporting oocytes along the oviductal lumen, propelling the spermatozoa in the opposite direction, and moving the viscous liquid that masks their surface (Junqueira and

Carneio 2003). On the other hand, the non-ciliated secretory cells are elongated and have well developed microvilli as their prominent feature (El-Mestrah and Kan 1999a). The prominent morphological characteristic of the non-ciliated cells seen under the electron microscope is the presence of variable sized secretory granules at the supranuclear region

(El-Mestrach and Kan 1999a). Well-developed Golgi complexes are associated with these secretory granules, indicating an active protein secretory process. Based on the staining intensity examined on the electron microscope, there are two types of secretory granules:

1) the homogeneous electron-dense, and 2) the electron-lucent granules. These granules

17 are present in the non-ciliated secretory oviductal epithelial cells of several species that have been investigated, including the mouse (Komatsu and Fujita 1978), rabbit (Jansen and Bajpai 1982), cow (Abe et al. 1993), golden hamster (Abe and Oikawa 1991;

El-Mestrah and Kan 1999a), monkey (Odor et al. 1983) and human (Clyman 1966).

In addition, ovarian hormones are involved in the differentiation of oviductal epithelial cells and the formation of secretory granules within the non-ciliated secretory cells (Young and Heath 2000; Abe and Oikawa 1990a). Estrogen has critical and specific effects on inducing the differentiation of secretory cells and ciliated cells (Abe and Oikawa

1992) whereas progesterone induces the regression of the oviductal epithelial cells

(Verhage et al. 1990). In summary, the cyclic variations of the number of ciliated and non-ciliated cells and the morphological features of these two cell types can contribute to the reproductive events that occur in different regions of the oviduct at different periods.

Figure 6. Electron photomicrograph showing two types of oviductal epithelial cells. The ciliated cells (CC) are columnar shaped with the nucleus (Nu) located basally. Non-ciliated secretory cells (SC) are characterized by the presence of microvilli (Mi) at their apical surface and numerous secretory granules (SG) at their apical pole. Lu: Lumen; Ci: Cilia; M: Mitochondria. X12,500. (Slightly modified from El-Mestrah and Kan 1999a).

19 2.4 Oviductal Fluid

The oviduct provides an optimized environment for a series of critical events in the reproductive process by presence of the oviductal fluid within its lumen. Chemical analysis of oviductal fluid revealed that it is a complex mixture of constituents derived from the plasma and some proteins secreted by the oviductal epithelial cells (Leese 1988). First, water is the largest component of oviductal fluid and appears to be transported by free

+ + − diffusion (Leese 1988). Second, the ions present in the fluid are mainly K , Na , HCO3 .

These electrolytes contribute not only to maintaining the osmolarity and pH of oviductal fluid (Leese 1988), but also modulating some specific events such as the development of embryos, the inhibition of sperm motility and the stimulation of sperm respiration (Roblero and Riffo 1986; Burkman et al. 1984; Hamner and Williams 1964). Third, non-electrolytes are also found in the oviductal fluid. These include glucose, lactate, pyruvate, and amino acids, which provide good substrates supporting the survival of gametes and embryos

(Leese 1988). The most abundant proteins in oviductal fluid are albumin and immunoglobulin G, which are derived from the blood; however, their embryonic functions are unknown (Leese 1988). Numerous enzymes and inhibitors, mainly derived from serum, have also been found and characterized in oviductal fluid of some species (Buhi et al.

2000). Among the proteins derived from the oviduct, a family of glycoproteins of high molecular weight has been detected and identified. These oviduct-specific glycoproteins, collectively known as oviductins, are synthesized and secreted into the oviductal lumen by the non-ciliated secretory cells of the columnar epithelium under the influence of steroid hormones (Buhi et al. 2000; for a review, see Buhi 2002).

20 2.5 General Review of Oviductin

Oviductin is a collective term designated for a family of glycoproteins that are synthesized and released to the oviductal lumen and later become associated with the ovulated oocytes and early embryos (Malette and Bleau 1993).

Oviductin has been identified and characterized in several species, including the baboon (Verhage et al. 1989; Verhage et al. 1997b; Jaffe et al. 1996), cow (Boice et al.

1990; Sendai et al. 1994), goat (Abe et al. 1995), hamster (Araki et al. 1987; Leveille et al.

1987; Kan et al. 1988; Oikawa et al. 1988; Kan et al. 1990; Suzuki et al. 1995), mouse

(Kapur and Johnson 1985; Sendai et al. 1995), monkey (Verhage et al. 1997a), pig (Brown and Cheng 1986; Buhi et al. 1990; Buhi et al. 1996), rabbit (Oliphant and Ross 1982), sheep (Sutton et al. 1984; Gandolfi et al. 1991; DeSouza and Murray 1995) and human

(Verhage et al. 1988; Arias et al. 1994; O'Day-Bowman et al. 1995).

These identified oviductins are highly conserved and display 70-78% identity and

76-87% similarity in their overall deduced amino acid sequence among species (Buhi

2002). Sequence comparisons showed that these oviductins display a high degree of conservation in the amino-terminal (N-terminal) region of the mature polypeptide compared with a low identity and similarity in the carboxyl-terminal (C-terminal) end

(Buhi 2002). The differences of oviductin sequences between species are due to exon 11 of the gene, which contains numerous Ser-Thr residues. For example, hamster and mouse oviductins have six tandem-repeat sequences, each made up of 15 amino acids and 21 tandem-repeat sequences composed of 7 amino acids in their C-terminus, respectively

(Paquette et al. 1995; Sendai et al. 1995). Oviductin in the human, baboon and rhesus

21 monkey contains only four tandem-repeat sequences (Verhage et al. 1998). In the cow, sheep and pig, tandem-repeat sequences are incomplete or absent (Buhi 2002).

Interestingly, there are multiple molecular weight variants and isoforms of oviductins among species. Under reducing condition, oviductins display a relatively high molecular weight ranging from about 70 to 130 kDa, with the exception of hamster (>160 kDa) (Buhi et al. 2000). Cow, pig, sheep, human, baboon and hamster oviductins have several molecular weight forms and at least two isoforms. Pig oviductin consists of two isoforms while bovine has seven isoforms (for a review, see Buhi 2002).

Oviductins possess domains that are similar to those of the chitinase-related proteins.

Alignment of the oviductins from different species with other gene products revealed that all oviductins are similar to the proteins from the chitinase gene family, such as human macrophage chitotriosidase, human cartilage glycoprotein 39 and a group of enzyme-chitinases (Malette et al. 1995a; Buhi 2002). Chitinase proteins share high identity and similarity in the N-terminal domain which is defined as the catalytic domain, and four residues in the chitinase appear to be critical for enzyme activity (Watanabe 1993).

However, since all oviductins examined lack the essential glutamic acid residues, they lack the chitinase activity (Buhi 2002).

Oviductins also enclose domains of significant similarity with a C-terminal mucin-type domain. The polypeptide sequences of oviductins show the very prominent feature that they contain Ser-Thr rich repeating units in their C-terminal end, which is a well-documented characteristic of mucin-like glycoproteins (Malette et al. 1995a). The specific enriched Ser-Thr residues are the potential acceptor sites for O-glycosylation and

22 the mucin proteins are negatively charged due to the presence of sialic acid and sulfate moieties (Strous and Dekker 1992). Although the sequences of tandem repeats in mucins are not highly conserved between species, both hamster and human oviductin show an unusual degree of similarity in the number of tandem repeats and the number of amino acids per repeat (Lapensee et al. 1997).

It is well established that gene expression and protein secretion of oviductins in all mammalian species examined to date are under hormonal control. However, this regulation seems to show species variations between those with a long estrous cycle and those having a shorter estrous cycle. Oviductin mRNA increases to a peak value in the late follicular phase in the human, but is very low or absent during the luteal phase (Arias et al. 1994).

Increased number of secretory granules containing oviductin in the non-ciliated oviductal epithelial cells and increased amount of oviductin in the oviductal fluid or cultured medium have been observed corresponding to the periods when maximum oviductin mRNA is expressed (Arias et al. 1994; Buhi et al. 1990, 1996; Buhi 2002). Furthermore, ovariectomy and hormonal replacement experiments revealed that estrogen maintains or restores oviductin mRNA expression whereas additional progesterone strongly suppresses mRNA and protein expression. In the hamster, there is no significant difference in the level of hamster oviductin mRNA throughout the five stages of the estrous cycle as indicated by experiments carried out with semi-quantitative RT-PCR (McBride et al. 2004), Northern blot analysis (Komiya et al. 1996) and RNA protection essay (Paquette et al. 1995). In contrast, immunocytochemical studies using a monoclonal antibody specific for the glycosidic epitope of hamster oviductin revealed different concentrations of glycoprotein

23 levels in the five different stages of the estrous cycle (Roux and Kan 1995), suggesting that hormonal regulation of oviductin in the hamster occurs post-translationally. It is noteworthy to point out that LH may also have a direct effect on oviductin expression since the treatment with human chorionic gonadotropin (hCG) on bovine epithelium in vitro increases the half-life of oviductin expression (Sun et al. 1997).

Oviductins have been immunolocalized exclusively in the oviduct of the female reproductive tract and share a universal characteristic that they are associated with the zona pellucida of ovulated oocytes and embryos except the mouse. The distribution of oviductin along the oviduct, however, displays regional differences. In bovine (Abe et al.

1993), sheep (Gandolfi et al. 1991) and pigs (Buhi et al. 1990), oviductin is localized within the infundibulum and ampullary region whereas rabbit (Oliphant et al. 1984) oviductin is restricted to the isthmus. In rodents, golden hamster oviductin was primarily produced in the non-ciliated secretory cells of the ampulla and isthmus (Abe and Oikawa

1991) compared to the mouse where oviductin is also found in the infundibulum (Kapur and Johnson 1988). The reason behind this phenomenon and the functional significance of the presence of oviductin at the particular regions remain to be elucidated. In addition, hamster oviductin has been localized to the surface of some non-ciliated epithelial cells lining the endometrium of the uterus during the estrous cycle and early gestation, suggesting a potential role for oviductin in the pre-implantation process (Martoglio and

Kan 1996; Roux et al. 1997).

One common characteristic shared by all oviductins with the exception of that of the mouse is that they are uniformly distributed throughout the zona pellucida of the ovulated

24 oocytes and early embryos (Abe and Oikawa 1990b; Kan et al. 1993; Gandolfi et al. 1991;

Boice et al. 1992; Buhi et al. 1993; Boice et al. 1991; O’Day-Bowman et al. 1996; Kapur and Johnson 1988). The association of oviductin with vitelline/blastomere membrane and perivitelline space was also observed. A detailed description of the association of hamster oviductin with post-ovulatory oocytes and embryos will be given in 2.6.4. The consistent distribution of oviductin in the post-ovulatory oocytes and early embryos across species indicates that oviductin may play important roles in regulating fertilization and early embryonic development.

Studies have revealed that oviductin can bind to the sperm. Indirect immunoflourescence labeling has shown that hamster oviductin can bind to the acrosomal region of homologous sperm before and after capacitation (Boatman and Magnoni 1995).

Bovine oviductin has been shown to bind to sperm head and midtail regions (King and

Killian 1994). However, O’Day-Bowman et al. (1993) revealed that human oviductin does not appear to bind to homologous spermatozoa. Functional studies have indicated that bovine oviductin has positive effect on motility and viability of bovine spermatozoa

(Abe et al. 1995) and that it can enhance bovine sperm capacitation and fertilizing ability in vitro (King et al. 1994). It has also been shown that oviductin can increase sperm-egg binding and penetration rate in the hamster and human in vitro (Boatman and Magnoni

1995; O’Day-Bowman et al. 1996) and decrease polyspermy in the pig (Kouba et al.,

2000).

25 2.6 Hamster Oviductin

2.6.1 Ontogeny of Hamster Oviductin

Indirect immunofluorescence labeling used to study the hamster oviductin ontogeny revealed that immunostaining of secretory oviductal epithelial cells starts to appear in

7-day-old female hamsters. The majority of the cells showed immunofluorescence at day

10 with a gradual increase in staining intensity until all the cells showed an immunoreaction at day 14 (Malette et al. 1995b). A maximum intensity of the labeling appeared at day 28 just prior to the first ovulatory cycle (Malette et al. 1995b). It has been suggested that the ontogeny of hamster oviductin expression parallels the hormonal changes that lead to sexual maturation, and that its maximum secretion is already established at the time of the first estrous cycle (Malette et al. 1995b).

2.6.2 Biochemical Characterization of Hamster Oviductin

Hamster oviductin, also known as hamster oviductin-1, is a family of glycoproteins with a molecular weight of 160-350kDa (Malette and Bleau 1993). Hamster oviductin consists of two major isoforms: the acidic form α and the β form (Malette and Bleau 1993).

The acidic form with an isoelectric point (PI) of 3.5-4.5 has a lower molecular weight of

160-210kDa compared to that of the β form (~210-350kDa) (Malette and Bleau 1993). The

β form displays a sharp band in the isoelectric focusing (IEF) dimension. The poor migration of the β form might be due to its highly glycosylation nature or non-covalent aggregation or both (Malette and Bleau 1993). However, peptide-mapping experiments revealed a similar pattern of peptide fragments from both forms of hamster oviductin; a later study showed that the N-terminal sequences of the two isoforms are the same and

26 there is no match to other proteins reported. Thus, the two isoforms of hamster oviductin are indeed composed of the same protein core but differ from each other in the post-translational modification (Malette and Bleau 1993). As mentioned before, by comparison to other reported proteins, hamster oviductin cDNA displays many common properties with the family of chitinase and mucin proteins (for a review, see Malette et al.

1995a).

Lectin probe approaches provide evidence showing that hamster oviductin contains terminal α-D-GalNAc and either terminal α-D-NeuAc or non terminal α-β-D-(GlcNAc)2 residue, because it reacts with HPA and WGA but not ConA and UEA-1 (Malette and

Bleau 1993).

After Malette and Bleau (1993) reported the sequence of the N-terminal portion of hamster oviductin, Suzuki and his colleagues (1995) generated the longest clone of hamster oviductin having 2.4kbp. The computer-calculated molecular weight of this whole length hamster oviductin cDNA is 70890 Da. Eight potential N-glycosylation sites and numerous

Ser/Thr residues are located in the C-terminal end between amino acid 432 and 605. Many putative sites for O-glycans are also found in this region (Suzuki et al. 1995). It has been suggested that only a few N-glycan chains are present in the hamster oviductin whereas

O-linked glycans make up 60% or more of the whole molecular weight of this glycoprotein

(Malette and Bleau 1993). Metabolic study revealed that hamster oviductin is a sulphated glycoprotein and the sulphation occurs on the carbohydrates, especially on O-glycosylated side chains (Malette and Bleau 1993).

27 2.6.3 Hormonal Regulation of Hamster Oviductin Synthesis and Glycosylation

As mentioned above, hormones are involved in the synthesis and serection of oviductins of all mammalian species examined to date; however, the pattern varies among species. In the hamster, it is likely that hormonal regulation occurs at the post-translational modification level, not at the mRNA expression level. Immunolabeling studies using a monoclonal antibody specific for glycosidic epitope of hamster oviductin resulted in the detection of a variation in labeling intensity in the secretory granules of oviductal cells during the estrous cycle with the highest concentration at the estrus stage and lowest at diestrus 1 and 2 stages (Roux and Kan 1995). Since hamster oviductin is a highly glycosylated protein, lectins become very useful tools to study its sugar moieties. A quantitative lectin-binding analysis conducted by El-Mestrah and Kan (1999b) showed that the concentration of WGA-binding sites in the secretory granules of non-ciliated secretory oviductal cells were highest at estrus and proestrus, and lowest at diestrus 2. Furthermore, the same study revealed an increase in GlcNAc and/or sialic acid residues on the sugar side chains at the proestrus and estrus stages (El-Mestrah and Kan 1999b). Based on the above studies and the previous report that hamster oviductin mRNA does not change during the estrous cycle, it was hypothesized that the hormonal regulation of hamster oviductin occurs at the post-translational level and studies had been carried out to study the glycosyltransferase activities in the hamster oviduct. Tulsiani et al. (1996) were the first to report a temporal surge of glycosyltransferase activities in the oviductal fluid of hamster at the onset of ovulation suggesting that these enzyme activities may be directly or indirectly regulated by hormones. In a quantitative immunocytochemical study using a polyclonal

28 antibody specific for the protein core of hamster oviductin, McBride et al. (2004) demonstrated that there were no significant differences in labeling intensity in the secretory granules of the non-ciliated oviductal cells between five different stages of the estrous cycle. A parallel semiquantitative RT-PCR study also showed that there were no significant differences in the relative mRNA concentrations of hamster oviductin between the five stages of the estrous cycle (McBride et al. 2004). The results from the latter study were consistent with those reported by other investigators studying mRNA expression in the hamster using Northern blot analysis (Komiya et al. 1996) and RNase protection assay

(Paquette et al. 1995). A novel finding arising from the study of McBride et al. (2004) was that specific glycosyltransferase activities such as mucin-type core 2

β6-GlcNAc-transferase (C2GnT2) displayed both significantly high level of gene expression and high enzyme activities at the proestrus and estrus stages, and low level of gene expression and enzyme activities at the diestrus 1 stage. Thus, the pattern of the gene expression and enzyme activities of C2GnT2 parallels the pattern of the hamster oviductin glycosylation during the estrous cycle.

2.6.4 Localization of Hamster Oviductin

Localization of hamster oviductin in the female reproductive tract, in the gametes and the embryos has been studied extensively in our laboratory and by others.

Oviductin was detected in the oviductal epithelium at the microscopic level using indirect immunofluorescent staining technique (St-Jacques and Bleau 1988). At the ultrastructural level, labeling of oviductin was observed in the secretory granules and Golgi complexes of the non-ciliated secretory epithelial cells while no reaction was detected in

29 the ovary (Abe and Oikawa 1991; Kan et al. 1989). These electron microscopic observations indicated that oviductin is primarily produced and secreted by the non-ciliated secretory cells in the oviductal epithelium. Previous studies have shown that hamster oviductin can gain access to the lumen of the uterus. Martoglio and Kan (1996) reported that by the use of silver enhancement immunohistochemical technique, hamster oviductin was detected at the cell surface of the endometrium throughout the estrous cycle. The labeling started to decrease from day 1 of gestation and disappeared after day 5 at which time the blastocyst is mature and ready to implant. In a later study on the fate of hamster oviductin in the uterus, biochemical analyses demonstrated that a degradation of oviductin occurred in the uterus, and a loss of immunoreactivity was also observed by the time of implantation (Roux et al. 1997). The authors suggested that hamster oviductin binds to some uterine epithelial cells via specific receptors expressed at the cell surface and is then endocytosed via coated pits and vesicles for degradation (Roux et al. 1997). Thus, hamster oviductin and its potential receptors may function in modulating the implantation process, determining the “receptive” or “non-receptive” state of the uterus.

Once hamster oviductin is released into the oviductal lumen, it becomes associated with and distributed throughout the zona pellucida of post-ovulatory oocytes (Abe and

Oikawa 1990b; Kan et al. 1988; Kan et al. 1989; Kan et al. 1990). It was demonstrated that immunolabeling for hamster oviductin was exclusively found in the zona pellucida encompassing the oocyte. Other structures within the oocyte, the perivitelline space, and neighboring cumulus cells were not labeled (Kan et al. 1988). Hamster oviductin was also detected in developing embryos. Using immunogold labeling in conjunction with

30 monoclonal antibody against the hamster oviductin, Kan and colleagues (Kan et al. 1993;

Kan and Roux 1995) reported that immunolabeling was detected not only in the zona pellucida and perivitelline space, but also over the plasma membranes of young fertilized oocytes, 2-cell, 4-cell, and 8-cell embryos. Hamster oviductin has also been localized in endocytic structures inside the cytoplasm including coated pits, coated vesicles, and multivesicular bodies (Kan et al. 1993), indicating that the oviductin that binds to the zona pellucida of the oocytes is internalized by the embryos and further processed for degradation (Kan et al. 1993).

It has also been demonstrated that hamster oviductin can bind to the spermatozoa prior to capacitation and that redistribution of hamster oviductin occurs on the sperm head plasma membrane (Boatman and Magnoni 1995). Recently, a more thorough immunocytochemical study carried out by Kan and Esperanzate (2006) showed that hamster oviductin bound specifically to the acrosomal cap and post-acrosomal region of non-capacitated sperm and that its binding to both regions was significantly increased during capacitation.

2.6.5 Possible Roles of Hamster Oviductin

The intimate association of hamster oviductin with ovulated oocytes, embryos and sperm indicate that it may exert some effects on the gametes and early embryo development. Schmidt et al. (1997) reported that in vitro sperm-egg binding rate was increased approximately 2-fold after hamster oviductin was added to the medium whereas human oviductin significantly reduced the binding rate between hamster sperm and hamster oviductal oocytes (Schmidt et al. 1997) indicating that hamster oviductin can

31 enhance the sperm-egg binding early in fertilization and that it has species-specific recognition property (Schmidt et al. 1997; Verhage et al. 1998). It has also been shown that hamster oviductin can enhance zona pellucida-induced acrosomal reaction and significantly facilitate sperm penetration of oviductal oocytes (Boatman et al. 1994). Not much progress has been done in the hamster oviductin in relation to sperm capacitation, although bovine oviductin has been shown to enhance sperm capacitation and fertilization capabilities in vitro (King et al. 1994). It is noteworthy to mention that recent preliminary studies conducted in our laboratory show that hamster oviductin can enhance the sperm capacitaion in a time dependent manner (unpublished observations).

It is widely accepted that hamster oviductin is a mucin-type of secretory glycoprotein and that it displays some properties characteristic of mucin (Malette and Bleau 1993;

Paquette et al. 1995). It covers the oviductal epithelium and is negatively charged due to the presence of sulphate and sialic acid groups (Malette and Bleau 1993). It is resistant to proteolysis by protease such as trypsin and chymotrypsin and is hardly digested by other proteases (Malette and Bleau 1993). Thus, hamster oviductin may serve as a protective barrier for oocytes and/or embryos in the female reproductive tract (Malette and Bleau

1993). Hamster oviductin is also present at the cell surface of the uterine epithelium until it disappears on day 5 and 6 when the hamster blastocyst is ready to implant (Martoglio and Kan 1996). This indicates that hamster oviductin may play a role in the implantation process in regulating uterine receptivity.

32 2.7 Rat Oviductin

Oviductin has been identified and characterized in many mammalian species as described above (see CHAPTER 1, 2.5 General Review of Oviductin); however, studies on oviductin in the rat are lacking. To date, a study conducted by Abe and Abe (1993) was the only one that reported a glycoprotein immunologically similar to hamster oviductin in the rat oviduct using a monoclonal antibody (mAb C8B11) raised against hamster oviductin.

But no further immunocytochemical evidence of the distribution of this glycoprotein in the female rat reproductive tract was provided. A subsequent Northern blot study conducted by

Arias et al. (1994) showed that the human oviductin cDNA probe hybridized to mRNA from oviductal tissue of the baboon, hamster, mouse, rabbit, cow, and cat but not to rat oviductal sample. As a result, whether an oviductin exists in the female rat reproductive tract needs to be resolved. Interestingly, earlier preliminary results obtained in our laboratory showed that the monoclonal antibody (F4.12) recognizing hamster oviductin cross-reacts with the rat oviductal tissue. This monoclonal antibody provides us with a useful tool to clarify the presence of oviduct-specific glycoprotein in the rat. Investigation of the localization of oviductin in the female rat reproductive tract would aid in broadening the information available on oviductin and its possible roles in the process of reproduction.

Moreover, since the mouse is the only species studied so far where, in the postovulatory oocytes, oviductin is not found in the zona pellucida but in the perivitelline space, it will be of interest to find out the precise site of localization of oviductin in the rat ovulated oocytes.

CHAPTER 2.

IN SITU AND IN VITRO IMMUNOLOCALIZATION OF OVIDUCTIN

BINDING SITES ON HAMSTER UTERINE EPITHELIAL CELLS

34 1. ABSTRACT

Oviductin is a high-molecular-weight oviduct-specific glycoprotein which plays an important role in sperm capacitation, sperm-egg binding and early embryo development.

Previous studies have shown that hamster oviductin can gain access to the uterus and bind to a certain population of cells lining the uterine epithelium. Based on the hypothesis that oviductin receptors are present on the surface of some uterine epithelial cells and are therefore likely to be involved in regulating uterine receptivity, a primary hamster uterine epithelial cell culture system has been developed in our laboratory. This cell culture system can provide us with a useful in vitro model to study the role of oviductin in the uterus and an approach by which the putative oviductin receptor can be isolated and identified. The present study was undertaken to investigate the presence of oviductin binding sites in the uterine epithelial cells both in situ and in vitro. Immunohistochemical localization of oviductin in the hamster uterus revealed certain uterine cells reactive to the monoclonal antibody against hamster oviductin. Confocal microscopy of isolated uterine cells previously exposed to purified oviductin and then labeled with the anti-hamster oviductin antibody followed by anti-mouse IgG (Alexa Fluor®555) showed a similar labeling pattern. The epithelial nature of the cultured cells was confirmed by localization of cytokeratin and oviductin in the same cells. Pre-embedding immunolabeling of cultured uterine cells also showed gold particles associated with the plasma membrane and microvilli. Taken together, results in the present investigation demonstrated that hamster oviductin can bind to the plasma membrane of certain hamster uterine epithelial

35 cells, suggesting the presence of a putative oviductin receptor on the uterine epithelial cell surface, which might play a role in regulating uterine receptivity.

36 2. INTRODUCTION

In mammals, the lumen of the oviduct provides a suitable milieu for gamete transportation and maturation, fertilization, and early embryo development (Leese 1988).

The oviductal fluid within the oviduct is a complex mixture of constituents derived from the plasma and proteins synthesized and secreted by the oviductal epithelium (Leese

1988). Among these macromolecules secreted by the oviductal cells, a family of high-molecular-weight glycoproteins has been identified. These oviduct-specific glycoproteins, collectively known as oviductins, are synthesized and released into the oviductal lumen by the non-ciliated, secretory epithelial cells (Buhi 2002).

Oviductin has been identified and characterized in many mammalian species including the human (Buhi 2002). Except for the mouse, mammalian oviductins share one common characteristic; it becomes intimately associated with the zona pellucida of post-ovulatory oocytes during their transit through the oviduct. Oviductin has many important functions during the reproductive process including enhancement of sperm capacitation (King et al.

1994) and fertilization as well as early embryonic development (Buhi 2002). Hamster oviductin is a 160~350kDa glycoprotein (Malette and Bleau 1993). The site of localization of hamster oviductin in the female reproductive tract has been extensively studied in our laboratory and others. We have previously shown that hamster oviductin can gain access to the uterus and bind to a certain population of cells lining the uterine epithelium (Martoglio and Kan 1996). Thus, it is hypothesized that oviductin receptors reside on the surface of the endometrium and therefore may play a role in uterine receptivity. The present study was undertaken to further elucidate the presence of hamster

37 oviductin and its binding site in the uterus by use of L.M. immunohistochemistry and

E.M. pre-embedding immunocytochemistry. Primary cultures of hamster uterine cells were used in conjunction with pre-embedding immunolabeling technique and confocal microscopy to verify the site of localization of hamster oviductin on the cell surface of cultured uterine epithelial cells. The primary hamster uterine cell culture system serves as a useful model for us to study the potential role of hamster oviductin in uterine receptivity and provides a good approach by which the putative oviductin receptor can be isolated and identified.

38 3. HYPOTHESIS

A putative oviductin receptor exists on the surface of certain uterine epithelial cells in the hamster that may regulate uterine receptivity.

4. OBJECTIVES

1) To develop and optimize the conditions for culturing primary hamster uterine

epithelial cells in vitro.

2) To verify the site of localization of hamster oviductin on the cell surface of

cultured uterine epithelial cells using a variety of approaches including

immunohistochemistry, confocal microscopy and pre-embedding immunolabeling

technique.

39 5. MATERIALS AND METHODS

5.1 Animals

Mature female golden hamsters (Mesocricetus auratus) (Harlan Sprague Dawley,

Indiana, USA), approximately 7~8 weeks old, were used in the present study. The animals were housed in a constant-temperature room with 12-h light: 12-h dark intervals and allowed free access to food and water. The 4-day estrous cycle of hamster includes 5 stages: metestrus, diestrus 1, diestrus 2, proestrus and estrus. These stages are extrapolated based on the date of metestrus which is indicated by the presence of viscous vaginal discharge following ovulation (Lisk 1985). The animals were subsequently cycled for 2~3 weeks to ensure regularity. All animal care practices and experimental procedures complied with the guidelines of the Canadian Council on Animal Care and were approved by the University Animal Use Committee of Queen’s University.

5.2 Monoclonal Anti-Hamster Oviductin Antibody

In the present study, the mouse anti-hamster oviductin monoclonal 4.12F6 antibody was generously donated by Dr. Gilles Bleau (Department of Obstetrics and Gynecology,

University of Montreal and Maisonneuve-Rosemont Hospital Research Center). The antigen specificity of the monoclonal antibody was previously assessed by immunoprecipitaion, Western-blot and immunofluorescence (St Jacques and Bleau 1988).

40 5.3 Lectin-Affinity Purification of Oviductin from Hamster Oviducts

Twelve animals were separated into different cages according to the stages of the estrous cycle (Lisk 1985; Roux and Kan 1995). They were sacrificed by cervical dislocation on the day of proestrus. Their oviducts were dissected from the reproductive tract under a dissecting microscope and immediately stored in liquid nitrogen until use.

For lectin-affinity purification of oviductin, six or more oviducts were homogenized (6 oviducts/mL) in 50mM Tris/HCL, 150mM NaCl, containing 0.02% NaN3, 35µg/mL aprotinin, 0.2mM PMSF, 1µM leupeptin, 100µM EDTA and 0.2mM benzamidine

(homogenization buffer, pH8.0) (Malette and Bleau 1993). The total homogenized proteins were centrifuged at 105,000g in an ultracentrifuge (OptimaTMXL-100k, Beckman,

USA) for 1 hour at 4°C to separate insoluble contents (Malette and Bleau 1993). The supernatant containing 750µg of total protein was loaded and incubated for 30min with a

5mL Helix-pomatia agglutinin (HPA) conjugated agarose (Sigma; Oakville, ON, Canada) column which was equilibrated with 10 bed volume of homogenization buffer. Then the column was washed with 10 bed volume of washing buffer containing 50mM Tris/HCL,

300mM NaCl, 0.02% NaN3 (washing buffer, pH 8.0) to remove the unbound proteins

(Malette and Bleau 1993). Oviductin prepared from the proestrus stage was further eluted with 3 bed volume of the elution buffer, pH2.5, containing 200mM glycine/HCL, 150mM

NaCl, 0.02% NaN3, and 200nM α-D-N-acetyl-galactosamine (α-D-GalNAc). All fractions were neutralized and concentrated by using Amicon® Ultra Centrifugal Filter Device

(Millipore; Ottawa, ON, Canada) in the centrifuge (CS-15R, Beckman, USA) for 10 min at 4300g. The total purified protein samples were pooled and the protein concentration

41 was determined using the Bio-Rad Lowry Protein Assay (Bio-Rad; Mississauga, ON,

Canada).

5.4 Immunodetection of Purified Stage-Specific Hamster Oviductin

One microgram of purified hamster oviductin (proestrus stage) was loaded on 8% one-dimensional SDS-PAGE. The protein was separated in the gel under reducing conditions and then transferred to a PVDF membrane electrophoretically. The membrane was blocked in 5% non-fat milk blocking solution (in TBS-T) for 1 hour at room temperature and followed by incubation at 4°C overnight with monoclonal anti-hamster oviductin antibody (1:100 in blocking solution). After washing with TBS-T thrice, 10 min each, HRP-conjugated rabbit anti-mouse IgG (1:10000 in blocking solution) was applied to the membrane and incubated for one hour at room temperature. The membrane was then washed in TBS-T and immunodetected using the Western LightingTM

Chemiluminescence Reagent Plus (Perkin-Elmer; Waltham, MA, USA).

5.5 Immunohistochemical Labeling of Hamster Oviductin in the Female Reproductive

Tract

Sexually mature female hamsters (n=3) were randomly selected. The animals were sacrificed by cervical dislocation and the ventral abdominal wall was cut open. The female reproductive tracts including ovaries, uteri, and oviducts were excised and rinsed in 0.01M PBS, pH 7.4. For immunohistochemistry (IHC) studies, the uteri, ovaries, and oviducts were fixed overnight by immersion in 4% paraformaldehyde (in 0.1M phosphate

42 buffer) followed by paraffin embedding according to routine procedures. Five-micron thick sections were prepared and placed on microscope slides for labeling. The sections were first deparaffinized in Hemo-D and rehydrated in decreasing concentrations of ethanol before being rinsed in 0.01M PBS. The sections were then incubated with 1% ovalbumin in 0.01M PBS for 20 min at room temperature to block the non-specific binding sites of the primary antibody. The excess ovalbumin was removed after 20 min and the anti-hamster oviductin monoclonal antibody (1:4 dilution) was applied to the sections for 2 hours at room temperature. The sections were then washed thrice in 0.01M

PBS. Endogenous peroxidase activity was removed by incubating sections in 1% H2O2 in double distilled water (ddH2O) for 1 hour. Subsequently, sections were washed with

0.01M PBS, and incubated with horseradish peroxidase (HRP)-conjugated rabbit anti-mouse IgG antibody (1:200 dilution) for 20 min at room temperature. The immunoperoxidase reaction (brown colour) was developed by incubation of the sections in 3,3’-diaminobenzidine tetrahydrochloride (DAB) substrate chromogen solution

(Zymed; South San Francisco, CA, USA) for 2.5 min. The sections were counterstained with hematoxylin (Zymed; South San Francisco, CA, USA), mounted with

Histomount-mounting solution (Zymed; South San Francisco, CA, USA) and were examined under a light microscope.

5.6 Primary Hamster Uterine Epithelial Cell Culture

Six female golden hamsters at proestrus stage were sacrificed by cervical dislocation for each experiment. Uteri were excised above the cervix and below the oviducts, stripped

43 of excess connective tissue and placed in sterilized Hank’s Buffered Saline Solution

(HBSS) (GIBCO; Burlington, ON, Canada) containing 10% penicillin (100I.U./mL) and streptomycin (100µg /mL) (GIBCO; Burlington, ON, Canada) on ice. Under the sterile condition, the exterior of the uterus was washed with HBSS containing 10% penicillin and streptomycin (100µg /mL) and the interior was washed by injecting HBSS from one end of the uterine lumen to the other. After one end was tied with sterilized thread, HBSS containing 2.5mg/mL Collagenase Type I (Sigma; Oakville, ON, Canada) was gently injected into the lumen with the sample of uterus held vertically. The upper end was immediately tightened to keep the collagenase solution inside the lumen. Then the uterus was placed in HBSS in a 15mL polyethylene conical tube kept in a water bath for 45 min at 37°C to dissociate the uterine epithelial cells. After incubation, both ends of the tied uteri were cut and the lumen was washed by injecting HBSS with 10% penicillin and streptomycin to flush out the dissociated epithelial cells. The washing HBSS was collected into a 15 mL polyethylene conical tube and centrifuged at 400g for 10 min in a centrifuge (IEC Centra® MP4R). The pellets containing epithelial cells and stromal cells were washed thrice with 6~8mL HBSS. The pellets were then resuspended in 10mL culture medium CMRL 1066 (Sigma; Oakville, ON, Canada) which was supplemented with 10mM L-glutamine, 5g/L glucose, 5mg/L insulin, 0.85g/L NaHCO3, 20mM Hepes,

10-9 M oestradiol-17β, 15ng/mL epidermal growth factor (EGF), 10% fetal bovine serum

(FBS) and 100 I.U./mL penicillin-100µg/mL streptomycin. After being centrifuged at

400g, the supernatant was discarded and the pellet was resuspended in CMRL 1066 medium as described above and then transferred to a sterilized, pre-coated T-25 culture

44 flask. The flask was incubated at 37°C to allow stromal cells to attach to the bottom.

After 40~50 min, the top layer of the medium in the flask containing mainly uterine epithelial cells was transferred and further processed for subsequent experiments: 1) For detection of cytokeratin of cultured cells, the medium was transferred to a 24-well plate

(1mL/well); 2) For immunolabeling to be examined by confocal microscopy, the medium was transfered to a Collagen Type I 8-well culture slide (1mL/well) (BD BiocoatTM;

Mississauga, ON, Canada); 3) For morphological examination and immunocytochemistry, the epithelial cells were transferred to another 3 flasks (approximately 3mL/flask) and

CMRL 1066 culture medium was added to a total of 12mL/flask. All cells were maintained under the standard culture environment (37°C and 5% CO2). Cells were examined daily and the medium was replaced with fresh medium every two days. The cells were harvested at confluence.

5.7 Immunoperoxidase Labeling of Cytokeratin in Cultured Primary Uterine Epithelial

Cells

The 24-well plate was rinsed with 0.01M PBS once and the cells were fixed with

100% methanol at -20℃ for 30min. Then the cells were treated, in a sequential manner, with 10%, 20%, 40%, 60%, 80% ethanol in 100mM phosphate buffer (pH 7.4), five minutes each followed by rinsing with 100mM phosphate buffer and 100% ethanol. The plate was left to-dried at room temperature. The sections were then incubated with 10% normal goat serum (NGS in 0.01M PBS) for 10 min at room temperature to block the non-specific binding sites. The rabbit anti-cytokeratin polyclonal antibody (DAKO;

45 Mississauga, ON, Canada) was applied to the 24-well plate at 1:500 dilution for 1 hour at room temperature. The plate was then washed thoroughly in 0.01M PBS followed by incubation with biotinylated goat anti-rabbit antibody (Inter Medico; Markham, ON,

Canada) in 1:100 dilution for 1 hour at room temperature. Endogenous peroxidase activity was removed by incubating sections in 1% H2O2 in ddH2O for 1 hour.

Subsequently, the plate was washed with 0.01M PBS, and incubated with streptavidin-enzyme conjugate (Inter Medico; Markham, ON, Canada) at 1:300 dilution for 30 min at room temperature. The immunoperoxidase reaction was developed by incubation of the plate in DAB substrate chromogen solution (Zymed; South San

Francisco, CA, USA) for 2.5 min. The sections were counterstained with methylene blue

(Zymed; South San Francisco, CA, USA) and further processed according to routine procedure. Finally, the cells were examined under an inverted light microscope. Mouse embryonic fibroblasts (3T3 cells) applied with the same anti-cytokeratin polyclonal antibody were used as a negative control.

5.8 Double Immunofluorescence Labeling in Cultured Primary Uterine Eipthelial Cells

For confocal microscopy, confluent uterine epithelial cells in the 8-well culture slide were incubated with purified hamster oviductin (10µg/mL in culture medium) for 1 hour at 37°C. After three washes with PBS, the cells were fixed in 4% paraformaldehyde in

0.1M PBS for 10 min at room temperature. The cells were washed with PBS for 3 min and then permeabilized with 0.2% Triton X-100 on ice for 5 min. The cells were rinsed with PBS and left in 10% NGS in PBS for 20 min before they were labeled with a

46 mixture of mouse anti-hamster oviductin monoclonal antibody (1:10) and rabbit anti-cytokeratin polyclonal antibody (1:500) (DAKO; Mississauga, ON, Canada) for 1.5 hour at room temperature. The cells were washed with TBS-T. After washing, the following steps were carried out in the darkroom. The cells were incubated with a mixture of diluted fluorescence conjugated secondary antibodies: Alexa Fluor®555 goat anti-mouse IgG (H+L) (10µg/mL, Molecular Probes; Burlington, ON, Canada) and Alexa

Fluor®488 goat anti-rabbit IgG (H+L) (10µg/mL, Molecular Probes; Burlington, ON,

Canada) for 50 min at room temperature. The nuclei were stained by DAPI (1:1000) for

10 min. After washing with PBS, the chambers of the 8-well slide were removed according to the manufacturer’s instruction. The slides were mounted by fluorescence mounting medium (DAKO; Mississauga, ON, Canada) and then examined under a Leica

TCS SP2 confocal microscope. In control sections, PBS was used in place of the purified hamster oviductin.

5.9 Preparation of Cultured Cells for Morphological Analysis under the Electron

Microscope

Cell fixation: Cultured hamster uterine epithelial cells were collected by gently scraping the bottoms of the flasks with a rubber cell scraper (Sigma; Okaville, ON,

Canada), transferred to a 15-mL polyethylene conical tube and washed by HBSS once.

The cells were then fixed by immersion in 2% glutaraldehyde in 0.1M PBS (pH 7.2) for 2 hours at room temperature. The sample was centrifuged at 400g and washed thrice with

0.1M PBS, 10 min each.

47 Osmication: The fixed cells were post-fixed by resuspension in 1% osmium tetroxide in ddH2O for 1 hour at room temperature. The cells were then centrifuged at 400 g and washed with HBSS thrice.

Dehydration: The dehydration procedure was performed by centrifugation and resuspension of the cells in a series of increasing concentrations of acetone: 30% (2 changes, 5 min each), 50% (2 changes, 5 min each), 70% (2 changes, 5 min each), 80% (2 changes, 5 min each), 95% (2 changes, 5 min each) and 100% (3 changes, 5 min each) at room temperature.

Epon infiltration: After dehydration, the cells were transferred to beam capsules and were ready to be infiltrated with bisphenol A/epichlorohydrin based epoxy resin (Epon)

(CANEMCO Inc; Montreal, QC, Canada). Cells were infiltrated with Epon/Acetone (1:1) for 1 hour with caps on, in Epon/Acetone (2:1) for 1 hour with caps on, and in

Epon/Acetone (3:1) overnight with caps off on a rotator.

Epon embedding: Epon/Acetone (3:1) was replaced with pure Epon and the pure

Epon was de-aerated for 4 hours at room temperature with the capsule lid off. After the cells were centrifuged to form pellets at the bottom of the beam capsules, Epon embedded cells in the beam capsules were incubated at 60°C in a Precision Thelco Model oven for

48 hours with the capsule lids on. Once the Epon was polymerized, one-micron thick sections were prepared to locate the cells. Ultra-thin sections of the selected regions were then prepared, counterstained, and examined under a Hitachi 7000 electron microscope operated at 75 kV.

48 5.10 Pre-Embedding Immunocytochemical Labeling of Oviductin on Cultured

Primary Hamster Uterine Epithelial Cells

Cultured cells were dissociated in the flask by incubation with 3mL 0.5X trypsin-EDTA (Invitrogen; Burlington, ON, Canada) at room temperature for 2~3 min.

Under an inverted microscope, as soon as the cells were observed to become round and start to float, equal volume of fresh culture medium with FBS was added to the flask in order to stop the activity of the trypsin-EDTA. The content in each flask was transferred to a 15mL polyethylene conical tube and centrifuged at 400 g for 10 min. The cell pellet was washed with culture medium once and then resuspended in culture medium containing 10µg/mL purified hamster oviductin and kept at 37°C for 45 min. The cells were then washed with 0.01M PBS twice and fixed with a fixative solution (3% paraformaldehyde and 0.1% glutaradehyde in 0.1M cacodylate buffer) for 45 min at room temperature.

After fixation, the cells were washed thrice with 0.01M PBS, followed by incubation in 1% ovalbumin for 10 min at room temperature. The cells were then labeled by incubation in anti-hamster oviductin monoclonal antibody (1:5 dilution in 0.01M PBS) at

4°C overnight. After labeling, the cells were washed with 0.01M PBS twice, incubated with 1% ovalbumin for 10 min followed by incubation with protein A-gold complex

(11nm, 1:3 dilution in 0.01M PBS) for 3 hours at room temperature. The cells were washed with 0.01M PBS twice after the protein A-gold incubation and kept in 2% glutaradehyde in 0.01M PBS at 4°C. The cells were post-fixed by resuspension in 1% osmium tetroxide in ddH2O and then embedded in Epon according to routine procedures.

49 6. RESULTS

6.1 Lectin-Affinity Purification of Hamster Oviductin

Recent studies carried out in our laboratory showed that the glycosylation of hamster oviductin is differentially regulated during the estrous cycle whereas the hamster oviductin mRNA and protein expressions remain constant through the estrous cycle (McBride et al.

2004). Hamster oviductin becomes fully glycosylated at the proestrus and estrus stages suggesting that full glycosylation of oviductin may be necessary for the glycoprotein to exert its full function during the early process of reproduction (McBride et al. 2005).

Oviductin prepared from the proestrus stage was therefore chosen for the present study.

The oviductin purified from the proestrus stage was used throughout the course of the present investigation also because larger amount of oviductin was obtained from this stage compared to that purified from the estrus stage.

Due to the presence of terminal α-D-GalNAc residues on the side chains of hamster oviductin, a single-step lectin-affinity purification method was successfully applied to obtain sufficient oviductin for the present study. Using the monoclonal antibody specific for hamster oviductin, a high-molecular-weight, polydispersed band ranging from

160~350kDa was detected in oviductin sample obtained from the proestrus stage (Fig. 7).

250 kDa

150 kDa

100 kDa

75 kDa

Figure 7. Western blot analysis examining the lectin-affinity purified oviductin from the proestrus stage. One microgram of purified hamster oviductin from the proestrus stage was analyzed by electrophoresis on a 8% polyacrylamide gel under reducing condition followed by incubation with anti-hamster oviductin mAb. The mAb recognizes a high-molecular-weight polydispersed band with a molecular weight ranging from 160 kDa to 350 kDa.

51 6.2 Immunohistochemical Labeling of Hamster Oviductin in the Female Reproductive

Tract

The monoclonal anti-hamster oviductin antibody was applied to analyze the distribution of hamster oviductin in the female reproductive tract. As revealed by post-embedding immunohistochemical experiments, at low magnification, a strong immunoreaction was seen along the oviductal epithelium. Intense immunolabeling occurred in the isthmus (Fig. 8A) and ampulla region of the hamster oviduct. At higher magnifications, the immunoreaction was detected in both cytoplasm and apical region of the non-ciliated secretory cells. A weak immunoreaction was also detected at the cell surface of the uterine epithelium (Figs. 8B and 8C). Immunostaining was absent in the ovary (Fig. 8D). No immunoreaction was detected in the control sections onto which PBS was applied instead of anti-hamster oviductin monoclonal antibody (Fig. 8E). From the observations above, in the female hamster reproductive tract, it appears that oviductin is secreted by non-ciliated secretory cells of the oviduct and trace of this glycoprotein is also present at the cell surface of the epithelium lining the uterus.

Figure 8. Immunohistochemical localization of oviductin in the female hamster reproductive tract using monoclonal anti-hamster oviductin antibody. (A) Immunolabeling of oviductin in the oviduct using anti-hamster oviductin monoclonal antibody displays a heterogeneous labeling pattern over the epithelium, especially in the isthmus of the oviduct (arrows). (B) In the uterus, immunoreaction is detected over the cell surface of some uterine epithelial cells (arrows). (C) Higher magnification micrgraph of boxed area in Fig. 8B shows immunoreaction on the surface of some uterine epithelial cells (arrows). (D) Immunoreaction is not detected within the ovary using the monoclonal anti-hamster oviductin antibody. (E) Control section shows the absence of immunoreaction in the oviductal epithelial cells. (Lu: lumen; Ep: epithelium; St: stroma) Bar = 50 μm

53 A B

C Ep

St Ep Lu

D E

54 6.3 Primary Cell Culture of Hamster Uterine Epithelial Cells

In the hope of elucidating the function of oviductin in the uterus, a method for culturing primary hamster uterine epithelial cells has been successfully established in this study. Under the light microscope, cultured hamster uterine epithelial cells displayed a polygonal shape and tended to grow in clusters (Figs. 9A, 9B and 9C) whereas the stromal cells appeared more elongated in shape and much smaller in size (Fig 9D). In each cell culture experiment, it was observed that approximately 80~85% of the cells at confluence were uterine epithelial cells with the rest made up of stromal cells. From our experience, uterine epithelial cells required passaging every 5~7 days whereas growth of stromal cells was more rapid and required passaging every 1~3 days. Both cell types senescenced completely after 7~10 passages.

Ep A B

Str Ep

C D

Figure 9. Light micrographs of primary cultured hamster uterine epithelial cells. (A) The primary uterine epithelial cells display polygonal shape (arrows) and tend to grow in clusters (48-hour culture). (B) A high magnification micrograph shows the typical uterine epithelial cells (Ep) with several dispersed stromal cells (Str) (48-hour culture).(C) A light micrograph shows a group of uterine epithelial cells after 72 hours in culture. (D) Most cells within the field are stromal cells which appear more elongated in shape and much smaller in size (48-hour culture). Bar = 50 μm.

56 6.4 Detection of Cytokeratin in the Cultured Hamster Uterine Epithelial Cells

To further ensure the epithelial nature of the cultured primary uterine cells, an immnuoperoxidase labeling was performed using a polyclonal rabbit anti-cytokeratin antibody. As expected, an intense dark-brownish immunoreaction was detected in most of the cells which displayed polygonal shape under the light microscope (Fig. 10A) confirming the epithelial nature of majority (i.e. 80~85%) of the cultured cells.

Immunolabeling was not detected in the fibroblast cells used as a negative control (Fig.

10B).

A B

Figure 10. Immunodetection of cytokeratin in the cultured hamster uterine epithelial cells. (A) Immunodetection for cytokeratin in the cultured cells using anti-cytokeratin polyclonal antibody results in an intense immunoreaction (brown colour) in most of the cells, confirming the epithelial nature of majority of the cells. (B) Fibroblast cells are used as negative control. Immunolabeling for the cytokeratin is absent. Bar = 50 μm

57 6.5 Double Immunofluorescence Labeling of Oviductin and Cytokeratin in Cultured

Primary Uterine Epithelial Cells

We reasoned that if a putative oviductin receptor is present on the cell surface of some uterine epithelial cells, purified oviductin should bind to their cell surface and these presumptive oviductin-binding sites can be detected by use of a monoclonal anti-oviductin antibody. To verify our prediction, a double immunofluorescence labeling study to detect the presence of oviductin binding sites on uterine epithelial cells in vitro was conducted.

The cultured cells were exposed to medium containing purified oviductin (10 μg/mL) for 1 hour at 37°C. After being fixed, the cells were labeled with monoclonal anti-hamster oviductin antibody. As shown in Figs. 11A-E, staining for cytokeratin (green) was intense throughout the cytoplasm except for the nucleus which was labeled by DAPI.

Immunolabeling for hamster oviductin (red) was mainly restricted to the cell in the peripheral areas of clusters (Figs. 11C and 11D). The labeling pattern of oviductin was different from that of cytokeratin. An x-z section obtained by confocal microscopy confirmed that oviductin only bound to the surface of the uterine epithelial cells (Fig. 11E).

In the two control groups (Figs. 11A and 11B), immunostaining for oviductin was absent from the cultured epithelial cells. Thus, results obtained with confocal microscopy confirmed not only the nature of the cultured uterine epithelial cells, but also the in vitro localization of oviductin binding sites on the uterine epithelial cell surfaces.

Figure 11. Confocal microscopy revealing the association of oviductin with cultured hamster uterine epithelial cells. In all groups (A-E), polyclonal anti-cytokeratin antibody was applied to the cultured cells and the cells show the green immunofluorescence labeling in the cytoplasma, confirming the epithelial nature. The cell nuclei are labeled by DAPI (blue). (A) The primary cultured cells, without prior incubation with purified oviductin and the monoclonal anti-hamster oviductin antibody, were incubated in Alexa Fluorescence®555 goat anti-mouse IgG alone. No immunoreaction for oviductin is detected. (B) The primary cultured cells were not incubated with purified hamster oviductin (10µg/mL) followed by incubation in the monoclonal anti-hamster oviductin antibody and Alexa Fluorescence®555 goat anti-mouse IgG. No immunoreaction for oviductin is detected. (C-E) The primary cultured cells were incubated with purified hamster oviductin (10µg/mL) followed by incubation with the anti-hamster oviductin antibody and Alexa Fluorescence®555 goat anti-mouse IgG; (C and D) The pripheral area of the cultured cells is labeled indicating that oviductin bind to the cell surface. (E) An *x-z section shows that oviductin only binds to the free surface of the uterine epithelial cell.

* An x-z section is produced by scanning a single line across the specimen (the x-axis) at different depths (z-axis) by stepper motor control of focus changes, then displaying the series as a merged image.

59 A B

Nucleus Nucleus

Cytokeratin Cytokeratin

Oviductin Oviductin

Merged Merged

60 C D E

Nucleus Nucleus Nucleus

Cytokeratin Cytokeratin Cytokeratin

Oviductin Oviductin Oviductin

bottom

top

Merged Merged Merged 40μm

61 6.6 Ultrastructural Morphology of Cultured Hamster Primary Uterine Epithelial Cells and

Stromal Cells

It became apparent that it was critical to assess the ultrastructural morphology of both uterine epithelial and stromal cells before we proceeded to localize oviductin on the cultured cells. Transmission electron micrographs of cultured primary cells after being fixed with glutaraldehyde showed that uterine epithelial cells displayed cellular polarity with the apical cell surface endowed with many microvilli (Fig. 12A). In the cytoplasm, well-developed Golgi complexes and numerous mitochondria could be easily seen. On the other hand, the uterine stromal cells were much smaller, more slender and elongated. The stromal cells contained a very prominent nucleus that occupies the majority of the cytoplasm within the cell. More importantly, the stromal cells did not have microvilli on their surface (Fig. 12B), a histological feature that can be used to distinguish the uterine epithelial cells from the stromal cells.

Figure 12. Transmission electron micrographs of cultured primary hamster uterine epithelial cells and stromal cells. (A) Without trypsinization, epithelial cells (X 5,000) display polarity with the apical cell surface endowed with many microvilli (arrows) while (B) stromal cells (X 2,500) appear smaller and more elongated without any microvilli.

63 6.7 Pre-Embedding Immunocytochemical Labeling of Oviductin on Cultured Hamster

Primary Uterine Epithelial Cells

To precisely locate oviductin binding sites in the cultured uterine epithelial cells in vitro, a pre-embedding immunocytochemical study at the electron microscopic level was carried out on cultured cells that had been pre-incubated with oviductin. Pre-embedding immunolabeling results showed that gold particles were detected over the apical surface of the cultured hamster uterine epithelial cells, especially in association with the microvilli

(Figs. 13A-C). Interestingly, some gold particles were observed in the proximity of coated pits of some epithelial cells (Fig. 13D), indicating that oviductin might be in the process of being internalized by some uterine epithelial cells. The content of the Golgi saccules and the rest of the organelles were devoid of labeling. This is expected since the pre-embedding approach is essentially a cell surface labeling technique. Immunolabeling was absent in control sections labeled only with protein A-gold complex.

Figure 13. Electron micrographs of Epon sections of cultured hamster uterine epithelial cells previously incubated with purified oviductin followed by labeling with monoclonal antibody. (A) Specific immunolabeling is detected over the apical cell surface of the hamster uterine epithelial cell and over the microvilli (arrows) projecting from the surface membrane (X 27,000). (B) Gold particles (arrows) are associated with the microvilli of an uterine epithelial cell (X 41,000). (C) A high magnification micrograph shows the immunogold-labeled microvilli (arrows) (X 60,000). (D) Gold particles were observed to be associated with coated pits (arrows) (X 34,000).

65 A

66 C

67 7. DISCUSSION

Our in situ immunohistochemical results clearly showed that oviductin is present in the non-ciliated secretory oviductal epithelial cells and on the surface of some uterine epithelial cells. It is known that, for most mammalian species studied, the oviductal epithelium consists of two types of cells: ciliated and non-ciliated secretory columnar epithelial cells. The above immunohistochemical labeling study showing the presence of oviductin in the non-ciliated secretory cells but not in the ciliated cells of the oviductal epithelium indicated that hamster oviductin is originally produced and secreted by the non-ciliated oviductal epithelial cells.

At the electron microscopical level, immunogold labeling was associated with the plasma membrane and microvilli of some uterine epithelial cells. Immunogold particles were also associated with what appeared to be coated pits along the membrane of uterine epithelial cells that were reactive to the anti-hamster oviductin antibody. These results are consistent with the findings previously reported (Martoglio and Kan 1996; Roux et al.

1997). Interestingly, Martoglio and Kan (1996) observed a heterogeneous labeling pattern of hamster oviductin over the uterine epithelium during the gestation period by using immunohistochemistry in conjunction with silver enhancement technique. In their study, hamster uterine luminal epithelial cells exhibited a high intensity immunolabeling throughout the estrous cycle and during day 1-3 of gestation. However, a marked reduction in staining intensity was found at day 5 and 6 when attachment of the blastocyst to the uterine wall normally occurs (Martoglio and Kan 1996). Studies were also conducted by

Martoglio and Kan (1996) to detect hamster oviductin in the oviductal epithelium during

68 early gestation. Results from the latter study showed an intense and consistent immunolabeling of oviductin from day 1 to day 6 of gestation which indicated that the hamster oviducts continuously synthesize and secret oviductin throughout early gestation.

Thus, the decreased labeling of the oviductin observed in the uterus cannot be due to the decreased secretion of the oviductin in the oviduct. Taken together, the present results and findings from previous study indicate the presence of a putative oviductin receptor on some of the uterine epithelial cells that may play a role in modulating uterine receptivity.

Implantation is a critical and complex process by which the blastocyst comes to make an intimate physical and physiological contact with the uterine endometrium.

However, due to ethical and practical reasons, the implantation process itself has been hardly observed directly in vivo in humans (Lee and DeMayo 2004). Therefore, animal models become the major resources that investigators can use to gain insight into the human implantation process. In the present study, we have successfully established and optimized a method for culturing hamster uterine epithelial cells in the hope of using the culture system as an in vitro model to study the interaction between oviductin and the uterine epithelium. A preliminary study carried out in our laboratory with Western blot analysis using a polyclonal anti-cytokeratin antibody on hamster uterine tissue from each of the five stages of the estrous cycle showed that the most intense immunoreaction was obtained from the uterine tissue at the proestrus stage (data not shown). This result was taken as an indication that epithelial cells of the hamster endometrium at the proestrus and estrus stages are the most robust. Therefore, in the present study, female hamsters at the proestrus stage were used for isolating and culturing uterine epithelial cells.

69 Our isolation method of hamster uterine epithelial cells was primarily based on procedures developed by Chaminadas et al. (1986) with minor modifications. We found the type of the collagenase used to dissociate uterine epithelial cells and the incubation time of the uterine horns with the collagenase are the main factors for obtaining a high yield of uterine epithelial cells. When freshly made collagenase (prepared on the same day of the experiment) was used for incubation of the uterine tissue at 37°C in a water bath, the resulting cultured cells appeared to be more contaminated by the presence of stromal cells. However, when the incubation time with the freshly made collagenase was reduced, this resulted in a lower yield of uterine epithelial cells, leading to insufficient seeding of cells for cell culture. Thus, after several trials, the condition was optimized by preparing a 0.25% collagenase solution and storing it at -20°C for 24 hours prior to use.

The cells were then isolated by incubation of the collagenase at 37°C in a water bath for

45 min.

In the present study, in order to separate epithelial cells and stromal cells more efficiently, a separation technique, pre-culturing the collected cells in the incubator (5%

CO2, 37°C) for 40 min, was applied. Taking advantage of the higher attachment rate of stromal cells to the bottom matrix, the top layer of the medium mainly containing uterine epithelial cells can then be transferred to a new flask for subsequent processing. Results showed that the percentage of the epithelial cells was increased to 85%. Compared to other methods used for separating epithelial cells and stromal cells such as membrane filtration and sedimentation, this pre-culturing method is very convenient and efficient.

70 In our cell culture system, a small number of stromal cells were found growing in the midst of uterine epithelial cells. Using a polyclonal anti-cytokeratin antibody in conjunction with either immunohistochemical or immunofluorescent techniques, we were able to distinguish epithelial cells from stromal cells and estimated that 80%~85% of the total number of cells present in our culture system were epithelial in nature. It has been widely accepted that stromal cells play a regulatory role for the overlying epithelium to grow and differentiate in the female reproductive tract (Donjacour and Cunha 1991;

Cunha et al. 1985). We did notice that with less stromal cells in the cultured cell population, the primary uterine epithelial cells grew slower and it took more time for the cells to become confluent in the culture flasks. Thus, in our culture system, the presence of a small number of stromal cells appeared to be advantageous to the culture environment.

Considering that oviductin receptor may function in mediating uterine receptivity, it would be important to detect its presence and localize it on the surface of uterine epithelial cells. The present study was limited by the unavailability of an antibody against the hamster oviductin receptor since the existence of the latter is still being sought.

Consequently, a different approach was attempted to verify the presence of oviductin-binding sites on the surface of uterine epithelial cells. A previous post-embedding immunocytochemical labeling study carried out in our laboratory showed that oviductin was associated with the apical plasma membrane and microvilli of some uterine epithelial cells. Immunolabeling was also found in the multivesicular bodies and over the limiting membrane of apical vesicles. With the successful establishment of a

71 uterine cell culture system, it became important to show also the presence of oviductin at the cell surface of primary cultured epithelial cells similar to what had been observed in situ. Therefore, both confocal microscopy and pre-embedding immunolabeling approach were used in the present investigation. Confocal microscopy of uterine epithelial cells previously incubated with purified oviductin prior to labeling with anti-hamster oviductin antibody revealed an intense immunoreaction on their cell surface. Similar to what was observed over the cell surface of some uterine epithelial cells in situ, the present in vitro study also showed that, among the uterine epithelial cells incubated with purified oviductin and labeled by its corresponding antibody, only some cells exhibited immunoreactivity.

Our pre-embedding immunolabeling results showed that oviductin was only present on the apical plasma membrane of the cultured uterine epithelial cells and associated with the surface microvilli. Moreover, in some epithelial cells, some coated pits were found labeled with gold particles indicating that oviductin was being internalized in some uterine epithelial cells possibly via receptor-mediated endocytosis. A confocal microscopy study of isolated, cultured uterine cells previously exposed to purified oviductin followed by labeling with the monoclonal antibody against hamster oviductin and anti-cytokeration antibody not only confirmed the epithelial nature of the cultured uterine cells but also revealed the oviductin binding sites on the cell surface of the labeled epithelial cells, confirming the results obtained with both light microscopic and electron microscopic immunocytochemistry.

It is apparent that our in vitro oviductin localization studies demonstrated a labeling

72 pattern similar to what was obtained in situ, and also this cell culture system exhibits morphological and immunohistochemical characteristics corresponding to those of the in vivo situation. Therefore, it represents a valuable tool for investigating the functions of oviductin and its receptor at the cell surface of uterine epithelium lining the endometrium.

The next step would be to immortalize these hamster uterine epithelial cells for further studies. If the immortalized cell line is proven to possess the same morphological and biochemical characteristics of the primary uterine cells, it will be a very useful in vitro system for studying the oviductin receptor and other macromolecules involved in the uterus-trophoblast interaction.

The present in vitro localization studies of hamster oviductin on the cultured uterine cells, incorporating results from previous biochemical and radioautographic studies, support the proposed hypothesis of the presence of a hamster oviductin receptor on the luminal surface of the uterine epithelial cells. Previous studies conducted by Martoglio and Kan (1997) revealed a decreasing immunofluorescence labeling of hamster oviductin on the cell surface of the endometrium during early gestation and the labeling disappeared after day 6 of gestation during which time the “window of implantation” of the uterus occurs. The functional significance of binding of oviductin to its receptor in the endometrium remains to be elucidated.

In the future, studies will be conducted to identify and characterize the potential receptor for hamster oviductin on the uterine epithelial surface. A preliminary far Western study conducted in our laboratory showed a putative hamster oviductin binding protein on the uterine membrane with a molecular weight of approximately 57kDa. Once the

73 sequence of oviductin receptor is identified, the cDNA can be cloned. A vector containing oviductin receptor cDNA can be further transfected into E. coli. The expressed protein will be very useful for further functional characterization and studies.

CHAPTER 3

DETECTION OF A HAMSTER OVIDUCTIN HOMOLOGUE IN THE

FEMALE RAT REPRODUCTIVE TRACT

75 1. ABSTRACT

In the present study, using the monoclonal anti-hamster oviductin antibody that cross-reacts with the rat tissue, experiments were conducted to detect and localize oviductin in the female rat reproductive system. Western blot analysis detected the presence of oviductin in the rat oviduct with a molecular weight in the range of

180~300kDa. Immunohistochemical labeling of the female rat reproductive tract revealed localization of oviductin in non-ciliated secretory oviductal epithelial cells and a discontinuous immunoreaction at the surface of the uterine epithelium. Immunolabeling was not found in the ovary. Under the electron microscope, immunogold labeling was found to be restricted to the secretory granules and microvilli of non-ciliated secretory cells of the oviduct. In tissue sections of LR-White embedded uterus, gold particles were found to be associated with some non-ciliated epithelial cells, indicating immunoreaction for oviductin is distributed unevenly along the uterine epithelial lining. Electron micrographs of ultra-thin LR-White sections of superovulated rat oviductal oocytes labeled by the monoclonal anti-hamster oviductin antibody demonstrated an intense immunolabeling within the perivitelline space surrouding the oocyte. The findings of the present study demonstrated that an oviduct-specific glycoprotein is present in the female rat reproductive tract and that the glycoprotein is specifically associated with flocculent material found in the perivitelline space of oocytes in passage through the oviduct. The function of the rat oviduct-specific glycoprotein remains to be elucidated.

76 2. INTRODUCTION

Oviductin is a collective term for a family of glycoproteins that are synthesized by the mammalian oviduct and released into the oviductal lumen and later become associated with the ovulated oocytes and early embryos (Malette and Bleau 1993). As mentioned earlier, oviductin has been identified and characterized in many species, including the baboon (Verhage et al. 1989; Jaffe et al. 1996), bull (Boice et al. 1990;

Sendai et al. 1994), goat (Abe et al. 1995), hamster (Araki et al. 1987; Kan et al. 1988;

Kan et al. 1990; Leveille et al. 1987; Oikawa et al. 1988; Suzuki et al. 1995), mouse

(Kapur and Johnson 1985; Sendai et al. 1995), monkey (Verhage et al. 1997b), pig

(Brown and Cheng 1986; Buhi et al. 1990; Buhi et al. 1996), rabbit (Oliphant and Ross

1982), sheep (Sutton et al. 1984; Gandolfi et al. 1991; DeSouza and Murray 1995) and human (Arias et al. 1994; O'Day-Bowman et al. 1995; O'Day-Bowman et al. 1996;

Verhage et al. 1988). However, no extensive studies have been conducted on oviductin in the rat species and oviductin has not been identified and characterized in the rat. Thus, it is of interest and worthwhile to examine if oviductin exists in the rat species, and if it is, to find out whether it has the same characteristics shared by other oviductins reported in literature. As a result of an unexpected preliminary observation that the monoclonal anti-hamster oviductin antibody cross-reacts with the rat oviductal tissue, we hypothesized that a hamster oviductin protein homologue exists in the rat. In the present study, experiments were conducted to detect the presence of oviductin in the female rat reproductive tract and the ovulated oocytes and to verify its specific sites of localization.

Results of the present investigation demonstrated that the immunohistochemical labeling

77 pattern of the rat oviduct-specific glycoprotein in the oviduct and uterus is very similar to what was observed in the hamster whereas labeling of oviductal oocytes resembles what was previously reported in the mouse. In the present study, an attempt was also made, for the first time, to clone and sequence the oviduct-specific glycoprotein in the rat which has not been previously reported.

78 3. HYPOTHESIS

A hamster oviductin homologue is present in the oviduct and uterus of the female rat reproductive tract and is associated with specific structures of ovulated oocytes.

4. OBJECTIVES

1) To test for the presence of oviductin in the rat oviduct, uterus and ovary using

Western blot analysis and/or immunohistochemistry.

2) To identify the cellular localization of a hamster oviductin homologue in the rat

reproductive tract and ovulated oocytes using electron microscopic

immunocytochemistry.

3) To clone and sequence the hamster oviductin homologue in the rat.

79 5. MATERIALS AND METHODS

5.1 Animals

Mature female Sprague Dawley rats (Harlan Sprague Dawley, Indiana, USA), approximately 250g weight, were used in the present study. They were housed and allowed free access to food and water in the Animal Care Facility at Queen’s University prior to the experiments. All animal care practices and experimental protocol followed the guidelines of the Canadian Council on Animal Care and were approved by the Animal

Use Committee of Queen’s University.

5.2 Monoclonal Anti-Hamster Oviductin Antibody

See details in Materials and Methods in Chapter 2.

5.3 Western Blot Analysis of Rat Oviduct-Specific Glycoprotein in the Female Rat

Reproductive Tract

Two animals were exposed to sodium pentobarbital prior to being sacrificed by decapitation. The ovaries, oviducts, and uteri were retrieved immediately and stored in liquid nitrogen until use. To homogenize the samples of ovary, oviduct and uterus, tissue samples were submerged (2 ovaries/mL; 4 oviducts/0.67mL; 1 uterus/mL) in homogenization buffer (50mM Tris, 150mM NaCl, 0.02% NaN3, 35µg/mL Aprotinin,

0.2mM PMSF, 1µM Leupeptin, 100µM EDTA, 0.2mM Benzamidine). The samples were centrifuged at 400g for 15 min at 4°C. The supernatant was collected and the total protein

80 concentration was determined using the Bio-Rad Lowry Protein Assay (Bio-Rad;

Mississauga, ON, Canada). The supernatant containing 60µg total protein each from the ovary, oviduct and uterus tissue was separated by 8% SDS-PAGE under reducing conditions. The separated proteins were transferred to a PVDF membrane at 200mA for 2 hours. When transfer was completed, the membrane was blocked with 5% non-fat milk in

TBS-T for 30 min, followed by incubation with monoclonal anti-hamster oviductin antibody (1:100 in blocking solution) at 4°C overnight. The membrane was washed in

TBS-T thrice, 10 min each before it was incubated with the polyclonal rabbit anti-mouse antibody (1:10000 in blocking solution) at room temperature for 1 hour. The membrane was washed thoroughly and visualized by chemiluminescence according to the manufacturer’s instructions. Purified hamster oviductin was used as a positive control while the tissue from lung, liver and stomach of both male and female rats was each used as the negative control. This experiment was repeated for five times.

5.4 Immunohistochemical Labeling of Rat Oviduct-Specific Glycoprotein in the Female

Rat Reproductive Tract

Tissue samples of rat ovaries, oviducts and uteri were dissected from three female rats, fixed in 4% paraformaldehyde in 0.1M Cacodylate buffer and embedded in paraffin.

Paraffin embedding and sectioning were done in the Department of Pathology and

Molecular Medicine on a fee-for-service basis. Paraffin sections were cut and placed on microscope glass slides for labeling. The sections were first deparaffinized in Hemo-D

(Fisher; Ottawa, ON, Canada) and rehydrated in decreasing concentrations of ethanol.

81 After being rinsed once in 0.01M PBS, the sections were incubated with 1% ovalbumin in

0.01M PBS for 20 min at room temperature to block the non-specific binding sites of the primary antibody. Then the sections were incubated with the anti-hamster oviductin antibody (1:4) for 2 hours at room temperature. The sections were washed thrice in 0.01M

PBS. Endogenous peroxidase activity was removed by incubating sections in 1% H2O2 in ddH2O for 1 hour. Sections were washed with 0.01M PBS, and HRP-conjugated rabbit anti-mouse IgG antibody (1:200) was applied to the sections for 20 min at room temperature. The immunoperoxidase reaction was developed by incubation of the sections in DAB solution (Zymed; South San Francisco, CA, USA) for 2.5 min. The sections were counterstained with hematoxylin (Sigma; Oakville, ON, Canada) and the sections were covered with glass cover-slips with histomount-mounting solution (Zymed;

South San Francisco, CA, USA) and examined under a light microscope. Negative controls were labeled only with polyclonal rabbit anti-mouse secondary antibody.

5.5 Immunocytochemical Labeling of Rat Oviduct-Specific Glycoprotein in the Female

Rat Reproductive tract

After tissue samples of rat oviducts and uteri were collected, they were embedded in

LR-White in the Department of Pathology and Molecular Medicine at Queen’s University.

Ultra-thin sections of oviducts and uteri were cut and placed onto formvar/carbon-coated nickel grids (Ted Pella Inc; Redding, CA, USA) for electron microscopic immunocytochemistry. Sections were labeled with monoclonal anti-hamster oviductin antibody (1:3) for 2 hours at room temperature followed by incubation with goat

82 anti-mouse IgG conjugated with gold particles (1:5) (10nm; Sigma; Oakville, ON,

Canada) for 30 min at room temperature. The sections were counterstained with uranyl acetate and lead citrate, and examined under an electron microscope (Hitachi 7000) operated at 75 kV.

5.6 Superovulation of Female Rats and Collection of Rat Oviductal Oocytes

Five mature female rats weighing approximately 250~300g were each superovulated following an intraperitoneal (i.p.) injection with approximately 42I.U. (150I.U./kg) of pregnant mare’s serum gonadotropin (PMSG; Sigma; Okaville, ON, Canada) in 250μl of

0.9% NaCl between 1100 am and 1130 am on the day of metestrus. Forty-eight hours later, each rat was intraperitoneally injected with 21I.U. (75I.U./kg) of human chorionic gonadotropin (hCG; Sigma; Okaville, ON, Canada) in 100μl of 0.9% NaCl (Mukumoto et al. 1995).

For collection of oviductal oocytes, superovulated female rats were anesthetized by i.p. injection with 150μl of sodium pentobarbital (2.27kg/1mL) 21 hours after injection with hCG (Mukumoto et al. 1995). Their ventral abdominal wall was opened immediately.

The oviducts were excised and rinsed with PBS. Under the dissection microscope, the ampullary portion of the oviduct was identified and torn open using a pair of fine tweezers. The cumulus-oocyte-complex was released into a porcelain well containing

PBS. The cumulus masses containing the ovulated oocytes were then transferred to another well containing PBS and washed. The cumulus masses were fixed in 2.5% glutaraldehyde in 0.1M Cacolydate buffer for 1 hour at room temperature. After fixation,

83 the masses were washed in 0.01M PBS thrice and embedded in 10% gelatin in 0.01M

PBS for 30 min to 1 hour at 4°C. When the gelatin became solidified, 2.5% glutaraldehyde in 0.01M PBS was added to the solidified gelatin containing cumulus masses for final fixation. Under the dissecting microscope, the excess gelatin surrounding the cumulus masses was trimmed off. The gelatin containing the cumulus masses was embedded in LR-White for immunocytochemistry.

5.7 Immunocytochemical Labeling of Oviduct-Specific Glycoprotein in the Rat Oviductal

Oocytes

One-micron thick sections of LR-white embedded cumulus-masses were first prepared, stained with toluidine blue and examined under a light microscope to locate the ovulated oocytes. Ultra-thin sections of the selected areas were subsequently prepared for immunolabeling. In brief, ultra-thin LR-White sections of cumulus masses containing the oocytes were incubated with monoclonal anti-hamster oviductin antibody (diluted 1:4 in

0.01M PBS) for 2 hours followed by labeling with 10nm gold conjugated anti-mouse IgG antibody (diluted 1:5 in 0.01M PBS; 10nm; Sigma, ON, Canada) for 45 min at room temperature. The sections were counterstained with uranyl acetate and lead citrate before examination on a Hitachi 7000 Electron Microscope operated at 75kV. Control experiments were performed using 0.01M PBS alone instead of anti-hamster oviductin antibody.

84 5.8 cDNA Cloning of Putative Rat Oviductin (NP_997469)

Using hamster oviductin amino acid sequence, the BLAST (Basic Local Alignment and Search Tool) was utilized to search for candidates of a putative rat oviductin.

Subsequently, the protein NP_997469 (genbank accession number) showing the highest level of identity to hamster oviductin was selected. In an attempt to clone the cDNA of

NP_997469, total oviduct RNA was extracted from female rat oviducts according to the manufacturer’s instruction (Sigma; Okaville, ON,Canada) and two primers were designed as the following: Forward: AATGCTAGCATGGCCAAGCTCATTCTTGTCACAGG; and Reverse: AATTGAATTCGTGGCCAGTTGCAGCAATTACAGCTG. A one-step

RT-PCR was performed following the protocol described in Qiagen OneStep RT-PCR Kit

Handbook (Qiagen; Mississauga, ON, Canada). In brief, reverse transcription was performed at 50°C for 30min to specifically synthesize the first strand of cDNA followed by a 95°C treatment for 15min to heat inactivate the reverse transcriptases and activate the hot start DNA polymerase. A PCR was followed immediately to amplify the double stranded cDNA. As predicted, a 1.4 kb band corresponding to the gene product of

NP_997469, the putative rat oviductin, was observed on the DNA gel. The PCR product was further supplemented with additional Taq polymerase and dATP at 72°C for 10min to add a single 3'-A overhang to each end of the PCR product so as to facilitate the ligation efficiency for the TA cloning. This PCR products with dA overhang was purified through

Wizard PCR Purification Kit (Promega; Nepean, ON, Canada) and ligated into the pGEM-T vector (Promega; Nepean, ON, Canada) using the Rapid DNA Ligation Kit

(Fermentas; Burlington, ON, Canada). The ligation product was transformed into E. coli

85 Top10 competent cells, and the cells were plated onto LB media (1% peptone, 0.5% yeast extract, 1% NaCl) plates supplemented with 50ng/ml Ampicillin. Individual colonies were inoculated into liquid 2XTY media (1.6% peptone, 1% yeast extract, and 0.5% NaCl) with 50ng/ml Ampicillin. Plasmids were purified as per the protocol provided by

Molecular Cloning (Sambrook et al. 1989). Positive candidate plasmids were digested with EcoRI+NheI to further confirm the full-length insert and then sent for sequencing

(ACGT Corporation; Toronto, ON, Canada).

86 6. RESULTS

6.1 Western Blot Detection of Oviductin Expression in the Female Rat Reproductive Tract

Western blot of the homogenized tissues from rat ovaries, oviducts and uteri with the anti-hamster oviductin antibody revealed a broad band in the range of 180~300kDa for oviduct only (Fig. 14). No apparent immunoreaction was observed in the ovary or uterus.

The control group of tissues, with the exception of the stomach, showed a negative reaction (Fig. 14).

♂rat ♀rat hamster

250

150

100

1 2 3 4 56 7 8 9 10

Figure 14. Western blot analysis examining the protein expression of hamster oviductin homologue in the rat. Different tissue samples of both female and male rats were subjected to SDS-PAGE followed by immunodetecion with monoclonal anti-hamster oviductin antibody (1:100) and HRP-conjugated rabbit anti-mouse IgG (1:10,000). Lane 1: total protein from homogenized ♂rat lung tissue (60μg/lane); Lane 2: total protein from homogenized ♂rat liver tissue (60μg/lane); Lane 3: total protein from homogenized ♂rat stomach tissue (60μg/lane); Lane 4: total protein from homogenized ♀rat ovary tissue (60μg/lane); Lane 5: total protein from homogenized ♀rat oviduct tissue (60μg/lane); Lane 6: total protein from homogenized ♀ rat uterus tissue (60μg/lane); Lane 7: total protein from homogenized ♀ rat stomach (60μg/lane); Lane 8: total protein from homogenized ♀ rat liver tissue (60μg/lane); Lane 9: total protein from homogenized ♀ rat lung tissue (60μg/lane); Lane 10: lectin-affinity purified hamster oviductin (1μg/lane).

87 6.2 Immunohistochemical Labeling of Oviductin in the Female Rat Reproductive Tract

Immunohistochemical labeling of paraffin sections of the rat reproductive tract revealed that the monoclonal anti-hamster oviductin antibody reacted specifically with the epithelium of the rat oviduct. A strong immunoreaction was observed in the isthmus region (Fig. 15A) and in the ampulla (not shown). In contrast, in the fimbriae area where the epithelial cells are almost exclusively non-secretory ciliated cells, the reaction was hardly seen (Fig. 15B). The labeling pattern of the epithelial cells was heterogeneous, indicating not all the cells lining the epithelium were labeled. Furthermore, a strong immunoreaction was detected in the cytoplasm of these immunoreacted cells (Fig. 15C), indicating that these cells are responsible for producing the glycoprotein. Interestingly, a relatively weak immunostaining occurred along the surface of the uterine epithelium. The labeling was scattered and discontinuous among the non-ciliated epithelial cells (Fig.

15D). Some of the glandular epithelial cells were labeled as well (not shown). The ovary was not labeled (Fig. 15E). No immunoreaction was observed in control sections with the primary antibody omitted in the incubation process (Fig. 15F).

Figure 15. Immunohistochemical localization of oviduct-specific glycoprotein in the rat reproductive tract using monoclonal anti-hamster oviductin antibody. (A) Immunohistochemical labeling reveals that the monoclonal anti-hamster oviductin antibody reacts specifically with the isthmus (Isth) region. (B) In the fimbriae (Fim) area where the epithelial cells are mainly non-secretory ciliated cells, the immunoreaction is hardly seen. (C) The epithelium of the oviduct within an isthmus region is intensively labeled by the monoclonal antibody. The labeling pattern is heterogeneous. The cytoplasm of some epithelial cells (arrows) is immunostained. (D) A weak immunoreaction (arrows) is detected over the surface of some cells of the uterine epithelium lining the endometrium. (E) The ovary is not labeled by the monoclonal anti-hamster oviductin antibody. (F) No immunoreaction is observed in the control sections when the anti-hamster oviductin antibody is omitted in the incubation process. (Isth: Isthmus; Fim: fimbriae) Bar = 50 µm

89 A B

Isth Fim

Isth

C D

E F

90 6.3 Immunocytochemical Labeling of Oviductin in the Female Rat Reproductive Tract

An immunocytochemical labeling study was conducted to localize oviductin in the rat oviduct at the ultrastructural level. Within the non-ciliated epithelial cells of the oviductal epithelium, gold particles were seen to be associated with the microvilli that projected into the oviductal lumen. The gold particles appeared as a halo surrounding the microvilli (Figs. 16A and 16B) whereas the ciliated cells of the oviductal epithelium were devoid of labeling and so were their cilia (Fig. 16A). A high density of the gold particles was also observed in the secretory granules located in the apical cytoplasm of the non-ciliated secretory cells (Figs. 16B and 16C). In addition, gold particles within the secretory granules displayed a distinct lamella-like, beaded-chain appearance (Fig. 16C).

In some secretory granules, a dense core was found to be located at one pole of the secretory granule (Fig. 16C, inset). The content of the Golgi complex was also labeled

(Fig. 16D). However, the nucleus, the cytoplasm (Fig. 16D) and the lateral cell membrane

(Fig. 16C) were devoid of labeling. Control sections of rat oviducts showed very few, randomly distributed gold particles at the background level.

Electron microscopic immunocytochemistry was carried out to localize the rat oviductin in the uterus using the anti-hamster oviductin antibody. Labeling by gold particles was observed associated with the microvilli of some cells of the uterine epithelium (Fig. 17A). Labeling was absent in the Golgi apparatus of the uterine epithelial cells. Labeling of control sections of rat uterus with gold-conjugated rabbit anti-mouse

IgG antibody alone resulted in a negative reaction (Fig. 17B).

Figure 16. Electron micrographs of thin-sections of LR-White embedded rat oviduct labeled with monoclonal anti-hamster oviductin antibody. (A) Gold particles are seen to be associated with the non-ciliated epithelial cells, especially with the microvilli (Mi) that project into the lumen. The adjacent ciliated cell together with their cilia (Ci) is devoid of labeling (X 19,000). (B) The microvilli (Mi) of a non-ciliated secretory epithelial cell are heavily labeled by gold particles (X 27,000). (C) High density of gold particles is observed in the secretory granules (SG) that are located in the apical cytoplasm of the cell. Gold particles within a secretory granule display a distinct lamella-like, beaded-chain appearance (arrowheads). In contrast, the lateral cell membrane (LCM) is devoid of labeling (X 29,000). Inset: High magnification micrograph showing a dense core at one pole of the granule (arrow) (X 77,000). (D) The content of the Golgi complex is labeled (arrowheads) by a few gold particles, however, the cytoplasm does not show any immunoreaction (X 110,000). (Mi: Microvilli; Ci: Cilia; SG: Secretory granules; LCM: Lateral cell membrane)

92 A Mi Ci

93 C

LCM D

Figure 17. Electron micrographs of thin-sections of LR-White embedded rat uteri labeled with monoclonal anti-hamster oviductin antibody. (A) Gold particles are found only at the cell surface of some uterine cells. Labeling is associated with the microvilli (arrows) (X 32,000). (B) Control section displays no immunoreaction (X 46,000).

95 A

96 6.4 Immunodetection of Oviductin in Rat Oviductal Oocytes

Application of anti-hamster oviductin monoclonal antibody to ultra-thin sections of cumulus masses containing post-ovulatory oocytes demonstrated an intense immunolabeling within the perivitelline space (PVS) between the zona pellucida (ZP) and the oolemma surrounding the oocyte (Figs. 18A and 18B). Gold particles were observed over flocculent material found within the PVS (Fig. 18B). At high magnification, the PVS appeared as a loose meshwork of palely stained flocculent material and darkly stained particulate which can be easily distinguished from the ZP matrix. The distribution of gold particles within the PVS was consistent and the labeling appeared to be associated with palely stained flocculent material (Figs. 18B and 18C). Few gold particles were also observed in the ZP (Fig. 18B). Interestingly, some cytoplasmic projections from the cumulus cells were seen extending into the ZP and PVS (Fig. 18A). The cumulus cells, cumulus matrix (Fig. 18D) and the oocyte proper (Fig. 18E) were devoid of any labeling.

Control sections showed an absence of labeling over the cumulus matrix, zona pellucida, perivitelline space and microvilli of the oocyte (Fig. 18F).

Figure 18. Electron micrographs of ultra-thin LR-White sections of superovulated rat oviductal oocytes labeled with monoclonal anti-hamster oviductin antibody. (A) Low magnification electron micrograph shows the disposition of the perivitelline space (PVS) in relation to the oocyte (O), zona pellucida (ZP), cumulus cell (CC) and cumulus matrix (CM) (X 15,000). (B) At a higher magnification, gold particles are observed throughout the PVS. A few gold particles is also observed in the zona pellucida (ZP) (arrows) (X 48,000). (C) The PVS appears as a loose meshwork made up of lightly stained flocculent material and darkly stained particulates, which can be easily distinguished from the zona pellucida (ZP). The labeling appears to be associated with the palely stained amorphous structure (arrows) (X 96,000). (D) The cumulus cell (CC) and cumulus matrix (CM) are devoid of any labeling (X 15,000). (E) No gold particles are found inside the oocyte (X 30,000). (F) Control section shows an absence of labeling over the PVS and microvilli of the oocyte (X 30,000). (PVS: Perivitelline space; O: Oocyte; ZP: Zona pellucida; CC: Cumulus cell; CM: Cumulus matrix; La: Laminar structures; Mi: Mitochondria)

98 A CM ZP

O PVS

B C O PVS

PVS

99 D E La CM

CC Mi

ZP PVS

100 6.5 Search for Candidate Protein of Rat Oviductin and cDNA Cloning of the Putative Rat

Oviductin (NP_997469)

Considering that oviductin is highly conserved across species, it was hypothesized in the present investigation that the rat genome also encodes oviductin. To our knowledge, such oviductin has not been identified in the rat species.

In an attempt to search for the gene of rat oviductin, a “candidate approach” was used. To begin with, the BLAST (http://www.ncbi.nlm.nih.gov/blast/Blast.cgi) was utilized to search for candidates that show similarity to hamster oviductin at the DNA level from the rat genome database (Altschul et al. 2005). However, no candidates were found. Subsequently, the BLAST was performed at the protein level and six candidates were identified and listed in Table 2 with a more detailed description. To further search for the conserved domains that these candidates and hamster oviductin share, SMART

(Simple Modular Architecture Research Tool) (http://smart.embl-heidelberg.de/) (Schultz et al. 1998) was used to predict putative domains for all the candidate proteins listed in

Table 2. The prediction results obtained with SMART indicated that all six candidates encode a Glyco_18 domain that belongs to glycoside hydrolase family 18 (Fig. 19). Since one of the candidate, NP_997469, shows the highest level of identity and similarity to hamster oviductin, a pilot experiment was carried out to clone the gene. Total rat RNA was extracted from two rat oviducts and used as a template to amplify the corresponding cDNA in a one-step RT-PCR reaction. A 1.4 kb DNA was successfully amplified and further cloned into pGEM-T vector (Promega) (Fig. 20). This construct was confirmed through sequencing (Fig. 21).

101

Table 2. BLAST results using hamster oviductin amino acid sequence in search of a putative rat oviductin in the rat genome.

Gene (Genbank accession number) Description Identity % Positive % NP_997469 Chitinase precursor-like 53 69

NP_001073157 Chitinase 1 (chitotriosidase) 48 65 102

XP_001069894 Similar to chitinase, acidic 46 62

EDL81837 Similar to hypothetical protein MGC58999 46 62

EDM09739 Chitinase 3-like 1, isoform CRA_a 46 62

XP_227567 Similar to chitinase 3-like protein 3 precursor (ECF-L) 46 63

Glyco_18 Hamster oviductin

GlGlyycoco_ 118 Cht BD2 Rat NP_997469

Glyco_18 Cht BD2 Rat NP_001073157

Glyco_18 Glyco_18 Rat XP_001069894

Glyco_18 Rat EDL81837

Glyco_18 Rat EDM09739

Glyco_18 Rat XP_227567

Figure 19. Domain of hamster oviductin homologue in the rat predicted by SMART. The blue triangles stand for Glyco_18 domain, and the green hexagons stand for chitin-binding domain type 2 (Cht BD2). The orange rectangle represents the transmembrane segment.

103

5.0kb 3.0kb pGEM-T 3.0kb

1.4kb cDNA 1.5kb

Figure 20. Agarose gel electrophoresis analysis of cloned cDNA of putative rat oviductin (NP_997469) in pGEM-T vector. Lane 1: DNA ladder; Lane 2: Plasmid (pGEM-T/cDNA) digested with EcoRI and NheI. The upper band represents the pGEM-T vector and the lower band represents the cloned cDNA.

104

1 acgcgggggc agtactgaag acaacatggc caagctcatt cttgtcacag gtctggttct

61 tctgctgaat gttcagctgg ggtctgccta caatctggta tgctacttca ccaactgggc

121 ccagtatcgg ccaggcctgg ggagtttcaa gcctgatgac attaatccct gcctgtgtac

181 tcacctgatc tatgcctttg ctgggatgca gaacaatcaa atcaccacca tagaatggaa

241 tgatgttact ctctataaag ctttcaatga cttgaaaaac aggaacagca aactgaaaac

301 tctcctggca attggaggtt ggaactttgg aactgctcct ttcaccacca tggtttccac

361 ttctcagaac cgccagacct ttattacctc agtcatcaaa ttcctgcgac agtatgggtt

421 tgatggactg gacctggact gggaataccc aggttcacgt gggagccctc ctcaggacaa

481 gcatctcttc actgtcctgg tgaaggaact acgtgaagct tttgagcagg aggctattga

541 gagcaacagg cccagattga tggttactgc tgctgtagct gctggcattt ccaacatcca

601 ggctggctac gagatccctg aactttctca gtacctggat ttcatccacg tcatgacata

661 tgacctccat ggctcctggg atggctacac tggggagaac agtcctcttt acaaactccc

721 tactgagact ggcagcaatg cctacctcaa tgtggattat gtcatgaatt actggaagga

781 taatggagcc ccagctgaga agctcattgt tggatttcca gagtatggac acacctacat

841 cctaagcaac ccttctgata ctggaattgg tgcccctacc tctggtaatg gccctgctgg

901 gccctatacc aggcaggctg ggttctgggc ctactatgag atttgtacct ttctgagaaa

961 tggagccact caggactggg atgcccccca ggaagttccc tatgcctaca agggcaatga

1021 gtgggttggc tatgataata tcaagagctt cagtgtcaag gctcagtggc ttaagcagaa

1081 caactttggt ggtgccatga tctgggccat tgaccttgat gacttcactg gctctttctg

1141 tgaccaggga aaattccctc tgacttctac tttgaataaa gcccttgata tacccacagc

1201 aggttgcaca gcccctgact taccttctga accagtaact actcctccag gaagtgggag

1261 tggtggtgga agctctggtg gaggttctga aggcagtgga ttctgtgctg gcaaagcaga

1321 tggcctctac cctgtggcag atgacagaaa tgctttctgg cactgcatca atggaatcac

1381 atatcagcag cattgtcagg cagggttagt ttttgatacc agctgtaatt gctgcaactg

1441 gccatgaaac taatgttctt cacgaaattt ctgccttctc ccttaatcct caccaaaggt

1501 agcttgcttc catttaaagt tatgcaataa aattgatagc c

Figure 21. cDNA sequence of a putative rat oviductin (NP_997469).

105 7. DISCUSSION

Although various studies have shown that oviduct-specific glycoproteins are present in many mammalian species (for a review, see Buhi 2002), little information is available about the presence of oviduct-specific glycoprotein in the rat. Arias and co-workers (1994) conducted a Northern blot analysis using human oviductin cDNA to probe total oviductal

RNA samples from several mammalian species, and no message was detected in the rat oviduct. The reason might be due to the following: first, the coding region of the human oviductin and that of the rat have very low identity. Secondly, the expression of the message in the rat is too low to be detected (Arias et al. 1994). However, a previous study carried out by Abe and Abe (1993) reported that the rat oviduct-specific glycoprotein was detectable in the oviduct using a monoclonal antibody produced against oviductal secretions from the golden hamster. By the use of a monoclonal anti-hamster oviductin antibody through Western blot analysis, immunohistochemical and immunocytochemical experiments, results in the present investigation have shown that a hamster oviductin homologue is present in the female rat reproductive tract. Thus, the present findings provide further evidence showing the presence of oviductin in the rat species, reinforcing the notion that oviductin is probably ubiquitous among mammals.

In the present study, Western blot analysis using the monoclonal antibody against hamster oviductin revealed the expression of oviduct-specific glycoprotein in the rat with a molecular weight of 180~300kDa which is very close to the molecular weight of

160~350kDa reported in hamster oviductin. The monoclonal antibody used in the present study was assessed and proven to be capable of recognizing the glycosidic epitope of

106 hamster oviductin (St-Jacques and Bleau 1988) and detecting the mature form of hamster oviductin (McBride et al. 2004). In the present study, the broad band ranging from

180~300kDa detected through Western blot analysis probably represents the fully glycosylated form of rat oviduct-specific glycoprotein. In the controls, Western blot analysis of the stomach tissue prepared from the male and female rats revealed a single band at 250kDa and ~200kDa, respectively. This is probably due to the cross-reaction of the antibody with certain mucin of the stomach. Similar cross-reaction of the monoclonal antibody with the stomach tissue had also been reported in the hamster (St-Jacques and

Bleau 1988).

Since we detected the protein expression of hamster oviductin homologue in the rat oviduct, subsequent experiments were necessary to localize this glycoprotein in tissue sections. Our study with light microscopic immunohistochemistry showed a prominent immunoreaction in the oviduct. The mammalian oviductal epithelium is composed mainly of non-ciliated and ciliated epithelial cells. In all species examined to date, it has been shown that oviductin is secreted by the non-ciliated secretory oviductal cells (See

Literature Review). The present study demonstrated that the rat is no exception.

Interestingly, the labeling of rat oviduct-specific glycoprotein displayed a regional difference along the oviduct. We found an intense labeling along the epithelium lining the isthmus and ampulla regions but not at the fimbriae region. This phenomenon has also been reported in other species studied. For example, mouse oviductin was localized in non-ciliated epithelial cells lining the infundibulum and ampulla region (Kapur and

Johnson 1988); cow oviductin was distributed in most of the secretory cells of the

107 oviductal epithelia from the fimbriae to the ampulla (Abe et al. 1993) whereas in the sheep, oviductin (sOP 92) was present only in the non-ciliated secretory cells of the ampulla (Gandolfi et al. 1991). It was also reported that golden hamster oviductin is primarily produced by non-ciliated secretory cells in the ampulla and isthmus portions of the oviduct (Abe and Oikawa 1991). The functional significance of these regional differences is not yet understood. It is probably related to various developmental and reproductive events that occur along the length of the oviduct (Kapur and Johnson 1988;

Abe et al. 1993). Compared to what has been reported in literature, our findings in the rat are very similar to those reported in the golden hamster. The confined localization of oviductin to the ampulla and isthmus regions in the rat is probably due to the abundance of non-ciliated secretory cells in the isthmus and ampulla with the latter being considered as the site of fertilization in the rat. Thus, it is conceivable that the rat oviductin may play a role during the fertilization process.

These early results led us to conduct further electron microscopic immunocytochemical studies to localize rat oviductin in the female reproductive tract at the ultrastructural level. It was observed that, in the non-ciliated secretory oviductal cells, high concentrations of gold particles were evenly distributed in the secretory granules.

These secretory granules found in the rat oviductal cells resemble those observed in the golden hamster. Epon sections of secretory granules in the hamster oviductal epithelial cells showed that these granules contain internal distinct lamina-like structures, radiating from a dense core located at one pole of the secretory granule (El-Mestrah and Kan 1999).

Similarly, in the rat, gold particles in the secretory granules display a beaded-chain

108 appearance and the dense cores are not labeled. The nature and the functional role of the dense core as well as that of the lamella-like structures are still unknown. It is suggested that the presence of the dense core is a characteristic of mature secretory granules in the oviduct and that it may play a role in the secretory mechanism of the secretory granules

(El-Mestrah and Kan 1999). The present findings suggest that the non-ciliated oviductal epithelial cells are active in synthesizing rat oviductin. Extracellularly, gold particles were associated mainly with the microvilli at the apical cell surface. The labeling formed a halo pattern. The appearance of the labeling in the form of a halo may be due to the highly glycosylated side chains associated with the oviductin peptide cord. In contrast, the cytoplasm, nucleus and the lateral and basal membranes of the cell were devoid of labeling. The present immunocytochemical results indicate that the rat oviduct-specific glycoprotein is synthesized by non-ciliated secretory cells, transported to the apical surface of the cell via secretory granules, and is then secreted into the oviductal lumen.

Light microscopic and electron microscopic immunocytochemical studies were carried out to trace the fate of rat oviduct-specific glycoprotein in the uterus. Results obtained with L.M. immunohistochemistry showed a punctuated immunoreaction at the apical surface of the epithelium lining the uterus. At the electron microscopic level, gold particles were found to be associated with microvilli of some uterine epithelial cells, yet, not all of the non-ciliated epithelial cells were labeled. Unlike what was observed in the oviduct, labeling by gold particles was absent in the secretory granules in the uterine epithelial cells. The rest of the cell organelles including the nucleus, Golgi complex, cytoplasm, and lateral cell membrane were also devoid of labeling. These observations

109 suggest that the rat uterus does not synthesize the oviduct-specific glycoprotein. In other words, it has provided us with indirect evidence that rat oviduct-specific glycoprotein is exclusively produced by the oviductal secretory epithelial cells. Upon its secretion into the lumen of the oviduct, rat oviductin can gain access to the lumen of the uterus and bind to the cell surface of certain uterine epithelial cells.

The ultrastructural observations of the rat ovulated oocytes revealed that the hamster oviductin homologue in the rat is associated with the perivitelline space (PVS) of the oocyte. The rat PVS examined under E.M. was very similar to that of the mouse. It appeared to be composed of two distinct components: the palely stained amorphous material and the darkly stained particulate components. The amorphous material was also found in the PVS of the mouse, pig, and opossum oocytes (Talbot and DiCarlantonio

1984; Kapur and Johnson 1988). The authors suggested that the electron microscopic image of the PVS of the mouse oviductal oocyte is likely to have been changed from its in situ state by the fixation, dehydration, and embedding procedures (Kapur and Johnson

1988). Our results showed that the hamster oviductin homologue in the rat is associated with the palely stained amorphous material within the PVS which has the same labeling pattern as in the mouse (Kapur and Johnson 1988). Thus, the rat is likely to be another species other than the mouse, among the many mammalian species studied to date, where oviductin is associated with the PVS and not the zona pellucida. A few gold particles were also seen in the zona pellucida. This might be due to the background labeling. In addition, the finding of lack of labeling over the cumulus cells and cumulus matrix rules out the possibility that the cumulus cells are a contributor of this glycoprotein to the PVS of

110 the rat oocyte. Secretion products coming from the oviductal fluid have to first pass through the cumulus matrix and the ZP before they can reach the PVS, therefore we believe that rat oviductin can penetrate the ZP and specifically bind to the material in the PVS.

However, the significance of this close association of oviductin with the PVS in the rat oviductal oocyte is unclear.

Based on the results obtained with Western blot analysis and both L.M. and E.M. immunocytochemical studies, respectively, on the detection and localization of a hamster oviductin homologue in the female rat reproductive tract, it is speculated that a similar oviductin protein is encoded by the rat genome. Although extensive research has been conducted on oviductin in many mammalian species, little has been done on oviductin in the rat. The discovery of rat oviductin will potentially open up a new avenue for oviductin research. This prompted us to search for the sequence of rat oviductin. Thus, a “candidate approach” was employed in an attempt to determine the sequence of this protein. Since the antibody cross-reaction implies that the rat oviductin has high similarity to the hamster oviductin, BLAST analysis was performed to search for proteins encoded in the rat genome that share high identity to hamster oviductin, and resulted in six potential candidates, all of which contain a putative Glyco_18 domain. In a pilot experiment, a putative rat oviductin, protein NP_997469, which has the highest similarity to hamster oviductin, has been successfully cloned.

This construct offers a promising candidate for future studies of the putative rat oviductin. This construct can be subcloned into expression vector and then transfected into certain mammalian cell line for overexpression. Transfected cells can then be tested

111 with Western blot using the monoclonal anti-hamster oviductin as primary antibody. If the transfected cells show cross-reaction, this candidate is very likely to be the oviductin homologue encoding the protein that shows the cross-reaction as detected in situ in the present investigation. This construct also lays the foundation for developing antibodies against rat oviductin in the future.

In summary, the present study was carried out to investigate the presence of an oviduct-specific glycoprotein in the female rat. To our knowledge, this is the first detailed report of the presence of an oviduct-specific glycoprotein in the rat species and the first report of the site of localization of oviductin in the rat ovulated oocytes. However, questions regarding the function of the oviductin in the rat remain to be answered.

112 SUMMARY AND CLOSING REMARKS

1. The present study has confirmed that, in the female hamster reproductive system, oviductin secreted by the non-ciliated secretory cells can gain access to the lumen of the uterus and bind to the cell surface of the epithelium lining the uterus.

2. The optimized conditions for culturing primary hamster uterine epithelial cells have been established. This culture system will be useful for further study of the interaction between oviductin and the uterine epithelial cells in vitro.

3. With confocal microscopy, immunofluorescence labeling of hamster oviductin in the cultured uterine epithelial cells showed that the binding is restricted to the peripheral area of the cultured cells, which is consistent with the previous results obtained with immunohistochemistry. Pre-embedding labeling study further showed that oviductin is peripherally associated with the microvilli of immunolabeled cultured uterine epithelial cells. Together, our results indicate that there are potential receptor sites for oviductin present on the cell surface of uterine epithelial cells.

4. The present investigation demonstrated the presence of a hamster oviductin homologue in the rat female reproductive tract. Western Blot analysis indicated that the hamster oviductin homologue has a molecular weight ranging from 180~300 kDa.

113 5. Further L.M. immunohistochemical and E.M. immunocytochemical results showed that hamster oviductin homologue in the rat is synthesized and secreted by the oviductal non-ciliated cells. In addition, immunolabeling is also observed on the cell surface of some uterine epithelial cells.

6. Surprisingly, immunolabeling of rat oviduct-specific glycoprotein in the ovulated oocytes was found in the perivitelline space and not in the zona pellucida. This finding was unlike what was previously observed in the hamster ovulated oocytes but similar to the distribution of oviduct-specific glycoprotein previously reported in the mouse ovulated oocytes. To our knowledge, this is the first detailed report of the presence of oviduct-specific glycoprotein in the rat species and the first report of the site of localization of rat oviductin in the ovulated oocytes.

7. The cDNA of a putative rat oviductin has been successfully cloned.

114 REFERENCES

Abe H, Abe M. (1993) Immunological detection of an oviductal glycoprotein in the rat. J Exp Zool 266: 328-335

Abe H, Numazawa C, Onodera M, Katsumi A. (1993) Immunological localization of oviduct-specific glycoproteins in the bovine oviductal epithelium at follicular and luteal phases. Cell Tissue Res 274: 41-47

Abe H, Oikawa T. (1992) Effects of estradiol and progesterone on the cytodifferentiation of epithelial cells in the oviduct of the newborn golden hamster. The Anat Rec 235(3): 390-398

Abe H, Oikawa T. (1991) Regional differences in the ultrastructural features of secretory cells of the golden hamster (Mesocricetus auratus) oviductal epithelium. J Anat 175: 147-158.

Abe H, Oikawa T. (1990a) Study of the differentiation of secretory cells in the golden hamster oviductal epithelium by use of a monoclonal antibody. J Exp Zool 254: 97-106.

Abe H, Oikawa T. (1990b) Ultrastructural evidence for an association between an oviductal glycoprotein and the zona pellucida of the golden hamster egg. J Exp Zool 256: 210-221.

Abe H, Onodera M, Sugawara S. (1995) Immunological detection and characterization of an estrus-associated antigen in the goat oviduct. J Exp Zool 272: 134-141.

Abe H, Sendai Y, Satoh T, Hoshi H. (1995) Secretory products of bovine oviductal epithelial cells support the viability and motility of bovine spermatozoa in culture in vitro. J Exp Zool 272(1): 54-61

Altschul SF, Wootton JC, Gertz EM, Agarwala R, Morgulis A, Schäffer AA, Yu YK. (2005) Protein database searches using compositionally adjusted substitution matrices. FEBS. J 272 (20): 5101-5109

Anderson TL, Hoffman LH. (1984) Alteraction in epithelia glycocalyx of rabbit uteri during early pseudopregnancy and pregnancy, and following ovariectomy. Am J Anat 171(3): 321-334

Aplin JD, Kimber SJ. (2004) Trophoblast-uterine interactions at implantation. Reprod Biol Endocrinol 5(2): 48-60

Araki Y, Kurata S, Oikawa T, Yamashita T, Hiroi M, Naiki M, Sendo F. (1987) A monoclonal antibody reacting with the zona pellucida of the oviductal egg but not with that of the ovarian egg of the golden hamster. J Reprod Immunol 11: 193-208.

Arias EB, Verhage HG, Jaffe RC. (1994) Complementary deoxyribonucleic acid cloning

115 and molecular characterization of an estrogen-dependent human oviductal glycoprotein. Biol Reprod 51: 685-694.

Bagchi IC, Li Q, Chenon YP. (2001) Role of steroid hormone-regulated gene in implantation. Ann N Y Acad Sci 943: 68-76

Baggish M, Valle RF, Guedj V. (2007) Physiology of the uterus and cervix. In: Hysteroscopy-Visual perspectives of uterine anatomy, physiology and pathology, 3rd ed, Lippincott Williams & Wilkins, Philadelphia

Barberini F, Sartori S, Motta P, Blerkom JV. (1978) Changes in the surface morphology of the rabbit endometrium related to the estrus and progestational stages of the reproductive cycle. Cell and Tissue Res 190(2): 207-222

Boatman DE, Felson SE, Kimura J. (1994) Changes in morphology, sperm penetration and fertilization of ovulated hamster egg induced by oviductal exposure. Hum Reprod 9(3): 519-526

Boatman DE, Magnoni GE. (1995) Identification of a sperm penetration factor in the oviduct of the golden hamster. Biol Reprod 52: 199-207.

Boice ML, Geisert RD, Blair RM, Verhage HG. (1990) Identification and characterization of bovine oviductal glycoproteins synthesized at estrus. Biol Reprod 43: 457-465.

Boice ML, Mavrogianis PA, Murphy CN, Prather RS, Day BN. (1992) Immunocytochemical analysis of the association of bovine-specific glycoproteins with early embryos. J Exp Zool 263(2): 225-229

Boice ML, McCarthy TJ, Mavrogianis PA, Fazleabas AT, Verhage HG. (1991) Localization of oviductal glycoproteins within the zona pellucida and perivitelline space of ovulated ova and early embryos in baboon (Papio anubis). Biol. Reprod. 43: 340-346

Bowen JA, Newton GR, Weise DW, Bazer FW, Burghardt RC. (1996) Characterization of a polarized porcine uterine epithelial model system. Biol of Reprod 55: 613-619

Brenner CA, Adler RR, Rappolee DA, Pedersen RA, Werb Z. (1989) Genes for extracellular matrix-degrading enzymes and their inhibitor, TIMP, are expressed during early mammalian development. Gene Dev 3: 848-859

Brown CR, Cheng WK. (1986) Changes in composition of the porcine zona pellucida during development of the oocyte to the 2- to 4-cell embryo. J Embryol Exp Morphol 92: 183-191.

Buhi WC. (2002) Characterization and biological roles of oviduct-specific, oestrogen-dependent glycoprotein. Reprod 123: 355-362.

Buhi WC, Alvarez IM, Choi I, Cleaver BD, Simmen FA. (1996) Molecular cloning and

116 characterization of an estrogen-dependent porcine oviductal secretory glycoprotein. Biol Reprod 55: 1305-1314.

Buhi WC, Alvarez IM, Kouba AJ. (2000) Secreted proteins of the oviduct. Cells Tissues Organs 166: 165-179.

Buhi WC, Alvarez IM, Sudhipong V, Dones-Smith MM. (1990) Identification and characterization of de novo-synthesized porcine oviductal secretory proteins. Biol Reprod 43: 929-938.

Buhi WC, O’Brien B, Alvarez IM, Erdos G, Dubois D. (1993) Immunogold localization of porcine oviductal secretory proteins within the zona pellucida, perivitelline space, and plasma membrane of oviductal and uterine oocytes and early embryo. Biol Reprod 48(6): 1274-1283

Bulletti C, Flamigni C, Giacomucci E. (1996) Reproductive failure due to spontaneous abortion and recurrent miscarriage. Hum Reprod Update 2:118-136.

Burkman LJ, Overstreet JW, Katz DF. (1984) A possible role for potassium and pyruvate in the modulation of sperm motility in the rabbit oviductal isthmus, J Reprod Fert 71: 367-376.

Carson DD. (2002) The glycobiology of implantation. Front Biosci 7: d1535-d1544.

Carson DD, DeSouza MM, Kardon R, Zhou X, Lagow E, Julian J. (1998) Mucin expression and function in the female reproductive tract. Hum Reprod Update 4(5): 459-64 Review

Chaminadas G, Propper AY, Rovez M, Prost O, Remy-Martin JP, Adessi GC. (1986) Culture of epithelial and stromal cells of guinea-pig endometrium and the effect of oestradiol-17 beta on the epithelial cells. J Reprod Fertil 77(2): 547-558

Classen-Linke I, Kusche M, Knauthe R, Beier HM. (1997) Establishment of a human endometrial cell culture system and characterization of its polarized hormone responsive epithelial cells. Cell Tissue Res 287(1): 171-185

Clyman MJ. (1966) Electron microscopy of the human fallopian tube. Fertil Steril 17: 281-301

Collins JA, Feeny D, Gunby J. (1997) The cost of infertility diagnosis and treatment in Canada in 1995. Hum Reprod 12: 951-958

Cunha GR, Bigsby RM, Cooke PS, Sugimura Y. (1985) Stromal-epithelial interactions in adult organs. Cell Differ 17: 137-148.

117 DeSouza MM, Mani SK, Julian J, Carson DD. (1998) Reduction of mucin-1 expression during the receptive phase in the rat uterus. Biol Reprod 58: 1503-1507

DeSouza MM, Murray MK. (1995) An estrogen-dependent secretory protein, which shares identity with chitinases, is expressed in a temporally and regionally specific manner in the sheep oviduct at the time of fertilization and embryo development. Endocrinology 136: 2485-2496

Donjacour AA, Cunha GR. (1991) Stromal regulation of epithelial function. Cancer Treat Res 53: 335-364.

El-Mestrah M, Kan FWK. (1999a) Ultrastructural and ultracytochemical features of secretory granules in the ampullary epithelium of the hamster oviduct. The Anat Rec 255: 227-239

El-Mestrah M, Kan FWK. (1999b) Differential distribution of lectin-binding glycoproteins in the secretory granules of hamster oviductal ampulla during the estrous cycle: a quantitative cytochemical analysis. Histochem Cell Biol 111(1): 23-32

Enders AC, Schlafke S. (1967) A morphological analysis of the early implantation stages in the rat. Am J Anat 120:185-226

Gandolfi F, Modina S, Brevini TA, Galli C, Moor RM, Lauria A. (1991) Oviduct ampullary epithelium contributes a glycoprotein to the zona pellucida, perivitelline space and blastomeres membrane of sheep embryos. Eur J Basic Appl Histochem 35: 383-392.

Glasser SR, Julian J, Decker GL, Tang JD, Carson DD. (1988) Development of morphological and functional polarity in primary culture of immature rat uterine epithelial cells. J Cell Biol 107: 2409-2423

Hamner CE, Williams W. (1964) Identification of sperm stimulating factor of rabbit oviduct fluid. Proc Soc Exp Biol Med 117: 240-243

Jaffe RC, Arias EB, O’Day-Bowman MB, Donnelly KM, Mavrogianis PA, Verhage HG. (1996) Regional distribution and hormonal control of estrogen-dependent oviduct-specific glycoprotein messenger ribonucleic acid in the baboon (Papio anubis). Biol of Reprod 55: 421-426

Jansen RPS, Bajpai VK. (1982) Oviduct and mucus glycoproteins in the estrous rabbit: ultrastructure and histochemistry. Biol Reprod 26: 155-168

Junqueira LC, Carneiro J. (2003) The female reproductive system. In: Basic Histology, 10th ed, McGraw-Hill Companies, Inc. Medical Publishing Division.

Kan FW, Esperanzate PW. (2006) Surface mapping of binding of oviductin to the plasma membrane of golden hamster spermatozoa during in vitro capacitation and acrosome

118 reaction. Mol Reprod Dev 73: 756-766.

Kan FW, Roux E. (1995) Elaboration of an oviductin by the oviductal epithelium in relation to embryo development as visualized by immunocytochemistry. Microsc Res Tech 31: 478-487.

Kan FW, Roux E, Bleau G. (1993) Immunolocalization of oviductin in endocytic compartments in the blastomeres of developing embryos in the golden hamster. Biol Reprod 48: 77-88.

Kan FW, Roux E, St Jacques S, Bleau G. (1990) Demonstration by lectin-gold cytochemistry of transfer of glycoconjugates of oviductal origin to the zona pellucida of oocytes after ovulation in hamsters. Anat Rec 226: 37-47.

Kan FW, St Jacques S, Bleau G. (1988) Immunoelectron microscopic localization of an oviductal antigen in hamster zona pellucida by use of a monoclonal antibody. J Histochem Cytochem 36: 1441-1447.

Kan FW, St Jacques S, Bleau G. (1989) Immunocytochemical evidence for the transfer of an oviductal antigen to the zona pellucida of hamster ova after ovulation. Biol Reprod 40: 585-598.

Kapur RP, Johnson LV. (1985) An oviductal fluid glycoprotein associated with ovulated mouse ova and early embryos. Dev Biol 112: 89-93.

Kapur RP, Johnson LV. (1988) Ultrastructral evidence that specialized regions of the murine oviduct contribute a glycoprotein to the extracellular matrix of mouse oocytes. Anat Rec 221(3): 720-729

King RS, Anderson SH, Killian GJ. (1994) Effect of bovine oviductal estrus-associated protein on the ability of sperm to capacitate and fertilize oocytes. J Androl 15(5): 468-478

King RS, Killian GJ. (1994) Purification of bovine estrus-associated protein and localization of binding on sperm. Biol Reprod 51(3): 34-42

Kirk D, King RJB, Heyes J, Peachey L, Hirsch PJ, Taylor RWT. (1978) Normal human endometrium in cell culture. I. Separation and characterization of epithelial and stromal components in vitro. In Vitro 14: 651-662

Komatsu M, Fugita H. (1978) Electron-microscopic studies on the development and aging of the oviduct epithelium of mice. Anat Embryol 153: 243-259

Komiya H, Onuma T, Hiroi M, Araki Y. (1996) In situ localization of messenger ribonucleic acid for an oviduct-specific glycoprotein during various hormonal conditions in the golden hamster. Biol Reprod 55: 1107-1118.

119 Kouba AJ, Abeydeera LR, Alvarez IM, Day BN, Buhi WC. (2000) Effects of porcine oviduct-specific glycoprotein on fertilization, polyspermy, and embryonic development in vitro. Biol of Reprod 63: 242-250

Lapensee L, Paquette Y, Bleau G. (1997) Allelic polymorphism and chromosomal localization of the human oviductin gene (MUC9). Fertil Steril 68:702-708.

Leese HJ. (1988) The formation and function of oviduct fluid. J Reprod Fertil 82: 843-856.

Lee KY, DeMayo FJ. (2004) Animal models of implantation. Reproduction 128: 679-695.

Leveille MC, Roberts KD, Chevalier S, Chapdelaine A, Bleau G. (1987) Uptake of an oviductal antigen by the hamster zona pellucida. Biol Reprod 36: 227-238.

Lisk RD. (1985) The estrous cycle. In: The Hamster: Reproduction and Behaviour. New York: Olenum.

Liszczak TM, Richardson GS, MacLaughlin DT, Kornblith P. (1977) Ultrastructure of human endometrial epithelium in monolayer culture with and without steroid hormones. In Vitro 13: 334-356

Malette B, Bleau G. (1993) Biochemical characterization of hamster oviductin as a sulphated zona pellucida-binding glycoprotein. Biochem J 295(2): 437-445.

Malette B, Filion B, St-Jacques S, Kan FWK, Bleau G. (1995b) Hormonal control of the biosynthesis of hamster oviductin. Micros Res Tech 31: 470-477.

Malette B, Paquette Y, Bleau G. (1995a) Oviductin possesses chitinase- and mucin-like domains: a lead in the search for the biological function of these oviductin-specific zp associating glycoproteins. Mol Reprod Dev 41(3): 384-397 Review

Martoglio AM, Kan FW (1996) Immunohistochemical localization of oviductin in the endometrial lining of the golden hamster (Mesocricetus auratus) during the estrous cycle and early gestation. Histochem J 28: 449-459.

McBride DS, Boisvert C, Bleau G, Kan FW. (2004) Evidence for the regulation of glycosylation of golden hamster (Mesocricetus auratus) oviductin during the estrous cycle. Biol Reprod 70: 198-203.

McBride DS, Brockhausen I, Kan FW. (2005) Detection of glycosyltransferases in the golden hamster (Mesocricetus auratus) oviduct and evidence for the regulation of O-glycan biosynthesis during the estrous cycle. Biochim Biophys Acta 1721: 107-115

Morris JE, Potter SW. (1984) A comparison of developmental changes in mouse blastocyst and uterine epithelium using DEAE beads and dextran sulfate in vitro. Dev Biol 103: 190-199

120 Mukumoto S, Mori K, Ishikawa H. (1995) Efficient induction of superovulation in adult rats by PMSG and hCG. Exp Anim 44(2): 111-118.

Mulholland J, Winterhager E, Beier HM. (1988) Changes in proteins synthesized by rabbit endometrial epithelial cells following primary culture. Cell Tissue Res 252(1): 123-132

O'Day-Bowman MB, Mavrogianis PA, Fazleabas AT, Verhage HG. (1993) In vitro incubation of golden (syrian) hamster ovarian oocytes and human sperm with a human oviduct specific glycoprotein. Mol Reprod Dev 38(2):160-169

O'Day-Bowman MB, Mavrogianis PA, Fazleabas AT, Verhage HG. (1995) A human oviduct-specific glycoprotein: synthesis, secretion, and localization during the menstrual cycle. Microsc Res Tech 32: 57-69.

O'Day-Bowman MB, Mavrogianis PA, Reuter LM, Johnson DE, Fazleabas AT, Verhage HG. (1996) Association of oviduct-specific glycoproteins with human and baboon (Papio anubis) ovarian oocytes and enhancement of human sperm binding to human hemizonae following in vitro incubation. Biol Reprod 54: 60-69.

Odor DL, Gaddum-Rosse P, Pumery RE. (1983) Secretory cells of the oviduct of the pig-tailed monkey, Macaca nemestrina, during the menstrual cycle and after estrogen treatment. Am J Anat 166:149-172

Oikawa T, Sendai Y, Kurata S, Yanagimachi R (1988) A glycoprotein of oviductal origin alters biochemical properties of the zona pellucida of hamster egg. Gamete Res. 19: 113-122.

Oliphant G, Reynolds AB. (1984) Immunocytochemical localization and determination of hormone-induced synthesis of the sulfated oviductal glycoproteins. Biol Reprod 31(1): 165-174

Oliphant G, Ross PR (1982) Demonstration of production and isolation of three sulfated glycoproteins from the rabbit oviduct. Biol Reprod 26: 537-544.

Paquette Y, Merlin Y, Malette B, Bleau G. (1995) Allelic polymorphism in the hamster oviductin gene is due to a variable number of mucin-like tandem repeats. Mol Reprod Dev 42:388-396

Paria BC, Lim H, Das SK, Reese J, Dey SK. (2000) Molecular signaling in uterine receptivity for implantation. Semin Cell Dev Biol 11:67–76.

Richardson JM, Kaushic C, Wira CR. (1995) Polymeric Immunoglobin (Ig) receptor production and IgA transcytosis in polarized cultures of mature rat uterine epithelial cells. Biol of Reprod 53: 488-498

121 Roblero LS, Riffo MD. (1986) High potassium concentration improves perimplantation development of mouse embryos in vitro. Fert Steril 45: 412-416

Roux E, Bleau G, Kan FW. (1997) Fate of hamster oviductin in the oviduct and uterus during early gestation. Mol Reprod Dev 46: 306-317.

Roux E, Kan FW. (1995) Stage-specific immunolabeling for oviductin in the secretory granules of the oviductal epithelium of the golden hamster during the estrous cycle. Anat Rec 241: 369-376

Rugh B. (1968) Reproductive systems of adult mice. In: The mouse. Its reproduction and development, Burgess Publishing Co.

Sambrook J, Fritsch EF, Maniatis T. (1989) Molecular Cloning. 2nd edition. Cold Spring Harbor Laboratory Press, NY.

Schmidt A, Mavroqianis PA, O’Day-Bowman MB, Verhage HG. (1997) Species-specific effect of oviductal glycoproteins on hamster sperm binding to hamster oocytes. Mol Reprod Dev 46(2): 201-207

Schultz J, Milpetz F, Bork P, Ponting CP. (1998) SMART, a simple modular architecture research tool: identification of signalling domains. Proc Natl Acad Sci U S A. 95(11): 5857-5864

Sendai Y, Abe H, Kikuchi M, Satoh T, Hoshi H. (1994) Purification and molecular cloning of bovine oviduct-specific glycoprotein. Biol Reprod 50: 927-934.

Sendai Y, Komiya H, Suzuki K, Onuma T, Kikuchi M, Hoshi H, Araki Y. (1995) Molecular cloning and characterization of a mouse oviduct-specific glycoprotein. Biol Reprod 53: 285-294.

Sharkey A. (1998) Cytokines and implantation. Rev Reprod 3: 52-61

Sharpe KL, Zimmer RL, Khan RS, Penny LL. (1992) Proliferative and morphogenic changes induced by the co-culture of rat uterine and peritoneal cells: a cell culture model for endometriosis. Fertil Steril 58(6): 1220-1229

St-Jacques S, Bleau G. (1988) Monoclonal antibodies specific for an oviductal component associated with the hamster zona pellucida. J Reprod Immunol 12: 247-261

Strous GJ, Dekker J. (1992) Mucin-type glycoproteins. Crit Rev Biochem Mol Biol 27: 57-92

Sun T, Lei ZM, Rao CV. (1997) A novel regulation of the oviductal glycoprotein gene expression by luteinizing hormone in bovine tubal epithelial cells. Mol and Cel Endor 131: 97-108

122

Sutton R, Nancarrow CD, Wallace AL, Rigby NW. (1984) Identification of an oestrus-associated glycoprotein in oviducal fluid of the sheep. J Reprod Fertil 72: 415-422.

Suzuki K, Sendai Y, Matsuda Y, Hoshi H, Hiroi M, Araki Y. (1995) Molecular characterization of a hamster oviduct-specific glycoptrotein. Biol of Reprod 53: 345-354

Talbot P, DiCarlantonio G. (1984) The oocyte-cumulus complex: Ultrastructure of the extracellular matrix in hamsters and mice. Gamete Res 10: 127-142

Thibodeaux JK, Goodeaux LL, Roussel JD, Menezo Y, Amborski GF, Moreau JD, Godke RA. (1991) Effects of stage of the bovine oestrous cycle on in-vitro characteristic of uterine and oviductal epithelial cells. Hum Reprod 6(6): 751-760

Thibodeaux JK, Myers MW, Goodeaux LL, Menezo Y, Roussel JD, Broussard JR, Godke RA. (1992) Evaluating an in vitro culture system of bovine uterine and oviduct epithelial cells for subsequent embryo co-culture. Reprod Fertil Dev 4: 573-583

Tomooka Y, DiAugustine RP, McLachlan JA. (1986) Proliferation of mouse uterine epithelial cells in vitro. Endocrinology 118(3): 1011-1018

Tulsian DR, Chayko CA, Orgebin-Crist MC, Araki Y. (1996) Temporal surge of glycosyltransferase activities in the genital tract of the hamster during the estrous cycle. Biol Reprod 54(5): 1032-1037

Varma VA, Melin SA, Adamec TA, Dorman BH, Siegfried JM, Walton LA, Carney CN, Norton CR, Kaufman DG. (1982) Monolayer culture of human endometrium: methods of culture and identification of cell types. In Vitro 18: 911-918

Verhage HG, Boice ML, Mavrogianis P, Donnelly K, Fazleabas AT (1989) Immunological characterization and immunocytochemical localization of oviduct-specific glycoproteins in the baboon (Papio anubis). Endocrinology 124: 2464-2472.

Verhage HG, Fazleabas AT, Donnelly K (1988) The in vitro synthesis and release of proteins by the human oviduct. Endocrinology 122: 1639-1645.

Verhage HG, Fazleabas AT, Mavrogianis PA, O’Day-Bowman MB, Donnelly KM, Arias EB, Jaffe RC. (1997b) The baboo oviduct: characteristics of an oestradiol-dependent oviduct-specific glycoprotein. Hum Reprod Update 3(6): 541-552

Verhage HG, Mavrogianis PA, Boice ML, Li W, Fazleabas AT. (1990) Oviductal epithelium of the baboon: hormonal control and the immuno-gold localization of oviduct-specific glycoproteins. Am J Anat 187(1): 81-90

123 Verhage HG, Mavrogianis PA, Boomsma RA, Schmidt A, Brenner RM, Slayden OV, Jaffe RC. (1997a) Immunologic and molecular characterization of an estrogen-dependent glycoprotein in the rhesus (Macaca mulatta) oviduct. Biol Reprod 57: 525-531.

Verhage HG, Mavrogianis PA, O’Day-Bowman MB, Schmidt A, Arias EB, Donnelly KM, Boomsma, Thibodeaus JK, Fazleabas, Jaffe RC. (1998) Characteristic of an oviductal glycoprotein and its potential role in the fertilization process. Biol Reprod 58(5): 1098-1101

Watanabe T, Kobori K, Miyashita K, Fujii T, Sakai H, Uchida M, Tanaka H. (1993) Identification of glutamic acid 204 and aspartic acid 200 in chitinase A1 of Bacillus circulans WL-12 as essential residues for chitinase activity. J Biol Chem 268: 18567-18572

Young B, Heath JW. (2000) Female reproductive system. In: Wheater’s Functional Histology, 4th ed, Churchill Livingstone.

124