<%BANNER%>

Studies on the Effects and Mechanisms of Action of Eicosapentaenoic Acid on PGF2alpha Production in Cultured Bovine Endo...


PAGE 1

STUDIES ON THE EFFECTS AND MECHANISMS OF ACTION OF EICOSAPENTAENOIC ACID ON PGF2 PRODUCTION IN CULTURED BOVINE ENDOMETRIAL CELLS By CRISTINA CALDARI-TORRES A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Cristina Caldari-Torres

PAGE 3

Le dedico esta tesis a Mami, Abuela y Lola por ser el ejem plo de lo que una mujer hecha y derecha debe ser.

PAGE 4

iv ACKNOWLEDGMENTS I would like to thank my advisor, Dr. L okenga Badinga, for allowing me to be a part of his laboratory. I am grateful for all his help and support, and for teaching me how to conduct research and encouraging me to perform well academically. I would also like to thank my committee members, Dr. Daniel C. Sharp and Dr. Charles R. Staples, for attending the committee meetings and for their advice on how to improve my work. I also extend many thanks to Liz S. Green e for her help with laboratory techniques, and for her support as a lab mate and as a frie nd. I would like to thank Dr. Alan Ealy for allowing me the use of his computer. Special thanks go to Carlos J. Rodrguez-Sallaberry for helping me improve as a researcher and for his patience, friendship and companionship. I am grateful to him for be ing a great source of st rength and for showing me the right way when I most needed it.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF FIGURES..........................................................................................................vii ABBREVIATION KEY....................................................................................................ix ABSTRACT......................................................................................................................x ii CHAPTER 1 INTRODUCTION........................................................................................................1 2 LITERATURE REVIEW.............................................................................................4 Estrous Cycle................................................................................................................4 Regulation of the Estrous Cycle............................................................................5 Characteristics of the Bovine Estrous Cycle.........................................................5 Phases of the Estrous Cycle...................................................................................6 Stages of the Estrous Cycle...................................................................................7 Anestrus.................................................................................................................8 Ovarian Follicular Dynamics in Ruminants...............................................................10 Hypothalamic-pituitary-ovarian axis...................................................................11 Follicular Waves in Cattle...................................................................................12 Follicular Growth Factors....................................................................................13 Ovulation.............................................................................................................15 Development and Function of the Corpus Luteum in Ruminants..............................16 Promoters of Angiogenesis..................................................................................17 Endocrine Regulators of CL Function.................................................................19 Autocrine and Paracrine Regulators....................................................................19 Luteolysis in Dome stic Ruminants.............................................................................21 Hormonal Regulation of Uterine PGF2 Synthesis.............................................23 Intracellular Signaling.........................................................................................24 Blood Flow and Vascular Changes.....................................................................25 Morphological Changes.......................................................................................26 PGF2 Inhibition of P4 Synthesis.........................................................................26 Immune-Mediated Events...................................................................................27 Tissue Metalloproteinases...................................................................................28 Apoptosis.............................................................................................................28

PAGE 6

vi Pregnancy Establishment in Ruminants.....................................................................28 Placentation in Ruminants...................................................................................30 Pregnancy and the Immune Response.................................................................30 PGHS-2 Gene Expression...................................................................................32 Shift in PG Production from PGF2 to PGE2......................................................33 Effects of Polyunsaturated Fatty Acids on Reproduction in Cattle............................34 Roles of Peroxisome Proliferator-A ctivated Receptors in Reproduction...................36 Roles of Mitogen-activated Pr otein Kinases in Reproduction...................................38 3 STUDIES ON THE EFFECTS OF EI COSAPENTAENOIC ACID ON PGF2 PRODUCTION IN CULTURED BOVINE ENDOMETRIAL CELLS....................41 Introduction.................................................................................................................41 Materials and Methods...............................................................................................43 Materials..............................................................................................................43 Cell Culture and Treatment.................................................................................43 PGF2 Radioimmunoassay..................................................................................45 RNA Isolation and Analysis................................................................................45 Statistical Analyses..............................................................................................46 Results........................................................................................................................ .47 Discussion...................................................................................................................48 Summary.....................................................................................................................50 4 STUDIES ON THE MECHANISMS OF ACTION OF EICOSAPENTAENOIC ACID ON PGF2 PRODUCTION BY CULTURED BOVINE ENDOMETRIAL CELLS........................................................................................................................58 Introduction.................................................................................................................58 Materials and Methods...............................................................................................59 Cell Culture and Treatment.................................................................................59 PGF2 Radioimmunoassay..................................................................................60 RNA Isolation and Analysis................................................................................61 Statistical Analyses..............................................................................................61 Results........................................................................................................................ .62 Discussion...................................................................................................................63 Summary.....................................................................................................................66 5 GENERAL DISCUSSION.........................................................................................81 LIST OF REFERENCES...................................................................................................88 BIOGRAPHICAL SKETCH...........................................................................................108

PAGE 7

vii LIST OF FIGURES Figure page 3-1 Experimental manipulations to exam ine the effect of eicospaentaenoic acid (EPA) on prostaglandin F2 (PGF2 ) response to phorbol-12,13-dibutyrate (PDBu) in bovine endometrial (BEND) cells..........................................................51 3-2 Experimental manipulations to study the effect of linoleic acid (LA) on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells........................................................................................52 3-3 Effect of eicosapentaenoic acid (EPA) on prostaglandin F2 (PGF2 ) response to phorbol-12,13-dibutyrate (PDBu) in bovine endometrial (BEND) cells.................53 3-4 Effect of eicosapentaenoic acid (E PA) on prostaglandin endoperoxide synthase (PGHS-2) mRNA response to phorbol-12,13dibutyrate (PDBu) in bovine endometrial (BEND) cells........................................................................................54 3-5 Effect of eicosapentaenoic acid (E PA) on peroxisome pro liferator-activated receptor (PPAR ) mRNA response to phorbol-12,13dibutyrate (PDBu) in bovine endometrial (BEND) cells............................................................................55 3-6 Effects of increasing n-6/n-3 fa tty acid ratios on prostaglandin F2 (PGF2 ) response to phorbol-12-13-dibutyrate (PDBu).........................................................56 3-7 Effects of increasing n-6/n-3 fatty acid ratios on prosta glandin endoperoxide synthase (PGHS-2) response to phorbol-12,13-dibutyrate (PDBu).........................57 4-1 Experimental manipulations to de termine if eicosapentaenoic acid (EPA)mediated inhibition of prostaglandin F2 (PGF2 ) secretion involves peroxisome proliferator-activated receptor (PPAR ) activation in bovine endometrial (BEND) cells............................................................................................................67 4-2 Experimental manipulations to determin e if peroxisome proliferator-activated receptor (PPAR inhibition may abolish eicosa pentaenoic acid (EPA) effect on prostaglandin F2 (PGF2 ) secretion in bovine endometrial (BEND) cells........68 4-3 Experimental manipulations to examin e the role of mitogen-activated protein (MAP) kinases in eicosapentaenoic acid (EPA) regulation of prostaglandin F2 (PGF2 ) production in bovine e ndometrial (BEND) cells........................................69

PAGE 8

viii 4-4 Effect of a peroxisome pr oliferator-activ ated receptor (PPAR agonist (L165,041) on prostaglandin F2 (PGF2 ) secretion in bovine endometrial (BEND) cells.......................................................................................................................... .70 4-5 Effect of a peroxisome pr oliferator-activ ated receptor (PPAR agonist (L165,041) on prostaglandin endoperoxide s ynthase (PGHS-2) mRNA abundance in bovine endometrial (BEND) cells........................................................................71 4-6 Effect of a peroxisome pr oliferator-activ ated receptor (PPAR agonist (L165,041) on PPAR mRNA response to PDBu in bovine endometrial (BEND) cells.......................................................................................................................... .72 4-7 Effect of sulindac sulfide on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.........................73 4-8 Effect of sulindac sulfide on prostagl andin endoperoxide synthase (PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.......................................................................................................................... .74 4-9 Effect of sulindac sulfide on peroxi some proliferator-activated receptor (PPAR mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells........................................................................................75 4-10 Effect of sulindac sulfone on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovi ne endometrial (BEND) cells.........................76 4-11 Effect of sulindac sulfone on prostagl andin endoperoxide synthase (PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.......................................................................................................................... .77 4-12 Effect of sulindac sulfone on peroxi some proliferator-activated receptor (PPAR mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells........................................................................................78 4-13 Effect of SB203580 and PD98059 on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.........................79 4-14 Effect of SB203580 and PD98059 on pr ostaglandin endoperoxide synthase (PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells........................................................................................80 5-1 Proposed model for eicosapentaenoic aci d (EPA) and linoleic acid (LA) regulation of prostaglandin F2 (PGF2 ) biosynthesis in bovine endometrial (BEND) cells6 5 -2 Signaling cascade leading to synthe sis and activation of prostaglandin endoperoxide synthase (PGHS-2) in bovine endometrial (BEND) cells.87

PAGE 9

ix ABBREVIATION KEY 9k-PGR PGE2-9-keto reductase AA Arachidonic acid AP-1 Activator protein-1 BEND cells Bovine endometrial cells CA Corpus albicans CL Corpus luteum CRE cAMP response element CREB cAMP response element binding protein DAG 1,2-diacylglycerol DGLA dihomo-linolenic DPA docosapentaenoic acid E2 Estrogen EPA Eicosapentaenoic acid ERK Extracellular-regulated kinase ET-1 Endothelin-1 FGF Fibroblast growth factor FSH Follicle-stimulating hormone FO Fish oil GH Growth hormone

PAGE 10

x GnRH Gonadotropin-releasing hormone GLA -linolenic acid IGF Insulin-like growth factor IGFBP Insulin-like grow th factor-binding protein IGFR Insulin-like growth factor receptor IFNInterferon IP3 Inositol-1,4,5-triphosphate JAK/STAT Janus kinase signa l transducer and activator of transcription LA Linoleic acid LH Luteinizing hormone LNA Linolenic acid LXR Liver X receptor MAP kinase Mitogen-ac tivated protein kinase MCP-1 Monocyte chemoattractant protein-1 MHC Major histocompatibility complex NE Norepinephrine OT Oxytocin P4 Progesterone PDBu Phorbol-12,13-dibutyrate PG Prostaglandin PGE2 Prostaglandin E2 PGF2 Prostaglandin F2

PAGE 11

xi PGFS Prostaglandin F synthase PGFM PGF2 metabolite (13,14-dihydro-15-keto prostaglandin F2 ) PGH2 Prostaglandin H2 PGHS-2 Prostaglandin endoperoxide synthase-2 PGI2 Prostaglandin I2 (Prostacyclin) PKC Protein kinase C PLA2 Phospholipase A2 PLC Phospholipase C PPARs Peroxisome prolif erators-activated receptors PPRE PPAR response element PUFAs Polyunsaturated fatty acids RAR Retinoic acid receptor RXR Retinoid X receptor StAR Steroidogenic acute regulatory protein TIMP Tissue inhibitor of metalloproteinases TNFTumor necrosis factor TPA Tissue-type plasminogen activator VEGF Vascular en dothelial growth factor

PAGE 12

xii Abstract of Thesis Presen ted to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science STUDIES ON THE EFFECTS AND MECHANISMS OF ACTION OF EICOSAPENTAENOIC ACID ON PGF2 PRODUCTION IN CULTURED BOVINE ENDOMETRIAL CELLS By Cristina Caldari-Torres December, 2005 Chair: Lokenga Badinga Major Department: Animal Sciences Recent studies have implicated n-3 pol yunsaturated fatty acids (PUFAs) in the reduction of prostaglandin F2 (PGF2 ) synthesis in the bovin e uterus. Although cattle diets contain a mixture of n-3 and n-6 fatty acids, currently th ere is a lack of information as to how these fatty acids may interact to alter PGF2 biosynthesis in the uterus. The objective of this thesis was to examine th e physiological effects and mechanisms of action of eicosapentaenoic acid (EPA; 20:5, n-3) on PGF2 production in phorbol-12,13dibutyrate (PDBu)-stimulated bovine endometrial (BEND) cells. Pre-incubation of confluent BEND cells with EPA for 24 h decreased PGF2 response to PDBu, but had no detectable effect on prosta glandin endoperoxide synthase -2 (PGHS-2) or peroxisome proliferator-activated receptor (PPAR ) mRNA abundance. The inhibitory effect of EPA on PGF2 secretion was reverted when increas ing amounts of linoleic acid (LA; 18:2, n-6) were added to the incubation medium.

PAGE 13

xiii Activation and inhibition of PPAR greatly reduced PGF2 and PGHS-2 mRNA responses to PDBu. The PPAR agonist, L-165,041, and sulindac sulfide decreased PGHS-2 mRNA response to PDBu, whereas EP A and sulindac sulfone had no detectable effects on PGHS-2 mRNA abundance in PDBu-stimulated BEND cells. Selective inhibition of p38 and ER K MAP kinases did not affect PGF2 and PGHS-2 mRNA responses to EPA, sugge sting that EPA may affect the PGF2 cascade at a site or sites that are dist al to these two MAP kinases. In summary, this study presents direct evidence that the i nhibition of uterine endometrial PGF2 biosynthesis by n-3 fatty acids de pends on the amount of n-6 fatty acids in the uterus, and that EPA and PPAR affect uterine PGF2 synthesis through complex mechanisms which may or may not involve PGHS-2 gene regulation. Further studies are needed to fully document th e role of p38 and ERK MAP kinases in endometrial PGF2 biosynthesis.

PAGE 14

1 CHAPTER 1 INTRODUCTION One of the most important aspects of the dairy cattle industry is the ability to maintain reproductively efficient animals. Th e inability of a dairy cow to have a normal reproductive cycle and become pregnant will lead to the culling of the animal, since failure to have a pregnant cow will resu lt in no milk production. Because of this, researchers have dedicated a lot of time a nd effort in studying the bovine reproductive cycle and investigating ways to make dairy cows as reproduc tively efficient as possible, which would help contribute to maximal milk revenue. Embryonic mortality is one of the reproducti ve problems associated with increased milk production by dairy cattle. It has been estimated that up to 40% of total embryonic losses occur between days 8 and 17 of pregnanc y, which is also the period of time when conceptus inhibition of uterine prostaglandin F2 (PGF2 ) occurs. Luteolysis in domestic ruminants is caused by an episodic release of PGF2 from the uterus that reaches the corpus luteum (CL) and causes its regression. If th e developing conceptu s is not able to inhibit the production of PGF2 the CL regresses, progesterone (P4) production decreases, and pregnancy cannot be mainta ined. Many researchers have focused on in vivo and in vitro studies to produce and evaluate syst ems that allow for increased embryo survival (Oldick et al., 1997; Moreira et al ., 2000; Binelli et al ., 2001; Badinga et al., 2002; Mattos et al., 2002, 2003). Results from se veral of these studies have shown that manipulating the polyunsaturated fatty acid (PUFA) content of the cows diet could help

PAGE 15

2 inhibit PGF2 production, but the mechanism of acti on of these fatty acids is not known. Polyunsaturated fatty acids are fatty acids with two or more double bonds. Fatty acids are unsubstituted monocarboxylic acids that occu r mainly as esters in natural fats and oils, and may also exist as non-esterifi ed forms known as free fatty acids. Polyunsaturated fatty acids can be divided in n-3 or n-6 fatty acids, depending on the position of the first double bond along the hydroc arbon chain. Those PUFAs with their first double bond at the third position of the hydrocarbon chain counting from the methyl end are classified n-3, while those with th eir first double bond at the sixth position are classified as n-6 PUFA s. Prostaglandin F2 is derived from the action of prostaglandin endoperoxide synthase-2 (PGHS-2) on arachi donic acid (AA), a PUFA that is cleaved from membrane phospholipids by the action of phospholipase A2 (PLA2). A study by Mattos et al. (2003) with bovine endometrial (BEND) cells s howed that n-3 fatty acids induced greater PGF2 inhibition than n-9 or n-6 fatty acids. It has been suggested th at PUFAs may inhibit PGF2 synthesis by decreasing the availability of AA, increasing the concentra tion of fatty acids that compete with AA for processing by PGHS-2, inhibiting PGHS-2 synt hesis and or activity, or affecting gene expression through activation of nuclear tran scription factors (Ma ttos et al., 2000). Peroxisome proliferator-activated receptors (PPARs) are an example of nuclear hormone receptors that mediate the effects of fatty acids and their derivatives at the level of gene expression. Three isoforms, each encoded by a different gene, have been identified: PPAR PPAR and PPAR Of these isoforms, PPAR is the only PPAR that has been associated with reproductive f unction (Berger et al., 2002).

PAGE 16

3 The objective of this thesis was to examine the physiological effects and mechanisms of action of eicosapentaenoic acid (EPA), an n-3 fatty acid, on PGF2 production in phorbol-12,13-dibutyrate (PDBu) -stimulated BEND cells. Phorbol-12,13dibutyrate (PDBu) is a phorbol-e ster that activates protein kinase C (PKC), a step necessary for prostaglandin (PG) production, and is therefore used to stimulate PGF2 production by cultured BEND cells. If EPA affects PGF2 production through the activation of PPAR then activation of this nuclear receptor would mimic the effect of EPA on PGF2 production in cultured BEND cells. Alternatively, if the aforementioned hypothesis is correct, inhibition of PPAR should block the effect of EPA on PGF2 production in cultured BEND cells. Also, if EPA exerts its PGF2 -inhibitory effect through the activation of th e mitogen-activated protein (MAP) kinases ERK or p38, which have been implicated in the activati on of PGHS-2, then inhi bition of the ERK and p38 MAP kinases should block the effect of EPA on PGF2 production in cultured BEND cells. This thesis will consist of a literature review (Chapter 2) that will focus on ruminants, specifically discussing their estr ous cycle, ovarian follicular dynamics, CL development and function, luteolysis, pregnanc y establishment, the effects of PUFAs on reproduction, and the role of PPARs in reproduction. Chapter 3 will present the experiments that were conducted to study the physiological e ffects of EPA on PGF2 production in cultured BEND cells, while Chapte r 4 will center on the studies conducted to elucidate the mechanisms of actio n of EPA in the inhibition of PGF2 production in cultured BEND cells. Chapter 5 will consis t of a discussion and concluding remarks.

PAGE 17

4 CHAPTER 2 LITERATURE REVIEW Estrous Cycle Estrous cycles provide animals with numerous opportunities for copulation and conception, allowing them to fulfill the ultima te goal in life, proc reation. This cycle varies in length from species to species, a nd the different types of estrous cycles can be categorized based on the number of estrous cycles per year and by whether they are regulated seasonally. In cattle, selection ha s favored a strategy of polyestrous cycles. This means that an animal has a uniform distribution of estrous cycles that occur regularly throughout the year. Seasonally po lyestrous females, like sheep and mares, display numerous estrous cycles only during ce rtain seasons of the year. These females can be divided further into long-day and short-day breeders. Long-day breeders begin to cycle as day length increases, while short-day breeders cycle when the length of daylight decreases. This is practical since long-day breeders, like ma res, tend to have a gestation length of almost a year, while a short-day breed ers gestation length is closer to 4 Mo. The gestation length and the seas onal regulation ensure that pa rturition occurs when there are enough natural resources and the season is favorable for the offspring. Animals such as dogs and wolves have monoestrous cycles meaning that they only have one cycle per year. To ensure that mating does occur, the estrus (heat) phase of the estrous cycle can last up to several days. Regardless of speci es, the estrous cycle is characterized by a series of predictable events that begin at es trus and end at the subsequent estrus. Among these events are alternating periods of se xual receptivity and non-receptivity. Sexual

PAGE 18

5 receptivity and mating are characteristic behavior of a female in the estrus stage of the estrous cycle. If concepti on does not occur, a new estrous cycle begins, but if it does occur, the animal goes into an extended period of sexual non-receptivity known as anestrus. Diestrus is the pe riod of sexual non-receptivity that characterizes the interval between periods of estrus (Heape, 1900). Regulation of the Estrous Cycle The estrous cycle is regulated by endocrin e and neuroendocrine mechanisms. This regulatory network includes gonadotropin releasing hormone (GnRH), folliclestimulating hormone (FSH), lutein izing hormone (LH), progesterone (P4) and estradiol (E2). These hormones are interconnected by the hypothalamic-pituitary-ovarian axis, which controls follicular growth, ovulation, and luteal function. Follicle-stimulating hormone and LH bind to specific receptors on the ovary and stimulate follicular development to produce a mature and compet ent egg (Driancourt, 2001). Gonadotropins released by the pituitary gland are influenced by changes in rate of GnRH synthesis and degradation. These endocr ine changes lead to marked morphologic and secretory changes in the ovaries and tubular genitalia of the animal such as (i) changes in vaginal and uterine cytology, (ii) changes in cervical tonus and water content, (iii) increases in blood supply, and (iv) changes in follicular development. Characteristics of the Bovine Estrous Cycle The length of the bovine estrous cycle is 20-21 d with heat or estrus occurring on day 1 and ovulation occurring 10-12 h after the end of standing estrus, or 24-30 hours after the onset of estrus (H afez and Hafez, 2000). Mean dur ation of estrus is 15 hours, but can range anywhere from 6 to 24 hours (Senger, 1997). Ovulation in bovids is spontaneous, meaning that the fe male periodically becomes se xually receptive. Selection

PAGE 19

6 has favored polyestrus females, which have estrous cycles that occur with a uniform distribution throughout the year, allowing them to become pr egnant without regard to season. In bovids, copulation usually occurs before ovulation, provi ding an opportunity for conception. Commonly only one follicle ovulates, although about 10% of the time two follicles may ovulate (Hafez and Hafez, 2000). Follicles ovulate on the right ovary about 60% of the time and on the left ovary about 40% of the time (Hafez and Hafez, 2000). Phases of the Estrous Cycle The estrous cycle is divided into a short fo llicular phase and a longer luteal phase. Each of these phases is dominated by one of the gonadotropic hormones, P4 or E2. The period of time from the regression of the corpus luteum (CL) to ovulation is known as the follicular phase. During this phase, pre ovulatory Graafian follicles are the primary ovarian structures. Th ese produce estradiol-17 E the dominant reproductive hormone. High E2 concentrations lead to behavioral alterations and major physiological changes in the reproductive tract. This st eroid is responsible for the behavioral characteristics observed in cows that are appr oaching or have reached sexual receptivity. High E2 concentrations at estrus lead to a de crease in the viscosity of cervical mucus, dilation of the cervix, improved contractility or tonicity of the uterus and increase in vascular growth of the endometrium (Haf ez and Hafez, 2000). These characteristic physiological changes favor c opulation and fertilization. The luteal phase of the estrous cycle is longer than the follicular phase and takes place from the time of ovulation to regression of the CL. The CL the dominant structure of the luteal phase, produces P4. Progesterone has the opposite physiological and

PAGE 20

7 behavioral effects of E2, causing the cow to be in a phase of sexual non-receptivity. This steroid is essential for pregnancy and, ther efore, is known as the pregnancy hormone. Progesterone induces physiol ogical changes that prepare a suitable environment for embryo development and attachment to the endometrium. When P4 is high, cervical mucus is thick, the cervical canal is tightly closed and the myometrium is relaxed (Hafez and Hafez, 2000). Thick cervical mucus and a ti ghtly closed cervical canal help restrict microorganisms from the upper reproductive tract to maintain a clean environment for the embryo, while the quiescent uterus allows implantation of the embryo. Stages of the Estrous Cycle The estrous cycle can be divided further in to the following four stages: proestrus, estrus, metestrus, and diestrus. Proestrus is the stage that begins after luteolysis as P4 starts to decrease and ends at the onset of estrus. This stage is characterized by the formation of ovulatory fo llicles that produce E2 and lead to a shift from the P4-dominated to the E2-dominated phase. During proestrus, th e females reproductive tract undergoes physiological changes that prepare it for the onset of estrus and mating. Estrus is an easily recognizable stage since the high E2 levels in blood lead to visible behavioral symptoms in the animal. As the animal enters estrus, she displays a behavior which indicates that she is appr oaching sexual receptivity such as increased locomotion, mounting of other females, nervousness and phonation (Senger, 1997). When fully sexually receptive, the female di splays standing estrus, and in some cases lordosis (arching of back in preparation for mounting). From the time of ovulation to the formation of a functional CL, the animal is in the metestrus stage of the estrous cycle. This stage is characterized by a decrease in E2 levels and an increase in P4 levels. Once the follicle ovulates, it undergoes cellular and

PAGE 21

8 structural remodeling that re sult in the formation of the CL, the dominant intraovarian structure of the luteal phase of the estrous cycle. This endocrine gland produces P4, causing this hormone to be de tected soon after ovulation. The longest stage of the estrous cycle is diestrus. During this stage, the CL (or corpora lutea in the rare case of more than one fo llicle ovulating) is fully functional and produces the maximal amount of P4. The duration of diestrus depends directly on the amount of time that the CL remains functional, meaning that if conception occurs, the CL will persist throughout pregnancy and so will dies trus. On the other hand, if there is no mating or conception, luteolysis will take pl ace, diestrus will e nd, and a new estrous cycle will begin. Anestrus A female that does not exhibit regular estr ous cycles is referred to as being in anestrus. Anestrus can occur if there is a failure to stimulate and maintain gonadotropin secretion due to insufficient GnRH releas e from the hypothalamus (Senger, 1997). The lack of gonadotropins prevents the formation of ovulatory follicles or a functional CL. Anestrus can be caused by pregnancy, pr esence of offspri ng, stress, pathology, undernutrition, or season (Senger, 1997). Anestrus caused by pregnancy is a direct result of endocrine factors, such as continuous P4 secretion mainly by the placenta (Hafez and Hafez, 2000). Once parturition occurs, P4 concentrations decrease dramatically, and a new estrous cycle can take place. It is important to note that for anestrus to end there needs to be not only parturition, but uterine involuti on, since the animal needs to have a normal uterus before it goes into estrus again and becomes pregnant.

PAGE 22

9 The extent of postpartum anestrus can al so depend on mammary stimulation by the offspring. Frequent nursing leads to an incr ease in serum concentra tions of prolactin. Serum prolactin concentration is inversely re lated to FSH and LH c oncentrations (Hafez and Hafez, 2000), meaning that frequent nursi ng leads to a decrease in the concentration of the hormones responsible for follicle grow th and ovulation. Weaning of offspring can result in short estrous cycles in postpar tum anestrous females. Almost 80% of postpartum anestrous cows that had their calf weaned and exhibited estrus within 10 days of weaning had estrous cycles of about 7-12 days in length (Ramrez-Godinez et al., 1982). There is a very close relationship between fertility and body condition and nutritional status of females, especially in dairy cattle. Undern utrition and a prolonged negative energy balance can si gnificantly extend the length of postpartum anestrus in these animals (Beam and Butler, 1999). Studies with cycling heifers showed that severe restriction of energy intake can induce nutr itional anestrus when body weight decreases by 15% (Rhodes et al., 1996). In high pr oducing dairy cows, the amount of milk production is related to the amount of time fr om parturition to th e cows first ovulation (Hafez and Hafez, 2000). This is related to the energy balance of the animal, since to maintain high milk production, the cow need s to mobilize fat from its body reserves. If the cow is in a state of negative energy bala nce, it will use its energy primarily for milk production and reproductive processes will be negatively affected. Undernutrition, caused by under consumption of nutrients befo re or after calving, may interfere with follicle maturation, ovulation, or other mechanisms regulating ovarian follicular dynamics (Jolly et al., 1995).

PAGE 23

10 It is also worth noting that after parturition, the incide nce of ovulations where the cow does not exhibit any behavioral charac teristics of estrus approximates 50-95% (Hafez and Hafez, 2000). Most dairy cows have normal fertile estrous cycles by around 35 days postpartum. In cows that ovulate ear ly after calving (such as pluriparous cows), the conception rate is lower when bred at fi rst postpartum estrus than at subsequent estrous cycles (Hafez and Hafez, 2000). Ovarian Follicular Dynamics in Ruminants In ruminants, ovarian follicular dyna mics are regulated by the hypothalamicpituitary-ovarian axis. This axis initia tes and regulates the reproductive processes through neuro-endocrine systems. Hormones produced by the hypothalamus stimulate (or inhibit) the pituitary gland which releas es hormones in an e ndocrine fashion, which bind to receptors in the ovaries leading to morphological an d endocrine changes in these tissues. These changes cause the ovarie s to go through several follicular waves, ultimately selecting a dominant follicle that will produce more E2 than the subordinate follicles, and which will eventually ovulate. After ovulation, a CL is formed, and the luteal phase of the estrous cycle begins. Th e presence or absence of an embryo creates a hormonal environment that leads to either the maintenance or regression of the CL, respectively. Regression of the CL (luteo lysis) permits the ovulation of the dominant follicle of the existing follicular wave, and the cycle starts again. Although much of the literature focuse s on the gonadotropins as the major hormonal players in follicular dynamics, ne w research has established that polypeptide growth factors such as insulin, insulin-like gr owth factor (IGF), and insulin-like growth factor binding proteins (I GFBPs) are key for oocyte maturation and ovulation, and for establishment of the dominant follicle.

PAGE 24

11 Hypothalamic-pituitary-ovarian axis Regulation of follicle development includes both positive and negative endocrine feedback, occurring at the leve l of the pituitary and hypotha lamus. The hypothalamus is connected to the pituitary gland throu gh the hypothalamic-hypophyseal portal system, which consists of a capillary plexus that can transport GnRH released by specialized hypothalamic nerve cells into the anterior pitu itary, and also a neur al connection to the posterior pituitary, where the hormone oxyt ocin (OT) is stored. The hypothalamic hormone GnRH is released in pulses into the portal system in response to neural signals, where it signals the release of FSH and LH from the anterior pituitary (Hafez and Hafez, 2000). The gonadotropins FSH and LH are resp onsible for the growth and selection of the dominant follicle, estrogen secretion from the follicle, and for the ovulation of the follicle, respectively. Through the hypothalami c-pituitary-ovarian axis, the pituitary hormones can exert a negative feedback to regulate the f unctions of the hypothalamus (Hafez and Hafez, 2000). Studies performed by Karsch et al. (1980; 198 7) suggest that in ovids E2 and P4 have a negative feedback effect on LH pulse amplitude and frequency, respectively, but that if E2 levels become sufficiently hi gh the negative feedback effect becomes interrupted, and a positive feedback effect is observed causing an LH surge to occur. The negative feedback effects of E2 and P4 on gonadotropin secretion in the ewe include an action on the brain and a consequent inhibiti on of pulsatile GnRH secretion (Karsch et al., 1987). The positive feedback effect observed when E2 levels become sufficiently high allows for pulsatile Gn RH secretion from the hypothalamus which results in the surge of LH n ecessary for follicle ovulation.

PAGE 25

12 As follicles grow, they attain steroidogenic capabilities, which allow them to produce estrogens, E2 being the primary biologically ac tive estrogen produced (Hafez and Hafez, 2000). Luteal cells of the CL are also steroidogenic, but produce P4 primarily. The CL is composed of large a nd small luteal cells. The la rge luteal cells spontaneously secrete P4 at a high rate, whereas the sma ll luteal cells secrete little P4 in the absence of LH stimulation (Hafez and Hafez, 2000). Proge sterone has a negative feedback effect on the hypothalamus, which prevents it from releasing GnRH at a high amplitude and frequency (Hafez and Hafez, 2000), which in tu rn causes a very low release of LH, the hormone necessary for follicle ovulati on. When luteolysis occurs and P4 levels are low, the negative feedback effect on the hypothalamus is removed and a new follicular wave can occur. Follicular Waves in Cattle In bovids, selection of an antral follicle th at will eventually ov ulate usually occurs in 2 or 3 waves. The ovary has a primordial fo llicle reserve, formed during fetal life or soon after birth (Hafez and Hafez, 2000), fr om which follicles are continuously being selected to grow and mature. Follicle growth refers to the proliferation and differentiation of theca and granulosa cells, whic h leads to an increased ability to produce E2 and respond to gonadotropins (Hafez and Hafez, 2000). Follicle growth, which is slow and gonadotropin-independent initia lly (Scaramuzzi et al., 1993), becomes gonadotropin-dependent during the final stages of maturation. Transient FSH rises occurring at regular 8-10 day intervals in cattle (Mihm et al ., 2002) lead to the emergence of up to 24 small (3-5 mm) follicles that grow beyond 4 mm in diameter (Adams et al., 1992). Follicles that are smaller than 4 mm require FSH to grow, while large antral follicles (7-9 mm) require LH for maturati on (Hafez and Hafez, 2000). These sequential

PAGE 26

13 FSH rises associated with new follicle wave s occur not only during the estrous cycle but also in the postpartum period (Crowe et al ., 1998; Stagg et al., 1998), and before puberty in cattle (Evans et al., 1994). As FSH concentrations start to decline, one follicle becomes the dominant follicle, whereas the rest of the follicles undergo atresia via apoptosis (Austin et al., 2001). The dominant follicle exhib its enhanced growth and E2 production, becomes FSH-independent a nd LH-dependent, and maintains FSH concentrations low to prevent the growth of subordinate foll icles (Ginther et al., 1997; 1999). The dominant follicle co ntinues to grow and produce E2 for 3 to 4 days, but as the CL produces high levels of P4, LH pulses are altered to a low frequency, high amplitude pattern causing the dominant follicle to become atretic (Sunderland et al., 1994; Evans et al., 1997). As the follicle becomes at retic, it starts to secrete less E2 and eventually loses dominance. When the dominant follicle regresses, there is another transient FSH rise, which triggers the emergence of a new wave of follicles and selecti on of a new dominant follicle (Sunderland et al., 1994). Following luteolysis, P4 levels decrease, removing this hormones negative feedback effect on the hypotha lamus. This allows LH to be secreted from the pituitary gland. A frequent LH pul se pattern during the fo llicular phase supports the existing dominant follicle, allowing it to undergo final differentiation, induce the gonadotropin surge, ovulate and lu teinize (Mihm et al., 2002). Follicular Growth Factors Although FSH and LH have historically b een defined as the factors stimulating follicular and luteal growth and differentiat ion, more recent research has revealed that other factors are also involved. These include inhibins, activin, follistatin, IGF, and IGFBPs.

PAGE 27

14 Inhibins are proteins that exert an FSH-suppressive action on the pituitary. Bovine follicular fluid contains dimeric forms ( and subunits) of inhibins (Good et al., 1995). Activin is a dimer linking two inhibin subunits and the inte raction of activin and its receptor is regulated by follistatin, which neutra lizes activin functions in the pituitary and ovary (Robertson et al., 1987; Findlay, 1993). Recruited follicles produce inhibins, causing FSH levels to decrease and LH levels to increase. During the first 33 hours of FSH decline, an increase in the larger molecu lar weight inhibins is seen in the fastest growing cohort follicles, and activin increases in the largest follicle (Austin et al., 2001). In the next 24 hours, the two largest follicles maintain high levels of the high molecular weight inhibins, but the larger of the two shows a reduction in follistatin concentrations (Austin et al., 2001). The transient FSH rise will result in enhanced growth and E2 synthesis in successful follicles, while folli statin and the amounts of the native 34 kDa inhibin dimer are kept at low concentrations (Mihm et al., 2002). The effects of IGFs on folli cular granulosa cells are de pendent on the concentration of FSH. In vitro experiments have shown that IGFs stimulate proliferation of follicular granulosa cells, and enhance E2, inhibin and activin synthe sis (Glister et al., 2001). Insulin-like growth factor binding proteins bind IGF with high affinity, making IGF unavailable to the IGF recepto r. High intrafollicular amoun ts of IGFBP-2, -4 and -5 (lower molecular weight) is negatively co rrelated with follicle estrogen activity (Echternkamp et al., 1994), and atretic bovine follicles have been shown to have increased concentration of these IGFBPs (Mazerbourg et al., 2001). Low amounts of IGFBPs are maintained in the dominant follicle, and this follicle demonstrates enhanced IGF bioavailability as a result of higher IGF-II synthesis and IGF-I binding, while

PAGE 28

15 subordinate follicles show increased levels of IGFBP-4 and -5 (de la Sota et al., 1996; Stewart et al., 1996; Mihm et al., 1997; Aus tin et al., 2001). As the dominant follicle loses dominance, there is an increase in low molecular weight IGFBPs and an increase in the incidence of granulosa cell apoptosis (de la Sota et al., 1996; Stewart et al., 1996; Manikkam and Rajamahendran, 1997). Ovulation As the dominant follicle grows, it attains LH receptors, which allow it to bind LH released from the pituitary gland. When this gonadotropin binds to its receptor, a cascade of intracellular events is initiated whereby cholesterol is converte d to testosterone. Testosterone diffuses out of the theca interna cells and enters the gr anulosa cells where it is converted to E2 (Senger, 1997). The E2 concentrations increase until they reach a threshold that allows the LH surge to occur. After the LH surge, the theca interna cells start to produce P4 instead of testosterone, which i nduces the production of collagenase, gelatinase, and stromelysin, enzymes known as metalloproteinases, which aid in the remodeling of the extracellular matrix. Cons equently the outer c onnective tissu e of the ovary ( tunica albuginea ) is digested while granulosa cells increase secretion of follicular fluid, causing the apex of the follicle (stigma) to push outward and weaken (Senger, 1997). Prostaglandins produced by the ovary cau se contraction of sm ooth muscle of the ovary, increasing local pressure and forcing the stigma to pr otrude even more (Senger, 1997). During ovulation, rupture of follicular bl ood vessels leads to the formation of the corpus hemorrhagicum or bloody body. Once ovulati on occurs, prostaglandin E2 (PGE2) helps remodel the follicle into the CL, a process that starts with the dramatic infolding of the follicular wall that facilitate s migration of fibroblasts, endothelial cells, and theca interna cells into the central regions of the developing CL (OShea et al., 1980).

PAGE 29

16 The breakdown of the basement membrane th at separates the avascular granulosa layer from the theca interna layer further facilita tes tissue remodeling a nd cellular migration during the luteinization pro cess (OShea et al., 1980). Development and Function of th e Corpus Luteum in Ruminants The CL is a transient endocrine gland co mposed of endothelial cells, steroidogenic large and small luteal cells, fibroblasts, smoot h muscle cells and imm une cells (OShea et al., 1989). The major hormones produced by this reproductive gland are P4, and to a lesser extent OT, which is stored in secr etory granules of la rge steroidogenic cells (McCracken et al., 1999). Progesterone is ne cessary for the initiati on and maintenance of pregnancy because it induces a quiescent stat e of the myometrium (Csapo and Pulkkinen, 1978) and suppresses the maternal immune re sponse to fetal antig ens (Szekeres-Bartho, 1992). During early pregnancy, the CL is the primary source of this hormone. Oxytocin, on the other hand, contributes to the luteolytic process if preg nancy is not established. In domestic animals such as ruminants, the ma jor luteotropin (substance that promotes the growth of the CL and P4 production) is LH (McCracken et al., 1999), although during early CL development, growth hormone (GH) seems to be the major player affecting P4 secretion from the CL in vivo (Miyamoto et al., 1998) and in vitro (Kobayashi et al., 2001). The CL grows very rapidly, and a substa ntial increase in mass is apparent as early as two days after ovulation (Fields et al., 1996). This increase in mass is due to hypertrophy of granulosa and th eca cells and repeated mito sis of steroidogenic cells, endothelial cells, and fibroblasts (Acosta et al., 2004; McCracken et al., 1999). Even though the CL has been the subject of intens e research for decades, many of the luteal regulatory mechanisms involved are not comp letely understood. The gonadotropins and growth hormones are the primary regulators of CL function, although other factors of

PAGE 30

17 intraand extraovarian origin have the poten tial to modulate the lo cal response to these hormones or to have direct specific functions (Schams and Berisha, 2004). This literature review will focus on angiogenesis promoters necessary for CL formation, and endocrine, autocrine, and paracrine regulators of CL function. Promoters of Angiogenesis As final follicular growth, ovulation and CL development occur, ovarian tissue is remodeled, and this tissue remodeling is associ ated with the occurrence of hemodynamic changes. The developing CL is characteri zed by highly active vascul arization (Acosta et al., 2004), and the majority of the steroidogenic ce lls of the mature CL are in contact with one or more capillaries (Re ynolds et al., 1992). Angiogenesi s (the development of new blood vessels from pre-existing blood vessels) in the developing CL needs to be tightly controlled since this process maintains th e delicate balance be tween promoters and inhibitors of angiogenesis (Schams and Beri sha, 2004). Factors th at regulate luteal angiogenesis include vascular endothelial growth factor (VEGF), acidic and basic fibroblast growth factor (FGF-1 and FGF-2), IGF-I, and IGF-II. All components of the VEGF system ar e found in the bovine CL (Garrido et al., 1993; Berisha et al., 2000). The luteal tissue seems to express predominantly the smallest and secretory isoforms of th e VEGF system (Schams and Berisha, 2004). The highest VEGF and VEGF receptor (VEGFR) mRNA and pr otein expression were detected at the time of angiogenesis (early luteal developm ent) (Berisha, 2000). Schams and Berisha (2004) stated that these findings suggest th at VEGF may act as a chemoattractant for sprouting endothelial cells (EC) since angiogenesis i nvolves the migration and proliferation of EC from the pre-existing ve ssels. Luteinizing hormone, IGF-I, and tumor

PAGE 31

18 necrosis factor (TNF) are some of the hormones that stimulate VEGF mRNA and protein expression in bovine granulosa cells (Schams and Berisha, 2004). Fibroblast growth factor-1 (acidic) mRNA expression in the bovine CL increases significantly during the midluteal stage (Schams and Berisha, 2002), while FGF-2 (basic) and FGF receptor (FGFR) expression is highest during the early luteal stage (Schams and Berisha, 2004). Fibroblast growth factor-2 concentrations in the bovine CL decrease during the mid-luteal stage and in crease during the late-luteal stage and these changes are accompanied by changes in the pr otein localization from the cytoplasm of capillary EC and smooth muscle cells of arteri es to the cytoplasm of luteal cells (Schams and Berisha, 2004). The predominant localizatio n of FGF-2 in EC during the early stages of CL development suggests that this is a dominating factor fo r endothelial growth (Gospodarowicz et al., 1985). Many researchers have demonstrated that the bovine CL expr esses IGF-I mRNA (Einspanier et al., 1990; Kir by et al., 1996; Perks et al ., 1999; Schams et al., 1999; Schams and Berisha, 2002). The highes t mRNA expression for both IGF-I and IGF-II was observed during the early luteal phase, followed by a decrease to a lower cyclic plateau (Schams and Berisha, 2004). Insu lin-like growth fact or binding protein-3 expression is positively correlated with IG F-I, -II, and IGF receptor-1 (IGFR-1) expression, and it has even been postulated that IGFBP-3 serves as a carrier and storage reservoir for IGFs within the intravascular compartment (Schams and Berisha, 2004). As was mentioned before, IGFs have been show n to stimulate VEGF mRNA, suggesting that the IGF system may have an indirect effect on angiogenesis. The fact that IGF-II is distinctly localized in pericytes (undiffe rentiated mesenchymal-like cells that may

PAGE 32

19 become fibroblasts, macrophages, or smooth-musc le cells) suggests a direct role of the IGF system in angiogenesis and capillary st abilization (Amselgrube r et al., 1994), since pericytes play an important role in th e modulation of endothe lial migration and proliferation (Orlidge and DAmore 1987; Antonelli-Orlidge et al., 1989). Endocrine Regulators of CL Function An endocrine regulator is one that is produced by a gland and acts on a distant organ, tissue, or gland by tr aveling through the bloodstream. The classical and primary endocrine hormones that support the developm ent and function of the CL are LH and GH. The CL expresses receptors for both of these hormones. An increase in LH receptor mRNA during th e mid-luteal phase is followed by an increase in LH receptors (Kobayashi et al., 2001 ). Most of the LH receptors are located on small luteal cells (Schams and Berisha, 2004), and LH stimulates these cells to produce P4 (Niswender and Nett, 1988). In fact, a classic study by Hoffman et al. (1974) showed that treatment of heifers with an LH antiserum brought P4 levels back to basal levels within a short period of time. Receptors for GH are found mainly in large lu teal cells (Lucy et al., 1993; Koelle et al., 1998). These cells are responsible for 80% of P4 produced by the CL (Niswender et al., 1985). Growth hormone stimulates th e CL to produce and secrete OT and P4 (Lieberman and Schams, 1994), and it has been shown that the latt er hormone and the luteolysin, PGF2 are more strongly stimulated by GH than by LH (Kobayashi, et al., 2001). Autocrine and Paracrine Regulators Autocrine regulators are those that exert their e ffect on the same gland that produce them, while paracrine regulators refer to thos e that exert their eff ect on cells neighboring

PAGE 33

20 the gland that produce them. Autocrine and paracrine regulators of the CL include growth factors, peptides, st eroids and prostaglandins. Angiogenic growth factors can also stimulate CL function by stimulating secretion of P4 and OT. Insulin-like growth facto r-I and IGF-II are poten t stimulators of P4 and OT (Einspanier et al., 1990; Sauerwin et al., 1992). Fibroblast growth factor-2 mRNA and protein expression have been demo nstrated in luteal cells (Schams et al., 1994), and in vitro studies have shown that it stimulates P4 and OT in a dose-dependent manner (Miyamoto et al., 1992; Liebermann et al., 1996). Ovarian peptides regulating CL functi on are OT, angiotensi n II (Ang II), and endothelin-1 (ET-1). Oxyt ocin is localized in large and sm all luteal cells (Kruip et al., 1985) and it has been shown to have a modul ating action on the lute otrophic effects of LH on P4 secretion by the CL (Schams et al., 1995). Studies suggest that during the development of the CL, OT may play an impor tant role as a lutetropic factor (Schams, 1996). Ang II and other growth factors that induce angiogenesis and support P4 production by large luteal cells seem to modulate the sprout of new blood capillaries needed to support luteal cell developmen t (Kobayashi et al., 2001). Th ere are conflicting results regarding the effects of Ang II on P4 production by luteal cells. Studies by Kobayashi et al. (2001) suggested that this peptide together with PGF2 greatly stimulated P4 secretion by the early CL, while studies with bovine CLs (Girsh et al., 1996; Miyamoto et al., 1997; Hayashi and Miyamoto, 1999) s uggested that it inhibited P4 production. It is conceivable that the inte raction of Ang II and PGF2 is necessary to induce stimulation of P4 production and secretion.

PAGE 34

21 There is evidence of ET-1 expression in the bovine CL during the estrous cycle and pregnancy (Berisha et al., 2002). The st udies that noted an inhibition of P4 secretion by Ang II in the bovine CL suggest that ET-1 ma y also have an inhibitory effect on P4 secretion (Girsh et al., 1996; Miyamoto et al., 1997; Hayashi and Miyamoto, 1999). Both P4 and norepinephrine (NE) have effects on the synthesis of hormones that regulate the CL. Proge sterone inhibits PGF2 secretion by the CL during the mid-luteal phase (Pate, 1988; Skarzynski and Okuda, 1999), and it also plays a luteotrophic role by stimulating the synthesis of LH receptors (Jones et al., 1992). Norepinephrine is known to stimulate the synthesis and release of PGF2 and PGE2 in bovine luteal cells (Skarzynski et al., 2001). Luteal PGF2 has distinct effects on CL func tion depending on the stage of the luteal cycle. At the early and mid-luteal phases it has a luteotrophic effect, which ceases at the late luteal stage desp ite its high local production (M iyamoto et al., 1993). Ovarian estradiol, OT and P4 appear to be the major regulators of the synthesis and secretion of luteal PGF2 during the estrous cycle (Grazul et al., 1988; Okuda et al., 2001). Luteolysis in Domestic Ruminants Luteolysis is the process whereby the CL regresses and loses its capability to produce P4. Luteolysis in domestic ruminants is caused by an episodic release of PGF2 from the uterus that reaches the CL by a c ountercurrent system between the uterine vein and the ovarian artery (Schams and Beri sha, 2004). Through this system, PGF2 from one of the uterine horns reach es its ipsilateral ovary a nd causes the structural and functional demise of the CL (Ginther, 1974). This countercurrent system is especially necessary in ovids because without it PGF2 would have to travel through the pulmonary

PAGE 35

22 circulation where most of it (>99%) would be metabolized after one passage through the lungs (Davis et al., 1980). Because only about 65% of PGF2 is enzymatically inactivated after one passage through the lungs in bovids (Davis et al ., 1985), it is thought that this hormone may also work through the sy stemic circulation in cows. Luteolysis is generally divided into functional and structur al luteolysis. During functional luteolysis, the CL loses its ability to synthesize and secrete P4 (McGuire et al., 1994), whereas during structural luteolysis the cellular components of th e CL are lost (Pate, 1994). Historically, it has been reported that luteolysis is initiated when E2 released from the preovulatory follicle triggers the release of OT from the pituitary gland. This causes the release of small quantities of PGF2 from the uterus (Fai rclough et al., 1980), which has a positive feedback effect that leads to the release of additiona l luteal oxytocin and PGF2 from the CL and uterus (Silvia et al., 1991). However, evidence is accumulating that suggests that OT is not e ssential for the initiation of PGF2 output during luteolysis (Kotwica et al., 1997; Kotwica et al., 1998). In blood (Par kinson et al, 1992) and in intact (Parkinson et al, 1992) and microdi alyzed bovine CL (Dougl as et al., 2000), OT concentrations are extremely low at the ti me of luteolysis and a study performed by Kotwica et al. (1997) demonstrat ed that blocking uterine oxyt ocin receptors had no effect on luteolysis or the duration of the estrous cy cle in heifers. Estradiol stimulates the activity of enzymes that control prostaglandi n (PG) synthesis in the uterine endometrium (Ham et al., 1975), and priming the uterus with P4 enhances E2-stimulated production of PGF2 in ruminants (Barcikowski et al., 1974; Le wis and Warren, 1977). Luteolysis is a complex process that involves many morphol ogical, structural, and molecular changes that allow the degradation of the CL so that a new estrous cycle can take place. Some of

PAGE 36

23 the aspects of luteolysis that will be discu ssed in this section are hormonal regulation of uterine PGF2 synthesis, changes in intracellu lar signaling, changes in blood flow, changes in morphology, effect of PGF2 on P4 synthesis, immune-mediated events, tissue metalloproteinases in the CL, a nd apoptosis of luteal cells. Hormonal Regulation of Uterine PGF2 Synthesis The endometrium is the s ite of luteolytic PGF2 synthesis, specifically the intercaruncural region of the surface epithelium (Kim and Fortier, 1995; Asselin et al., 1998). Synthesis of this luteolysin is stimulated by E2, the hormone that stim ulates the activity of enzymes involved in PG synthesis and also by P4, which has been found to have a priming effect on the uterus when administered before E2. It is thought that the priming effect of P4 is due to the accumulation of lipids in the endometrium (Boshier et al., 1981). Lipid accumulation is important because PGF2 is derived from arachidonic acid (AA), an omega-6 (n-6) fatty acid found in the phospho lipid membrane of th e cell. Release of this fatty acid by phospholipase A2 (PLA2) is the rate-limiting step in PGF2 synthesis (Kunze and Vogt, 1971). The released AA is converted to PGH2 by the enzyme prostaglandin endoperoxide synthase-2 (PGHS-2; DeWitt and Smith, 1995), also known as COX-2, an enzyme that contains cy clooxygenase and endoperoxidase activities (Wlodawer et al., 1976). Prostaglandin H2 (PGH2) is an endoperoxide (unstable, biologically active molecule) that is further acted upon by the enzyme PGF synthase (PGFS) to yield PGF2 (Watanabe et al., 1985). An alternate pathway for the formation of PGF2 is through PGE2-9-keto reductase (9K-PGR), an NADPH-dependent enzyme that catalyzes the conversion of PGE2 to PGF2 (Asselin and Fortier, 2000). Progesterone is also thought to increase the concentration and activity of PGFS, further enhancing

PAGE 37

24 PGF2 synthesis (Eggleston et al., 1990). The positive feedback effect of the low levels of uterine PGF2 on luteal PGF2 can be attributed to the increase in availability of AA and PGHS-2 activity in th e CL in response to PGF2 (Niswender et al., 2000). The major point of control of E2 and P4 in PGF2 synthesis in the endometrium is through the regulation of endometrial OT receptor concentr ations, and it has been postulated that in the intact cycling sheep, E2 enhances the formation of OT receptors in the endometrium, but that during the luteal phase, P4 blocks the formation of endometrial OT receptors (McCracken et al., 1995). It al so has been postulated that P4 exerts a nongenomic regulation of OT receptors by binding them with high affinity, therefore preventing the binding of OT to its own receptor (Grazzini et al., 1998). At the end of the luteal phase, prolonged exposure of the uterus to P4 produced by the CL leads to a downregulation of the P4 receptor (Vu Hai et al., 1977), and the action of E2 is no longer suppressed. This leads to the E2-induced formation of endometrial OT receptors, and a large amplification of endometrial PGF2 secretion induced by OT at the e nd of the luteal phase which leads to the regression of the CL (Niswender et al., 2000). Intracellular Signaling High-affinity receptors for PGF2 are localized to the larg e luteal cells of the CL (Fitz et al., 1982). Binding of PGF2 to its receptor leads to the activation of membrane bound phospholipase C (PLC) (Berridge et al., 1 984), which catalyzes the hydrolysis of phosphatidylinositol-4,5-bisphosphate to inositol-1,4,5-tr isphosphate (IP3) (Davis et al., 1985) and 1,2-diacylglycerol (DAG) (Berridg e et al., 1984). As the amount of IP3 increases in the cytoplasm, it causes the release of free Ca2+ from the smooth endoplasmic reticulum of the cell into the cytoplasm (Berridge et al., 1984). Free

PAGE 38

25 intracellular Ca2+ activates phospholipase A2 (PLA2) (Flower and Blackwell, 1976), which cleaves AA from membrane phospholipids. Also, free Ca2+ and DAG in the plasma membrane stimulate the Ca2+-dependent protein kinase C (PKC) (Nishizuka, 1986), which is believed to mediate many of the antisteroidogenic actions of PGF2 through posttranslational modification of cellu lar proteins such as those involved in steroidogenesis (McGuire et al., 1994) and c holesterol biosynthesis (Behrman et al., 1971). It is also possible th at PKC mediates apoptosis of large luteal cells, because activation of this enzyme induces the expres sion and activation of proteins involved in apoptosis in other cell types (Sch wartzman and Cidlowski, 1993). Blood Flow and Vascular Changes A steep decrease in blood flow to the CL has been postulated as one of the main luteolytic actions of PGF2 Studies with ewes demonstr ated that administration of PGF2 caused a reduction in blood flow to the CL concurrent with a reduction in P4 secretion (Niswender et al., 1973; Nett et al., 1976). This de cline in blood flow caused by PGF2 can lead to luteolysis by means of a re duction in luteotropic support, nutrients, and substrates for steroidogenesis reachi ng the CL (Phariss et al., 1970). Also, vasoactive peptides like ET-1 have been implicated as having a direct luteolytic role. Endothelin-1 is a strong vasoc onstrictor (Huggins et al., 1993) that is thought to mediate the inhibitory effect of PGF2 on luteal cells. The PGF2 stimulates endothelial cells of the CL to produce ET-1 in vivo (Ohtani et al ., 1998) and ET-1 inhibits the steroidogenic activities of luteal cells in vitro (Girsh et al., 1996). Furthermore, a study by Meidan et al. (1999) showed that ET-1 inhibited P4 production by luteal cells in a dose-dependent manner. Being a strong vasoconsctrictor, ET1 may cause arteriole constriction (Ohtani

PAGE 39

26 et al., 1998) by reducing blood flow during early luteolys is, and the resulting hypoxia may cause further release of ET-1 (Rakugi et al., 1990). Morphological Changes Morphological changes associated with PGF2 are not evident unt il 24-36 h after exposure of the luteal cells (large and small) to the lu teolysin (Sawyer et al., 1990), although by this time the steroidogenic capabili ties of these cells are markedly reduced. Large luteal cell numbers decrease before sm all steroidogenic luteal cells, and a decrease in large luteal ce ll size is also evident at th is time (Braden et al., 1988). PGF2 Inhibition of P4 Synthesis The PGF2 decreases luteal synt hesis of progesterone in vivo in many species, including bovids and ovids, where it has been shown to decrease P4 secretion from purified preparations of large luteal cells (Wiltbank et al., 1991). The PGF2 negatively regulates P4 synthesis by decreasi ng cholesterol transport across the mitochondrial membranes. This is critical because P4 is derived from choles terol and steroidogenesis occurs in the mitochondria. The disruption in the transport of choles terol can occur at the two different following points: transport from the cytoso l to the outer mitochondrial membrane and from the outer to the inne r mitochondrial membrane where the enzymes required for side-chain cleavage reside. Transport of cholesterol in the cytoplasm is tightly coupl ed to interactions between sterol binding proteins and the cyto skeleton. In a variety of species, P4 secretion from luteal cells is decreased by disruption of the cytoskelet on (Silavin et al., 1980), and studies in ewes have shown that treatment with PGF2 dramatically reduces the number of luteal cells staining for tubulin, a critic al component of microtubular fibers (Murdoch,

PAGE 40

27 1996). This decrease in tubulin was followed by a decrease in P4, indicating that disruption of the cytoskeleton precedes the decreased synthesis of P4. Still, it has not been elucidated if disruption of the microtubul e network prevents transport of cholesterol to mitochondria or disturbs other as pects of luteal steroidogenesis. In vitro studies have also shown that PGF2 may inhibit transport of chol esterol across the mitochondrial membranes (Wiltbank et al., 1993). Treat ment of cows and ewes with PGF2 decreased mRNA encoding for steroidogeni c acute regulatory protein (StAR), which is followed by a decline in StAR protein (Pescador et al., 1996 ). This reduction in StAR may lead to a reduction in cholesterol transport ac ross the mitochondrial membranes. Immune-Mediated Events The immune system plays a pivotal ro le in both functional (decline in P4 secretory capacity of the CL) and structural luteolysis (involution or regression of the CL). During luteolysis, leukocytes, T-lymphocytes and m acrophages increase significantly in the CL (Penny et al., 1999). Macrophages perform se veral roles in the re gressing CL, but the major one is phagocytosis of degenerative luteal cells (Pepperell et al., 1992) and degradation of the extracellular matrix (Parker, 1991). They also assist in cytokinemediated inhibition of steroidogenesis a nd stimulation of CL production of PGF2 This is mediated by the macrophage-produced cytoki nes interleukin-1 (IL-1), which has been shown to be a potent stimulator of PGF2 in cultured bovine cells (Nothnick and Pate, 1990), and tumor necrosis factor(TNF), which stimulates PGF2 production and inhibits LH-stimulated P4 production in vitro (Fairchild-Benyo and Pate, 1992). It has been postulated that monocyte chemoattract ant protein-1 (MCP-1), a potent chemotactic

PAGE 41

28 agent for macrophages, may play an important role in structural luteolysis by increasing the migration of macrophages into the CL (Tsai et al., 1997). Tissue Metalloproteinases Tissue metalloproteinases have been pr oposed as mediators of structural luteolysis. As the CL becomes the corpus albicans (CA), substantial tissue remodeling occurs, and tissue metalloproteinases are though t to be involved in remodeling of the extracellular matrix (Birke dal-Hansen, 1995). The CL of sheep and cows expresses tissue inhibitor of metalloproteinases-1 (T IMP-1), which serves in maintaining the structural integrity of this gland (Smith et al., 1995; Smith et al., 1996). As the CL regresses, TIMP-1 levels decrease, allowi ng for an increased activity of the tissue metalloproteinases (McCracken et al., 1999). Apoptosis It is generally accepted that structural lute olysis is an example of apoptosis. This programmed cell death has been reported to occur during CL regression in ruminants (Juengel, 1993) and to be promoted by PGF2 (Sawyer et al., 1990). Morphological evidence that a cell is undergoing apoptosis is the appearance of nuclear fragment containing degenerate chromatin (Sawyer et al., 1990), cell shrinkage, and appearance of membrane-bound cytoplasmic fractions (Kerr et al., 1972). These cell fragments are phagocytized by cells of the immune syst em, like macrophages, which increase the apoptotic process in populations of luteal cells by phagocytosing membrane-enclosed fragments of these cells (Gemmell et al., 1976). Pregnancy Establishment in Ruminants Successful pregnancy in vi viparous animals depends on the development of an allogenic fetus within its mothers uterus. This presents the challenge of preventing the

PAGE 42

29 maternal immune system from recognizing the fetal antigens as foreign, since it presents paternal genes, and mounting a response agai nst it. Therefore, in most mammals, the conceptus has found ways to escape the dele terious effects of maternal rejection. Another barrier the embryo must preven t is the pulsatile release of PGF2 from the uterus, which would result in regression of the CL and a down regulation in P4 production. In ruminants, as in most other species, a si gnal is sent by the developing embryo with the purpose of allowing the mother to recognize the embryos presence. This prevents luteolysis from occurring and allows P4 secretion to continue, si nce this is the hormone that prepares and maintains the uterus as a suitable environment for the development of the conceptus. Luteal P4 is necessary to allow main tenance of pregnancy until the placenta can produce adequate amounts of this st eroid to complete the gestation interval (Niswender et al., 2000). Ruminants have developed a way of rescuing the CL during early pregnancy that seems to be unique to these species, in which implantation is both late in onset and displays limited inva siveness (Wooding, 1992). The anti-luteolytic signal is a protein, interferon(IFN), produced by the mononuclear cells of the trophectoderm, and it is released between da ys 14 and 17 of pregnancy (LaFrance and Goff, 1985; Okuda et al., 2002). This means that the antilu teolytic signal is released while the blastocyst is still elongating and be fore the trophoblast attaches to the uterine wall (Roberts et al., 1996). Interferoninhibits luteolysis by down-regulating OT receptors in the endometrium, therefore preventing OT-stimulated pulsatile PGF2 release (Okuda et al., 2002), and also by different OT receptor-independent mechanism (LaFrance and Goff, 1990; Bine lli et al., 2000). This is important to note since some studies suggest that OT is not mandatory, but supplementary, for the luteolytic cascade

PAGE 43

30 (Kotwica et al., 1998; Kotwica et al., 1997). The mechanisms by which IFNinhibits luteolysis that are independent of OT r eceptor downregulation, are the attenuation of PGHS-2 (also known as COX-2) gene expres sion (Binelli et al., 2000; Pru et al., 2001) and the shifting of PG production in the endometrium from PGF2 to PGE2, which has a luteotropic or luteoprotective effect on the CL (Xiao et al., 1998). Placentation in Ruminants Placentation in ruminants is of the synepitheliochorial type, which shows limited invasiveness, as opposed to hemochorial placen tation, which is seen in humans and mice, or endotheliochorial placenta tion seen in dogs and cats (R oberts et al., 1996). In synepitheliochorial placentation, binucleate cells migrate from the chorion and fuse with epithelial cells of the uterus (Roberts et al ., 1996). Cell to cell fusions lead to the formation of an extensive s yncytium between the maternal and fetal tissues in sheep, while cell fusion is less extreme in cows a nd the barrier layer becomes partly populated with trinucleate cells (Robert s et al., 1996). In cattle, th e areas of interface known as placentomes provide tight attachments between the maternal and fetal membranes, which allow for nutrient exchange. Placentomes are formed by the branching fetal cotyledonary villi, which grow down into the maternal caruncular crypts in a finger-in-glove arrangement (Schlafer et al., 2000). One area of intense research has been how the fetus escapes the maternal immune response in these areas of in timate attachment. Pregnancy and the Immune Response There is no doubt that one of the most intere sting areas of resear ch in pregnancy is the study of how the fetus escapes the mate rnal immune system. The major challenge that the fetus presents to the mothers uterus is that it expresses paternally derived genes,

PAGE 44

31 which are seen as foreign to the maternal immune system. There have been several theories regarding the maternal-conceptu s immunologic relationship (Hansen, 2000), but recent research has favored some of these more strongly than others. In cows and ewes, there is a decrease in the nu mber of lymphocytes, which se rve in antigen presentation, phagocytosis, antibody and cytoki ne secretion, among other thi ngs. This decrease is localized primarily to the placentomes of cows and the caruncular endometrium of sheep, the areas where the fetal and ma ternal tissues are in closes t attachment, suggesting that the maternal immune tolerance is both local and specific to the areas adjacent to fetal tissues (Leung et al., 2000). It has been suggested that this decrease is mediated by IFNwhich has been shown to regulate lym phocyte proliferation (Newton et al., 1989; Skopets et al., 1992), and to have antiviral act ivity in vitro (Pontzer et al., 1988). A study by Leung et al. (2000) suggested that in early pregnancy, the cytokine interleukin-1 (IL1) could act in concert with IFNto stimulate stromal PGE2 synthesis, which in turn may inhibit uterine expression of the pro-inflammato ry cytokine interleuki n-2 (IL-2). This is biologically relevant because st udies have shown an increase in concentrations of IL-2 in placental tissues of mice (T angri and Raghupathy, 1993) a nd women (Hill et al., 1995; Marzi et al., 1996) that have undergone a s pontaneous abortion. It has also been documented that early in pregnancy, a comp lete shutdown of majo r histocompatibility complex class I (MHC class I) expression by tr ophoblast cells appears to be critical for normal placental development and fetal survival (Davies et al., 2004). The MHC class I protein is expressed in most somatic cells and is used to present peptides derived from intracellular pathogens, or from the animals own proteins, to cytotoxic T lymphocytes, which lyse the target cell. The MHC class II molecules tend to be expressed on antigen-

PAGE 45

32 presenting cells such as macrophages, dendri tic cells and B lymphoc ytes, which use them to present antigens from extr acellular pathogens to helper T lymphocytes, which regulate the activity of other lymphoc ytes through cytokine secre tion. Trophoblasts of most species do not present either MHC class I or MHC class II antigen s (Hunt et al., 1987; Kydd et al., 1991; Davies et al., 2004), suggesting that this lack of expression protects the placenta from attack by the maternal immune sy stem (Davies et al., 2004). Bovids are an exception in that they do present MHC class I protein, but only in the interplacentomal region, which may be the mechanism used by the fetus to prevent rejection by the maternal immune system. Studies by Davies et al. (2000) showed that in bovids, as pregnancy progresses, there is a significant in crease in expression of class I antigens in interplacentomal trophoblast cel ls, while endometrial epithe lial cells of the maternal crypts lack detectable class I expression throughout pregnancy. PGHS-2 Gene Expression A study by Binelli et al. (2000) with cultu red bovine endometrial (BEND) cells demonstrated that bovine IFN(bIFN) suppresses PGF2 production, steady-state levels of PGHS-2 mRNA, and expression of PGHS-2 and PLA2 proteins, after these cells had been stimulated to produce PGF2 with phorbol-12,13-dibutyrate (PDBu), a protein kinase C (PKC) stimulator. A decrease in PGHS-2 activity would directly decrease PGF2 production since, as it was mentioned befo re, this is the enzyme that converts AA into PGH2, the precursor of PGF2 The authors hypothesize d that since PKC may stimulate PGF2 production via stimulation of synt hesis and/or activity of both PLA2 (Mayer and Marshall, 1993; Karimi and Lennartz, 1995) and PGHS-2 (DeWitt, 1991; Vezza et al., 1996), the effects of bIFNcould be at the level of PKC by inhibiting its

PAGE 46

33 ability to stimulate PGF2 synthesis through modulation of these two enzymes. Because there was still some PGF2 production in the presence of bIFN, it was suggested that bIFNlikely does not affect PDBU-induced PGF2 synthesis at sites upstream of PKC. Later experiments by the same group (B inelli et al., 2001) indicated that IFNactivates the Janus kinase (JAK)-signal transducer and activator of transcri ption (STAT) pathway regulating gene expression of PGHS-2 in a manner that decreases secretion of PGF2 Even though it is highly possibl e that activation of the JAKSTAT pathway is involved in the regulation of the anti -luteolytic effects of IFN, there is a whole host of signaling pathways activated by IFNs that cannot be excl uded from consideration (Binelli et al., 2001). Shift in PG Production from PGF2 to PGE2 Interferonmay also prevent luteolysis by shifting primary prostaglandin production in the endometrium from luteolytic PGF2 to luteotropic PGE2. Studies have established that IFNmay inhibit PGF2 production in the endometrium by downregulating PGFS and 9K-PGR (Asselin and Fortier, 2000; LaFrance and Goff, 1990). This would prevent the conversion of PGH2 or PGE2 into PGF2 respectively. Additionally, studies by Xiao et al. (1998) showed that IFNhad an inhibitory effect on PGHS-2 mRNA expression in epithelial cells of the endometrium, which is the primary site of PGF2 production, while it has a stimulat ory effect on PGHS-2 mRNA expression and PG synthesis in stromal cells, which are the primary source of PGE2 (Kim and Fortier, 1995). More recent studi es have demonstrated that IFNspecifically increases one of the PGE2 receptor subtypes in the myometrium and endometrial stroma, but not in luminal or glandular epithelium (Arosh et al., 2004).

PAGE 47

34 Effects of Polyunsaturated Fatty Acids on Reproduction in Cattle High early embryonic mortality is one of the major problems affecting the dairy cattle industry. Increased milk production in dairy cattle is associated with a decrease in reproductive performance. Selective breeding has contributed to th e propagation of high producing cows that exhibit a reduction in occu rrence and intensity of estrus as well as embryo survival, due to alterations in meta bolic rate associated with lactation and management (Thatcher et al 2003). It is estimated that up to 40% of embryonic losses occur between day 8 and day 17 of pregnancy (T hatcher et al., 1995), which is coincident with the period of conceptus inhibition of uterine PGF2 secretion. It is conceivable that some of these early embryo losses may be due to lack of or suboptimal production of IFNby the developing conceptus. Polyunsaturated fatty acids are fatty acids with two or more double bonds. Fatty acids are unsubstituted monocarboxy lic acids that occur mainly as esters in natural fats and oils, and may also exist in non-est erified forms known as free fatty acids (Abayasekara et al., 1999). Unsaturated fatty acids can be divided into n-3, n-6, or n-9 fatty acids, depending on the position of the first double bond from the methyl end of the hydrocarbon chain. Polyunsaturat ed fatty acids that contain their first double bond at the third bond counting from the omega end are n-3 PUFAs, while those th at have their first double bond at the sixth position ar e n-6 fatty acids. Essentia l fatty acids that cannot be endogenously produced by ruminants include lin oleic acid (LA; 18:2 n-6), which can be obtained from plant oils, and -linolenic acid (LNA; 18:3 n-3), which predominates in forage lipids and in linseed (Cheng et al., 2001). Metabolic convers ions can occur only within the same PUFA family, meaning that LA can only be converted to other n-6 fatty

PAGE 48

35 acids such as -linolenic acid (GLA; 18:3), dihomo-linolenic (DGLA; 20:3), AA (20:4), and docosapentaenoic acid (DPA; 22:5), whil e LNA can be converted to other members of the n-3 family such as eicosapenaenoi c acid (EPA; 20:5) and docosahexaenoic acid (DHA; 22:6) (Bezard et al., 1994) These processes occur by the action of desaturase and elongase enzymes present in the animal. Thes e longer chain PUFAs can also be obtained directly from the diet. Using lactating dairy cows, Oldick et al. (1997) showed that OT-induced prostaglandin F2 metabolite (PGFM) concen trations in plasma were greatly reduced in cows infused abomasally with 0.45 kg/d of yellow grease, compared with infusion of tallow, glucose or water. Additionally, Mattos et al. (2002) demonstrated that supplementation of postpartum dairy cows with fish meal containing EPA and DHA considerably attenuated serum PGF2 response to OT injection. Bovine endometrial (BEND) cells are a good model for the study of endometrial regulation of PGF2 These are a line of spontaneous ly replicating e ndometrial cells originating from cows on day 14 of their es trous cycle (Staggs et al., 1998) that can be stimulated with phorbol-12,13dibutyrate (PDBu; a phorbo l ester) to produce PGF2 A study by Mattos et al. (2003) using BEND cells reported that the n-3 fatty acids, LNA, DHA, and EPA, induced greater PGF2 inhibition than the n-9 fatty acid, oleic acid (OA), or the n-6 fatty acids, LA and AA. Although the mechanism of action by which supplemental n-3 fatty acids inhibit PGF2 production is not fully understood, Mattos et al. (2000) suggested that PUFAs may inhibit PGF2 synthesis by decreasing the availability of AA, increasing the concentration of fatty acids that compete with AA for processing by PGHS-2, inhibiting PGHS-2

PAGE 49

36 synthesis and or activity, or affecting ge ne expression through activation of nuclear transcription factors. In cultured BEND cells, n-3 PUFAs were shown to suppress PGF2 production without altering PGHS -2 mRNA synthesis (Mattos et al., 2003). This would suggest that supplemental fa tty acids may affect PGF2 biosynthesis through mechanisms which do not require PGHS-2 gene regulation. The n-3 polyunsaturated fatty acids may decrease AA synthesis by inhibiting the desaturation and/or elongation processes necessary for the conversion of LA to AA (Bezard et al., 1994), or by replacing AA in tissue phospholipids, which has be en demonstrated in animals fed diets rich in n-3 fatty acids (Trujillo and Broughton, 1995). Another possible mechanism for dietary reduction in prostaglandin synthesis is the competiti on of PUFAs such as EPA for PGHS-2 activity, which would result in production of prosta glandins of the 3 series, which are less biologically active (Mattos et al., 2000). A study by Leaver et al. (1991) showed that feeding rats a diet rich in n3 fatty acids resulted in increa sed secretion of prostaglandins of the 3 series from ut erine explants cultured in vitro Dietary fatty acids may repress genes involved in the synthesis of pros taglandins through activation of nuclear transcription factors such as peroxisome proliferator-activated receptors (PPARs). Roles of Peroxisome Proliferator-Ac tivated Receptors in Reproduction Polyunsaturated fatty acids and their various metabolites can act at the level of the nucleus, in conjunction with nuc lear receptors and transcript ion factors, to affect the transcription of a variety of genes. Peroxi some proliferator-activated receptors are an example of nuclear hormone receptors that me diate the effects of fatty acids and their derivatives at the level of gene expression. Peroxisome proliferator-activated receptors belong to the steroid hormone nuclear re ceptor superfamily of ligand-activated

PAGE 50

37 transcription factors that al so includes the retinoic acid r eceptor (RAR), liver X receptor (LXR), and the ubiquitous retinoid X recep tor (RXR) (Sampath and Ntambi, 2005). Three isoforms, PPAR PPAR and PPAR each encoded by a different gene, have been identified. Peroxisome pr oliferator-activated receptor is expressed in numerous metabolic sites such as liver, kidney, heart, sk eletal muscle, and brown fat (Braissant et al., 1996), while PPAR is highly expressed in adipocyt es, where it regulates adipocyte differentiation, lipid storage a nd insulin sensitivity (Chawla et al., 1994; Schiffrin et al., 2003). Peroxisome proliferator-activated receptor is expressed in a wide range of tissues and cells, with relative higher levels of expression noted in brain, adipose, and skin (Braissant et al., 1996), and in the hum an placenta and large intestine (Mukherjee et al., 1997; Auboeuf et al., 1997). Each of these receptors binds to the peroxisome proliferator response element (PPREs) of regul ated genes as a heterodimer with a RXR. The PPRE comprises a direct repeat of 6 nuc leotides, separated by one spacer nucleotide and has the consensus sequence AGGTCA n A GGTCA (van Bilsen et al., 2002). The PPAR/RXR dimer is known to inte ract with various coactivators and corepressors which are thought to modulate transcriptional activity by interacting both with nuclear receptors and basal transcription factors (v an Bilsen et al., 2002). It is generally believed that the PPARs are constitutively localized in the nuc leus, as opposed to other ligand-activated transcription factors which are translocated to the nucleus after binding to their cognate ligand (van Bilsen et al., 2002). Studies ha ve shown that PUFAs can activate PPARs at micromolar concentration ranges (Hihi et al., 2002). To date, the only PPAR that has been a ssociated with reproductive function is PPAR (Berger et al., 2002). Some fatty acid agonists for PPAR are DGLA, AA, and

PAGE 51

38 EPA, of which the latter may be the most pot ent activator of this chemical class (Xu et al., 1999). Prostacyclin (PGI2), derived from the action of PGHS-2, can also interact with PPAR (Forman et al., 1997). Studies with mice and rats have shown that PPAR plays a specific role in embryo development, specifically duri ng implantation and decidualization (Barak et al., 2001; Ding et al., 2003). Furthermore, a study by Lim et al. (2000) demonstrated that PGI2 is the primary PG that is essential for implantation and decidualization, and suggested that the effects of PGI2 are mediated by activation of PPAR This suggests that PGs produced by P GHS-2 may exert their effects directly on the nucleus via the activati on of PPARs. Moreover, in PGHS-2 deficient mice, decidualization and implantation failures can be reversed by the administration of a PPAR -selective agonist (Lim et al., 1997). The bovine endometrium, as well as BEND cells, have been shown to express PPAR (MacLaren et al., 2003) and a study by Bala guer et al. (2005) showed an inverse relationship between PPAR and uterine expression of estrogen receptor alpha and PGHS-2 genes, suggesting that this nuclear receptor may play an important role in the control of reproductive processes in mammalian species. Activation of PPAR by its agonists may lead to the transc ription of genes that are esse ntial to repro ductive events, therefore affecting reproductive performance. Roles of Mitogen-activated Prot ein Kinases in Reproduction The mitogen-activated protein (MAP) kina se signaling cascade is believed to function as an important regulator of pros taglandin biosynthesis (Guan et al., 1998), and components of this pathway have been imp licated as mediators of phosphorylation of intracellular substrates such as protein kina ses and transcription factors (Karin, 1994).

PAGE 52

39 Three major classes of MAP kinases have been identified, each having several isoforms. They include the extracellular signal re gulated protein kina se (ERK), c-Jun Nterminal/stress-activated pr otein kinase (JNK/SAPK), and p38 MAP kinase (Robinson and Cobb, 1997). These MAP kinases are act ivated by distinct upstream MAPK/ERK kinases (MEKs, MKKs), which recognize a nd phosphorylate threonine and tyrosine residues within a tripeptide motif (Thr-X -Tyr) required for MAP kinase activation (Kriakis and Avruch, 1996). There is evidence that the MAP kinase path way is involved in the activation of the PG biosynthetic pathway. For example, Lin et al. (1998) demonstrated that activation of cytosolic PLA2 was mediated by MAP kinase, while Guan et al. (1997) showed that inhibition of p38 MAP kinase resulted in the decrease of IL-1 mediated PGHS-2 expression and PGE2 production. Studies with the ovine endometrium have shown that this tissue expresses all cla sses of MAP kinase, and that OT induces phosphorylation of ERK1/2, suggesting this MAP ki nase mediates OT-induced PGF2 synthesis in ovine endometrium (Burns et al., 2001) A recent study performed at the University of Florida demonstrated that activation of p38 MAP ki nase by PDBu is required for continued presence of PGHS-2 mRNA and secretion of PGF2 in BEND cells (Guzeloglu et al., 2004). In fact, several studies show evidence for a regulator y effect of p38 MAP kinase on PGHS-2 mRNA stability. Inhibition of this MAP kinase in several cell types resulted in reduced expression and/or increased turnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbarama iah et al., 2003). Sc herle et al. (2000) studied the involvement of various signali ng pathways leading to PGHS-2 and PPAR expression in the mouse uterus and demons trated that inhibiti on of p38 MAP kinase

PAGE 53

40 blocked PPAR expression and decreased PGHS-2 ge ne expression in luminal epithelial and stromal cells. Ait-Said et al. (2003) recently showed that EPA suppresses p38 MAP kinase phosphorylation in stimul ated human microvascular endot helial cells, resulting in a down-regulation of PGHS-2 gene expression. This report suggests a novel regulatory point for EPA in PGF2 inhibition.

PAGE 54

41 CHAPTER 3 STUDIES ON THE EFFECTS OF EI COSAPENTAENOIC ACID ON PGF2 PRODUCTION IN CULTURED BOVINE ENDOMETRIAL CELLS Introduction Pregnancy establishment in ruminants is dependent on the attenuation of the pulsatile release of PGF2 from the endometrium (McCracken et al., 1970). Prostaglandin F2 is the luteolytic agent in many species, and it is produced in the endometrium through the action of several en zymes. The rate-limiting step in PG synthesis is the cleavage of sn-2 fatty acyl ester bond of membrane phospholipids by cytosolic phospholipase A2 (Van den Bosh, 1980; Irvine, 198 2). The AA that is released by phospholipid hydrolysis is acted on by PGHS-2 to form PGH2, which then is converted to PGF2 Regression of the CL through the action of PGF2 leads to a decrease in plasma concentrations of P4, which is needed for maintenance of pregnancy. Early embryonic mortality is one of the major problems affecting the dairy cattle industry, and it is estimated that up to 40% of embryonic losses occu r between d 8 and d 17 of pregnancy (Thatcher et al 1994), which is coincident with the period of conceptus inhibition of uterine PGF2 secretion. Many researcher s have focused on identifying ways to increase embryo survival in domestic species (Oldick et al., 1997; Moreira et al., 2000; Mattos et al., 2002, 2003; Badinga et al ., 2002). Several of these studies have shown that manipulating the fat content of th e ruminants diets can have a beneficial effect on reproduction (Sta ples et al., 1998).

PAGE 55

42 Eicosapentaenoic acid and DHA have been shown to inhibit PGF2 synthesis in various tissue and cell models (Olsen et al., 1992; Bagu ma-Nibasheka et al., 1999; Mattos et al., 2002). For example, infusing ew es with 3 ml/kg of body weight per day of an emulsion of fish oil (FO) cont aining 30% EPA and 20% DHA blocked a betamethason-induced increase in plasma concentration of PGF2 and delayed the occurrence of parturi tion (Baguma-Nibasheka et al., 1999). In humans, consumption of large quantities of FO resulted in dela yed parturition, conceiva bly due to reduced secretion of PGF2 (Olsen et al., 1992). Recently, Mattos et al. (2002) reported that supplementing the diet of lactating dairy cows with fish meal (rich in n-3 fatty acids) reduced plasma PGFM concentrations after an OT injection. The same authors showed that EPA and DHA were pot ent inhibitors of PGF2 secretion in BEND cells and that LA, an n-6 fatty acid, did not affect PGF2 response to PDBu in cultured BEND cells (Mattos et al., 2003). It has been postulated that supplemental fatty acids may inhibit PGF2 secretion by decreasing the availability of the prec ursor AA, increasing the concentrations of fatty aci ds that compete with AA for processing by PGHS, or by inhibiting PGHS synthesis and or activity (M attos et al., 2000). The University of Florida investigators recently repor ted that n-3 PUFAs suppressed PGF2 production without altering PGHS-2 mRNA synthesis in cultured BEND cells (Mattos et al., 2003), suggesting that supplemental fatty acids may affect PGF2 biosynthesis through mechanisms which do not require PGHS-2 gene regulation. The objective of this study was to study the physiological effects of EPA on PGF2 production in PDBu-stimulated BEND cells We hypothesized that incubation of BEND cells with EPA would l ead to a reduction in PGF2 production by these cells, and

PAGE 56

43 that increasing the concentration of the n-6 fatty acid, LA, in these cells may abolish the inhibitory effect of EPA on endometrial PGF2 production. Materials and Methods Materials Polystyrene tissue culture dishes (100 x 20 mm) were purch ased from Corning (Corning Glass Works, Corning, NY). The Ham F-12 medium, antibiotic/antimycotic (ABAM), phorbol 12,13-dibutyrate (PDBu), horse serum, D-valine, insulin, and fatty acid-free bovine serum albumin (BSA) were fr om Sigma Chemical Co. (St. Louis, MO). The Minimum Essential medium (MEM) and fetal bovine serum (FBS) were from US Biologicals (Swampscott, MA) and Atlanta Biologicals (Norcross, GA), respectively. Eicosapentaenoic acid, LA and PGF2 standard were from Cayman Chemicals (Ann Arbor, MI). Hanks Balanced Salt Solution (HBSS) and TriZol reagent were from GIBCO BRL (Carlsbad, CA). Isotopically-labeled PGF2 (5, 6, 8, 9, 11, 12, 14, 15 [n-3H] PGF2 ; 208 Ci/mmol) was from Amersham Bios ciences (Piscataway, NJ). The antiPGF2 antibody was purchased from Oxford Biom edicals (Oxford, MI). BioTrans nylon membrane and [ -32P]deoxycytidine triphosphate (SA 3000 Ci/nmol) were from MP Biolomedicals (Atlanta, GA). The PGHS-2 cDNA probe was cloned from an ovarian follicular cDNA library (Liu et al., 1999). Cell Culture and Treatment Immortalized BEND cells (America n Type Culture Collection # CRL-2398, Manassas, VA) were cultured as described by Mattos et al. (2003). Cells were suspended (0.5 x 106 cells/mL) in growth medium (40% Ham F-12, 40% MEM, 1% ABAM, 200 U/L of insulin, 0.343 g/L of D-valine, 10% he at-inactivated FBS a nd 10% horse serum)

PAGE 57

44 and incubated at 37oC in a 95% air-5% CO2 environment. Each culture was replenished with fresh medium every 2 days until cells reached confluence. To examine the effect of EPA on PDBu-stimulated PGF2 secretion, confluent BEND cells were rinsed twice with HBSS and incubated in serum-free medium with or without EPA (100 M) for 24 h (Figure 3-1). This concentration and incubation time were selected since previous studies have show n that this particular concentration of EPA and incubation time have the mo st inhibitory effect on PGF2 in cultured BEND cells (Mattos et al., 2003). The fatty acid was co mplexed with BSA at a molar ratio of 2:1 before being added to the cultures. Th e 24 h incubation period with fatty acid was chosen based on reported effects of EP A and DHA on phorbol ester-induced PGF2 secretion by BEND cells after 24 h (Mattos et al., 2003). Medium was then removed and the cells incubated in serum-free medium with or without PDBu (100 ng/mL) for an additional 6 h (Figure 3-1). After PDBu ch allenge, samples (0.5 mL) of cell-conditioned media were collected and store at -20oC until assayed for PGF2 concentration. The remaining cell monolayers were rinsed in ice-co ld HBSS, lysed with TriZol, and stored at -80oC until PGHS-2 and PPAR mRNA quantifications. To determine the effect of LA on EPA-mediated inhibition of PGF2 secretion, confluent BEND cells were incubated with 6 molar ratios of EPA to LA for 24 h (Figure 3-2). The molar n-6:n-3 ratios were 0 (all EPA), 1, 4, 9, 19, and (all LA). The total concentration of fatty acids in the culture medium was maintained at 100 M. After incubation, the medium was removed and the ce lls were incubated in fresh serum-free medium with PDBu (100 ng/mL) for an additional 6 h. Aliquots (0.5 mL) of cellconditioned media then were collected and stored at -20oC for subsequent PGF2

PAGE 58

45 radioimmunoassay. The remaining cell monolayer s were rinsed in ice-cold HBSS, lysed with 1 mL of TRIzol Reagent, and stored at -80oC until PGHS-2 mRNA quantification. PGF2 Radioimmunoassay Prostaglandin F2 concentration in cell-conditioned media was measured in duplicates as described by Dane t-Desnoyers et al. (1994) and modified by Binelli et al. (2000). Assay sensitivity was 5 ng/mL and intr aand inter-assay coefficients of variation were 8.2 and 15.8%, respectively. To adjust fo r between well differences in cell density, final PGF2 concentrations were expressed as picograms per million cells. Cell numbers in individual dishes were determined usi ng a hemocytometer (Sigma Chemical Co., St. Louis, MO). RNA Isolation and Analysis Total cellular RNA was isolated from control and treated BEND cell cultures using TRIZol reagent. Two hundred L of chloroform were added to each tube containing 1 mL of TRIzol. Tubes were sh aken vigorously by hand for 15 seconds and incubated at room temperature for 3 minutes. Samples were centrifuged at no more than 12,000 x g for 15 minutes at 4C. After centrif ugation the aqueous phase was transferred to a fresh tube, and RNA was precipitated using 500 L of isopropanol. Samples were incubated at room temperature for 10 mi nutes and centrifuged at no more than 12,000 x g for 10 minutes at 4C. The supernatant was removed and the RNA pellet was washed once by adding 1 mL of 75% ethanol a nd centrifuging at no more than 7,500 x g for 3 minutes at 4C. At the end of the proce dure the RNA pellet was air dried for 5 to 10 minutes, after which time the RNA was dissolved in RNase-free water. Ten micrograms of RNA was fractionated in 1.0% agarose-formaldehyde gel and blotted to a BioTrans

PAGE 59

46 nylon membrane by capillary action. The R NA was cross-linked to the membrane by UV irradiation and baked at 80oC for 1 h. The RNA filters were hybridized consecutively with random primer-labeled PGHS-2 and PPAR cDNA probes (Balaguer et al., 2005). After hybridization, RNA filters were washed for 20 min in 50 mL 2X SSC, 0.1% SDS at room temperature, followed by two 15-min washes in 0.1X SSC, 0.1% SDS at 42oC. The filters were blotted dry and exposed to X-ray films for 6 to 24 h at -80oC. Hybridization signals for each target gene were quantified by densitometric analysis. Statistical Analyses Concentrations of PGF2 in cell culture medium and mRNA responses were analyzed using the General Linear Model pr ocedure of the SAS software package (SAS Institute Inc., Cary, NC). For PGF2 concentration, the sources of variation included experiment, treatment, experiment x trea tment interaction, and well (experiment x treatment). One experiment constituted one run in which all treatments ( i.e. control, PDBu, PDBu+EPA) were tested. The well, nest ed within experiment and treatment, was considered a random variable and, therefore, the well variance was used as error term to test the effects of experiment, treatment and experiment x treatment interaction. The statistical model used was the following: yijk = + Ei + Tj + ETij + W(ET)ijk + ijkl where = overall mean; Ei = effect of the ith experiment; Tj = effect of the jth treatment; ETij = the interaction of the ith experiment and the jth treatment; W(ET)ijk = is the well nested within the ith experiment and the jth treatment; and ijkl = random error. For PGHS-2 and PPAR mRNA responses, the mathematical models included experiment, treatment and experiment x treatme nt interaction. Dens itometric values for

PAGE 60

47 each target gene were expressed as ratios ove r the values for 18S ribosomal RNA. When treatment effects were detected (P < 0.05), means were separated using orthogonal contrasts. Results Prostaglandin F2 production was negligible in cont rol BEND cells (Figure 3-3). Treatment with PDBu for 6 h stimulated ( P < 0.01) PGF2 secretion 10-fold (Figure 3-3). Pre-incubation of BEND cells with EPA for 24 h decreased ( P < 0.01) PGF2 response to PDBu by 75% (Figure 3-3). Ho wever, concentration of PGF2 remained higher (+ 2.7fold) in EPA-treated than c ontrol cells (Figure 3-3). Compared to untreated cells, PDBu increased ( P < 0.01) steady-state PGHS-2 mRNA concentrations by 20-fold (Figure 3-4) Eicosapentaenoic acid had no detectable effect on PGHS-2 mRNA responses to PDBu (Figure 3-4). PDBu increased ( P < 0.05) steady-state PPAR mRNA concentrations by 1.3-fold, as compared to untreated cells (Figure 3-5) Eicosapentaenoic acid had no detectable effect on PPAR mRNA response to PDBu (Figure 3-5). The inhibitory effect of EPA on PGF2 production decreased ( P < 0.01) from 88 to 40% as the n-6/n-3 fatty acid ratio in the culture medi um increased from 0 to 19 (Figure 3-6). Addition of LA alone to the culture medium had no de tectable effect on PGF2 response to PDBu (Figure 3-6). Steady-state PGHS-2 mRNA c oncentrations increased ( P < 0.05) from 18 to 93%, as the n-6/n-3 fatty acid ratio in the culture medium increased from 0 to 19 (Figure 3-7). Complete substitution of EPA by LA in the medium further enhanced ( P < 0.05) PGHS-2 mRNA response to PDBu (+ 149%; Figure 3-7).

PAGE 61

48 Discussion Evidence is rapidly accumulating that supplemental fatty acids can have major effects on eicosanoid synthesi s in domestic animals (Ba guma-Nibasheka et al., 1999; Mattos et al., 2003; Cheng et al., 2001; Char trand et al., 2003). Depending on the amount and type of particular fatty acids reaching th e target tissues, supplemental fatty acids can either stimulate (Burke et al., 1996; Filley et al., 1999) or inhibit (Baguma-Nibasheka et al., 1999; Mattos et al., 2003; Cheng et al., 2001) prostanoid synthesis. Results of this study extend previous observat ions that n-3 fatty acids ar e potent inhibitors of PG secretion in mammalian species (Baguma-Ni basheka et al., 1999; Olsen et al., 1992; Mattos et al., 2000; Mattos et al., 2002; Mattos et al., 2003; Cheng et al., 2001). Although the exact mechanism by which suppl emental n-3 fatty acids inhibit PGF2 production is not fully understood, it is concei vable that increased concentration of EPA in membrane phospholipids as a result of tr eating BEND cells with EPA could displace AA, leading to increased synthesis of PGs of the 3 series at the expense of PGs of the 2 series (Mattos et al., 2003). This hypothe sis does not rule out the possibility that supplemental EPA also may inhibit PGHS-2 activity in cultured BEND cells. In fact, incubation of rat hepatoma cells with AA, EPA, DHA, or heineicosapentaenoic acid (C21:5 n-3) inhibited the PGHS-2 enzyme activit y (Larsen et al., 1997). Eicosapentaenoic acid inactivated the enzyme almost completely when added 30 sec before addition of AA. The rate-limiting step in PG synthesis involves the cleavage of sn-2 fatty acyl ester bond of membrane phospholipids by cytosolic PLA2 (Van den Bosh, 1980; Irvine, 1982). The AA that is released by phospholipid hydrol ysis is acted on by PGHS-2 to form PGH2, which then is converted to PGF2 Consistent with a previous observation (Mattos

PAGE 62

49 et al., 2003), the present study provided no evidence for EPA regulation of PGHS-2 mRNA abundance in BEND cells. As disc ussed above, supplemental EPA may alter endometrial PGF2 production through competitive disp lacement of AA from membrane phospholipids and/or through alteration of the PGHS-2 enzymatic activity. Polyunsaturated fatty acids elicit se veral physiological changes through the alteration of the activity or s ynthesis of nuclear PPARs (Boc her et al., 2002). In mice, PPAR deficiency leads to placental defects and results in frequent mid gestational lethalities (Barak et al., 2002), suggesting that this nuclear receptor may play an important role in the control of reproductive processes in mammalian species. Consistent with a recent in vivo experiment (Palin et al., 2005), supplemental EPA had no detectable effects on PPAR response to PDBu in cultured BEND cells. Results suggest that supplemental n-3 fatty acids may alter endometrial PGF2 production through a mechanism which does not require induction of PPAR gene. However, whether and how these fatty acids may control the activity of this nuclear receptor warrants further investigation. Conventional cattle diets contain a mixture of n-6 and n-3 fatty acids. Therefore, we examined the effect of EP A on endometrial production of PGF2 in the presence of increasing concentrations of LA. The inhib itory effect of EPA on uterine endometrial PGF2 production decreased from 88 to 40%, as th e n-6/n-3 fatty acid ratio in the culture medium increased from 0 to 19. These findi ngs are consistent with previous reports (Trujillo et al., 1995; Achard et al., 1997), a nd suggest that the net inhibition of uterine PGF2 synthesis by n-3 fatty acids may depend on the amount of n-6 fatty acids reaching the target tissue. Increasing c oncentrations of LA in the cell culture system may increase

PAGE 63

50 the availability of AA in membrane phospholip ids and therefore decrease the competition by n-3 fatty acids for the PGHS-2 enzyme. Whether and how incr easing n-6/n-3 fatty acid ratios alter the PGHS-2 activ ity is yet to be elucidated. Summary Phorbol ester stimulated PGF2 production and up-regulated PGHS-2 gene expression within 6 h in cultured BEND cells. Pre-incubation of confluent BEND cells with EPA for 24 h decreased PGF2 response to PDBu, but had no detectable effect on PGHS-2 mRNA abundance. The inhi bitory effect of EPA on PGF2 response to PDBu was reverted when increasing amounts of LA and decreasing amounts of EPA were added to the culture medium. These findings indicate that the net in hibition of uterine endometrial PGF2 synthesis by n-3 fatty acids may depend on the availability of n-6 fatty acids within the target tissue.

PAGE 64

51 Figure 3-1. Experimental manipulations to ex amine the effect of eicospaentaenoic acid (EPA) on prostaglandin F2 (PGF2 ) response to phorbol-12,13-dibutyrate (PDBu) in bovine endometrial (BEND) cells.

PAGE 65

52 Figure 3-2. Experimental manipulations to study the effect of li noleic acid (LA) on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. PDBu + + + + + + + LA, M 0 0 50 80 90 95 100 EPA, M 0 100 50 20 10 5 0 n-6/n-3 0 0 1 4 9 19

PAGE 66

53 Figure 3-3. Effect of eicosapentae noic acid (EPA) on prostaglandin F2 (PGF2 ) response to phorbol-12,13-dibutyrate (PDB u) in bovine endometrial (BEND) cells. Data represents least square means SEM of three independent experiments. When treatment effects were detected (P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (Control) vs. (PDBu), (PDBu + EPA), P < 0.0001; Contrast 2: (PDBu) vs. (PDBu + EPA), P < 0.0001.

PAGE 67

54 Figure 3-4. Effect of eico sapentaenoic acid (EPA) on prostaglandin endoperoxide synthase (PGHS-2) mRNA response to phorbol-12,13-dibutyrate (PDBu) in bovine endometrial (BEND) cells. Te n micrograms of total cellular RNA isolated from control and treated BEND ce lls were subjected to Northern blot analysis (A), and resul ting densitometric values were analyzed by the GLM procedure of SAS (B). The top panel s hows a representative Northern blot, whereas the bottom panel represents means SEM calculated over two experiments (n = 4 for each treatment). When treatment effects were detected (P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (Control) vs. (PDBu), (PDBu + EPA), P < 0.0001; Contrast 2: (PDBu) vs. (PDBu + EPA), P = 0.4.

PAGE 68

55 Figure 3-5. Effect of eicosape ntaenoic acid (EPA) on peroxi some proliferator-activated receptor (PPAR ) mRNA response to phorbol-12,13dibutyrate (PDBu) in bovine endometrial (BEND) cells. Te n micrograms of total cellular RNA isolated from control and treated BEND ce lls were subjected to Northern blot analysis (A), and resul ting densitometric values were analyzed by the GLM procedure of SAS (B). The top panel s hows a representative Northern blot, whereas the bottom panel represents means SEM calculated over two experiments (n = 4 for each treatment). When treatment effects were detected (P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (Control) vs. (PDBu), (PDBu + EPA), P = 0.04; Contrast 2: (PDBu) vs. (PDBu + EPA), P = 0.3.

PAGE 69

56 Figure 3-6. Effects of increasing n-6/n3 fatty acid ratios on prostaglandin F2 (PGF2 ) response to phorbol-12-13-dibutyrate (PDBu) Data represents least square means SEM of three independent expe riments. When treatment effects were detected (P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (PDBu), ( ) vs. (0), (1), (4), (9), ( 19), P < 0.0001; Contrast 2: (0) vs. (1), (4), (9), (19), ( ), P < 0.0001; Contrast 3: (1) vs. (4), (9), (19), ( ), P = 0.0001; Contrast 4: (4) vs. (9), (19), ( ), P = 0.0003; Contrast 5: (9) vs. (19), ( ), P = 0.3; Contrast 6: (PDBu) vs. ( ), P = 0.6.

PAGE 70

57 Figure 3-7. Effects of increas ing n-6/n-3 fatty acid ratios on prostaglandin endoperoxide synthase (PGHS-2) response to phor bol-12,13-dibutyrate (PDBu). Ten micrograms of total cellular RNA isolated from control and treated BEND cells were subjected to Northern blot analysis (A), and re sulting densitometric values were analyzed by the GLM proce dure of SAS (B). The top panel shows a representative Northern blot, wherea s the bottom panel represents means SEM calculated over two experiments. When treatment effects were detected (P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (PDBu), ( ) vs. (0), (1), (4), (9), (19), P = 0.1; Contrast 2: (0) vs. (1), (4), (9), (19), ( ), P = 0.3; Contrast 3: (1 ) vs. (4), (9), (19), ( ), P = 0.03; Contrast 4: (4) vs. (9), (19), ( ), P = 0.2; Contrast 5: (9) vs. (19), ( ), P = 0.3; Contrast 6: (PDBu) vs. ( ), P = 0.0004.

PAGE 71

58 CHAPTER 4 STUDIES ON THE MECHANISMS OF ACTION OF EICOSAPENTAENOIC ACID ON PGF2 PRODUCTION BY CULTURED BOVINE ENDOMETRIAL CELLS Introduction Polyunsaturated fatty acids and their me tabolites can act at the level of the nucleus to affect the transcri ption of a variety of genes. One potential mechanism by which PUFAs affect mammalian gene expression is through the intera ction with nuclear receptors commonly referred to as PPARs (Berge r et al., 2002). Peroxisome proliferatoractivated receptors are members of the nucle ar hormone receptor superfamily of liganddependent transcription factor s (Ding et al., 2003). Three PPAR subtypes that have been recognized include PPAR PPAR and PPAR Of these three different subtypes, PPAR is the only one that has been associated with reproductive processes (Berger et al., 2002). Recently Balaguer et al. (2005) showed an inverse relationship between PPAR and uterine expression of estrogen recep tor alpha and PGHS-2 genes, suggesting that this nuclear receptor may play an impor tant role in the control of reproductive processes in mammalian species. The MAP kinase signaling cascade is beli eved to function as an important regulator of PG biosynthesis (Guan et al., 1998), and compon ents of this pathway have been implicated as mediators of phosphoryl ation of intracellular substrates such as protein kinases and transcription factors (Kar in, 1994). In mammalian cells, at least three different subfamilies of MAP kinases have b een identified (ERK, JNK/SAPK, and p38), each having several isoforms (Guan et al ., 1998). Recent investigations have

PAGE 72

59 demonstrated the importance of ERK and p38 MAP kinases as regu lators of the PG biosynthetic pathway. Specifically, ERK ha s been shown to mediate OT-induced PGF2 synthesis in ovine endometriu m (Burns et al., 2001), while p38 regulates the stability of PGHS-2 mRNA (Dean et al., 1999; Guan et al ., 1998; Jang et al., 2000; Subbaramaiah et al., 2003). Whether and how supplemental PUFAs interact with the MAP kinases signaling pathway has not been fully documented. The objective of this investigation was to study the mechanisms of action of EPA on PGF2 production in PDBu-stimulated BEND cel ls. We hypothesized that if EPA exerts its PGF2 -inhibitory effect th rough activation of PPAR then activation of this nuclear receptor would mimic the effect of EPA on PGF2 production. Alternatively, if EPA acts through the ERK or p38 MAP kinase signal transduction pathways, then blocking these kinases would block the PGF2 -inhibitory effect of EPA. Materials and Methods Cell Culture and Treatment Immortalized BEND cells (American Type Culture Collection # CRL-2398, Manassas, VA) were cultured as described by Mattos et al. (2003) with the following modifications. Cells were suspended (0.5 x 106 cells/mL) in growth medium (40% Ham F-12, 40% MEM, 1% ABAM, 200 U/L of insulin, 0.343 g/L of D-valine, 10% heatinactivated FBS and 10% horse serum) and incubated at 37oC in a 95% air-5% CO2 environment. Each culture was replenished with fresh medium every 2 days until cells reached confluence. Experiment 1. To determine if EPA-mediated inhibition of PGF2 secretion involves PPAR activation, confluent BEND cells were rinsed with HBSS and then

PAGE 73

60 assigned randomly to one of the following treatments: 1) control (medium alone), 2) PDBu (100 ng/mL), 3) PPAR agonist (L-165,041; 1 M), or 4) PDBu + L-165,041 (Figure 4-1). Experiment 2. In another set of experiments, confluent BEND cells were treated with 1) medium alone (control), 2) PDBu, 3) PDBu + EPA, or 4) PDBu + EPA + PPAR antagonist (sulindac sulfide or sulindac sulfone ; Figure 4-2). In each experiment, aliquots (0.5 mL) of cell conditioned media were collected at the end of the cu lture and stored at 20C until analyzed for PGF2 concentration. The remaining cell monolayers were lyzed with Trizol and stored at -80C until PGHS-2 and PPAR mRNA quantifications. Experiment 3. To determine whether EPA-mediated PGF2 secretion involves p38 or ERK MAP kinase activation, confluent BEND cells were treated with EPA (100 ng/mL), a combination of EPA and p38 MAP kinase inhibitor (SB203580 [1 M]), or a combination of EPA and ERK MA P kinase inhibitor (PD98059 [20 M]) for 24 h, and then challenged with PDBu (100 ng/mL) fo r an additional 6 h (Figure 4-3). After incubation, aliquots (0.5 mL) of cell-conditioned media were collected and stored at 20oC for subsequent PGF2 radioimmunoassay. The remaining cell monolayers were rinsed in ice-cold HBSS, lysed with TriZol, and stored at -80oC until PGHS-2 mRNA quantification. PGF2 Radioimmunoassay Prostaglandin F2 concentration in cell-conditioned media was measured in duplicates as described by Dane t-Desnoyers et al. (1994) and modified by Binelli et al. (2000). Assay sensitivity was 5 ng/mL and intr aand inter-assay coefficients of variation were 8.2 and 15.8%, respectively. To adjust fo r between well differences in cell density,

PAGE 74

61 final PGF2 concentrations were expressed as picograms per million cells. Cell numbers in individual dishes were determined usi ng a hemocytometer (Sigma Chemical Co., St. Louis, MO). RNA Isolation and Analysis Total cellular RNA was isolated from control and treated BEND cell cultures using TriZol reagent according to the manufacturers instru ctions. Ten micrograms of RNA was fractionated in 1.0% agarose-formal dehyde gel and blotted to a BioTrans nylon membrane by capillary action. The RNA was cross-linked to the membrane by UV irradiation and baked at 80oC for 1 h. The RNA filters were hybridized consecutively with random primer-labeled PGHS-2 and PPAR cDNA probes (Balaguer et al., 2005). After hybridization, RNA filters were washed for 20 min in 50 mL 2X SSC, 0.1% SDS at room temperature, followed by two 15-min washes in 0.1X SSC, 0.1% SDS at 42oC. The filters were blotted dry and exposed to X-ray films for 6 to 24 h at -80oC. Hybridization signals for each target gene were quantified by densitometric analysis. Statistical Analyses Concentrations of PGF2 in cell culture medium and mRNA responses were analyzed using the General Linear Model pr ocedure of the SAS software package (SAS Institute Inc., Cary, NC). For PGF2 concentration, the sources of variation included experiment, treatment, experiment x trea tment interaction, and well (experiment x treatment). The well, nested within experi ment and treatment, was considered a random variable and, therefore, the well variance was used as error term to test the effects of experiment, treatment and experiment x tr eatment interaction. For PGHS-2 and PPAR mRNA responses, the mathematical mode ls included experiment, treatment and

PAGE 75

62 experiment x treatment interaction. Densito metric values for each target gene were expressed as ratios over the values for 18S ri bosomal RNA. When treatment effects were detected (P < 0.05), means were se parated using orthogonal contrasts. Results To test the hypothesis that EP A may attenuate endometrial PGF2 secretion through activation of nuclear PPAR we examined endometrial PGF2 response to PDBu in the absence or presence of L-165,410, a specific PPAR agonist. The PPAR agonist completely abolished (P < 0.01) PGF2 response to PDBu (Figure 4-4). The decrease in PGF2 secretion coincided with a significant reduction in PGHS-2 mRNA abundance in PDBu + L-165,410-treated cells (Figure 4-5). Basal and PDBu-induced PPAR mRNA concentrations were unaffected by supplemental PPAR agonist (Figure 4-6). When added in combination with EPA, the PPAR inhibitor sulindac sulfide decreased (P < 0.01) PGF2 response to PDBu to a greate r extent (-94%) than did EPA alone (-78%; Figure 4-7). The decrease in PGF2 secretion due to sulindac sulfide coincided with a reduction (P < 0.01) in st eady-state PGHS-2 mRNA concentration in sulindac sulfide-treated cells (Figure 4-8). The PPAR mRNA response to PDBu was unaffected by supplemental suli ndac sulfide (Figure 4-9). Because sulindac sulfide decreased PGHS-2 mRNA concentration in PDBUtreated BEND cells, it was unclear whether this PPAR inhibitor decreased endometrial PGF2 secretion via repression of PGHS-2 gene and or through a specific inhibition of nuclear PPAR To test a PGHS-2-independent mechanism of PGF2 inhibition in endometrial cells, we examined PGF2 response to PDBu in the presence of EPA and sulindac sulfone, a PPAR inhibitor that has been reporte d to have no effects on PGHS-2

PAGE 76

63 gene expression (Babbar et al., 2003). Su lindac sulfone decreas ed (P < 0.01) PGF2 response to PDBu to a greater extent (-86%) than did EPA alone (-71%; Figure 4-10). However, unlike sulindac sulfide, treatment of BEND cells with su lindac sulfone had no detectable effects on stea dy-state PGHS-2 mRNA concentration (Figure 4-11). Interestingly, priming of BEND cells with EPA alone or EPA and sulindac sulfone increased (P < 0.05) PPAR response to PDBu (Figure 4-12). To test whether EPA i nhibits endometrial PGF2 production through the p38 or ERK MAP kinase signal transducti on pathways, we examined PGF2 response to PDBu in the absence or presence of SB203580, a p38 inhibitor, or PD98059, an ERK inhibitor. Pre-incubation of BEND cells for 24 h with ei ther inhibitor had no detectable effect on PGF2 production as compared to EPA + PDBu treated BEND cells (Figure 4-13). There was no difference in PGHS-2 mRNA level among any of the treatments tested (Figure 414). Discussion Polyunsaturated fatty acids elicit severa l metabolic changes through alteration of the activity or synthesis of nucl ear PPARs, all of which appear to have distinct patterns of expression and functional roles (Brassant a nd Wahli, 1998). Our laboratory recently detected an inverse relationship between PPAR and uterine estrogen receptor alpha and PGHS-2 genes (Balaguer et al., 2005), suggesti ng that this nuclear receptor may play an important role in the control of reproductive processes in mammalian species. In the present study, PPAR activation greatly reduced PGF2 and PGHS-2 mRNA responses to PDBu in cultured BEND cells (F igure 4-4 and Figure 4-5). This is in contrast with a previous study in which PPAR activation induced PGHS-2 gene

PAGE 77

64 expression in human hepatocellular carcinoma cells (Glinghammar et al., 2003). Consistent with our findings, Inoui e et al. (2000) reported that PPAR activation reduced the PGHS-2 promoter activity in lipopolys accharide-stimulated monocytic cells. Furthermore, Subbamaiah et al. (2001) show ed that PPAR ligands could suppress tissuetype plasminogen activator (TPA)-driven PGHS-2 transcription in hum an epithelial cells via activator protein-1 (AP-1) and cAMP re sponse element binding protein (CREB) at the cAMP response element (CRE) site in the pr oximal promoter. These studies collectively indicate that the net effect of PPAR activation may vary depending on the cell type and likely depends on the presence and or activati on of other transcrip tion co-factors in a given cell system (Lim et al., 2004). To further characterize the role of PPAR in EPA-induced attenuation of endometrial PGF2 secretion, we examined the effects of two PPAR inhibitors on PGF2 response to EPA in cultured BEND cells. Sulindac sulfide a nd sulindac sulfone decreased PGF2 secretion to greater extents than did EPA alone. The decrease in PGF2 secretion was associated w ith a significant reduction in PGHS-2 mRNA abundance in sulindac sulfidetreated, but not sulindac sulfone-treated cells. Results are consistent with previous reports that su lindac sulfide is a potent in hibitor of cyclooxygenase gene expression and PG synthesis in several ma lignant cells (Marnett, 1992; Meade et al., 1993; Lim et al., 1999). The observation that sulindac sulf one decreased PGF2 secretion without altering endometrial PGHS -2 mRNA content also is cons istent with the literature data (Thompson et al., 1995,1997; Babbar et al., 2003), and would suggest that the sulfone derivative of suli ndac regulates uterine PGF2 secretion through a mechanism that is PGHS-2-independent.

PAGE 78

65 Previous studies have shown that the MAP kinase signaling cascade functions as an important regulator of prosta glandin biosynthesis (Guan et al., 1998), and in ruminants, there is evidence that ERK and p38 MAP kinase s are involved in activation of the PG biosynthetic pathway (Guan et al., 1997; Lin et al., 1998; Burns et al ., 2001; Guzeloglu et al., 2004). Inhibition of p38 MAP kinase in several cell types resulted in reduced expression and/or increased tu rnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al ., 2003). Most im portantly, a recent investigation by Ait-Said et al. (2003) showed that EPA suppressed p38 MAP kinase phosphorylation and down-regulated PGHS-2 gene expression in stimulated human microvascular endothelial cells, suggesting a novel regulatory point for EPA in PGF2 inhibition. The present study indicated that p38 or ER K inhibitor had no detectable effect on PGF2 or PGHS-2 mRNA responses to EPA in cu ltured BEND cells. This is in contrast with previous observations which showed that inhibiti on of the p38 MAP kinase in several cell types resulted in reduced expression and/or incr eased turnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al., 2003). It is conceivable that the use of PDBu in BEND cells leads to an activation of PKC and, subsequently, of PGHS -2, which may override the e ffect of p38 inhibition on PGHS-2 gene expression. Collectively, these findings indicate that EPA and PPAR alter uterine endometrial PGF2 secretion through complex mechanisms which may or may not involve PGHS-2 regulation. Whether and how EPA and PPAR affect the activitie s of several enzymes involved in endometrial PGF2 cascade remains to be elucidated. Further studies are

PAGE 79

66 necessary to investigate the role of p38 and ERK in EPA-mediated PGF2 inhibition in cultured BEND cells. Summary Phorbol ester stimulated PGF2 production and up-regulated PGHS-2 gene expression within 6 h in cultured BEND cells. Pre-incubation of confluent BEND cells with a PPAR agonist (L-165,041) for 24 h decreased PGF2 and steady-state PGHS-2 mRNA responses to PDBu, but had no detectable effect on PPAR mRNA concentrations. Pre-incuba tion of BEND cells with PPAR inhibitors (sulindac sulfide and sulindac sulfone) decreased PGF2 response to PDBu to a greater extent than did EPA alone. Treatment of BEND cells with sulindac sulfide resulted in a reduction in steady-state PGHS-2 mRNA levels, while us e of sulindac sulfone had no detectable effects on steady-state PGHS-2 mRNA a bundance. Basal and PDBu-induced PPAR mRNA concentrations were unaffected by suppl emental sulindac sulfide, while sulindac sulfone increased PPAR response to PDBu. Inhibiton of p38 and ERK MAP kinases had no detectable effects on the PGF2 or PGHS-2 response to EPA. Results of these studies indicate that EPA and PPAR alter uterine endometrial PGF2 secretion through complex mechanisms which may or may not involve PGHS-2 regulation. Further studies are necessary to elucidate the ro le of MAP kinases in EPAmediated PGF2 inhibition in the bovine uterus.

PAGE 80

67 Figure 4-1. Experimental manipulations to determine if eicosapentaenoic acid (EPA)mediated inhibition of prostaglandin F2 (PGF2 ) secretion involves peroxisome proliferator-activated receptor (PPAR ) activation in bovine endometrial (BEND) cells.

PAGE 81

68 Figure 4-2. Experimental manipulations to determine if peroxisome proliferatoractivated receptor (PPAR inhibition may abolish eicosapentaenoic acid (EPA) effect on prostaglandin F2 (PGF2 ) secretion in bovine endometrial (BEND) cells.

PAGE 82

69 Figure 4-3. Experimental manipulations to ex amine the role of mitogen-activated protein (MAP) kinases in eicosapentaenoic acid (EPA) regulation of prostaglandin F2 (PGF2 ) production in bovine e ndometrial (BEND) cells.

PAGE 83

70 Figure 4-4. Effect of a peroxisome proliferator-ac tivated receptor (PPAR agonist (L165,041) on prostaglandin F2 (PGF2 ) secretion in bovine endometrial (BEND) cells. Data represents least square means SEM of two independent experiments. When treatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. C ontrast 1: (Control), (Ag) vs. (PDBu), (PDBu + Ag), P < 0.0001; Contrast 2: (PDBu) vs. (PDBu + Ag), P < 0.0001; Contrast 3: (Control) vs. (Ag), P = 0.2.

PAGE 84

71 Figure 4-5. Effect of a peroxisome proliferator-ac tivated receptor (PPAR agonist (L165,041) on prostaglandin endoperoxide synthase (PGHS-2) mRNA abundance in bovine endometrial (BEND) cells. Ten micrograms of total cellular RNA isolated from control and treated BEND cells were subjected to Northern blot analysis (A), and resulti ng densitometric values were analyzed by the GLM procedure of SAS (B). The top panel shows a representative Northern blot, whereas the bottom panel represents means SEM calculated over two experiments (n = 4 for each treatment). When treatment effects were detected ( P < 0.05), means were separate d using orthogonal contrasts. Contrast 1: (Control), (Ag) vs. (PDBu), (PDBu + Ag), P = 0.0001; Contrast 2: (PDBu) vs. (PDBu + Ag), P = 0.004; Contrast 3: (Control) vs. (Ag), P = 0.9.

PAGE 85

72 Figure 4-6. Effect of a peroxisome proliferator-ac tivated receptor (PPAR agonist (L165,041) on PPAR mRNA response to PDBu in bovine endometrial (BEND) cells. Ten micrograms of total cellular RNA isolated from control and treated BEND cells were subjected to Northern blot analysis (A), and resulting densitometric values were analyzed by the GLM procedure of SAS (B). The top panel shows a representative Nort hern blot, whereas the bottom panel represents means SEM calculated over two experiments (n = 4 for each treatment). When treatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. C ontrast 1: (Control), (Ag) vs. (PDBu), (PDBu + Ag), P = 0.008; Contrast 2: (PDBu) vs. (PDBu + Ag), P = 0.9; Contrast 3: (Control) vs. (Ag), P = 0.2.

PAGE 86

73 Figure 4-7. Effect of sulind ac sulfide on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Data represents least square m eans SEM of five independent experiments. When treatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. Contra st 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA + I) P < 0.0001; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P < 0.0001.

PAGE 87

74 Figure 4-8. Effect of sulind ac sulfide on prostaglandin endo peroxide synthase (PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Ten micrograms of tota l cellular RNA isolated from control and treated BEND cells were subjected to No rthern blot analysis (A), and resulting densitometric values were analyzed by the GLM procedure of SAS (B). The top panel shows a representative No rthern blot, whereas the bottom panel represents means SEM calculated over two experiments (n = 4 for each treatment). When treatme nt effects were detected ( P < 0.05), means were separated using orthogonal contrast s. Contrast 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA + I) P = 0.2; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P = 0.0005.

PAGE 88

75 Figure 4-9. Effect of sulindac sulfide on pe roxisome proliferator-activated receptor (PPAR mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Ten microgr ams of total cellular RNA isolated from control and treated BEND cells were subjected to Northern blot analysis (A), and resulting densitometric values were analyzed by the GLM procedure of SAS (B). The top panel s hows a representative Northern blot, whereas the bottom panel represents means SEM calculated over two experiments (n = 4 for each experiment).

PAGE 89

76 Figure 4-10. Effect of suli ndac sulfone on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Data represents least square means SEM of three independent experiments. When treatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. Contra st 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA + I) P < 0.0001; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P = 0.0005.

PAGE 90

77 Figure 4-11. Effect of suli ndac sulfone on prostaglandin endoperoxide synthase (PGHS2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Ten micrograms of tota l cellular RNA isolated from control and treated BEND cells were subjected to Nort hern blot analysis (A), and resulting densitometric values were analy zed by the GLM procedure of SAS (B). The top panel shows a representative Northern blot whereas the bottom panel represents means SEM calcula ted over two experiments (n = 4 for each treatment). When tr eatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA + I) P = 0.07; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P = 0.4.

PAGE 91

78 Figure 4-12. Effect of sulindac sulfone on pe roxisome proliferator-activated receptor (PPAR mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Ten microgr ams of total cellular RNA isolated from control and treated BEND cells were subjected to Northern blot analysis (A), and resulting densitometric values were analyzed by the GLM procedure of SAS (B). The top panel shows a representa tive Northern blot, whereas the bottom panel represents means SEM calcula ted over two experiments (n = 4 for each treatment). When tr eatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. Contrast 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA + I) P = 0.01; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P = 0.8.

PAGE 92

79 Figure 4-13. Effect of SB 203580 and PD98059 on prostaglandin F2 (PGF2 ) response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Data represents least square means SEM of three independent experiments. When treatment effects were detected ( P < 0.05), means were separated using orthogonal contrasts. Contra st 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA + SB), (PDBu + EPA + PD) P < 0.0001; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + SB), (PDBu + EPA + PD), P < 0.0001, Contrast 3: (PDBu + EPA + SB) vs. (PDBu + EPA + PD), P = 0.9.

PAGE 93

80 Figure 4-14. Effect of SB203580 and PD98059 on prostaglandin endoperoxide synthase (PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Ten microgr ams of total cellular RNA isolated from control and treated BEND cells were subjected to Northern blot analysis (A), and resulting densitometric values were analyzed by the GLM procedure of SAS (B).

PAGE 94

81 CHAPTER 5 GENERAL DISCUSSION Recent studies have implicated n-3 pol yunsaturated fatty acids (PUFAs) in the reduction of prostaglandin F2 (PGF2 ) synthesis in the bovin e uterus. Although cattle diets contain a mixture of n-3 and n-6 fatty acids, currently th ere is a lack of information as to how these fatty acids may interact to alter PGF2 biosynthesis in the uterus. Results of this study extend previous observations that n-3 fatty acid s are potent inhibitors of PG secretion in mammalian species (Baguma-Ni basheka et al., 1999; Olsen et al., 1992; Mattos et al., 2000; Mattos et al., 2002; Mattos et al., 2003; Cheng et al., 2001). Although the exact mechanism by which suppl emental n-3 fatty acids inhibit PGF2 production is not fully understood, it is conceiva ble that increased availability of EPA in membrane phospholipids as a result of treating BEND cells with EPA could displace AA, leading to increased synthesis of PGs of the 3 series at the ex pense of PGs of the 2 series (Mattos et al., 2003). This hypothesis does not rule out the possibility that supplemental EPA also may inhibit PGHS-2 activity in cultu red BEND cells. In fact, incubation of rat hepatoma cells with AA, EPA, D HA, or heineicosapentaenoic acid (C21:5 n-3) inhibited the PGHS-2 enzyme activity (Larsen et al., 1997 ). Eicosapentaenoic acid inactivated the enzyme almost completely when added 30 sec before addition of AA. The rate-limiting step in PG synthesis involves the cleavage of sn-2 fatty acyl ester bond of membrane phospholipids by cytosolic PLA2 (Van den Bosh, 1980; Irvine, 1982). The AA that is released by phospholipid hydrol ysis is acted on by PGHS-2 to form

PAGE 95

82 PGH2, which then is converted to PGF2 Consistent with a previous observation (Mattos et al., 2003), the present study provided no evidence for EPA regulation of PGHS-2 mRNA abundance in BEND cells. As disc ussed above, supplemental EPA may alter endometrial PGF2 production through competitive disp lacement of AA from membrane phospholipids and/or through alteration of the PGHS-2 enzymatic activity. Polyunsaturated fatty acids elicit se veral physiological changes through the alteration of the activity or s ynthesis of nuclear PPARs (Boc her et al., 2002). In mice, PPAR deficiency leads to placental defects and results in frequent mid gestational lethalities (Barak et al., 2002), suggesting that this nuclear receptor may play an important role in the control of reproductive processes in mammalian species. Consistent with a recent in vivo experiment (Palin et al., 2005), supplemental EPA had no detectable effects on PPAR response to PDBu in cultured BEND cells. Results suggest that supplemental n-3 fatty acids may alter endometrial PGF2 production through a mechanism which does not require induction of PPAR gene. However, whether and how these fatty acids may control the activity of this nuclear recep tor warrants further investigation. Conventional cattle diets contain a mixture of n-6 and n-3 fatty acids. Therefore, we examined the effect of EP A on endometrial production of PGF2 in the presence of increasing concentrations of LA. The inhib itory effect of EPA on uterine endometrial PGF2 production decreased from 88 to 40%, as th e n-6/n-3 fatty acid ratio in the culture medium increased from 0 to 19. These findi ngs are consistent with previous reports (Trujillo et al., 1995; Achard et al., 1997), a nd suggest that the net inhibition of uterine PGF2 synthesis by n-3 fatty acids may depend on the amount of n-6 fatty acids reaching

PAGE 96

83 the target tissue. Increasing c oncentrations of LA in the cell culture system may increase the availability of AA in membrane phospholip ids and therefore decrease the competition by n-3 fatty acids for the PGHS-2 enzyme. Whether and how incr easing n-6/n-3 fatty acid ratios alter the PGHS-2 activ ity is yet to be elucidated. For the experiments that focused on studyi ng the mechanisms of action of EPA on PGF2 production in BEND cells, we hypothe sized that if EPA affects PGF2 production through the activation of PPAR or of p38 or ERK, then activation of this nuclear receptor, or of these MAP kinases, would mimic the effect of EPA on PGF2 production in cultured BEND cells. In the present study, PPAR activation greatly reduced PGF2 and PGHS-2 mRNA responses to PDBu in cultu red BEND cells. This is in contrast with a previous study in which PPAR activation induced PGHS-2 gene expression in human hepatocellular carcinoma cells (Glinghammar et al., 2003). Cons istent with our findings, Inouie et al. (2000) reported that PPAR activation significantly reduced the PGHS-2 promoter activity in lipopolysaccharide-stim ulated monocytic cells. Furthermore, Subbamaiah et al. (2001) showed that PPAR ligands could suppress TPA-driven PGHS-2 transcription in human epithelial cells via AP-1 and CREB binding proteins at the CRE site in the proximal promoter. These studies collectively indicate that the net effect of PPAR activation may vary depending on the ce ll type and likely depends on the presence and or activation of other tr anscription co-factors in a gi ven cell system (Lim et al., 2004). To further characterize the role of PPAR in EPA-induced attenuation of endometrial PGF2 secretion, we examined the effects of two PPAR inhibitors on PGF2 response to EPA in cultured BEND cells. Sulindac sulfide a nd sulindac sulfone

PAGE 97

84 decreased PGF2 secretion to greater extents than did EPA alone. The decrease in PGF2 secretion was associated w ith a significant reduction in PGHS-2 mRNA abundance in sulindac sulfidetreated, but not sulindac sulfone-treated cells. Results are consistent with previous reports that su lindac sulfide is a potent in hibitor of cyclooxygenase gene expression and PG synthesis in several ma lignant cells (Marnett, 1992; Meade et al., 1993; Lim et al., 1999). The observation that sulindac sulf one decreased PGF2 secretion without altering endometrial PGHS -2 mRNA content also is cons istent with the literature data (Thompson et al., 1995,1997; Babbar et al., 2003), and would suggest that the sulfone derivative of suli ndac regulates uterine PGF2 secretion through a mechanism that is PGHS-2-independent. Previous studies have shown that the MAP kinase signaling cascade functions as an important regulator of prosta glandin biosynthesis (Guan et al., 1998), and in ruminants, there is evidence that ERK and p38 MAP kinase s are involved in activation of the PG biosynthetic pathway (Guan et al., 1997; Lin et al., 1998; Burns et al., 2001, Guzeloglu et al., 2004). Inhibition of p38 MAP kinase in several cell types results in reduced expression and/or increased tu rnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al ., 2003). Most importantly, a recent investigation by Ait-Said et al. (2003) showed that EPA suppressed p38 MAP kinase phosphorylation and down-regulated PGHS-2 gene expression in stimulated human microvascular endothelial cells, suggesting a novel regulatory point for EPA in PGF2 inhibition. The present study indicated that p38 or ER K inhibitor had no detectable effect on PGF2 or PGHS-2 mRNA responses to EPA in cu ltured BEND cells. This is in contrast

PAGE 98

85 with previous observations which showed that inhibiti on of the p38 MAP kinase in several cell types resulted in reduced expression and/or incr eased turnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al., 2003). It is conceivable that the use of PDBu in BEND cells leads to an activation of PKC and, subsequently, of PGHS -2, which may override the e ffect of p38 inhibition on PGHS-2 gene expression. In summary, this study presents direct evidence that the i nhibition of uterine endometrial PGF2 biosynthesis by n-3 fatty acids de pends on the amount of n-6 fatty acids in the uterus, and that EPA and PPAR affect uterine PGF2 synthesis through complex mechanisms which may or may not involve PGHS-2 gene regulation. Further studies are needed to fully document th e role of p38 and ERK MAP kinases in endometrial PGF2 biosynthesis.

PAGE 99

86 Figure 5-1. Proposed model for eicosa pentaenoic acid (EPA) and linoleic acid (LA) regulation of prostaglandin F2 (PGF2 ) biosynthesis in bovine endometrial (BE ND) cells. Increased availability of EPA in membrane phospholipids as a result of treating BEND cells with EP A displaces arachidonic acid (AA), leadi ng to increased synthesis of PGs of the 3 series at the expense of PGs of the 2 series. In creasing concentrations of LA in the incubation medium increase the availability of AA in me mbrane phospholipids and, as a result, decrease the competition by n-3 fatty acids for the prostaglandin enoperoxide synthase (PGHS-2) enzyme, ultimately leading to increased synthesis of PGF2 by BEND cells.

PAGE 100

87 Figure 5-2. Signaling cascade lead ing to synthesis and activati on of prostaglandin endoperoxi de synthase (PGHS-2) in bovine endometrial (BEND) cells. Tr eatment of BEND cells with phorbol12,13-dibutyrate (PDBu) results in activation of protein kinase C (PKC). Activated PKC acts directly or through mitogen-activated protein (MAP) kinase pathways to regulate PGHS-2 mRNA tr anscription. Eicosapentae noic acid (EPA) has no detectable effect on PGHS-2 gene transcription, but ma y affect prostaglandin synt hesis through inhibition of PGHS-2 activit y in BEND cells.

PAGE 101

88 LIST OF REFERENCES Abayasekara, D.R., Wathes, D.C., 1999. E ffects of altering dietary fatty acids composition on prostaglandin synthesis and fertility. Prostaglanding Leukot. Essent. Fatty Acids. 61, 275-287. Achard, F., Gilbert, M., Benistant, C., Ben Slama, S., DeWitt, D.L., Smith, W.L., Lagarde, M., 1997. Eicosapentaenoic a nd docosahexaenoic acids reduce PGH synthase 1 expression in bovine aortic endothelial cel ls. Biochem. Biophys. Res. Commun. 241, 513-518. Acosta, T.J., Miyamoto, A., 2004. Vascular cont rol of ovarian function: ovulation, corpus luteum formation and regression. Anim. Reprod. Sci. 82-83, 127-140. Adams, G.P., Matteri, R.L., Kastelic, J.P ., Ko, J.C.H., Ginther, O.J., 1992. Association between surges of follicle-stimulating hormone and the emergence of follicular waves in heifers. J. Reprod. Fertil. 94, 177-188. Ait-Said, F., Elalamy, I., Werts, C., Gomar d, M.T., Jacquemin, C., Couetil, J.P., Hatmi, M., 2003. Inhibition by eicosapentaenoic ac id of IL-1beta-induced PGHS-2 expression in human microva scular endothelial cells: involvement of lipoxygenasederived metabolites and p38 MAPK path way. Biochim. Biophys. Acta. 1631, 7784. Amselgruber, W., Sinowatz, F., Schams, D ., Skottner, A., 1994. Immunohistochemical aspects of insulin-like grow th factors I and II in the bovine corpus luteum. J. Reprod. Fertil. 101, 445-451. Antonelli-Orlidge, A., Smith, S.R., DAmor e, P.A., 1989. Influence of pericytes on capillary endothelial cell growt h. Am. Rev. Respir. Dis. 140, 1129-1131. Arosh, J.A., Banu, S.K., Kimmins, S., Chapdela ine, P., Maclaren, L.A., Fortier, M.A., 2004. Effect of interferonon prostaglandin biosynthesi s, transport, and signaling at the time of maternal recognition of pre gnancy in cattle: evidence of polycrine actions of prostaglandin E2. Endocrinology. 145, 5280-5293.

PAGE 102

8 Asselin, E., Drolet, P., Fortier, M.A., 1998. In vitro response to oxytocin and interferonin bovine endometrial cells from carunc ular and inter-carun cular areas. Biol. Reprod. 59, 241-247. Asselin, E., Fortier, M.A., 2000. Detection a nd regulation of the messenger for a putative bovine endometrial 9-keto-prostaglandin E2 reductase: effect of oxytocin and interferon. Biol. Reprod. 62, 125-131. Auboeuf, D., Rieusset, J., Fajas, L., Vallier, P., Frering, V., Riou, J.P ., Staels, B., Auwer, J., Laville, M., Vidal, H., 1997. Tissue distribution and quan tification of the expression of mRNAs of peroxisome prolif erator-activated re ceptors and liver X receptor-alpha in humans: no alteration in adipose tissue of obese and NIDDM patients. Diabetes. 46, 1319-1327. Austin, E.J., Mihm, M., Evans, A.C.O., Knight P.G., Ireland, J.L.H., Ireland, J.J., Roche, J.F., 2001. Alterations in in trafollicular regulatory f actors and apoptosis during selection of follicles in the first follicular wave of the bovine estrous cycle. Biol. Reprod. 64, 839-848. Babbar, N., Ignatenko, N.A., Casero, R. A. Jr, Gerner, E.W., 2003. Cyclooxygenaseindependent induction of apoptosis by suli ndac sulfone is mediated by polyamines in colon cancer. J. Biol. Chem. 278, 47762-47775 Badinga, L., Guzeloglu, A., Thatcher, W.W ., 2002. Bovine somatotropin attenuates phorbol ester-induced prosataglandin F2 production in bovine endometrial cells. J. Dairy Sci. 85, 537-543. Baguma-Nibasheka, M., Brenna, J.T., Nathanie lsz, P.W., 1999. Delay of preterm delivery in sheep by omega-3 long-chain po lyunsaturates. Biol Reprod. 60, 698-701. Balaguer, S.A., Pershing, R.A., RodrguezSallaberry, C., Thatcher, W.W., Badinga, L., 2005. Effects of bovine somatotropin on uterin e genes related to the prostaglandin cascade in lactating dairy co ws. J. Dairy Sci. 88, 543-552. Barak, Y., Liao, D., He, W., Ong, E.S., Nelson, M.C., Olefsky, J.M., Boland, R., Evans, R.M., 2002. Effects of peroxisome pr oliferator-activated receptor on placentation, adiposity, and colorectal cancer. Proc. Natl. Acad. Sci. 99, 303-308. Barcikowski, K., Carlson, J.C., Wilson, L., McCracken, J.A., 1974. The effect of endogenous and exogenous estradiol-17 on the release of Prostaglandin F2 from the ovine uterus. Endocrinology 95, 1340-1349. Beam, S.W., Butler, W.R., 1999. Energy balan ce effects on follicular development and first ovulation in postpartum cows. J. Reprod. Fertil. Suppl. 54, 411-424. Behrman, H.R., MacDonald, G.J., Greep, R.O ., 1971. Regulation of ovarian cholesterol esters: evidence for enzymatic sites of pros taglandin-induced loss of corpus luteum function. Lipids 6, 791-796.

PAGE 103

90 Berger J., Moller, D.E., 2002. The mechanisms of action of PPARs. Annu. Rev. Med. 53, 409-435. Berisha, B., Schams, D., Kosmann, M., Amselgruber, W., Einspanier, R., 2000. Expression and tissue concentration of va scular endothelial growth factor, its receptors and localization in the bovine corpus luteum during estrous cycle and pregnancy. Biol. Reprod. 63, 1106-1114. Berisha, B,. Schams, D., Miyamoto, A., 2002. The expression of angiotensin and endothelin system members in bovine co rpus luteum during estrous cycle and pregnancy. Endocrine. 19, 305-312. Berridge, M.J., Irvine, R.F., 1984. Inositol tr iphosphate: a novel second messenger in cellular signal trans duction. Nature 312, 315-321. Bezard, J., Blond, J.P., Bernard, A., Clouet, P ., 1994. The metabolism and availability of essential fatty acids in animal and human tissues. Reprod. Nutr. Dev. 34, 539-568. Binelli, M., Guzeloglu, A., Badi nga, L., Arnold, D.R., Sirois, J., Hansen, T.R., Thatcher, W.W., 2000. Interferonmodulates phorbol ester-induced production of prostaglandin and expression of cyclooxygenase-2 and phospholipase-A2 from bovine endometrial cells. Biol. Reprod. 63, 417-424. Binelli, M., Subramaniam, P., Diaz, T ., Johnson, G.A., Hansen, T.R., Badinga, L., thatcher, W.W., 2001. Bovine interferon-ta u stimulates the Janus-kinase signal transducer and activator of transcription pathway in bovine endometrial epithelial cells. Biol. Reprod. 64, 654-665. Birkedal-Hansen, H., 1995. Proteolytic remode ling of the extracellular matrix. Curr. Opin. Cell Biol. 7, 728-735. Bocher, V., Pineda-Torra, I., Fruchart, J.C ., and Staels, B., 2002. PPARs: Transcription factors controlling lipid and lipoprotein metabolism. Ann. NY. Acad. Sci. 967, 718. Boshier, D. P., Holloway, H., Millener, N. M., 1981. Triacylglycerols in the rat uterine epithelium during the oestru s cycle and early pregnanc y. J. Reprod. Fertil. 62, 441446. Braden, T.D., Gamboni, F., Niswender, G.D., 1988. Effects of prostaglandin F2 -induced luteolysis on the population of cells in the ovine corpus luteum. Biol. Reprod. 39, 245-253. Braissant, O., Foufelle, F., Scotto, C., Dauca, M., Wahli, W., 1996. Differential expression of peroxisome proliferato r-activated receptors (PPARs): tissue distribution of PPAR, and in the adult rat. Endocrinology. 137, 354-366.

PAGE 104

91 Braissant, O., Wahli, W., 1998. Differential expression of peroxisome proliferatoractivated receptor-alpha, -beta, and -g amma during rat embryonic development. Endocrinology. 139, 2748-2754. Burke, J.M., Carroll, D.J., Rowe, K.E ., Thatcher, W.W., and Stormshak, F., 1996. Intravascular infusion of lipid into ewes stimulates pr oduction of progesterone and prostaglandin. Biol. Reprod. 55, 169-175. Burns, P.D., Mendes, J.O. Jr, Yemm, R.S., Cl ay, C.M., Nelson, S.E., Hayes, S.H., Silvia, W.J., 2001. Cellular mechanisms by which oxytocin mediates ovine endometrial prostaglandin F2 synthesis: role of G(i) proteins and mit ogen-activated protein kinases. Biol. Reprod. 65, 1150-1155. Chartrand, A., Matte, J.J., Lessard, M., Chouina rd, P.Y., Giguere, A., and Laforest, J.P., 2003. Effect of dietary fat sources on sy stemic and intrauterine synthesis of prostaglandins during early pregnancy in gilts. J. Anim. Sci. 81, 726-734 Chawla, A., Schwartz, E.J., Dimaculanga n, D.D., Lazar, M.A., 1994. Peroxisome proliferator-activated receptor (PPAR) : adipose-predominant expression and induction early in adipocyte diffe rentiation. Endocrinology. 135, 798-800. Cheng, Z., Robinson, R.S., Pushpakumara, P. G., Mansbridge, R.J., wathes, D.C., 2001. Effect of dietary polyunsaturated fatty aci ds on uterine prostaglandin synthesis in the cow. J. Endocrinol. 171, 463-473. Crowe, M.A., Padmanabhan, V., Mihm, M., Beitins, I.Z., Roche, J.F., 1998. Resumption of follicular waves in beef cows is not associated with periparturient changes in Follicle-Stimulating Hormone heterogeneity despite major changes in steroid and gonadotropin secretions. Biol. Reprod. 58, 1445-1450. Csapo, A.I., Pulkkinen, M., 1978. Indispensabili ty of the human corpus luteum in the maintenance of early pregnancy. Luteectom y evidence. Obstet. Gynecol. Surv. 33, 69-81. Danet-Desnoyers, G., Wetzels, C., and That cher, W.W., 1994. Natural and recombinant bovine interferonregulate basal and oxytocin-induced secretion of prostaglandin F2 and E2 by epithelial cells and stromal cells in the endometrium. Reprod. Fertil. Dev. 6, 193-202. Davis, A.J., Fleet, I.R., Harrison, F. A., Maule-Walker, F.M., 1980. Pulmonary metabolism of prostaglandin F2 in the conscious non-pregnant ewe and sow. J. Physiol. 301, 86P. Davis, A.J., Fleet, I.R., Harrison, F. A., Maule-Walker, F.M., 1985. Pulmonary metabolism of prostaglandin F2 in the conscious non-pregnant cow. J. Physiol. 358, 107P.

PAGE 105

92 Davies, C.J., Fisher, P.J., Schlafer, D.H ., 2000. Temporal and regional regulation of major histocompatibility complex class I e xpression at the bovine uterine/placental interface. Placenta 21, 194-202. Davies, C.J., Hill, J.R., Edwards, J.L., Schric k, F.N., Fisher, P.J., Eldridge, J.A., Schlafer, D.H., 2004. Major histocompatib ility antigen expression on the bovine placenta: its relationship to abnormal pregnancies a nd retained placenta. Anim. Reprod. Sci. 82/83, 267-280. Dean, J.L., Brook, M., Clark, A.R., Saklatva la, J., 1999. p38 mitogen-activated protein kinase regulates cyclooxygenase-2 mRNA stability and transcription in lipopolysaccharide-treated human monocytes. J. Biol. Chem. 274, 264-269. de la Sota, R.L., Simmen, F.A., Diaz, T ., Thatcher, W.W., 1996. Insulin-like growth factor system in bovine first-wave do minant and subordinate follicles. Biol. Reprod. 55, 803-812. DeWitt, D.L., 1991. Prostaglandin endoperoxi de synthase: regulation of enzyme expression. Biochim. Biophys. Acta. 1083, 121-134. DeWitt, D., Smith, W.L., 1995. Yes, but do they still get headaches? Cell. 83, 345-348. Ding, N.Z., Teng, C.B., Ma, H., Ni, H., Ma, X.H., Xu, L.B., Yang, Z.M., 2003. Peroxisome proliferator-activated receptor expression and regulation in mouse uterus during embryo implantation a nd decidualization. Mol. Reprod. Dev. 66, 218224. Douglas, W.S., Britt, J.H., 2000. In vivo oxytoc in release from microdialyzed bovine corpora lutea during spontaneous and pr ostaglandin-induced regression. Biol. Reprod. 62, 726-730. Driancourt, M.A., 2001. Regulation of ovarian follicular dynamics in farm animals: implications for manipulation of reproduction. Theriogenology. 55, 1211-1239. Echternkamp, S.E., Howard, H.J., Roberts, A.J ., Grizzle, J., Wise, T., 1994. Relationships among concentrations of steroids, insulin -like growth factor-I, and insulin-like growth factor binding proteins in ovarian follicular fl uid of beef cattle. Biol. Reprod. 51, 971-981. Eggleston, D.L., Wilken, C., Van Kirk, E.A ., Slaughter R.G., Ji, T.H., Murdoch, W.J., 1990. Progesterone induces expression of endometrial messenger RNA encoding for cyclooxygenase (sheep). Prostaglandins 39, 675-683. Einspanier, R., Miyamoto, A., Schams, D., Muller, M., Brem, G., 1990. Tissue concentration, mRNA expression and stimul ation of IGF-I in luteal tissue during the oestrous cycle and pregnancy of cows. J. Reprod. Fertil. 90, 439-445.

PAGE 106

93 Evans, A.C.O., Adams, G.P., Rawlings, N. C., 1994. Endocrine and ovarian follicular changes leading up to the first ovulation in prepubertal heifers. J. Reprod. Fertil. 100, 187-194. Evans, A.C.O., Fortune, J.E., 1997. Selection of the dominant follicle in cattle occurs in the absence of differences in the expr ession of messenger ribonucleic acid for gonadotropin receptors. Endocrinology 138, 2963-2971. Fairchild-Benyo, D.L., Pate, J.L ., 1992. Tumor necrosis factor alters bovine luteal cell synthetic capacity and viab ility. Endocrinology 130, 854-860. Fairclough, R.J., Moore, L.G., McGowan, L.T., Peterson, A.J., Smith, J.F., Tervit, H.R., Watkins, W.B., 1980. Temporal relationshi p between plasma concentrations of 13,14-dihydro-15-keto-prostaglandin F and ne urophysisn I/II around luteolysis in sheep. Prostaglandins 20, 199-208. Fields, M.J., Fields, P.A., 1996. Morphologica l characteristics of the bovine corpus luteum during the estrous cycle an d pregnancy. Theriogenology 45, 1295-1325. Filley, S.J., Turner, H.A., and St ormshack, F., 1999. Prostaglandin F2 concentrations, fatty acid profiles, and fertility in lipid-i nfused postpartum b eef heifers. Biol. Reprod. 61, 1317-1323. Findlay, J.K., 1993. An update in the roles of inhibin, activin, and follistatin as local regulators of folliculogenesis. Biol. Reprod. 48, 15-23. Fitz, A., Mayan, M.H., Sawyer, H.R., Nisw ender, G.D., 1982. Characterization of two steroidogenic cell types in the ovine corpus luteum. Biol. Reprod. 27, 703-722. Flower, R.J., Blackwell, G.J., 1976. The importance of phospholipase A2 in prostaglandin synthesis. Biochem. Pharmacol. 25, 285-291. Forman, B.M., Chen, B.J., Evans, R.M., 1997. The peroxisome pro liferator-activated receptors: ligands and activators. Ann. N.Y. Acad. Sci. 27, 266-275. Garrido, C., Saule, S., Gospodarowicz, D., 1993. Transcriptional regulation of vascular endothelial growth factor gene expression in ovari an bovine granulosa cells. Growth Factors 8, 109-117. Gemmell, R.T., Stacy, B.D., Thornburn, G.D., 1976. Morphology of the regressing corpus luteum in the ewe. Biol. Reprod. 14, 270-279. Ginther, O.J., 1974. Internal regulation of physiological processes through venoarterial pathways: a review. J Anim. Sci. 39, 550-564. Ginther, O.J., Kulick, L.J., Kot, K., Wiltbank, M.C., 1997. Emergence and deviation of follicles during the development of follicu lar waves in cattle. Theriogenology 48, 75-87.

PAGE 107

94 Ginther, O.J., Bergfelt, D.R., Kulick, L.J ., Kot, K., 1999. Selection of the dominant follicle in cattle: establishment of fol licle deviation in less than 8 hours through depression of FSH concentra tions. Theriogenology. 52, 1079-1093. Girsh, E., Wang, W., Mamluk, R., Arditi, F., Friedman, A., Milvae, R.A., Meidan, R., 1996. Regulation of endothelin-1 expression in the bovine corpus luteum: elevation by prostaglandin F2 Endocrinology 137, 5191-5196. Glinghammar, B., Skogsberg, J., Hamste n, A., Ehrenborg, E., 2003. PPAR activation induces COX-2 gene expression and cell pr oliferation in human hepatocellular carcinoma cells. Biochem. Biophys. Res. Commun. 308, 361-368. Glister, C., Tannetta, D.S., Groome, N.P., Knight, P.G., 2001. Interactions between follicle-stimulating hormone and growth fact ors in modulating secretion of steroids and inhibin-related peptid es by nonluteinized bovine gr anulosa cells. Biol. Reprod. 65, 1020-1028. Good, T.E.M., Weber, P.S.D., Ireland, J.L.H., Pulsaki, J., Padmanabhan, V., Schneyer, A.L., Lambert-Mersserlian, G., Ghosh, B.R ., Miller, W.L., Groom, N., Ireland, J.J., 1995. Isolation of nine different biologi cally and immunologically active molecular variants of bovine follicular inhibin. Biol. Reprod. 53, 1478-1488. Gospodarowicz, D., Cheng, J., Lui, G.M., Baird, A., Esch, F., Bohlen, P., 1985. Corpus luteum Angiogenic factor is related to fibroblast growth fa ctor. Endocrinology 117, 2383-2391. Grazul, A.T., Kirsch, J.D., Slanger, W.D., Marchello, M.J., Redmer, D.A., 1988. Prostaglandin F2, oxytocin and progesterone secret ion by bovine lu teal cells at several stages of luteal development: effects of oxyt ocin, luteinizing hormone, prostaglandin F2 and estradiol-17 Prostaglandins 38, 307-318. Grazzini, E., Guillon, B., Mouillac, B., Zingg, H. H., 1998. Inhibiton of oxytocin receptor function by direct binding of progesterone. Nature 392, 509-512. Guan, Z., Baier, L.D., Morrison, A.R., 1997. p38 mitogen-activated protein kinase downregulates nitric oxide and up-regulates prostaglandin E2 biosynthesis stimulated by interleukin-1beta. J. Biol. Chem. 272, 8083-8089. Guan, Z., Buckman, S.Y., Pentland, A.P., Templeton, D.J., Morrison, A.R., 1998. Induction of cyclooxygenase-2 by the act ivated MEKK1 --> SEK1/MKK4 --> p38 mitogen-activated protein kinase pathway. J. Biol. Chem. 273, 12901-12908. Guzeloglu, A., Subramaniam, P., Michel F., Thatcher, W.W., 2004. Interferoninduces degradation of prostaglandin H synthase -2 messenger RNA in bovine endometrial cells through a transcription-depend ent mechanism. Biol. Reprod. 71, 170-176. Hafez, E.S.E., Hafez, B., 2000. Reproduction in Farm Animals, seventh ed. Lippincott Williams & Wilkins, Philadelphia, PA.

PAGE 108

95 Ham, E.A., Cirillo, V.J., Zanetti, M.E., Ke uhl, F.A., 1975. Estrogen-directed synthesis of specific prostaglandins in the uter us. Proc. Natl. Acad. Sci. 72, 1420-1424. Hansen, P.J. Immunology of Reproduction. 2000. In: Hafez, E.S.E., Hafez, B. Reproduction in Farm Animals, seventh ed. Lippincott Williams & Wilkins, Philadelphia, PA. Hayashi, K., Miyamoto, A., 1999. Angioten sin II interacts with prostaglandin F2 and endothelin-1 as a local luteolytic factor in the bovine corpus luteum in vitro. Biol. Reprod. 60, 1104-1109. Heape, W., 1900. The sexual season of mamm als and the relation of the pro-oestrum to menstruation. Proceed. of the Royal Soc. 44, 1-69. Hihi, A.K., Michalik, L., Wahli, W., 2002. PPARs : transcriptional effectors of fatty acids and their derivatives. Cell Mol. Life Sci. 59, 790-798. Hill, J.A., Polgar, K., Anderson, D.J., 1995. Thelper 1-type immunity to trophoblast in women with recurrent spontan eous abortion. JAMA 273, 1933-1936. Hoffmann, B., Schams, D., Bopp, R., Ende r, M.L., Gimenez, T., Karg, H., 1974. Luteotrophic factors in the cow: evidence for LH rather than prolactin. J. Reprod. Fertil. 40, 1619-1623. Huggins, J.P., Pelton, J.T., Miller, R.C., 1993. Th e structure and specificity of endothelin receptors: their importance in physiology and medicine. Pharmacol. Ther. 59, 55123. Hunt, J.S., Andrews, G.U., Wood, G.W., 1987. Normal trophobl asts resist induction of class I HLA. J. Immunol. 138, 2481-2487. Inoue, H., Tanabe, T., and Umesono, K., 2000. Feedback control of cyclooxygenase-2 expression through PPAR gamma. J. Biol. Chem. 275, 28028-28032. Irvine, R.F., 1982. How is the level of free arachidonic acid controlled in mammalian cells? Biochem. J. 204, 3-16. Jang, B.C., Sanchez, T., Schaefers, H.J., Trifan, O.C., Liu, C.H., Creminon, C., Huang, C.K., Hla, T., 2000. Serum withdrawal-induced post-transcriptiona l stabilization of cyclooxygenase-2 mRNA in MDA-MB-231 ma mmary carcinoma cells requires the activity of the p38 stressactivated protein kinase J. Biol. Chem. 275, 3950739515. Jolly, P.D., McDougall, S., Fitzpatrick, L. A., Macmillan, K.L., Entwistle, K.W., 1995. Physiological effects of undernutrition on postpartum anoestrus in cows. J. Reprod. Fertil. Suppl. 49, 477-492.

PAGE 109

96 Jones, L.S., Ottobre, J.S., Pate, J.L., 1992. Progesterone regulation of luteinizing hormone receptors on cultured bovine lute al cells. Mol. Cell. Endocrinol. 85, 3339. Juengel, J.L., Garverick, H.A., Johnson, A.L., Young-Quist, R.S., Smith, M.F., 1993. Apoptosis during luteal regression in cattle. Endocrinology 132, 249-254. Karimi, K, Lennartz, M.R., 1995. Protein kina se C activation precedes arachidonic acid release during IgG-mediated ph agocytosis. J. Immunol. 15, 5786-5794. Karin, M., 1994. Signal transduction from th e cell surface to the nucleus through the phosphorylation of transcri ption factors. Curr. Opin. Cell Biol. 6, 415-424. Karsch, F.J., Legan, S.J., Ryan, K.D., Fost er, D.L., 1980. Importance of estradiol and progesterone in regulating LH secretion and estrous behavior during the sheep estrous cycle. Biol. Reprod. 23, 404-413. Karsch, F.J., Cummins, J.T., Thomas, G. B., Clarke, I.J., 1987. Steroid feedback inhibition of pulsatile secretion of gona dotropin-releasing hormone in the ewe. Biol. Reprod. 36, 1207-1218. Kerr, J.R.F., Wyllie, A.H., Currie, A.R., 1972. Apoptosis: a basic biological phenomenon with wide-ranging implications in tiss ue kinetics. Br. J. Cancer 26, 239-257. Kim, J.J., Fortier, M.A., 1995. Cell type speci ficity and protein ki nase C dependency on the stimulation of prostagla ndin E2 and prostaglandin F2 production by oxytocin and platelet-activating fact or in bovine endometrial cells. J. Reprod. Fertil. 103, 239-247. Kirby, C.J., Thatcher, W.W., Collier, R.J., Simmen, F.A., Lucy, M.C., 1996. Effects of growth hormone and pregnancy on expressi on of growth hormone receptor, insulinlike growth factor-I, and insulin-like grow th factor binding pr otein-2 and genes in bovine uterus, ovary and oviduct. Biol. Reprod. 55, 996-1002. Kobayashi, S., Miyamoto, A., Berisha, B ., Schams, D., 2001. Growth hormone, but not luteinizing hormone, acts with lut eal peptides on prostaglandin F2 and progesterone secretion by bovine corpora lutea in vitro. Prostaglandins. Lipid. Mediat. 63, 79-92. Koelle, S., Sinowatz, F., Boie, G., Linc oln, D., 1998. Developmental changes in the expression of the growth hormone recep tor messenger ribonucleic acid and protein in the bovine ovary. Biol. Reprod. 59, 836-842. Kotwica, J., Skarzynski, D., Bogacki, M., Melin, P., Starostka B., 1997. The use of an oxytocin antagonist to study the function of ovarian oxytocin during luteolysis in cattle. Theriogenology 48, 1287-1299.

PAGE 110

97 Kotwica, J., Skarzynski, D., Jaroszewski, J., Williams, G.L., Bogacki, M., 1998. Uterine secretion of prostaglandin F2 stimulated by different doses of oxytocin and released spontaneously during luteolysis in cattle. Reprod. Nutr. Dev. 38, 217-226. Kruip, T.A., Vullings, H.G., Schams, D., Jonis, J., Klarenbeek, A., 1985. Immunocytochemical demonstration of oxyt ocin in bovine ovarian tissues. Acta. Endocrinol. 109, 537-542. Kunze, H., Vogt, W., 1971. Significance of phospholipase A for prostaglandin formation. Ann. NY Acad. Sci. 180, 123-125. Kydd, J.H., Butcher, G.W., Antczak, D.F., Allen, W.R., 1991. Expression of major histocompatibility complex (MHC) cla ss 1 molecules on early trophoblast. J. Reprod. Fertil. 44 (Suppl.), 463-477. Kyriakis, J.M., Avruch, J., 1996. Sounding the al arm: protein kinase cascades activated by stress and inflammation. J. Biol. Chem. 271, 24313-24316. LaFrance, M., Goff, A.K., 1985. Effect of pregnancy in oxytocin-induced release of prostaglandin F2 in heifers. Biol. Reprod. 33, 1113-1119. LaFrance, M., Goff, A.K, 1990. Control of bovine uterine prostaglandin F2 release in vitro. Biol. Reprod. 42, 288-293. Larsen, L.N., Hovik,. K, Bremer, J., Holm K.H., Myhren, F., and Borretzen, B., 1997. Heneicosapentaenoate (21:5, n-3): its inco rporation into lipids and its effects on arachidonic acid (C20:4, n-6) and ei cosanoid synthesis. Lipids 32, 707-714. Leaver, H.A., Howie, A., Wilson, N.H., 1991. The biosynthesis of the 3-series prostaglandins in rat uterus after alphalinolenic acid feeding: mass spectroscopy of prostaglandins E and F produced by rat uter i in tissue culture. Prostaglandins Leukot. Essent. Fatty Acids. 42, 217-224. Leung, S.T., Derecka, K., Mann, G.E., Flint, A.P.F., Wathes, D.C., 2000. Uterine lymphocyte distribution and interleukin e xpression during early pregnancy in cows. J. Reprod. Fertil. 119, 25-33. Lewis, P.E., Warren, J.E., 1977. Effect of i ndomethacin on luteal function in ewes and heifers. J. Anim. Sci. 45, 763-767. Liebermann, J., Schams, D., 1994. Actions of somatotrophin on oxytocin and progesterone release from the microdialys ed bovine corpus luteum in vitro. J. Endocrinol. 143, 243-250. Liebermann, J., Schams, D., Miyamoto, A., 1996. Effects of local growth factors on the secretory function of bovine corpus luteum during the oestrous cycle and pregnancy in vitro. Reprod. Fertil. Dev. 8, 1003-1011.

PAGE 111

98 Lim, H., Paria, B.C., Das, S.K., Dinc huk, J.E., Langenbach, R., Trzaskos, J.M., Dey, S.K., 1997. Multiple female reproductive fail ures in cyclooxygenase 2-deficient mice. Cell. 91, 197-208. Lim, J.T., Piazza, G.A., Han, E.K., Delohe ry, T.M., Li, H., Finn, T.S., Buttyan, R., Yamamoto, H., Sperl, G.J., Brendel, K., Gross, P.H., Pamukcu, R., Weinstein, I.B., 1999. Sulindac derivatives inhibit growth a nd induce apoptosis in human prostate cancer cell lines. Biochem. Pharmacol. 58, 1097-1107. Lim, H., Dey, S.K., 2000. PPAR functions as a prostacyclin receptor in blastocyst implantation. Trends Endocrinol. Metab. 11, 137-142. Lim, H.J., Moon, I., Han, K., 2004. Transcri ptional cofactors exhibit differential preference toward peroxisome proliferato r-activated receptors alpha and delta in uterine cells. Endocrinology 45, 2886-2895. Lin, L.L., Wartmann, M., Lin, A.Y., Knopf, J.L., Seth, A., Davis, R.J., 1993. cPLA2 is phosphorylated and activated by MAP kinase. Cell. 72, 269-278. Liu, J., Antaya, M., Boerboom, D., Lussier, J. D., Silversides, D.W ., and Sirois, J., 1999. The delayed activation of the prostaglandi n G/H synthase-2 promoter in bovine granulosa cells is associated with down-regulation of truncated upstream stimulatory factor-2. J. Biol. Chem. 274, 35037-35045. Lucy, M.C., Collier, R.J., Kitchell, M.L., Dibner, J.J., Hauser, S.D., Krivi, G.G., 1993. Immunohistochemical and nucleic acid an alysis of somatotropin receptor populations in the bovine ova ry. Biol. Reprod. 48, 1219-1227. MacLaren, L.A., Guzeloglu, A., Bilby, T. R., Michel, F., Thatcher, W.W., 2003. Expression of peroxisome proliferator-a ctivated receptors (PPARs) in bovine endometrium. Biol. Reprod. 68 (Suppl. 1), 264. Manikkam, M., Rajamahendran, R., 1997. Progest erone-induced atresia of the proestrous dominant follicle in the bovine ovary: ch anges in diameter, insulin-like growth factor system, aromatase activity, steroid hormones and apoptotic index. Biol. Reprod. 57, 580-587. Marnett, L.J., 1992. Aspirin and the potential role of prostaglandins in colon cancer. Cancer Res. 52, 5575-5589. Marzi, M., Vigano, A., Trabattoni D., Villa, M.L., Salvaggio, A ., Clerici, E., Clerici, M., 1996. Characterization of type 1 and t ype 2 cytokine production profile in physiologic and pathologic human pr egnancy. Clin. Exp. Immun. 106, 127-133. Mattos, R., Staples, C.R., and Thatcher, W. W., 2000. Effects of dietary fatty acids on reproduction in ruminants. Rev. Reprod. 5, 38-45.

PAGE 112

99 Mattos, R., Staples, C.R., Williams, J., Am orocho, A., McGuire, M.A., Thatcher, W.W., 2002. Uterine, ovarian, and production res ponses of lactating dairy cows to increasing dietary concentra tions of menhaden fish meal. J. Dairy Sci. 85, 755-764. Mattos, R., Guzeloglu, A., Badinga, L., Staples, C.R., Thatcher, W.W., 2003. Polyunsaturated fatty acids and bovine inte rferon-tau modify phorbol-ester induced secretion of prostaglandin F2 alpha and expression of prostaglandin endoperoxide synthase-2 and phospholipase A2 in bovi ne endometrial cells. Biol. Reprod. 69, 780-787. Mayer, R.J., Marshall, L.A., 1993. New insights on mammalian phospholipase A2; comparison of arachidonyl-selective and nonselective enzymes. FASEB 7, 339348. Mazerbourg, S., Overgaard, M.T., Oxvig, C., Christiansen, M., Conover, C.A., Laurendeau, I., Vidaud, M., TosserKlopp, G., Zapf, J., Monget, P., 2001. Pregnancy-associated plasma protein-A (PAPP-A) in ovine, bovine, porcine, and equine ovarian follicles: involvement in IGF binding protein-4 proteolytic degradation and mRNA expression during follicular development. Endocrinology 12, 5243-5253. McCracken, J.A., Glew, M.E., Scaramuzzi, R. J. 1970. Corpus luteum regression induced by prostaglandin F2 J. Clin. Endocrinol. Metab. 30, 544-547. McCracken, J.A., Custer E.E., Lamsa, J. C., Robinson, A.G., 1995. The central oxytocin pulse generator: a pacemaker for lute olysis. Adv. Exp. Med. Biol. 395, 133-154. McCracken, J.A., Custer, E.E., Lamsa, J.C ., 1999. Luteolysis: a neuroendocrine-mediated event. Physiol. Rev. 79, 263-323. McGuire, W.J., Juengel, J.L., Niswender, G.D., 1994. Protein kinase C second messenger system mediates the antisteroidoge nic effects of prostaglandin F2 in the ovine corpus luteum in vivo. Biol. Reprod. 51, 800-806. Meade, E.A., Smith, W.L., and Dewitt, D.L., 1993. Differential inhibition of prostaglandin endoperoxide synthase (c yclooxygenase) isozymes by aspirin and other non-steroidal anti-inflammator y drugs. J. Biol. Chem. 268, 6610-6614. Meidan, R., Milvae, R.A., Weiss, S., Levy, N., Friedman, A., 1999. Intraovarian regulation of luteolysis. J. Reprod. Fertil. Suppl. 54, 217-228. Mihm, M., Good, T.E.M., Ireland, J.L.H., Irela nd, J.J., Knight, P.G., Roche, J.F., 1997. Decline in serum follicle-stimulating hormone concentrations alters key intrafollicular growth factors involved in selection of the dominant follicle in heifers. Biol. Reprod. 57, 1328-1337. Mihm, M., Crowe, M.A., Knight, P.G., Austi n, E.J., 2002. Follicle wave growth in cattle. Reprod. Dom. Anim. 37, 191-200.

PAGE 113

100 Miyamoto, A., Okuda, K., Schweigert, F.J., Sc hams, D., 1992. Effects of basic fibroblast growth factor, transforming growth factorand nerve growth factor on the secretory function of the bovine corpus luteum in vitro. J. Endocrinol. 135, 103114. Miyamoto, A., von Lutzow, H., Schams, D., 1993. Acute actions of prostaglandin F2 E2, and I2 in microdialyzed bovine cor pus luteum in vitro. Biol. Reprod. 49, 423430. Miyamoto, A., Kobayashi, S., Arata, S ., Ohtani, M., Fukui, Y., Schams, D., 1997. Prostaglandin F2 promotes the inhibitory actio n of endothelin-1 on the bovine luteal function in vitro. J. Endocrinol. 152, R7-R11. Miyamoto, A., Takemoto, K., Acosta, T.J ., Othani, M., Yamada, J., Fukui, Y., 1998. Comparative activities of gr owth hormone and luteinizi ng hormone in the direct stimulation of local release of progesterone from microdi alyzed ovine corpora lutea in vivo. J. Reprod. Dev. 44, 273-280. Moreira, F., Risco, C.A., Pires, M.F., Ambr ose, J.D., Drost, M ., Thatcher, W.W., 2000. Use of bovine somatotropin in lactating dairy cows receiving timed artificial insemination. J. Dairy Sci. 83, 1237-1247. Moreira, F., Badinga, L., Burnley, C., Th atcher, W.W., 2002. B ovine somatotropin increases embryonic development in superovulated cows and improves posttransfer pregnancy rates when given to lactating recipient cows. Theriogenology. 57, 1371-1387. Mukherjee, R., Jow, L., Croston, G.E ., Paterniti, J.R., 1997. Identification, characterization, and tissue di stribution of human peroxiso me proliferator-activated receptor (PPAR) isoforms PPAR 2 versus PPAR 1 and activation with retinoid X receptor agonists and antagonist s. J. Biol. Chem. 272, 8071-8076. Murdoch, W.J., 1996. Microtubular dynamics in granulose cells of periovulatory follicles and granulosa-derived (large) lutein cel ls of sheep: relationships to the steroidogenic folliculoluteal shift and f unctional luteolysis. Biol. Reprod. 54, 11351140. Nett, T.M., McClellan, M.C., Niswender, G. D., 1976. Effects of prostaglandins on the ovine corpus luteum: blood flow, secret ion of progesterone and morphology. Biol. Reprod. 15, 66-78. Newton, G.R., Vallet, J.L., Hansen, P.J., Bazer, F.W., 1989. Inhibition of lymphocyte proliferation by ovine trophoblast-1 and a high molecular weight glycoprotein produced by the peri-implantation sheep conceptus. Am. J. Reprod. Immun. 19, 99107. Nishizuka, Y., 1986. Studies and perspectives of protein kinase C. Science 233, 305-312.

PAGE 114

101 Niswender, G.D., Diekman, M. A., Nett, T.M., Akbar, A.M., 1973. relative blood flow to the ovaries of cycling and pre gnant ewes. Biol. Reprod. 9, 87. Niswender, G.D., Scwall, R.H., Fitz, T.A ., Farin, C.E., Sawyer, H.R., 1985. Regulation of luteal function in domestic ruminants: new concepts. Recent. Prog. Horm. Res. 41, 101-151. Niswender, G.D., Nett, T.M., 1988. The corpus lu teum and its control. The Physiology of Reproduction. Raven Press. NY. Niswender, G.D., Juengel, J.L., Silva, P.J., Rollyson, M.K., McIntush, E.W., 2000. Mechanisms controlling the f unction and life span of the corpus luteum. Physiol. Reviews. 80, 1-29. Nothnick, W.B., Pate, J.L., 1990. Interleuki n-1 beta is a potent stimulator of prostaglandin synthesis in bovine luteal cells. Biol. Reprod. 43, 898-903. Ohtani, M., Kobayashi, S., Miyamoto, A ., Hayashi, K., Fukui, Y., 1998. Real-time relationships between intral uteal and plasma concentra tions of endothelin, oxytocin and progesterone duri ng prostaglandin F2 -induced luteolysis in the cow. Biol. Reprod. 58, 103-108. Okuda, K., Uenoyama, Y., Berisha, B., Lange I.G., Taniguchi, H., Kobayashi, S., Miyamoto, A., Schams, D., 2001. Estradiol 17is produced in bovine corpus luteum. Biol. Reprod. 65, 1634-1639. Okuda, K., Miyamoto, Y., Skarzynski, D.J., 2002. Regulation of endometrial prostaglandin F2 synthesis during luteol ysis and early pregnancy in cattle. Domest. Anim. Endocrinol. 23, 255-264. Oldick, B.S., Staples, C.R., Thatcher, W. W., Gyawu, P., 1997. Abomasal infusion of glucose and fateffect on digestion, pr oduction, and ovarian and uterine function of cows. J. Dairy Sci. 80, 1315-1328. Olsen, S.F., Sorensen, J.D., Secher, N.J., He degaard, M., Henriksen, T.B., Hansen, H.S., Grant, A., 1992. Randomised controlled trial of effect of fish-oil supplementation on pregnancy duration. Lancet. 339, 1003-1007. Orlidge, A., DAmore, P.A., 1987. Inhibition of capillary endothelial cell growth by pericytes and smooth muscle ce lls. J. Cell. Biol. 105, 1455-1462. OShea, J.D., Cran, D.G., Hay, M.F., 1980. Fate of the theca interna following ovulation in the ewe. Cell Tissue Res. 210, 305-319. OShea, J.D., Rodgers, R.J., DOcchio, M. J., 1989. Cellular composition of the cyclic corpus luteum of the cow. J. Reprod. Fertil. 85, 483-487.

PAGE 115

102 Othani, M.S., Kobayashi, A., Miyamoto, A., Hayashi, K., Fukui, Y., 1998. Real-time relationships between intr aluteal and plasma concen trations of endothelin, oxytocin, and progesterone during prostaglandin F2 -induced luteolysis in the cow. Biol. Reprod. 58, 103-108. Palin, M.F., Brochu-Gaudreau, K., Beaudry, D ., Small, J., Petit, H.V., 2005. Effects of feeding flaxseed on cyclooxygenase 2 (C OX-2) and peroxisome proliferatoractivated receptors (PPAR) delta and gamma mRNA levels at the time of maternal recognition of pregnancy in Holstein cows. Biol. Reprod. Special Issue, 142. Parker, C.W., 1991. Neutrophil mechanisms. Am. Rev. Respir. Dis. 143, 559-560. Parkinson, T.J., Wathes, D.C., Jenner, L.J., Lamming, G.E., 1992. Plasma and luteal concentrations of oxytocin in cyclic and early-pregnant cattle. J. Reprod. Fertil. 94, 161-167. Pate, J.L., 1988. Regulation of prostaglandi n synthesis by progesterone in the bovine corpus luteum. Prostaglandins 36, 303-315. Pate, J.L., 1994. Cellular components involve d in luteolysis. J. Anim. Sci. 72, 1884-1890. Penny, L.A., Armstrong, D., Bramley, T.A., Webb, R., Collins, R.A., Watson, E.D., 1999. Immune cells and cyt okine production in the bovine corpus luteum throughout the oestrous cycle and after i nduced luteolysis. J. Reprod. Fertil. 115, 87-96. Pepperell, J.R., Wolcott, C., Behrman, H.R., 1992. Effects of neutr ophils in rat luteal cells. Endocrinology 130, 1001-1008. Perks, C.M., Peters, A.R., Wathes, D.C., 1999. Follicular and luteal expression of insulinlike growth factors I and II and the type I IGF receptor in the bovine ovary. J. Reprod. Fertil. 116, 157-165. Perks, C.M., Peters, A.R., Wathes, D.C., 1999. Follicular and luteal expression of insulinlike growth factors I and II and the type 1 IGF receptor in the bovine ovary. J. Reprod. Fertil. 116, 157-165. Pescador, N., Soumano, K., Stocco, D. M., Price, D.A., Murphy, B.D., 1996. Steroidogenic acute regulat ory protein in bovine cor pora lutea. Biol. Reprod. 55, 485-491. Phariss, B.B., Cornette, J.C., Gutnecht, G.D., 1970. Vascular control of luteal steroidogenesis. J. Repro d. Fertil. Suppl. 10, 97-103. Pontzer, C.H., Torres, B.A., Vallet, J.L ., Bazer, F.W., Johnson, H.M., 1988. Antiviral activity of the pregnanc y recognition hormone ovine trophoblast protein-1. Biochem. Biophys. Res. Comm. 152, 801-807.

PAGE 116

103 Pru, J.K., Rueda, B.R., Austin, K.J., That cher, W.W., Guzeloglu, A., Hansen, T.R., 2001. Interferonsuppresses pr ostaglandin F2 secretion independen tly of the mitogen activated protein kinase and nuclear factor B pathways. Biol. Reprod. 64, 965973. Rakugi, H., Tabuchi, Y., Nakamura, M., Na gano, M., Higashimori, K., Makami, H., Ogihara, T., Suzuki, N., 1990. Evidence for endothelin-1 release from resistance vessels of rats in response to hypoxi a. Biochem. Biophys. Res. Commun. 169, 973977. Ramrez-Godinez, J.A., Kiracofe, G.H ., Schalles, R.R., Niswender, G.D., 1982. Endocrine patterns in the postpartum beef cow associated with weaning: a comparison of the short and subsequent normal cycles. J. Anim. Sci. 55, 153-158. Reynolds, L.P., Killilea, S.D., Redmer, D.A., 1992. Angiogenesis in the female reproductive system. FASEB J 6, 886-892. Rhodes, F.M., Entwistle, K.W., Kinder, J. E., 1996. Changes in ovarian function and gonadotropin secretion preceding the onset of nutritionally induced anestrus in Bos indicus heifers. Biol. Reprod .55, 1437-1443. Roberts, R.M., Xie, S., Mathialagan, N., 1996. Maternal recognition of pregnancy. Biol. Reprod. 54, 294-302. Robertson, D.M., Klein, R., de Vos, F.L., McLachlan, R.I., Wettenhall, R.E.H., Hearn, M.T.W., Burger, H.G., de Kretser, D.M ., 1987. The isolation of polypeptides with FSH-suppressing activity from bovine follic ular fluid which are structurally different from inhibin. Biochem. Biophys. Res. Commun. 149, 744-749. Robinson, M.J., Cobb, M.H, 1997. Mitogen-acti vated protein kinase pathways. Curr. Opin. Cell Biol. 9, 180-186. Sampath, H., Ntambi, J.M., 2005. Polyunsaturated Fatty Acid regulation of genes of lipid metabolism. Annu. Rev. Nutr. 25, 317-340. Sauerwein, H., Miyamoto, A., Ginther, J ., Meyer, H.H.D., Schams, D., 1992. Binding and action of insulin-like growth factors and insulin in bovine luteal tissue during the oestrous cycle. J. Reprod. Fertil. 96, 103-115. Sawyer, H.R., Niswender, K.D., Braden, T. D., Niswender, G.D., 1990. Nuclear changes in ovine luteal cells in response to PGF2 Domest. Anim. Endocrinol. 7, 229-238. Scaramuzzi, R.J., Adams, N.R., Baird, D.T., Campbell, B.K., Downing, J.A., Findlay, J.K., Henderson, K.M., Martin, G.B., McNa tty, K.P., McNeilly, A.S., Tsonis, C.G., 1993. A model for follicle selection and the determination of ovul ation rate in the ewe. Reprod. Fertil. Dev. 5, 459-478.

PAGE 117

104 Schams, D., Amselgruber, W., Einspanier R., Sinowatz, F., Gospodarowicz, D., 1994. Localization and tissue concentr ation of basic fibroblast gr owth factor in the bovine corpus luteum. Endocrine. 2, 907-912. Schams, D., Liebermann, J., Miyamoto, A., 1995. Luteotropic role of bovine luteal oxytocin in vitro. Adv. Exp. Med. Biol. 395, 543-544. Schams, D., 1996. Paracrine/auto crine regulation of bovine co rpus luteum function. J. Reprod. Dev. 42, 54-57. Schams, D., Berisha, B., Kosmann, M., Einspanier, R., Amselgruber, W.M., 1999. Possible role of growth hormone, IGFs, a nd IGF-binding proteins in the regulation of ovarian function in large farm anim als. Domest. Anim. Endocrinol. 17, 279-285. Schams, D., Berisha, B., 2002. Va scular endothelial growth fa ctor (VEGF) and fibroblast growth factor (FGF) family members in corpus luteum of farm animals. Adv Reprod VI (No. 62). Schams, D., Berisha, B., 2004. Regulation of corpus luteum function in cattleand overview. Reprod. Dom. Anim. 39, 241-251. Scherle, P.A., Ma, W., Lim, H., Dey, S.K., Trzaskos, J.M., 2000. Regulation of cyclooxygenase-2 induction in the mouse ut erus during decidualization. An event of early pregnancy. J. Biol. Chem. 275, 37086-37092. Schiffrin, E.L., Amiri, F., Benkirane, K ., Iglarz, M., Diep, Q.N., 2003. Peroxisome proliferator-activated receptors: vasc ular and cardiac effects in hypertension. Hypertension. 4, 2664-2668. Schlafer, D.H., Fisher, P.J., Davies, C.J ., 2000. The bovine placenta before and after birth: placental development and function in health and disease. Anim. Reprod. Sci. 60/61, 145-160. Schwartzman, R.A., Cidlowski, J.A., 1993. A poptosis: the biochemistry and molecular biology of programmed cell death. Endocr. Rev. 14, 133-151. Senger, P.L., 1997. Pathways to Pregnanc y and Parturition, second ed. Current Conceptions, Inc., Pullman, WA. Silavin, S.L., Moss, G.E., Niswender, G.D., 1980. Regulation of ster oidogenesis in the ovine corpus luteum. Steroids 36, 229-241. Silvia, W.J., Lewis, G.S., McCracken, J. A., Thatcher, W.W., Wilson, L., 1991. Hormonal regulation of uterine secretion of prostaglandin F2 during luteolysis in ruminants. Biol. Reprod. 45, 655-663.

PAGE 118

105 Skarzynski, D.J., Okuda, K., 1999. Sensitivity of bovine corpora lute a to prostaglandin F2 is dependent on progesterone, oxytoc in, and prostaglandins. Biol. Reprod. 60, 1292-1298. Skarzynski, D.J., Jarosezewski, J.J., Okuda K., 2001. Luteotropic mechanisms in the bovine corpus luteum: role of oxitocin, prostaglandin F2 progesterone and noradrenaline. J. Reprod. Dev. 47, 125-137. Skopets, B., Li, J., Thatcher, W.W., Robert s, R.M., Hansen, P.J., 1992. Inhibition of lymphocyte proliferation by bovine tr ophoblast protein-1 (type 1 trophoblast interferon) and bovine in terferon-alpha 1. Vet. Immun. Immunopath. 34, 81-96. Smith, G.W., McCrone, S., Petersen, S.L ., Smith, M.F., 1995. Expression of messenger ribonucleic acid encoding tissue inhibitor of metalloproteinases-2 within ovine follicles and corpora lutea. Endocrinology 136, 570-576. Smith, G.W., Juengel, J.L., McIntush, E.W ., Young-Quist, R.S., Garverick, H.A., Smith, M.F., 1996. Ontogenies of messenger RNA encoding tissue inhibitor of metalloproteinases-1 and within bovine pe riovulatory follicles and luteal tissue. Dom. Anim. Endocrinol. 13, 151-160. Stagg, K., Spicer, L.J., Sreenan, J.M., Roch e, J.F., Diskin, M.G., 1998. Effect of calf isolation on follicular wave dynamics, gonadotropin and metabolic hormone changes, and interval to firs t ovulation in beef cows fed either of two energy levels postpartum. Biol. Reprod. 59, 777-783. Staggs, K.L., Austin, K.J., Johnson, G.A., Teix eira, M.G., Talbott, C.T., Dooley, V.A., Hansen, T.R., 1998. Complex induction of bovine uterine proteins by interferon. Biol. Reprod. 59, 293-297. Staples, C.R., Burke, J.M., Thatcher, W.W., 1998. Influence of supplemental fats on reproductive tissues and performance of l actating cows. J. Dairy Sci. 81, 856-871. Stewart, R.E., Spicer, L.J., Hamilton, T.D., Keefer, B.E., Dawson, L.J., Morgan, G.L., Echternkamp, S.E., 1996. Levels of insulin -like growth fact or (IGF) binding proteins, luteinizing hormone and IGF -I receptors, and steroids in dominant follicles during the first follicular wave in cattle exhibiting regular estrous cycles. Endocrinology 137, 2842-2850. Subbaramaiah, K., Lin, D.T., Hart, J. C., Dannenberg, A.J., 2001. Peroxisome proliferator-activated receptor gamma ligands suppress the transcriptional activation of cyclooxygenase-2. Evidence fo r involvement of activator protein-1 and CREB-binding protein/p300. J. Biol. Chem. 276, 12440-12448. Subbaramaiah, K., Marmo, T.P., Dixon, D.A ., Dannenberg, A.J., 2003. Regulation of cyclooxgenase-2 mRNA stability by taxane s: evidence for involvement of p38, MAPKAPK-2, and HuR. J. Biol. Chem. 278, 37637-37647.

PAGE 119

106 Sunderland, S.J., Crowe, M.A., Boland, M.P., Roche, J.F., Ireland, J.J., 1994. Selection, dominance and atresia of follicles during th e oestrous cycle of heifers. J Reprod. Fertil. 101, 547-555. Szekeres-Bartho, J., 1992. Immunosuppressi on by Progesterone in Pregnancy. CRC Press, Inc. Boca Ratn, FL. Tangri, S., Raghupathy, R., 1993. expre ssion of cytokines in mice undergoing immunologically mediated spontaneous fetal resorptions. Biol. Reprod. 49, 850865. Thatcher, W.W., Staples, C.R., Danet-De snoyers, G., Oldick, B., Schmitt, E.P., 1994. Embryo health and mortality in sheep a nd cattle. J. Anim. Sci. 72 (Suppl. 3), 16-30. Thatcher, W.W., Meyer, M.D., Danet-De snoyers, G., 1995. Maternal recognition of pregnancy. J. Reprod. Fertil. Suppl. 49, 15-28. Thatcher, W.W., Binelli, M., Burke, J., Staples, C.R., Ambrose, J.D., and Coelho, S., 1998. Antiluteolytic signals between the conceptus and endometrium. Theriogenology 47, 131-140. Thatcher, W.W., Guzeloglu, A., Meikle, A., Kamimura, S., Bilby, T., Kowalski, A.A., Badinga, L., Pershing, R., Bartolome, J ., Santos, J.E., 2003. Regulation of embryo survival in cattle. Reprod. Suppl. 61, 253-266. Thompson, H.J., Briggs, S., Paranka, N.S., Piazza, G.A., Brendel, K., Gross, P.H., Sperl, G.J., Pamukcu, R., Ahnen, D.J., 1995. Inhi bition of mammary carcinogenesis in rats by sulfone metabolite of sulinda c. J. Natl. Cancer Inst. 87, 1259-1260. Thompson, H.J., Jiang, C., Lu, J., Mehta, R. G., Piazza, G.A., Paranka, N.S., Pamukcu, R., Ahnen, D.J., 1997. Sulfone metabolit e of sulindac inhibits mammary carcinogenesis. Cancer Res. 57, 267-271. Trujillo, E.P., Broughton, K.S., 1995. Ingesti on of n-3 polyunsaturated fatty acids and ovulation in rats. J. Reprod. Fertil. 105, 197-203. Tsai, S., Wiltbank, M., 1997. prostaglandin F2 induces expression of prostaglandin G/H synthase-2 in the ovine corpus luteum: a potential positive feedback effect loop during luteolysis. Biol. Reprod. 57, 1016-1022. van Bilsen, M., van der Vusse, G.J., Gilde, A.J., Lindhout, M., van der Lee, K.A., 2002. Peroxisome proliferator-activated receptor s: lipid binding protei ns controling gene expression. Mol. Cell. Biochem. 239, 131-138. Van den Bosh, H., 1980. Intracellular phos pholipase A. Biochim. Biophys. Acta 604, 191-246.

PAGE 120

107 Vezza R., Habib, A., Li, H., Lawson, J.A ., Fitzgerald, G.A., 1996. Regulation of cyclooxygenase by protein kinase C. Evid ence against the importance of direct enzyme phosphorylation. J. Biol. Chem. 271, 30028-30033. Vu Hai, M.T., Logeat, F., Warembourg, M ., Milgrom, E., 1977. Hormonal control of progesterone receptors. Ann. NY Acad. Sci. 286, 199-209. Watanabe, K.R., Yoshida, T., Shimizu, T., Hayaishi, O., 1985. Enzymatic formation of prostaglandin F2 from prostaglandin H2 and D2. Purification and properties of prostaglandin F synthetase from bovi ne lung. J. Biol. Chem. 260, 7035-7041. Wiltbank, M.C., Belfiore, C.J., Niswender, G.D., 1993. Steroidogenic enzyme activity after acute activation of pr otein kinase (PK) A and PKC in ovine small and large luteal cells. Mol. Cell. Endocrinol. 97, 1-7. Wooding, F.B., 1992. Current topi c: the synepitheliochoria l placenta of ruminants: binucleate cell fusions and horm one production. Placenta 13, 101-113. Wlodawer, P., Kindahl, H., Hamberg, M., 1976. Biosynthesis of PGF2 from arachidonic acid and prostaglandin endoper oxides in the uterus. Biochem. Biophys. Acta 431, 603-614. Xiao, C.W., Liu, J.M., Sirois, J., Goff, A. K., 1998. Regulation of cyclooxygenase-2 and prostaglandin F synthase gene expres sion by steroid hormones and interferonin bovine endometrial cells. Endocrinology 139, 2293-2299. Xu, H.E., Lambert, M.H., Montana, V.G., Parks, D.J., Blanchard, S.G., Brown, P.J., Sternbach, D.D., Lehmann, J.M., Wisel y, G.B., Willson, T.M., Kliewer, S.A., Milburn, M.V., 1999. Molecular recognition of fatty acids by peroxisome proliferator-activated rece ptors. Mol. Cell. 3, 397-403. Yamada, O., Abe, M., Takehana, K., Hiraga T., Hiratsuka, H., 1994. Microvasculature of mature bovine follicles and its changes with ovulation. J. Reprod. Dev. 40, 307315.

PAGE 121

108 BIOGRAPHICAL SKETCH Cristina Francesca Lucia Caldari-Torres was born in San Juan, Puerto Rico, in 1980. She is daughter of Yvette Torres-Rivera and Dr. Pier Luigi Caldari-Collini. She graduated from University High School in 1997, and started her Bachelor of Science that same year at University of Puerto Ri co, Ro Piedras Campus. She graduated magna cum laude in 2002. After graduation she worked as a veterinary technician at a small animal clinic and as a working student at a horse farm. In fall 2003 she started her Master of Science at the University of Florida, Depart ment of Animal Sciences, under the guidance of Dr. Lokenga Badinga. Her research focu sed on the effects of nutrition on reproduction in bovids. Cristina has been a horse rider fo r 14 years and is currently working as a barn manager, training and showing hunter and jumper horses.


Permanent Link: http://ufdc.ufl.edu/UFE0013130/00001

Material Information

Title: Studies on the Effects and Mechanisms of Action of Eicosapentaenoic Acid on PGF2alpha Production in Cultured Bovine Endometrial Cells
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013130:00001

Permanent Link: http://ufdc.ufl.edu/UFE0013130/00001

Material Information

Title: Studies on the Effects and Mechanisms of Action of Eicosapentaenoic Acid on PGF2alpha Production in Cultured Bovine Endometrial Cells
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0013130:00001


This item has the following downloads:


Full Text












STUDIES ON THE EFFECTS AND MECHANISMS OF ACTION OF
EICOSAPENTAENOIC ACID ON PGF2a PRODUCTION IN CULTURED BOVINE
ENDOMETRIAL CELLS















By

CRISTINA CALDARI-TORRES


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2005



























Copyright 2005

by

Cristina Caldari-Torres
































Le dedico esta tesis a Mami, Abuela y Lola por ser el ejemplo de lo que una mujer hecha
y derecha debe ser.















ACKNOWLEDGMENTS

I would like to thank my advisor, Dr. Lokenga Badinga, for allowing me to be a

part of his laboratory. I am grateful for all his help and support, and for teaching me how

to conduct research and encouraging me to perform well academically. I would also like

to thank my committee members, Dr. Daniel C. Sharp and Dr. Charles R. Staples, for

attending the committee meetings and for their advice on how to improve my work.

I also extend many thanks to Liz S. Greene for her help with laboratory techniques,

and for her support as a lab mate and as a friend. I would like to thank Dr. Alan Ealy for

allowing me the use of his computer. Special thanks go to Carlos J. Rodriguez-Sallaberry

for helping me improve as a researcher and for his patience, friendship and

companionship. I am grateful to him for being a great source of strength and for showing

me the right way when I most needed it.
















TABLE OF CONTENTS



A C K N O W L E D G M E N T S ................................................................................................. iv

L IST O F F IG U R E S .... ...... ................................................ .. .. ..... .............. vii

A B B R E V IA T IO N K E Y ............................................ .................................................. ix

ABSTRACT .............. ............................................. xii

CHAPTER

1 IN TRODU CTION ................................................. ...... .................

2 LITERATURE REVIEW ........................................................................4

E strous C ycle .................................................................. ................................. 4
Regulation of the Estrous Cycle ................................ ......................... ........ 5
Characteristics of the Bovine Estrous Cycle ..................................................5
Phases of the E strous Cycle........................................................ ............... 6
Stages of the E strou s C ycle ....................................................................... .. ....7
A nestrus ........................ ............. ....................... ...........................
Ovarian Follicular Dynamics in Ruminants .................................... ............... 10
H ypothalam ic-pituitary-ovarian axis .................... .................. .................. 11
Follicular W aves in C attle ......................................... ................. ............... 12
Follicular Growth Factors.......................... ........................ ........... .... 13
O vulation ............................................................................................................ 15
Development and Function of the Corpus Luteum in Ruminants............. .............. 16
Prom others of Angiogenesis......... ................................................ ................ 17
Endocrine R regulators of CL Function...................................... ..................... 19
Autocrine and Paracrine Regulators ..................................... ...............19
Luteolysis in D om estic Rum inants.................................... .......................... ......... 21
Hormonal Regulation of Uterine PGF2a Synthesis ........................................23
Intracellular Signaling .................................................. ........ ....... ............24
Blood Flow and Vascular Changes .......................................... ............... 25
M orphological C hanges............................................................ .....................26
PGF2a Inhibition of P4 Synthesis..................................... ......... ............... 26
Im mune-M ediated Events ............................................................................27
Tissue M etalloproteinases ............................................................................28
A poptosis...............................................................28


v









Pregnancy Establishm ent in Rum inants ........................................ .....................28
Placentation in Ruminants .................................................... 30
Pregnancy and the Immune Response................................. ........................30
PGHS-2 Gene Expression ..................................................... 32
Shift in PG Production from PGF2a to PGE2 .......................................... 33
Effects of Polyunsaturated Fatty Acids on Reproduction in Cattle............................34
Roles of Peroxisome Proliferator-Activated Receptors in Reproduction...................36
Roles of Mitogen-activated Protein Kinases in Reproduction .............................. 38

3 STUDIES ON THE EFFECTS OF EICOSAPENTAENOIC ACID ON PGF2a
PRODUCTION IN CULTURED BOVINE ENDOMETRIAL CELLS....................41

Intro du action ...................................... ................................................ 4 1
M materials and M methods ....................................................................... ..................43
M materials ............... ................... ........................... ... .... ........ 43
Cell Culture and Treatm ent ........................................ ........................... 43
PGF2a Radioim m unoassay ............................................................................ 45
RN A Isolation and A nalysis.......................................... .......................... 45
Statistical A naly ses........... .................................................. ...... .. .. .... .... 46
R e su lts ...........................................................................................4 7
D isc u ssio n ............................................................................................................. 4 8
S u m m a ry ......................................................................................................5 0

4 STUDIES ON THE MECHANISMS OF ACTION OF EICOSAPENTAENOIC
ACID ON PGF2a PRODUCTION BY CULTURED BOVINE ENDOMETRIAL
C E L L S ...............................................................................5 8

In tro d u ctio n ........................................................................................5 8
M materials an d M eth od s ......................................................................................... 59
C ell Culture and Treatm ent ....................................................... 59
PGF2a Radioimmunoassay .................................................60
RN A Isolation and A nalysis................................ ................... 61
Statistical A nalyses................................................... 61
R e su lts ...........................................................................................6 2
D isc u ssio n ............................................................................................................. 6 3
S u m m a ry ......................................................................................................6 6

5 GENERAL DISCU SSION ....................................................................... 81

L IST O F R E F E R E N C E S ............................................................................................. 88

BIOGRAPHICAL SKETCH .............................................................. ...............108















LIST OF FIGURES


Figure page

3-1 Experimental manipulations to examine the effect of eicospaentaenoic acid
(EPA) on prostaglandin F2a (PGF2a) response to phorbol-12,13-dibutyrate
(PDBu) in bovine endometrial (BEND) cells. ................................................51

3-2 Experimental manipulations to study the effect of linoleic acid (LA) on
prostaglandin F2a (PGF2a) response to eicosapentaenoic acid (EPA) in bovine
endom trial (B EN D ) cells. ............................................... .............. ...............52

3-3 Effect of eicosapentaenoic acid (EPA) on prostaglandin F2a (PGF2a) response to
phorbol-12,13-dibutyrate (PDBu) in bovine endometrial (BEND) cells.................53

3-4 Effect of eicosapentaenoic acid (EPA) on prostaglandin endoperoxide synthase
(PGHS-2) mRNA response to phorbol-12,13-dibutyrate (PDBu) in bovine
endom trial (B E N D ) cells............................................... ..................................54

3-5 Effect of eicosapentaenoic acid (EPA) on peroxisome proliferator-activated
receptor 6 (PPAR6) mRNA response to phorbol-12,13-dibutyrate (PDBu) in
bovine endometrial (BEND) cells ................................................ ............... 55

3-6 Effects of increasing n-6/n-3 fatty acid ratios on prostaglandin F2a (PGF2a)
response to phorbol-12-13-dibutyrate (PDBu)...................................................... 56

3-7 Effects of increasing n-6/n-3 fatty acid ratios on prostaglandin endoperoxide
synthase (PGHS-2) response to phorbol-12,13-dibutyrate (PDBu).........................57

4-1 Experimental manipulations to determine if eicosapentaenoic acid (EPA)-
mediated inhibition of prostaglandin F2a (PGF2a) secretion involves peroxisome
proliferator-activated receptor 6 (PPAR6) activation in bovine endometrial
(B E N D ) cells. ...................................................... ................. 67

4-2 Experimental manipulations to determine if peroxisome proliferator-activated
receptor 6 (PPAR6) inhibition may abolish eicosapentaenoic acid (EPA) effect
on prostaglandin F2a (PGF2a) secretion in bovine endometrial (BEND) cells. .......68

4-3 Experimental manipulations to examine the role of mitogen-activated protein
(MAP) kinases in eicosapentaenoic acid (EPA) regulation of prostaglandin F2a
(PGF2a) production in bovine endometrial (BEND) cells.....................................69









4-4 Effect of a peroxisome proliferator-activated receptor 6 (PPAR6) agonist (L-
165,041) on prostaglandin F2a (PGF2a) secretion in bovine endometrial (BEND)
c e lls ........................................................................... 7 0

4-5 Effect of a peroxisome proliferator-activated receptor 6 (PPAR6) agonist (L-
165,041) on prostaglandin endoperoxide synthase (PGHS-2) mRNA abundance
in bovine endom trial (B EN D ) cells............................................. .....................71

4-6 Effect of a peroxisome proliferator-activated receptor 6 (PPAR6) agonist (L-
165,041) on PPAR6 mRNA response to PDBu in bovine endometrial (BEND)
c e lls ........................................................................... 7 2

4-7 Effect of sulindac sulfide on prostaglandin F2a (PGF2a) response to
eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.......................73

4-8 Effect of sulindac sulfide on prostaglandin endoperoxide synthase (PGHS-2)
mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND)
c e lls ........................................................................... 7 4

4-9 Effect of sulindac sulfide on peroxisome proliferator-activated receptor 6
(PPAR6) mRNA response to eicosapentaenoic acid (EPA) in bovine
endom trial (B EN D ) cells............................................... ............................. 75

4-10 Effect of sulindac sulfone on prostaglandin F2a (PGF2a) response to
eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.......................76

4-11 Effect of sulindac sulfone on prostaglandin endoperoxide synthase (PGHS-2)
mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial (BEND)
c e lls ........................................................................... 7 7

4-12 Effect of sulindac sulfone on peroxisome proliferator-activated receptor 6
(PPAR6) mRNA response to eicosapentaenoic acid (EPA) in bovine
endom trial (B EN D ) cells............................................... ............................. 78

4-13 Effect of SB203580 and PD98059 on prostaglandin F2a (PGF2a) response to
eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells.......................79

4-14 Effect of SB203580 and PD98059 on prostaglandin endoperoxide synthase
(PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine
endom trial (B EN D ) cells............................................... ............................. 80

5-1 Proposed model for eicosapentaenoic acid (EPA) and linoleic acid (LA) regulation
of prostaglandin F2a (PGF2a) biosynthesis in bovine endometrial (BEND) cells...86

5-2 Signaling cascade leading to synthesis and activation of prostaglandin
endoperoxide synthase (PGHS-2) in bovine endometrial (BEND) cells.............87
















ABBREVIATION KEY


9k-PGR

AA

AP-1

BEND cells

CA

CL

CRE

CREB

DAG

DGLA

DPA

E2

EPA

ERK

ET-1

FGF

FSH

FO

GH


PGE2-9-keto reductase

Arachidonic acid

Activator protein-1

Bovine endometrial cells

Corpus albicans

Corpus luteum

cAMP response element

cAMP response element binding protein

1,2-diacylglycerol

dihomo-y-linolenic

docosapentaenoic acid

Estrogen

Eicosapentaenoic acid

Extracellular-regulated kinase

Endothelin-1

Fibroblast growth factor

Follicle-stimulating hormone

Fish oil

Growth hormone









GnRH

GLA

IGF

IGFBP

IGFR

IFN-T'

IP3

JAK/STAT



LA

LH

LNA

LXR

MAP kinase

MCP-1

MHC

NE

OT

P4

PDBu

PG

PGE2

PGF2a


Gonadotropin-releasing hormone

y-linolenic acid

Insulin-like growth factor

Insulin-like growth factor-binding protein

Insulin-like growth factor receptor

Interferon c

Inositol-1,4,5-triphosphate

Janus kinase signal transducer and activator

of transcription

Linoleic acid

Luteinizing hormone

Linolenic acid

Liver X receptor

Mitogen-activated protein kinase

Monocyte chemoattractant protein-1

Major histocompatibility complex

Norepinephrine

Oxytocin

Progesterone

Phorbol-12,13-dibutyrate

Prostaglandin

Prostaglandin E2

Prostaglandin F2a









PGFS Prostaglandin F synthase

PGFM PGF2a metabolite (13,14-dihydro-15-keto

prostaglandin F2a)

PGH2 Prostaglandin H2

PGHS-2 Prostaglandin endoperoxide synthase-2

PGI2 Prostaglandin 12 (Prostacyclin)

PKC Protein kinase C

PLA2 Phospholipase A2

PLC Phospholipase C

PPARs Peroxisome proliferators-activated receptors

PPRE PPAR response element

PUFAs Polyunsaturated fatty acids

RAR Retinoic acid receptor

RXR Retinoid X receptor

StAR Steroidogenic acute regulatory protein

TIMP Tissue inhibitor of metalloproteinases

TNF-a Tumor necrosis factor a

TPA Tissue-type plasminogen activator

VEGF Vascular endothelial growth factor















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

STUDIES ON THE EFFECTS AND MECHANISMS OF ACTION OF
EICOSAPENTAENOIC ACID ON PGF2a PRODUCTION IN CULTURED BOVINE
ENDOMETRIAL CELLS

By

Cristina Caldari-Torres

December, 2005

Chair: Lokenga Badinga
Major Department: Animal Sciences

Recent studies have implicated n-3 polyunsaturated fatty acids (PUFAs) in the

reduction of prostaglandin F2a (PGF2a) synthesis in the bovine uterus. Although cattle

diets contain a mixture of n-3 and n-6 fatty acids, currently there is a lack of information

as to how these fatty acids may interact to alter PGF2a biosynthesis in the uterus. The

objective of this thesis was to examine the physiological effects and mechanisms of

action of eicosapentaenoic acid (EPA; 20:5, n-3) on PGF2a production in phorbol-12,13-

dibutyrate (PDBu)-stimulated bovine endometrial (BEND) cells. Pre-incubation of

confluent BEND cells with EPA for 24 h decreased PGF2a response to PDBu, but had no

detectable effect on prostaglandin endoperoxide synthase-2 (PGHS-2) or peroxisome

proliferator-activated receptor 6 (PPAR6) mRNA abundance. The inhibitory effect of

EPA on PGF2a secretion was reverted when increasing amounts of linoleic acid (LA;

18:2, n-6) were added to the incubation medium.









Activation and inhibition of PPAR6 greatly reduced PGF2a and PGHS-2 mRNA

responses to PDBu. The PPAR6 agonist, L-165,041, and sulindac sulfide decreased

PGHS-2 mRNA response to PDBu, whereas EPA and sulindac sulfone had no detectable

effects on PGHS-2 mRNA abundance in PDBu-stimulated BEND cells.

Selective inhibition of p38 and ERK MAP kinases did not affect PGF2a and

PGHS-2 mRNA responses to EPA, suggesting that EPA may affect the PGF2a cascade at

a site or sites that are distal to these two MAP kinases.

In summary, this study presents direct evidence that the inhibition of uterine

endometrial PGF2a biosynthesis by n-3 fatty acids depends on the amount of n-6 fatty

acids in the uterus, and that EPA and PPAR6 affect uterine PGF2a synthesis through

complex mechanisms which may or may not involve PGHS-2 gene regulation. Further

studies are needed to fully document the role of p38 and ERK MAP kinases in

endometrial PGF2a biosynthesis.














CHAPTER 1
INTRODUCTION

One of the most important aspects of the dairy cattle industry is the ability to

maintain reproductively efficient animals. The inability of a dairy cow to have a normal

reproductive cycle and become pregnant will lead to the culling of the animal, since

failure to have a pregnant cow will result in no milk production. Because of this,

researchers have dedicated a lot of time and effort in studying the bovine reproductive

cycle and investigating ways to make dairy cows as reproductively efficient as possible,

which would help contribute to maximal milk revenue.

Embryonic mortality is one of the reproductive problems associated with increased

milk production by dairy cattle. It has been estimated that up to 40% of total embryonic

losses occur between days 8 and 17 of pregnancy, which is also the period of time when

concepts inhibition of uterine prostaglandin F2z (PGF2a) occurs. Luteolysis in domestic

ruminants is caused by an episodic release of PGF2a from the uterus that reaches the

corpus luteum (CL) and causes its regression. If the developing concepts is not able to

inhibit the production of PGF2,, the CL regresses, progesterone (P4) production

decreases, and pregnancy cannot be maintained. Many researchers have focused on in

vivo and in vitro studies to produce and evaluate systems that allow for increased embryo

survival (Oldick et al., 1997; Moreira et al., 2000; Binelli et al., 2001; Badinga et al.,

2002; Mattos et al., 2002, 2003). Results from several of these studies have shown that

manipulating the polyunsaturated fatty acid (PUFA) content of the cow's diet could help









inhibit PGF2a production, but the mechanism of action of these fatty acids is not known.

Polyunsaturated fatty acids are fatty acids with two or more double bonds. Fatty acids

are unsubstituted monocarboxylic acids that occur mainly as esters in natural fats and

oils, and may also exist as non-esterified forms known as free fatty acids.

Polyunsaturated fatty acids can be divided in n-3 or n-6 fatty acids, depending on the

position of the first double bond along the hydrocarbon chain. Those PUFA's with their

first double bond at the third position of the hydrocarbon chain counting from the methyl

end are classified n-3, while those with their first double bond at the sixth position are

classified as n-6 PUFA's. Prostaglandin F2a is derived from the action of prostaglandin

endoperoxide synthase-2 (PGHS-2) on arachidonic acid (AA), a PUFA that is cleaved

from membrane phospholipids by the action of phospholipase A2 (PLA2). A study by

Mattos et al. (2003) with bovine endometrial (BEND) cells showed that n-3 fatty acids

induced greater PGF2z inhibition than n-9 or n-6 fatty acids.

It has been suggested that PUFAs may inhibit PGF2a synthesis by decreasing the

availability of AA, increasing the concentration of fatty acids that compete with AA for

processing by PGHS-2, inhibiting PGHS-2 synthesis and or activity, or affecting gene

expression through activation of nuclear transcription factors (Mattos et al., 2000).

Peroxisome proliferator-activated receptors (PPARs) are an example of nuclear hormone

receptors that mediate the effects of fatty acids and their derivatives at the level of gene

expression. Three isoforms, each encoded by a different gene, have been identified:

PPARac, PPAR6, and PPARy. Of these isoforms, PPAR6 is the only PPAR that has been

associated with reproductive function (Berger et al., 2002).









The objective of this thesis was to examine the physiological effects and

mechanisms of action of eicosapentaenoic acid (EPA), an n-3 fatty acid, on PGF2a

production in phorbol-12,13-dibutyrate (PDBu)-stimulated BEND cells. Phorbol-12,13-

dibutyrate (PDBu) is a phorbol-ester that activates protein kinase C (PKC), a step

necessary for prostaglandin (PG) production, and is therefore used to stimulate PGF2a

production by cultured BEND cells. If EPA affects PGF2a production through the

activation of PPAR6, then activation of this nuclear receptor would mimic the effect of

EPA on PGF2a production in cultured BEND cells. Alternatively, if the aforementioned

hypothesis is correct, inhibition of PPAR6 should block the effect of EPA on PGF2a

production in cultured BEND cells. Also, if EPA exerts its PGF2z-inhibitory effect

through the activation of the mitogen-activated protein (MAP) kinases ERK or p38,

which have been implicated in the activation of PGHS-2, then inhibition of the ERK and

p38 MAP kinases should block the effect of EPA on PGF2a production in cultured BEND

cells.

This thesis will consist of a literature review (Chapter 2) that will focus on

ruminants, specifically discussing their estrous cycle, ovarian follicular dynamics, CL

development and function, luteolysis, pregnancy establishment, the effects of PUFAs on

reproduction, and the role of PPARs in reproduction. Chapter 3 will present the

experiments that were conducted to study the physiological effects of EPA on PGF2a

production in cultured BEND cells, while Chapter 4 will center on the studies conducted

to elucidate the mechanisms of action of EPA in the inhibition of PGF2a production in

cultured BEND cells. Chapter 5 will consist of a discussion and concluding remarks.














CHAPTER 2
LITERATURE REVIEW

Estrous Cycle

Estrous cycles provide animals with numerous opportunities for copulation and

conception, allowing them to fulfill the ultimate goal in life, procreation. This cycle

varies in length from species to species, and the different types of estrous cycles can be

categorized based on the number of estrous cycles per year and by whether they are

regulated seasonally. In cattle, selection has favored a strategy of polyestrous cycles.

This means that an animal has a uniform distribution of estrous cycles that occur

regularly throughout the year. Seasonally polyestrous females, like sheep and mares,

display numerous estrous cycles only during certain seasons of the year. These females

can be divided further into long-day and short-day breeders. Long-day breeders begin to

cycle as day length increases, while short-day breeders cycle when the length of daylight

decreases. This is practical since long-day breeders, like mares, tend to have a gestation

length of almost a year, while a short-day breeder's gestation length is closer to 4 Mo.

The gestation length and the seasonal regulation ensure that parturition occurs when there

are enough natural resources and the season is favorable for the offspring. Animals such

as dogs and wolves have monoestrous cycles meaning that they only have one cycle per

year. To ensure that mating does occur, the estrus (heat) phase of the estrous cycle can

last up to several days. Regardless of species, the estrous cycle is characterized by a

series of predictable events that begin at estrus and end at the subsequent estrus. Among

these events are alternating periods of sexual receptivity and non-receptivity. Sexual









receptivity and mating are characteristic behavior of a female in the estrus stage of the

estrous cycle. If conception does not occur, a new estrous cycle begins, but if it does

occur, the animal goes into an extended period of sexual non-receptivity known as

anestrus. Diestrus is the period of sexual non-receptivity that characterizes the interval

between periods of estrus (Heape, 1900).

Regulation of the Estrous Cycle

The estrous cycle is regulated by endocrine and neuroendocrine mechanisms. This

regulatory network includes gonadotropin releasing hormone (GnRH), follicle-

stimulating hormone (FSH), luteinizing hormone (LH), progesterone (P4) and estradiol

(E2). These hormones are interconnected by the hypothalamic-pituitary-ovarian axis,

which controls follicular growth, ovulation, and luteal function. Follicle-stimulating

hormone and LH bind to specific receptors on the ovary and stimulate follicular

development to produce a mature and competent egg (Driancourt, 2001). Gonadotropins

released by the pituitary gland are influenced by changes in rate of GnRH synthesis and

degradation. These endocrine changes lead to marked morphologic and secretary

changes in the ovaries and tubular genitalia of the animal such as (i) changes in vaginal

and uterine cytology, (ii) changes in cervical tonus and water content, (iii) increases in

blood supply, and (iv) changes in follicular development.

Characteristics of the Bovine Estrous Cycle

The length of the bovine estrous cycle is 20-21 d with heat or estrus occurring on

day 1 and ovulation occurring 10-12 h after the end of standing estrus, or 24-30 hours

after the onset of estrus (Hafez and Hafez, 2000). Mean duration of estrus is 15 hours,

but can range anywhere from 6 to 24 hours (Senger, 1997). Ovulation in bovids is

spontaneous, meaning that the female periodically becomes sexually receptive. Selection









has favored polyestrus females, which have estrous cycles that occur with a uniform

distribution throughout the year, allowing them to become pregnant without regard to

season. In bovids, copulation usually occurs before ovulation, providing an opportunity

for conception. Commonly only one follicle ovulates, although about 10% of the time

two follicles may ovulate (Hafez and Hafez, 2000). Follicles ovulate on the right ovary

about 60% of the time and on the left ovary about 40% of the time (Hafez and Hafez,

2000).

Phases of the Estrous Cycle

The estrous cycle is divided into a short follicular phase and a longer luteal phase.

Each of these phases is dominated by one of the gonadotropic hormones, P4 or E2. The

period of time from the regression of the corpus luteum (CL) to ovulation is known as the

follicular phase. During this phase, preovulatory Graafian follicles are the primary

ovarian structures. These produce estradiol-173 (E2), the dominant reproductive

hormone. High E2 concentrations lead to behavioral alterations and major physiological

changes in the reproductive tract. This steroid is responsible for the behavioral

characteristics observed in cows that are approaching or have reached sexual receptivity.

High E2 concentrations at estrus lead to a decrease in the viscosity of cervical mucus,

dilation of the cervix, improved contractility or tonicity of the uterus and increase in

vascular growth of the endometrium (Hafez and Hafez, 2000). These characteristic

physiological changes favor copulation and fertilization.

The luteal phase of the estrous cycle is longer than the follicular phase and takes

place from the time of ovulation to regression of the CL. The CL, the dominant structure

of the luteal phase, produces P4. Progesterone has the opposite physiological and









behavioral effects of E2, causing the cow to be in a phase of sexual non-receptivity. This

steroid is essential for pregnancy and, therefore, is known as the pregnancy hormone.

Progesterone induces physiological changes that prepare a suitable environment for

embryo development and attachment to the endometrium. When P4 is high, cervical

mucus is thick, the cervical canal is tightly closed and the myometrium is relaxed (Hafez

and Hafez, 2000). Thick cervical mucus and a tightly closed cervical canal help restrict

microorganisms from the upper reproductive tract to maintain a clean environment for the

embryo, while the quiescent uterus allows implantation of the embryo.

Stages of the Estrous Cycle

The estrous cycle can be divided further into the following four stages: proestrus,

estrus, metestrus, and diestrus. Proestrus is the stage that begins after luteolysis as P4

starts to decrease and ends at the onset of estrus. This stage is characterized by the

formation of ovulatory follicles that produce E2 and lead to a shift from the P4-dominated

to the E2-dominated phase. During proestrus, the female's reproductive tract undergoes

physiological changes that prepare it for the onset of estrus and mating.

Estrus is an easily recognizable stage since the high E2 levels in blood lead to

visible behavioral symptoms in the animal. As the animal enters estrus, she displays a

behavior which indicates that she is approaching sexual receptivity such as increased

locomotion, mounting of other females, nervousness and phonation (Senger, 1997).

When fully sexually receptive, the female displays standing estrus, and in some cases

lordosis (arching of back in preparation for mounting).

From the time of ovulation to the formation of a functional CL, the animal is in the

metestrus stage of the estrous cycle. This stage is characterized by a decrease in E2 levels

and an increase in P4 levels. Once the follicle ovulates, it undergoes cellular and









structural remodeling that result in the formation of the CL, the dominant intraovarian

structure of the luteal phase of the estrous cycle. This endocrine gland produces P4,

causing this hormone to be detected soon after ovulation.

The longest stage of the estrous cycle is diestrus. During this stage, the CL (or

corpora lutea in the rare case of more than one follicle ovulating) is fully functional and

produces the maximal amount of P4. The duration of diestrus depends directly on the

amount of time that the CL remains functional, meaning that if conception occurs, the CL

will persist throughout pregnancy and so will diestrus. On the other hand, if there is no

mating or conception, luteolysis will take place, diestrus will end, and a new estrous

cycle will begin.

Anestrus

A female that does not exhibit regular estrous cycles is referred to as being in

anestrus. Anestrus can occur if there is a failure to stimulate and maintain gonadotropin

secretion due to insufficient GnRH release from the hypothalamus (Senger, 1997). The

lack of gonadotropins prevents the formation of ovulatory follicles or a functional CL.

Anestrus can be caused by pregnancy, presence of offspring, stress, pathology,

undernutrition, or season (Senger, 1997).

Anestrus caused by pregnancy is a direct result of endocrine factors, such as

continuous P4 secretion mainly by the placenta (Hafez and Hafez, 2000). Once

parturition occurs, P4 concentrations decrease dramatically, and a new estrous cycle can

take place. It is important to note that for anestrus to end there needs to be not only

parturition, but uterine involution, since the animal needs to have a normal uterus before

it goes into estrus again and becomes pregnant.









The extent of postpartum anestrus can also depend on mammary stimulation by the

offspring. Frequent nursing leads to an increase in serum concentrations of prolactin.

Serum prolactin concentration is inversely related to FSH and LH concentrations (Hafez

and Hafez, 2000), meaning that frequent nursing leads to a decrease in the concentration

of the hormones responsible for follicle growth and ovulation. Weaning of offspring can

result in short estrous cycles in postpartum anestrous females. Almost 80% of

postpartum anestrous cows that had their calf weaned and exhibited estrus within 10 days

of weaning had estrous cycles of about 7-12 days in length (Ramirez-Godinez et al.,

1982).

There is a very close relationship between fertility and body condition and

nutritional status of females, especially in dairy cattle. Undernutrition and a prolonged

negative energy balance can significantly extend the length of postpartum anestrus in

these animals (Beam and Butler, 1999). Studies with cycling heifers showed that severe

restriction of energy intake can induce nutritional anestrus when body weight decreases

by 15% (Rhodes et al., 1996). In high producing dairy cows, the amount of milk

production is related to the amount of time from parturition to the cow's first ovulation

(Hafez and Hafez, 2000). This is related to the energy balance of the animal, since to

maintain high milk production, the cow needs to mobilize fat from its body reserves. If

the cow is in a state of negative energy balance, it will use its energy primarily for milk

production and reproductive processes will be negatively affected. Undernutrition,

caused by under consumption of nutrients before or after calving, may interfere with

follicle maturation, ovulation, or other mechanisms regulating ovarian follicular

dynamics (Jolly et al., 1995).









It is also worth noting that after parturition, the incidence of ovulations where the

cow does not exhibit any behavioral characteristics of estrus approximates 50-95%

(Hafez and Hafez, 2000). Most dairy cows have normal fertile estrous cycles by around

35 days postpartum. In cows that ovulate early after calving (such as pluriparous cows),

the conception rate is lower when bred at first postpartum estrus than at subsequent

estrous cycles (Hafez and Hafez, 2000).

Ovarian Follicular Dynamics in Ruminants

In ruminants, ovarian follicular dynamics are regulated by the hypothalamic-

pituitary-ovarian axis. This axis initiates and regulates the reproductive processes

through neuro-endocrine systems. Hormones produced by the hypothalamus stimulate

(or inhibit) the pituitary gland which releases hormones in an endocrine fashion, which

bind to receptors in the ovaries leading to morphological and endocrine changes in these

tissues. These changes cause the ovaries to go through several follicular waves,

ultimately selecting a dominant follicle that will produce more E2 than the subordinate

follicles, and which will eventually ovulate. After ovulation, a CL is formed, and the

luteal phase of the estrous cycle begins. The presence or absence of an embryo creates a

hormonal environment that leads to either the maintenance or regression of the CL,

respectively. Regression of the CL (luteolysis) permits the ovulation of the dominant

follicle of the existing follicular wave, and the cycle starts again.

Although much of the literature focuses on the gonadotropins as the major

hormonal players in follicular dynamics, new research has established that polypeptide

growth factors such as insulin, insulin-like growth factor (IGF), and insulin-like growth

factor binding proteins (IGFBPs) are key for oocyte maturation and ovulation, and for

establishment of the dominant follicle.











Hypothalamic-pituitary-ovarian axis

Regulation of follicle development includes both positive and negative endocrine

feedback, occurring at the level of the pituitary and hypothalamus. The hypothalamus is

connected to the pituitary gland through the hypothalamic-hypophyseal portal system,

which consists of a capillary plexus that can transport GnRH released by specialized

hypothalamic nerve cells into the anterior pituitary, and also a neural connection to the

posterior pituitary, where the hormone oxytocin (OT) is stored. The hypothalamic

hormone GnRH is released in pulses into the portal system in response to neural signals,

where it signals the release of FSH and LH from the anterior pituitary (Hafez and Hafez,

2000). The gonadotropins FSH and LH are responsible for the growth and selection of

the dominant follicle, estrogen secretion from the follicle, and for the ovulation of the

follicle, respectively. Through the hypothalamic-pituitary-ovarian axis, the pituitary

hormones can exert a negative feedback to regulate the functions of the hypothalamus

(Hafez and Hafez, 2000). Studies performed by Karsch et al. (1980; 1987) suggest that in

ovids E2 and P4 have a negative feedback effect on LH pulse amplitude and frequency,

respectively, but that if E2 levels become sufficiently high the negative feedback effect

becomes interrupted, and a positive feedback effect is observed causing an LH surge to

occur. The negative feedback effects of E2 and P4 on gonadotropin secretion in the ewe

include an action on the brain and a consequent inhibition of pulsatile GnRH secretion

(Karsch et al., 1987). The positive feedback effect observed when E2 levels become

sufficiently high allows for pulsatile GnRH secretion from the hypothalamus which

results in the surge of LH necessary for follicle ovulation.









As follicles grow, they attain steroidogenic capabilities, which allow them to

produce estrogens, E2 being the primary biologically active estrogen produced (Hafez and

Hafez, 2000). Luteal cells of the CL are also steroidogenic, but produce P4 primarily.

The CL is composed of large and small luteal cells. The large luteal cells spontaneously

secrete P4 at a high rate, whereas the small luteal cells secrete little P4 in the absence of

LH stimulation (Hafez and Hafez, 2000). Progesterone has a negative feedback effect on

the hypothalamus, which prevents it from releasing GnRH at a high amplitude and

frequency (Hafez and Hafez, 2000), which in turn causes a very low release of LH, the

hormone necessary for follicle ovulation. When luteolysis occurs and P4 levels are low,

the negative feedback effect on the hypothalamus is removed and a new follicular wave

can occur.

Follicular Waves in Cattle

In bovids, selection of an antral follicle that will eventually ovulate usually occurs

in 2 or 3 waves. The ovary has a primordial follicle reserve, formed during fetal life or

soon after birth (Hafez and Hafez, 2000), from which follicles are continuously being

selected to grow and mature. Follicle growth refers to the proliferation and

differentiation of theca and granulosa cells, which leads to an increased ability to produce

E2 and respond to gonadotropins (Hafez and Hafez, 2000). Follicle growth, which is

slow and gonadotropin-independent initially (Scaramuzzi et al., 1993), becomes

gonadotropin-dependent during the final stages of maturation. Transient FSH rises

occurring at regular 8-10 day intervals in cattle (Mihm et al., 2002) lead to the emergence

of up to 24 small (3-5 mm) follicles that grow beyond 4 mm in diameter (Adams et al.,

1992). Follicles that are smaller than 4 mm require FSH to grow, while large antral

follicles (7-9 mm) require LH for maturation (Hafez and Hafez, 2000). These sequential









FSH rises associated with new follicle waves occur not only during the estrous cycle but

also in the postpartum period (Crowe et al., 1998; Stagg et al., 1998), and before puberty

in cattle (Evans et al., 1994). As FSH concentrations start to decline, one follicle

becomes the dominant follicle, whereas the rest of the follicles undergo atresia via

apoptosis (Austin et al., 2001). The dominant follicle exhibits enhanced growth and E2

production, becomes FSH-independent and LH-dependent, and maintains FSH

concentrations low to prevent the growth of subordinate follicles (Ginther et al., 1997;

1999). The dominant follicle continues to grow and produce E2 for 3 to 4 days, but as the

CL produces high levels of P4, LH pulses are altered to a low frequency, high amplitude

pattern causing the dominant follicle to become atretic (Sunderland et al., 1994; Evans et

al., 1997). As the follicle becomes atretic, it starts to secrete less E2 and eventually loses

dominance. When the dominant follicle regresses, there is another transient FSH rise,

which triggers the emergence of a new wave of follicles and selection of a new dominant

follicle (Sunderland et al., 1994). Following luteolysis, P4 levels decrease, removing this

hormone's negative feedback effect on the hypothalamus. This allows LH to be secreted

from the pituitary gland. A frequent LH pulse pattern during the follicular phase supports

the existing dominant follicle, allowing it to undergo final differentiation, induce the

gonadotropin surge, ovulate and luteinize (Mihm et al., 2002).

Follicular Growth Factors

Although FSH and LH have historically been defined as the factors stimulating

follicular and luteal growth and differentiation, more recent research has revealed that

other factors are also involved. These include inhibins, activin, follistatin, IGF, and

IGFBPs.









Inhibins are proteins that exert an FSH-suppressive action on the pituitary. Bovine

follicular fluid contains dimeric forms (a and 3 subunits) of inhibins (Good et al., 1995).

Activin is a dimer linking two 3 inhibin subunits and the interaction of activin and its

receptor is regulated by follistatin, which neutralizes activin functions in the pituitary and

ovary (Robertson et al., 1987; Findlay, 1993). Recruited follicles produce inhibins,

causing FSH levels to decrease and LH levels to increase. During the first 33 hours of

FSH decline, an increase in the larger molecular weight inhibins is seen in the fastest

growing cohort follicles, and activin increases in the largest follicle (Austin et al., 2001).

In the next 24 hours, the two largest follicles maintain high levels of the high molecular

weight inhibins, but the larger of the two shows a reduction in follistatin concentrations

(Austin et al., 2001). The transient FSH rise will result in enhanced growth and E2

synthesis in successful follicles, while follistatin and the amounts of the native 34 kDa

inhibin dimer are kept at low concentrations (Mihm et al., 2002).

The effects of IGFs on follicular granulosa cells are dependent on the concentration

of FSH. In vitro experiments have shown that IGFs stimulate proliferation of follicular

granulosa cells, and enhance E2, inhibin and activin synthesis (Glister et al., 2001).

Insulin-like growth factor binding proteins bind IGF with high affinity, making IGF

unavailable to the IGF receptor. High intrafollicular amounts of IGFBP-2, -4 and -5

(lower molecular weight) is negatively correlated with follicle estrogen activity

(Echternkamp et al., 1994), and atretic bovine follicles have been shown to have

increased concentration of these IGFBPs (Mazerbourg et al., 2001). Low amounts of

IGFBPs are maintained in the dominant follicle, and this follicle demonstrates enhanced

IGF bioavailability as a result of higher IGF-II synthesis and IGF-I binding, while









subordinate follicles show increased levels of IGFBP-4 and -5 (de la Sota et al., 1996;

Stewart et al., 1996; Mihm et al., 1997; Austin et al., 2001). As the dominant follicle

loses dominance, there is an increase in low molecular weight IGFBPs and an increase in

the incidence of granulosa cell apoptosis (de la Sota et al., 1996; Stewart et al., 1996;

Manikkam and Rajamahendran, 1997).

Ovulation

As the dominant follicle grows, it attains LH receptors, which allow it to bind LH

released from the pituitary gland. When this gonadotropin binds to its receptor, a cascade

of intracellular events is initiated whereby cholesterol is converted to testosterone.

Testosterone diffuses out of the theca internal cells and enters the granulosa cells where it

is converted to E2 (Senger, 1997). The E2 concentrations increase until they reach a

threshold that allows the LH surge to occur. After the LH surge, the theca internal cells

start to produce P4 instead of testosterone, which induces the production of collagenase,

gelatinase, and stromelysin, enzymes known as metalloproteinases, which aid in the

remodeling of the extracellular matrix. Consequently the outer connective tissue of the

ovary (tunica albuginea) is digested while granulosa cells increase secretion of follicular

fluid, causing the apex of the follicle (stigma) to push outward and weaken (Senger,

1997). Prostaglandins produced by the ovary cause contraction of smooth muscle of the

ovary, increasing local pressure and forcing the stigma to protrude even more (Senger,

1997). During ovulation, rupture of follicular blood vessels leads to the formation of the

corpus hemorrhagicum, or "bloody body." Once ovulation occurs, prostaglandin E2

(PGE2) helps remodel the follicle into the CL, a process that starts with the dramatic

infolding of the follicular wall that facilitates migration of fibroblasts, endothelial cells,

and theca internal cells into the central regions of the developing CL (O'Shea et al., 1980).









The breakdown of the basement membrane that separates the avascular granulosa layer

from the theca internal layer further facilitates tissue remodeling and cellular migration

during the luteinization process (O'Shea et al., 1980).

Development and Function of the Corpus Luteum in Ruminants

The CL is a transient endocrine gland composed of endothelial cells, steroidogenic

large and small luteal cells, fibroblasts, smooth muscle cells and immune cells (O'Shea et

al., 1989). The major hormones produced by this reproductive gland are P4, and to a

lesser extent OT, which is stored in secretary granules of large steroidogenic cells

(McCracken et al., 1999). Progesterone is necessary for the initiation and maintenance of

pregnancy because it induces a quiescent state of the myometrium (Csapo and Pulkkinen,

1978) and suppresses the maternal immune response to fetal antigens (Szekeres-Bartho,

1992). During early pregnancy, the CL is the primary source of this hormone. Oxytocin,

on the other hand, contributes to the luteolytic process if pregnancy is not established. In

domestic animals such as ruminants, the major luteotropin (substance that promotes the

growth of the CL and P4 production) is LH (McCracken et al., 1999), although during

early CL development, growth hormone (GH) seems to be the major player affecting P4

secretion from the CL in vivo (Miyamoto et al., 1998) and in vitro (Kobayashi et al.,

2001). The CL grows very rapidly, and a substantial increase in mass is apparent as early

as two days after ovulation (Fields et al., 1996). This increase in mass is due to

hypertrophy of granulosa and theca cells and repeated mitosis of steroidogenic cells,

endothelial cells, and fibroblasts (Acosta et al., 2004; McCracken et al., 1999). Even

though the CL has been the subject of intense research for decades, many of the luteal

regulatory mechanisms involved are not completely understood. The gonadotropins and

growth hormones are the primary regulators of CL function, although other factors of









intra- and extraovarian origin have the potential to modulate the local response to these

hormones or to have direct specific functions (Schams and Berisha, 2004). This literature

review will focus on angiogenesis promoters necessary for CL formation, and endocrine,

autocrine, and paracrine regulators of CL function.

Promoters of Angiogenesis

As final follicular growth, ovulation and CL development occur, ovarian tissue is

remodeled, and this tissue remodeling is associated with the occurrence of hemodynamic

changes. The developing CL is characterized by highly active vascularization (Acosta et

al., 2004), and the majority of the steroidogenic cells of the mature CL are in contact with

one or more capillaries (Reynolds et al., 1992). Angiogenesis (the development of new

blood vessels from pre-existing blood vessels) in the developing CL needs to be tightly

controlled since this process maintains the delicate balance between promoters and

inhibitors of angiogenesis (Schams and Berisha, 2004). Factors that regulate luteal

angiogenesis include vascular endothelial growth factor (VEGF), acidic and basic

fibroblast growth factor (FGF-1 and FGF-2), IGF-I, and IGF-II.

All components of the VEGF system are found in the bovine CL (Garrido et al.,

1993; Berisha et al., 2000). The luteal tissue seems to express predominantly the smallest

and secretary isoforms of the VEGF system (Schams and Berisha, 2004). The highest

VEGF and VEGF receptor (VEGFR) mRNA and protein expression were detected at the

time of angiogenesis (early luteal development) (Berisha, 2000). Schams and Berisha

(2004) stated that these findings suggest that VEGF may act as a chemoattractant for

sprouting endothelial cells (EC) since angiogenesis involves the migration and

proliferation of EC from the pre-existing vessels. Luteinizing hormone, IGF-I, and tumor









necrosis factor a (TNF-c) are some of the hormones that stimulate VEGF mRNA and

protein expression in bovine granulosa cells (Schams and Berisha, 2004).

Fibroblast growth factor-1 (acidic) mRNA expression in the bovine CL increases

significantly during the mid-luteal stage (Schams and Berisha, 2002), while FGF-2

(basic) and FGF receptor (FGFR) expression is highest during the early luteal stage

(Schams and Berisha, 2004). Fibroblast growth factor-2 concentrations in the bovine CL

decrease during the mid-luteal stage and increase during the late-luteal stage and these

changes are accompanied by changes in the protein localization from the cytoplasm of

capillary EC and smooth muscle cells of arteries to the cytoplasm of luteal cells (Schams

and Berisha, 2004). The predominant localization of FGF-2 in EC during the early stages

of CL development suggests that this is a dominating factor for endothelial growth

(Gospodarowicz et al., 1985).

Many researchers have demonstrated that the bovine CL expresses IGF-I mRNA

(Einspanier et al., 1990; Kirby et al., 1996; Perks et al., 1999; Schams et al., 1999;

Schams and Berisha, 2002). The highest mRNA expression for both IGF-I and IGF-II

was observed during the early luteal phase, followed by a decrease to a lower cyclic

plateau (Schams and Berisha, 2004). Insulin-like growth factor binding protein-3

expression is positively correlated with IGF-I, -II, and IGF receptor-1 (IGFR-1)

expression, and it has even been postulated that IGFBP-3 serves as a carrier and storage

reservoir for IGFs within the intravascular compartment (Schams and Berisha, 2004). As

was mentioned before, IGFs have been shown to stimulate VEGF mRNA, suggesting that

the IGF system may have an indirect effect on angiogenesis. The fact that IGF-II is

distinctly localized in pericytes (undifferentiated mesenchymal-like cells that may









become fibroblasts, macrophages, or smooth-muscle cells) suggests a direct role of the

IGF system in angiogenesis and capillary stabilization (Amselgruber et al., 1994), since

pericytes play an important role in the modulation of endothelial migration and

proliferation (Orlidge and D'Amore 1987; Antonelli-Orlidge et al., 1989).

Endocrine Regulators of CL Function

An endocrine regulator is one that is produced by a gland and acts on a distant

organ, tissue, or gland by traveling through the bloodstream. The classical and primary

endocrine hormones that support the development and function of the CL are LH and

GH. The CL expresses receptors for both of these hormones.

An increase in LH receptor mRNA during the mid-luteal phase is followed by an

increase in LH receptors (Kobayashi et al., 2001). Most of the LH receptors are located

on small luteal cells (Schams and Berisha, 2004), and LH stimulates these cells to

produce P4 (Niswender and Nett, 1988). In fact, a classic study by Hoffman et al. (1974)

showed that treatment of heifers with an LH antiserum brought P4 levels back to basal

levels within a short period of time.

Receptors for GH are found mainly in large luteal cells (Lucy et al., 1993; Koelle et

al., 1998). These cells are responsible for 80% of P4 produced by the CL (Niswender et

al., 1985). Growth hormone stimulates the CL to produce and secrete OT and P4

(Lieberman and Schams, 1994), and it has been shown that the latter hormone and the

luteolysin, PGF2,, are more strongly stimulated by GH than by LH (Kobayashi, et al.,

2001).

Autocrine and Paracrine Regulators

Autocrine regulators are those that exert their effect on the same gland that produce

them, while paracrine regulators refer to those that exert their effect on cells neighboring









the gland that produce them. Autocrine and paracrine regulators of the CL include

growth factors, peptides, steroids and prostaglandins.

Angiogenic growth factors can also stimulate CL function by stimulating

secretion of P4 and OT. Insulin-like growth factor-I and IGF-II are potent stimulators of

P4 and OT (Einspanier et al., 1990; Sauerwin et al., 1992). Fibroblast growth factor-2

mRNA and protein expression have been demonstrated in luteal cells (Schams et al.,

1994), and in vitro studies have shown that it stimulates P4 and OT in a dose-dependent

manner (Miyamoto et al., 1992; Liebermann et al., 1996).

Ovarian peptides regulating CL function are OT, angiotensin II (Ang II), and

endothelin-1 (ET-1). Oxytocin is localized in large and small luteal cells (Kruip et al.,

1985) and it has been shown to have a modulating action on the luteotrophic effects of

LH on P4 secretion by the CL (Schams et al., 1995). Studies suggest that during the

development of the CL, OT may play an important role as a lutetropic factor (Schams,

1996).

Ang II and other growth factors that induce angiogenesis and support P4 production

by large luteal cells seem to modulate the sprout of new blood capillaries needed to

support luteal cell development (Kobayashi et al., 2001). There are conflicting results

regarding the effects of Ang II on P4 production by luteal cells. Studies by Kobayashi et

al. (2001) suggested that this peptide together with PGF2a greatly stimulated P4 secretion

by the early CL, while studies with bovine CLs (Girsh et al., 1996; Miyamoto et al.,

1997; Hayashi and Miyamoto, 1999) suggested that it inhibited P4 production. It is

conceivable that the interaction of Ang II and PGF2a is necessary to induce stimulation of

P4 production and secretion.









There is evidence of ET-1 expression in the bovine CL during the estrous cycle and

pregnancy (Berisha et al., 2002). The studies that noted an inhibition of P4 secretion by

Ang II in the bovine CL suggest that ET-1 may also have an inhibitory effect on P4

secretion (Girsh et al., 1996; Miyamoto et al., 1997; Hayashi and Miyamoto, 1999).

Both P4 and norepinephrine (NE) have effects on the synthesis of hormones that

regulate the CL. Progesterone inhibits PGF2a secretion by the CL during the mid-luteal

phase (Pate, 1988; Skarzynski and Okuda, 1999), and it also plays a luteotrophic role by

stimulating the synthesis of LH receptors (Jones et al., 1992). Norepinephrine is known

to stimulate the synthesis and release of PGF2, and PGE2 in bovine luteal cells

(Skarzynski et al., 2001).

Luteal PGF2a has distinct effects on CL function depending on the stage of the

luteal cycle. At the early and mid-luteal phases it has a luteotrophic effect, which ceases

at the late luteal stage despite its high local production (Miyamoto et al., 1993). Ovarian

estradiol, OT and P4 appear to be the major regulators of the synthesis and secretion of

luteal PGF2a during the estrous cycle (Grazul et al., 1988; Okuda et al., 2001).

Luteolysis in Domestic Ruminants

Luteolysis is the process whereby the CL regresses and loses its capability to

produce P4. Luteolysis in domestic ruminants is caused by an episodic release of PGF2a

from the uterus that reaches the CL by a countercurrent system between the uterine vein

and the ovarian artery (Schams and Berisha, 2004). Through this system, PGF2a from

one of the uterine horns reaches its ipsilateral ovary and causes the structural and

functional demise of the CL (Ginther, 1974). This countercurrent system is especially

necessary in ovids because without it PGF2a would have to travel through the pulmonary









circulation where most of it (>99%) would be metabolized after one passage through the

lungs (Davis et al., 1980). Because only about 65% of PGF2a is enzymatically

inactivated after one passage through the lungs in bovids (Davis et al., 1985), it is thought

that this hormone may also work through the systemic circulation in cows. Luteolysis is

generally divided into functional and structural luteolysis. During functional luteolysis,

the CL loses its ability to synthesize and secrete P4 (McGuire et al., 1994), whereas

during structural luteolysis the cellular components of the CL are lost (Pate, 1994).

Historically, it has been reported that luteolysis is initiated when E2 released from

the preovulatory follicle triggers the release of OT from the pituitary gland. This causes

the release of small quantities of PGF2a from the uterus (Fairclough et al., 1980), which

has a positive feedback effect that leads to the release of additional luteal oxytocin and

PGF2a from the CL and uterus (Silvia et al., 1991). However, evidence is accumulating

that suggests that OT is not essential for the initiation of PGF2a output during luteolysis

(Kotwica et al., 1997; Kotwica et al., 1998). In blood (Parkinson et al, 1992) and in

intact (Parkinson et al, 1992) and microdialyzed bovine CL (Douglas et al., 2000), OT

concentrations are extremely low at the time of luteolysis and a study performed by

Kotwica et al. (1997) demonstrated that blocking uterine oxytocin receptors had no effect

on luteolysis or the duration of the estrous cycle in heifers. Estradiol stimulates the

activity of enzymes that control prostaglandin (PG) synthesis in the uterine endometrium

(Ham et al., 1975), and priming the uterus with P4 enhances E2-stimulated production of

PGF2a in ruminants (Barcikowski et al., 1974; Lewis and Warren, 1977). Luteolysis is a

complex process that involves many morphological, structural, and molecular changes

that allow the degradation of the CL so that a new estrous cycle can take place. Some of









the aspects of luteolysis that will be discussed in this section are hormonal regulation of

uterine PGF2a synthesis, changes in intracellular signaling, changes in blood flow,

changes in morphology, effect of PGF2a on P4 synthesis, immune-mediated events, tissue

metalloproteinases in the CL, and apoptosis of luteal cells.

Hormonal Regulation of Uterine PGF2a Synthesis

The endometrium is the site ofluteolytic PGF2a synthesis, specifically the inter-

caruncural region of the surface epithelium (Kim and Fortier, 1995; Asselin et al., 1998).

Synthesis of this luteolysin is stimulated by E2, the hormone that stimulates the activity of

enzymes involved in PG synthesis and also by P4, which has been found to have a

priming effect on the uterus when administered before E2. It is thought that the priming

effect of P4 is due to the accumulation of lipids in the endometrium (Boshier et al., 1981).

Lipid accumulation is important because PGF2a is derived from arachidonic acid (AA),

an omega-6 (n-6) fatty acid found in the phospholipid membrane of the cell. Release of

this fatty acid by phospholipase A2 (PLA2) is the rate-limiting step in PGF2a synthesis

(Kunze and Vogt, 1971). The released AA is converted to PGH2 by the enzyme

prostaglandin endoperoxide synthase-2 (PGHS-2; DeWitt and Smith, 1995), also known

as COX-2, an enzyme that contains cyclooxygenase and endoperoxidase activities

(Wlodawer et al., 1976). Prostaglandin H2 (PGH2) is an endoperoxide (unstable,

biologically active molecule) that is further acted upon by the enzyme PGF synthase

(PGFS) to yield PGF2a (Watanabe et al., 1985). An alternate pathway for the formation

of PGF2a is through PGE2-9-keto reductase (9K-PGR), an NADPH-dependent enzyme

that catalyzes the conversion of PGE2 to PGF2a (Asselin and Fortier, 2000). Progesterone

is also thought to increase the concentration and activity of PGFS, further enhancing









PGF2a synthesis (Eggleston et al., 1990). The positive feedback effect of the low levels

of uterine PGF2a on luteal PGF2a can be attributed to the increase in availability of AA

and PGHS-2 activity in the CL in response to PGF2a (Niswender et al., 2000). The major

point of control of E2 and P4 in PGF2a synthesis in the endometrium is through the

regulation of endometrial OT receptor concentrations, and it has been postulated that in

the intact cycling sheep, E2 enhances the formation of OT receptors in the endometrium,

but that during the luteal phase, P4 blocks the formation of endometrial OT receptors

(McCracken et al., 1995). It also has been postulated that P4 exerts a nongenomic

regulation of OT receptors by binding them with high affinity, therefore preventing the

binding of OT to its own receptor (Grazzini et al., 1998). At the end of the luteal phase,

prolonged exposure of the uterus to P4 produced by the CL leads to a downregulation of

the P4 receptor (Vu Hai et al., 1977), and the action of E2 is no longer suppressed. This

leads to the E2-induced formation of endometrial OT receptors, and a large amplification

of endometrial PGF2a secretion induced by OT at the end of the luteal phase which leads

to the regression of the CL (Niswender et al., 2000).

Intracellular Signaling

High-affinity receptors for PGF2a are localized to the large luteal cells of the CL

(Fitz et al., 1982). Binding of PGF2a to its receptor leads to the activation of membrane

bound phospholipase C (PLC) (Berridge et al., 1984), which catalyzes the hydrolysis of

phosphatidylinositol-4,5-bisphosphate to inositol-1,4,5-trisphosphate (IP3) (Davis et al.,

1985) and 1,2-diacylglycerol (DAG) (Berridge et al., 1984). As the amount of IP3

increases in the cytoplasm, it causes the release of free Ca2+ from the smooth

endoplasmic reticulum of the cell into the cytoplasm (Berridge et al., 1984). Free









intracellular Ca2+ activates phospholipase A2 (PLA2) (Flower and Blackwell, 1976),

which cleaves AA from membrane phospholipids. Also, free Ca2+ and DAG in the

plasma membrane stimulate the Ca2+-dependent protein kinase C (PKC) (Nishizuka,

1986), which is believed to mediate many of the antisteroidogenic actions of PGF2a

through posttranslational modification of cellular proteins such as those involved in

steroidogenesis (McGuire et al., 1994) and cholesterol biosynthesis (Behrman et al.,

1971). It is also possible that PKC mediates apoptosis of large luteal cells, because

activation of this enzyme induces the expression and activation of proteins involved in

apoptosis in other cell types (Schwartzman and Cidlowski, 1993).

Blood Flow and Vascular Changes

A steep decrease in blood flow to the CL has been postulated as one of the main

luteolytic actions of PGF2a. Studies with ewes demonstrated that administration of

PGF2a caused a reduction in blood flow to the CL concurrent with a reduction in P4

secretion (Niswender et al., 1973; Nett et al., 1976). This decline in blood flow caused

by PGF2a can lead to luteolysis by means of a reduction in luteotropic support, nutrients,

and substrates for steroidogenesis reaching the CL (Phariss et al., 1970). Also,

vasoactive peptides like ET-1 have been implicated as having a direct luteolytic role.

Endothelin-1 is a strong vasoconstrictor (Huggins et al., 1993) that is thought to mediate

the inhibitory effect of PGF2a on luteal cells. The PGF2a stimulates endothelial cells of

the CL to produce ET-1 in vivo (Ohtani et al., 1998) and ET-1 inhibits the steroidogenic

activities of luteal cells in vitro (Girsh et al., 1996). Furthermore, a study by Meidan et

al. (1999) showed that ET-1 inhibited P4 production by luteal cells in a dose-dependent

manner. Being a strong vasoconsctrictor, ET-1 may cause arteriole constriction (Ohtani









et al., 1998) by reducing blood flow during early luteolysis, and the resulting hypoxia

may cause further release of ET-1 (Rakugi et al., 1990).

Morphological Changes

Morphological changes associated with PGF2a are not evident until 24-36 h after

exposure of the luteal cells (large and small) to the luteolysin (Sawyer et al., 1990),

although by this time the steroidogenic capabilities of these cells are markedly reduced.

Large luteal cell numbers decrease before small steroidogenic luteal cells, and a decrease

in large luteal cell size is also evident at this time (Braden et al., 1988).

PGF2a Inhibition of P4 Synthesis

The PGF2a decreases luteal synthesis of progesterone in vivo in many species,

including bovids and ovids, where it has been shown to decrease P4 secretion from

purified preparations of large luteal cells (Wiltbank et al., 1991). The PGF2a negatively

regulates P4 synthesis by decreasing cholesterol transport across the mitochondrial

membranes. This is critical because P4 is derived from cholesterol and steroidogenesis

occurs in the mitochondria. The disruption in the transport of cholesterol can occur at the

two different following points: transport from the cytosol to the outer mitochondrial

membrane and from the outer to the inner mitochondrial membrane where the enzymes

required for side-chain cleavage reside.

Transport of cholesterol in the cytoplasm is tightly coupled to interactions between

sterol binding proteins and the cytoskeleton. In a variety of species, P4 secretion from

luteal cells is decreased by disruption of the cytoskeleton (Silavin et al., 1980), and

studies in ewes have shown that treatment with PGF2a dramatically reduces the number

of luteal cells staining for tubulin, a critical component of microtubular fibers (Murdoch,









1996). This decrease in tubulin was followed by a decrease in P4, indicating that

disruption of the cytoskeleton precedes the decreased synthesis of P4. Still, it has not

been elucidated if disruption of the microtubule network prevents transport of cholesterol

to mitochondria or disturbs other aspects of luteal steroidogenesis. In vitro studies have

also shown that PGF2a may inhibit transport of cholesterol across the mitochondrial

membranes (Wiltbank et al., 1993). Treatment of cows and ewes with PGF2a decreased

mRNA encoding for steroidogenic acute regulatory protein (StAR), which is followed by

a decline in StAR protein (Pescador et al., 1996). This reduction in StAR may lead to a

reduction in cholesterol transport across the mitochondrial membranes.

Immune-Mediated Events

The immune system plays a pivotal role in both functional (decline in P4 secretary

capacity of the CL) and structural luteolysis involutionn or regression of the CL). During

luteolysis, leukocytes, T-lymphocytes and macrophages increase significantly in the CL

(Penny et al., 1999). Macrophages perform several roles in the regressing CL, but the

major one is phagocytosis of degenerative luteal cells (Pepperell et al., 1992) and

degradation of the extracellular matrix (Parker, 1991). They also assist in cytokine-

mediated inhibition of steroidogenesis and stimulation of CL production of PGF2,. This

is mediated by the macrophage-produced cytokines interleukin-1 (IL-1), which has been

shown to be a potent stimulator of PGF2a in cultured bovine cells (Nothnick and Pate,

1990), and tumor necrosis factor-a (TNF-c), which stimulates PGF2a production and

inhibits LH-stimulated P4 production in vitro (Fairchild-Benyo and Pate, 1992). It has

been postulated that monocyte chemoattractant protein-1 (MCP-1), a potent chemotactic









agent for macrophages, may play an important role in structural luteolysis by increasing

the migration of macrophages into the CL (Tsai et al., 1997).

Tissue Metalloproteinases

Tissue metalloproteinases have been proposed as mediators of structural

luteolysis. As the CL becomes the corpus albicans (CA), substantial tissue remodeling

occurs, and tissue metalloproteinases are thought to be involved in remodeling of the

extracellular matrix (Birkedal-Hansen, 1995). The CL of sheep and cows expresses

tissue inhibitor of metalloproteinases-1 (TIMP-1), which serves in maintaining the

structural integrity of this gland (Smith et al., 1995; Smith et al., 1996). As the CL

regresses, TIMP-1 levels decrease, allowing for an increased activity of the tissue

metalloproteinases (McCracken et al., 1999).

Apoptosis

It is generally accepted that structural luteolysis is an example of apoptosis. This

programmed cell death has been reported to occur during CL regression in ruminants

(Juengel, 1993) and to be promoted by PGF2a (Sawyer et al., 1990). Morphological

evidence that a cell is undergoing apoptosis is the appearance of nuclear fragment

containing degenerate chromatin (Sawyer et al., 1990), cell shrinkage, and appearance of

membrane-bound cytoplasmic fractions (Kerr et al., 1972). These cell fragments are

phagocytized by cells of the immune system, like macrophages, which increase the

apoptotic process in populations of luteal cells by phagocytosing membrane-enclosed

fragments of these cells (Gemmell et al., 1976).

Pregnancy Establishment in Ruminants

Successful pregnancy in viviparous animals depends on the development of an

allogenic fetus within its mother's uterus. This presents the challenge of preventing the









maternal immune system from recognizing the fetal antigens as foreign, since it presents

paternal genes, and mounting a response against it. Therefore, in most mammals, the

concepts has found ways to escape the deleterious effects of maternal rejection.

Another barrier the embryo must prevent is the pulsatile release of PGF2a from the uterus,

which would result in regression of the CL and a down regulation in P4 production. In

ruminants, as in most other species, a signal is sent by the developing embryo with the

purpose of allowing the mother to recognize the embryo's presence. This prevents

luteolysis from occurring and allows P4 secretion to continue, since this is the hormone

that prepares and maintains the uterus as a suitable environment for the development of

the concepts. Luteal P4 is necessary to allow maintenance of pregnancy until the

placenta can produce adequate amounts of this steroid to complete the gestation interval

(Niswender et al., 2000). Ruminants have developed a way of rescuing the CL during

early pregnancy that seems to be unique to these species, in which implantation is both

late in onset and displays limited invasiveness (Wooding, 1992). The anti-luteolytic

signal is a protein, interferon-' (IFN-'), produced by the mononuclear cells of the

trophectoderm, and it is released between days 14 and 17 of pregnancy (LaFrance and

Goff, 1985; Okuda et al., 2002). This means that the antiluteolytic signal is released

while the blastocyst is still elongating and before the trophoblast attaches to the uterine

wall (Roberts et al., 1996). Interferon-T inhibits luteolysis by down-regulating OT

receptors in the endometrium, therefore preventing OT-stimulated pulsatile PGF2a release

(Okuda et al., 2002), and also by different OT receptor-independent mechanism

(LaFrance and Goff, 1990; Binelli et al., 2000). This is important to note since some

studies suggest that OT is not mandatory, but supplementary, for the luteolytic cascade









(Kotwica et al., 1998; Kotwica et al., 1997). The mechanisms by which IFN-' inhibits

luteolysis that are independent of OT receptor downregulation, are the attenuation of

PGHS-2 (also known as COX-2) gene expression (Binelli et al., 2000; Pru et al., 2001)

and the shifting of PG production in the endometrium from PGF2a to PGE2, which has a

luteotropic or luteoprotective effect on the CL (Xiao et al., 1998).

Placentation in Ruminants

Placentation in ruminants is of the synepitheliochorial type, which shows limited

invasiveness, as opposed to hemochorial placentation, which is seen in humans and mice,

or endotheliochorial placentation seen in dogs and cats (Roberts et al., 1996). In

synepitheliochorial placentation, binucleate cells migrate from the chorion and fuse with

epithelial cells of the uterus (Roberts et al., 1996). Cell to cell fusions lead to the

formation of an extensive syncytium between the maternal and fetal tissues in sheep,

while cell fusion is less extreme in cows and the barrier layer becomes partly populated

with trinucleate cells (Roberts et al., 1996). In cattle, the areas of interface known as

placentomes provide tight attachments between the maternal and fetal membranes, which

allow for nutrient exchange. Placentomes are formed by the branching fetal cotyledonary

villi, which grow down into the maternal caruncular crypts in a finger-in-glove

arrangement (Schlafer et al., 2000). One area of intense research has been how the fetus

escapes the maternal immune response in these areas of intimate attachment.

Pregnancy and the Immune Response

There is no doubt that one of the most interesting areas of research in pregnancy is

the study of how the fetus escapes the maternal immune system. The major challenge

that the fetus presents to the mother's uterus is that it expresses paternally derived genes,









which are seen as foreign to the maternal immune system. There have been several

theories regarding the maternal-conceptus immunologic relationship (Hansen, 2000), but

recent research has favored some of these more strongly than others. In cows and ewes,

there is a decrease in the number of lymphocytes, which serve in antigen presentation,

phagocytosis, antibody and cytokine secretion, among other things. This decrease is

localized primarily to the placentomes of cows and the caruncular endometrium of sheep,

the areas where the fetal and maternal tissues are in closest attachment, suggesting that

the maternal immune tolerance is both local and specific to the areas adjacent to fetal

tissues (Leung et al., 2000). It has been suggested that this decrease is mediated by IFN-

..which has been shown to regulate lymphocyte proliferation (Newton et al., 1989;

Skopets et al., 1992), and to have antiviral activity in vitro (Pontzer et al., 1988). A study

by Leung et al. (2000) suggested that in early pregnancy, the cytokine interleukin-1 (IL-

1) could act in concert with IFN-' to stimulate stromal PGE2 synthesis, which in turn may

inhibit uterine expression of the pro-inflammatory cytokine interleukin-2 (IL-2). This is

biologically relevant because studies have shown an increase in concentrations of IL-2 in

placental tissues of mice (Tangri and Raghupathy, 1993) and women (Hill et al., 1995;

Marzi et al., 1996) that have undergone a spontaneous abortion. It has also been

documented that early in pregnancy, a complete shutdown of major histocompatibility

complex class I (MHC class I) expression by trophoblast cells appears to be critical for

normal placental development and fetal survival (Davies et al., 2004). The MHC class I

protein is expressed in most somatic cells and is used to present peptides derived from

intracellular pathogens, or from the animal's own proteins, to cytotoxic T lymphocytes,

which lyse the target cell. The MHC class II molecules tend to be expressed on antigen-









presenting cells such as macrophages, dendritic cells and B lymphocytes, which use them

to present antigens from extracellular pathogens to helper T lymphocytes, which regulate

the activity of other lymphocytes through cytokine secretion. Trophoblasts of most

species do not present either MHC class I or MHC class II antigens (Hunt et al., 1987;

Kydd et al., 1991; Davies et al., 2004), suggesting that this lack of expression protects the

placenta from attack by the maternal immune system (Davies et al., 2004). Bovids are an

exception in that they do present MHC class I protein, but only in the interplacentomal

region, which may be the mechanism used by the fetus to prevent rejection by the

maternal immune system. Studies by Davies et al. (2000) showed that in bovids, as

pregnancy progresses, there is a significant increase in expression of class I antigens in

interplacentomal trophoblast cells, while endometrial epithelial cells of the maternal

crypts lack detectable class I expression throughout pregnancy.

PGHS-2 Gene Expression

A study by Binelli et al. (2000) with cultured bovine endometrial (BEND) cells

demonstrated that bovine IFN-' (bIFN-') suppresses PGF2a production, steady-state

levels of PGHS-2 mRNA, and expression of PGHS-2 and PLA2 proteins, after these cells

had been stimulated to produce PGF2a with phorbol-12,13-dibutyrate (PDBu), a protein

kinase C (PKC) stimulator. A decrease in PGHS-2 activity would directly decrease

PGF2a production since, as it was mentioned before, this is the enzyme that converts AA

into PGH2, the precursor of PGF2a. The authors hypothesized that since PKC may

stimulate PGF2a production via stimulation of synthesis and/or activity of both PLA2

(Mayer and Marshall, 1993; Karimi and Lennartz, 1995) and PGHS-2 (DeWitt, 1991;

Vezza et al., 1996), the effects of bIFN-' could be at the level of PKC by inhibiting its









ability to stimulate PGF2a synthesis through modulation of these two enzymes. Because

there was still some PGF2a production in the presence of bIFN-', it was suggested that

bIFN-' likely does not affect PDBU-induced PGF2a synthesis at sites upstream of PKC.

Later experiments by the same group (Binelli et al., 2001) indicated that IFN-' activates

the Janus kinase (JAK)-signal transducer and activator of transcription (STAT) pathway

regulating gene expression of PGHS-2 in a manner that decreases secretion of PGF2a.

Even though it is highly possible that activation of the JAK-STAT pathway is involved in

the regulation of the anti-luteolytic effects of IFN-', there is a whole host of signaling

pathways activated by IFNs that cannot be excluded from consideration (Binelli et al.,

2001).

Shift in PG Production from PGF2, to PGE2

Interferon-' may also prevent luteolysis by shifting primary prostaglandin

production in the endometrium from luteolytic PGF2a to luteotropic PGE2. Studies have

established that IFN-' may inhibit PGF2a production in the endometrium by down-

regulating PGFS and 9K-PGR (Asselin and Fortier, 2000; LaFrance and Goff, 1990).

This would prevent the conversion of PGH2 or PGE2 into PGF2a, respectively.

Additionally, studies by Xiao et al. (1998) showed that IFN-' had an inhibitory effect on

PGHS-2 mRNA expression in epithelial cells of the endometrium, which is the primary

site of PGF2a production, while it has a stimulatory effect on PGHS-2 mRNA expression

and PG synthesis in stromal cells, which are the primary source of PGE2 (Kim and

Fortier, 1995). More recent studies have demonstrated that IFN-' specifically increases

one of the PGE2 receptor subtypes in the myometrium and endometrial stroma, but not in

luminal or glandular epithelium (Arosh et al., 2004).









Effects of Polyunsaturated Fatty Acids on Reproduction in Cattle

High early embryonic mortality is one of the major problems affecting the dairy

cattle industry. Increased milk production in dairy cattle is associated with a decrease in

reproductive performance. Selective breeding has contributed to the propagation of high

producing cows that exhibit a reduction in occurrence and intensity of estrus as well as

embryo survival, due to alterations in metabolic rate associated with lactation and

management (Thatcher et al., 2003). It is estimated that up to 40% of embryonic losses

occur between day 8 and day 17 of pregnancy (Thatcher et al., 1995), which is coincident

with the period of concepts inhibition of uterine PGF2a secretion. It is conceivable that

some of these early embryo losses may be due to lack of or suboptimal production of

IFN-' by the developing concepts.

Polyunsaturated fatty acids are fatty acids with two or more double bonds. Fatty

acids are unsubstituted monocarboxylic acids that occur mainly as esters in natural fats

and oils, and may also exist in non-esterified forms known as free fatty acids

(Abayasekara et al., 1999). Unsaturated fatty acids can be divided into n-3, n-6, or n-9

fatty acids, depending on the position of the first double bond from the methyl end of the

hydrocarbon chain. Polyunsaturated fatty acids that contain their first double bond at the

third bond counting from the omega end are n-3 PUFAs, while those that have their first

double bond at the sixth position are n-6 fatty acids. Essential fatty acids that cannot be

endogenously produced by ruminants include linoleic acid (LA; 18:2 n-6), which can be

obtained from plant oils, and c-linolenic acid (LNA; 18:3 n-3), which predominates in

forage lipids and in linseed (Cheng et al., 2001). Metabolic conversions can occur only

within the same PUFA family, meaning that LA can only be converted to other n-6 fatty









acids such as y-linolenic acid (GLA; 18:3), dihomo-y-linolenic (DGLA; 20:3), AA (20:4),

and docosapentaenoic acid (DPA; 22:5), while LNA can be converted to other members

of the n-3 family such as eicosapenaenoic acid (EPA; 20:5) and docosahexaenoic acid

(DHA; 22:6) (Bezard et al., 1994). These processes occur by the action of desaturase and

elongase enzymes present in the animal. These longer chain PUFAs can also be obtained

directly from the diet.

Using lactating dairy cows, Oldick et al. (1997) showed that OT-induced

prostaglandin F2a metabolite (PGFM) concentrations in plasma were greatly reduced in

cows infused abomasally with 0.45 kg/d of yellow grease, compared with infusion of

tallow, glucose or water. Additionally, Mattos et al. (2002) demonstrated that

supplementation of postpartum dairy cows with fish meal containing EPA and DHA

considerably attenuated serum PGF2a response to OT injection.

Bovine endometrial (BEND) cells are a good model for the study of endometrial

regulation of PGF2,. These are a line of spontaneously replicating endometrial cells

originating from cows on day 14 of their estrous cycle (Staggs et al., 1998) that can be

stimulated with phorbol-12,13-dibutyrate (PDBu; a phorbol ester) to produce PGF2a. A

study by Mattos et al. (2003) using BEND cells reported that the n-3 fatty acids, LNA,

DHA, and EPA, induced greater PGF2a inhibition than the n-9 fatty acid, oleic acid (OA),

or the n-6 fatty acids, LA and AA.

Although the mechanism of action by which supplemental n-3 fatty acids inhibit

PGF2a production is not fully understood, Mattos et al. (2000) suggested that PUFAs may

inhibit PGF2a synthesis by decreasing the availability of AA, increasing the concentration

of fatty acids that compete with AA for processing by PGHS-2, inhibiting PGHS-2









synthesis and or activity, or affecting gene expression through activation of nuclear

transcription factors. In cultured BEND cells, n-3 PUFAs were shown to suppress PGF2a

production without altering PGHS-2 mRNA synthesis (Mattos et al., 2003). This would

suggest that supplemental fatty acids may affect PGF2a biosynthesis through mechanisms

which do not require PGHS-2 gene regulation. The n-3 polyunsaturated fatty acids may

decrease AA synthesis by inhibiting the desaturation and/or elongation processes

necessary for the conversion of LA to AA (Bezard et al., 1994), or by replacing AA in

tissue phospholipids, which has been demonstrated in animals fed diets rich in n-3 fatty

acids (Trujillo and Broughton, 1995). Another possible mechanism for dietary reduction

in prostaglandin synthesis is the competition of PUFAs such as EPA for PGHS-2 activity,

which would result in production of prostaglandins of the 3 series, which are less

biologically active (Mattos et al., 2000). A study by Leaver et al. (1991) showed that

feeding rats a diet rich in n-3 fatty acids resulted in increased secretion of prostaglandins

of the 3 series from uterine explants cultured in vitro. Dietary fatty acids may repress

genes involved in the synthesis of prostaglandins through activation of nuclear

transcription factors such as peroxisome proliferator-activated receptors (PPARs).

Roles of Peroxisome Proliferator-Activated Receptors in Reproduction

Polyunsaturated fatty acids and their various metabolites can act at the level of the

nucleus, in conjunction with nuclear receptors and transcription factors, to affect the

transcription of a variety of genes. Peroxisome proliferator-activated receptors are an

example of nuclear hormone receptors that mediate the effects of fatty acids and their

derivatives at the level of gene expression. Peroxisome proliferator-activated receptors

belong to the steroid hormone nuclear receptor superfamily of ligand-activated









transcription factors that also includes the retinoic acid receptor (RAR), liver X receptor

(LXR), and the ubiquitous retinoid X receptor (RXR) (Sampath and Ntambi, 2005).

Three isoforms, PPAR a, PPAR 6, and PPAR y, each encoded by a different gene, have

been identified. Peroxisome proliferator-activated receptor a is expressed in numerous

metabolic sites such as liver, kidney, heart, skeletal muscle, and brown fat (Braissant et

al., 1996), while PPAR y is highly expressed in adipocytes, where it regulates adipocyte

differentiation, lipid storage and insulin sensitivity (Chawla et al., 1994; Schiffrin et al.,

2003). Peroxisome proliferator-activated receptor 6 is expressed in a wide range of

tissues and cells, with relative higher levels of expression noted in brain, adipose, and

skin (Braissant et al., 1996), and in the human placenta and large intestine (Mukherjee et

al., 1997; Auboeuf et al., 1997). Each of these receptors binds to the peroxisome

proliferator response element (PPREs) of regulated genes as a heterodimer with a RXR.

The PPRE comprises a direct repeat of 6 nucleotides, separated by one spacer nucleotide

and has the consensus sequence AGGTCA n AGGTCA (van Bilsen et al., 2002). The

PPAR/RXR dimer is known to interact with various coactivators and corepressors which

are thought to modulate transcriptional activity by interacting both with nuclear receptors

and basal transcription factors (van Bilsen et al., 2002). It is generally believed that the

PPARs are constitutively localized in the nucleus, as opposed to other ligand-activated

transcription factors which are translocated to the nucleus after binding to their cognate

ligand (van Bilsen et al., 2002). Studies have shown that PUFAs can activate PPARs at

micromolar concentration ranges (Hihi et al., 2002).

To date, the only PPAR that has been associated with reproductive function is

PPAR 6 (Berger et al., 2002). Some fatty acid agonists for PPAR 6 are DGLA, AA, and









EPA, of which the latter may be the most potent activator of this chemical class (Xu et

al., 1999). Prostacyclin (PGI2), derived from the action of PGHS-2, can also interact with

PPAR6 (Forman et al., 1997). Studies with mice and rats have shown that PPAR 6 plays

a specific role in embryo development, specifically during implantation and

decidualization (Barak et al., 2001; Ding et al., 2003). Furthermore, a study by Lim et al.

(2000) demonstrated that PGI2 is the primary PG that is essential for implantation and

decidualization, and suggested that the effects of PGI2 are mediated by activation of

PPAR 6. This suggests that PGs produced by PGHS-2 may exert their effects directly on

the nucleus via the activation ofPPARs. Moreover, in PGHS-2 deficient mice,

decidualization and implantation failures can be reversed by the administration of a

PPAR 6-selective agonist (Lim et al., 1997).

The bovine endometrium, as well as BEND cells, have been shown to express

PPAR 6 (MacLaren et al., 2003) and a study by Balaguer et al. (2005) showed an inverse

relationship between PPAR 6 and uterine expression of estrogen receptor alpha and

PGHS-2 genes, suggesting that this nuclear receptor may play an important role in the

control of reproductive processes in mammalian species. Activation of PPAR 6 by its

agonists may lead to the transcription of genes that are essential to reproductive events,

therefore affecting reproductive performance.

Roles of Mitogen-activated Protein Kinases in Reproduction

The mitogen-activated protein (MAP) kinase signaling cascade is believed to

function as an important regulator of prostaglandin biosynthesis (Guan et al., 1998), and

components of this pathway have been implicated as mediators of phosphorylation of

intracellular substrates such as protein kinases and transcription factors (Karin, 1994).









Three major classes of MAP kinases have been identified, each having several isoforms.

They include the extracellular signal regulated protein kinase (ERK), c-Jun N-

terminal/stress-activated protein kinase (JNK/SAPK), and p38 MAP kinase (Robinson

and Cobb, 1997). These MAP kinases are activated by distinct upstream MAPK/ERK

kinases (MEKs, MKKs), which recognize and phosphorylate threonine and tyrosine

residues within a tripeptide motif (Thr-X-Tyr) required for MAP kinase activation

(Kriakis and Avruch, 1996).

There is evidence that the MAP kinase pathway is involved in the activation of the

PG biosynthetic pathway. For example, Lin et al. (1998) demonstrated that activation of

cytosolic PLA2 was mediated by MAP kinase, while Guan et al. (1997) showed that

inhibition of p38 MAP kinase resulted in the decrease of IL-1 mediated PGHS-2

expression and PGE2 production. Studies with the ovine endometrium have shown that

this tissue expresses all classes of MAP kinase, and that OT induces phosphorylation of

ERK1/2, suggesting this MAP kinase mediates OT-induced PGF2a synthesis in ovine

endometrium (Burns et al., 2001). A recent study performed at the University of Florida

demonstrated that activation of p38 MAP kinase by PDBu is required for continued

presence of PGHS-2 mRNA and secretion of PGF2a in BEND cells (Guzeloglu et al.,

2004). In fact, several studies show evidence for a regulatory effect of p38 MAP kinase

on PGHS-2 mRNA stability. Inhibition of this MAP kinase in several cell types resulted

in reduced expression and/or increased turnover of PGHS-2 mRNA (Dean et al., 1999;

Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al., 2003). Scherle et al. (2000)

studied the involvement of various signaling pathways leading to PGHS-2 and PPAR 6

expression in the mouse uterus and demonstrated that inhibition of p38 MAP kinase






40


blocked PPAR 6 expression and decreased PGHS-2 gene expression in luminal epithelial

and stromal cells. Ait-Said et al. (2003) recently showed that EPA suppresses p38 MAP

kinase phosphorylation in stimulated human microvascular endothelial cells, resulting in

a down-regulation of PGHS-2 gene expression. This report suggests a novel regulatory

point for EPA in PGF2a inhibition.














CHAPTER 3
STUDIES ON THE EFFECTS OF EICOSAPENTAENOIC ACID ON PGF2a
PRODUCTION IN CULTURED BOVINE ENDOMETRIAL CELLS

Introduction

Pregnancy establishment in ruminants is dependent on the attenuation of the

pulsatile release of PGF2a from the endometrium (McCracken et al., 1970).

Prostaglandin F2a is the luteolytic agent in many species, and it is produced in the

endometrium through the action of several enzymes. The rate-limiting step in PG

synthesis is the cleavage of sn-2 fatty acyl ester bond of membrane phospholipids by

cytosolic phospholipase A2 (Van den Bosh, 1980; Irvine, 1982). The AA that is released

by phospholipid hydrolysis is acted on by PGHS-2 to form PGH2, which then is

converted to PGF2a. Regression of the CL through the action of PGF2a leads to a

decrease in plasma concentrations of P4, which is needed for maintenance of pregnancy.

Early embryonic mortality is one of the major problems affecting the dairy cattle

industry, and it is estimated that up to 40% of embryonic losses occur between d 8 and d

17 of pregnancy (Thatcher et al., 1994), which is coincident with the period of concepts

inhibition of uterine PGF2a secretion. Many researchers have focused on identifying

ways to increase embryo survival in domestic species (Oldick et al., 1997; Moreira et al.,

2000; Mattos et al., 2002, 2003; Badinga et al., 2002). Several of these studies have

shown that manipulating the fat content of the ruminant's diets can have a beneficial

effect on reproduction (Staples et al., 1998).









Eicosapentaenoic acid and DHA have been shown to inhibit PGF2a synthesis in

various tissue and cell models (Olsen et al., 1992; Baguma-Nibasheka et al., 1999;

Mattos et al., 2002). For example, infusing ewes with 3 ml/kg of body weight per day of

an emulsion of fish oil (FO) containing 30% EPA and 20% DHA blocked a

betamethason-induced increase in plasma concentration of PGF2a and delayed the

occurrence of parturition (Baguma-Nibasheka et al., 1999). In humans, consumption of

large quantities of FO resulted in delayed parturition, conceivably due to reduced

secretion of PGF2a (Olsen et al., 1992). Recently, Mattos et al. (2002) reported that

supplementing the diet of lactating dairy cows with fish meal (rich in n-3 fatty acids)

reduced plasma PGFM concentrations after an OT injection. The same authors showed

that EPA and DHA were potent inhibitors of PGF2a secretion in BEND cells and that LA,

an n-6 fatty acid, did not affect PGF2a response to PDBu in cultured BEND cells (Mattos

et al., 2003). It has been postulated that supplemental fatty acids may inhibit PGF2a

secretion by decreasing the availability of the precursor AA, increasing the

concentrations of fatty acids that compete with AA for processing by PGHS, or by

inhibiting PGHS synthesis and or activity (Mattos et al., 2000). The University of

Florida investigators recently reported that n-3 PUFAs suppressed PGF2a production

without altering PGHS-2 mRNA synthesis in cultured BEND cells (Mattos et al., 2003),

suggesting that supplemental fatty acids may affect PGF2a biosynthesis through

mechanisms which do not require PGHS-2 gene regulation.

The objective of this study was to study the physiological effects of EPA on

PGF2a production in PDBu-stimulated BEND cells. We hypothesized that incubation of

BEND cells with EPA would lead to a reduction in PGF2a production by these cells, and









that increasing the concentration of the n-6 fatty acid, LA, in these cells may abolish the

inhibitory effect of EPA on endometrial PGF2a production.

Materials and Methods

Materials

Polystyrene tissue culture dishes (100 x 20 mm) were purchased from Corning

(Corning Glass Works, Corning, NY). The Ham F-12 medium, antibiotic/antimycotic

(ABAM), phorbol 12,13-dibutyrate (PDBu), horse serum, D-valine, insulin, and fatty

acid-free bovine serum albumin (BSA) were from Sigma Chemical Co. (St. Louis, MO).

The Minimum Essential medium (MEM) and fetal bovine serum (FBS) were from US

Biologicals (Swampscott, MA) and Atlanta Biologicals (Norcross, GA), respectively.

Eicosapentaenoic acid, LA and PGF2a standard were from Cayman Chemicals (Ann

Arbor, MI). Hanks Balanced Salt Solution (HBSS) and TriZol reagent were from

GIBCO BRL (Carlsbad, CA). Isotopically-labeled PGF2a (5, 6, 8, 9, 11, 12, 14, 15 [n-

3H] PGF2a; 208 Ci/mmol) was from Amersham Biosciences (Piscataway, NJ). The anti-

PGF2a antibody was purchased from Oxford Biomedicals (Oxford, MI). BioTrans nylon

membrane and [a-32P]deoxycytidine triphosphate (SA 3000 Ci/nmol) were from MP

Biolomedicals (Atlanta, GA). The PGHS-2 cDNA probe was cloned from an ovarian

follicular cDNA library (Liu et al., 1999).

Cell Culture and Treatment

Immortalized BEND cells (American Type Culture Collection # CRL-2398,

Manassas, VA) were cultured as described by Mattos et al. (2003). Cells were suspended

(0.5 x 106 cells/mL) in growth medium (40% Ham F-12, 40% MEM, 1% ABAM, 200

U/L of insulin, 0.343 g/L of D-valine, 10% heat-inactivated FBS and 10% horse serum)









and incubated at 370C in a 95% air-5% CO2 environment. Each culture was replenished

with fresh medium every 2 days until cells reached confluence.

To examine the effect of EPA on PDBu-stimulated PGF2a secretion, confluent

BEND cells were rinsed twice with HBSS and incubated in serum-free medium with or

without EPA (100 ptM) for 24 h (Figure 3-1). This concentration and incubation time

were selected since previous studies have shown that this particular concentration of EPA

and incubation time have the most inhibitory effect on PGF2a in cultured BEND cells

(Mattos et al., 2003). The fatty acid was completed with BSA at a molar ratio of 2:1

before being added to the cultures. The 24 h incubation period with fatty acid was

chosen based on reported effects of EPA and DHA on phorbol ester-induced PGF2a

secretion by BEND cells after 24 h (Mattos et al., 2003). Medium was then removed and

the cells incubated in serum-free medium with or without PDBu (100 ng/mL) for an

additional 6 h (Figure 3-1). After PDBu challenge, samples (0.5 mL) of cell-conditioned

media were collected and store at -200C until assayed for PGF2a concentration. The

remaining cell monolayers were rinsed in ice-cold HBSS, lysed with TriZol, and stored at

-80C until PGHS-2 and PPAR6 mRNA quantifications.

To determine the effect of LA on EPA-mediated inhibition of PGF2a secretion,

confluent BEND cells were incubated with 6 molar ratios of EPA to LA for 24 h (Figure

3-2). The molar n-6:n-3 ratios were 0 (all EPA), 1, 4, 9, 19, and oo (all LA). The total

concentration of fatty acids in the culture medium was maintained at 100 ptM. After

incubation, the medium was removed and the cells were incubated in fresh serum-free

medium with PDBu (100 ng/mL) for an additional 6 h. Aliquots (0.5 mL) of cell-

conditioned media then were collected and stored at -200C for subsequent PGF2a









radioimmunoassay. The remaining cell monolayers were rinsed in ice-cold HBSS, lysed

with 1 mL of TRIzol Reagent, and stored at -800C until PGHS-2 mRNA quantification.

PGF2a Radioimmunoassay

Prostaglandin F2a concentration in cell-conditioned media was measured in

duplicates as described by Danet-Desnoyers et al. (1994) and modified by Binelli et al.

(2000). Assay sensitivity was 5 ng/mL and intra- and inter-assay coefficients of variation

were 8.2 and 15.8%, respectively. To adjust for between well differences in cell density,

final PGF2a concentrations were expressed as picograms per million cells. Cell numbers

in individual dishes were determined using a hemocytometer (Sigma Chemical Co., St.

Louis, MO).

RNA Isolation and Analysis

Total cellular RNA was isolated from control and treated BEND cell cultures

using TRIZol reagent. Two hundred p.L of chloroform were added to each tube

containing 1 mL of TRIzol. Tubes were shaken vigorously by hand for 15 seconds and

incubated at room temperature for 3 minutes. Samples were centrifuged at no more than

12,000 x g for 15 minutes at 40C. After centrifugation the aqueous phase was transferred

to a fresh tube, and RNA was precipitated using 500 p.L of isopropanol. Samples were

incubated at room temperature for 10 minutes and centrifuged at no more than 12,000 x g

for 10 minutes at 40C. The supernatant was removed and the RNA pellet was washed

once by adding 1 mL of 75% ethanol and centrifuging at no more than 7,500 x g for 3

minutes at 40C. At the end of the procedure the RNA pellet was air dried for 5 to 10

minutes, after which time the RNA was dissolved in RNase-free water. Ten micrograms

of RNA was fractionated in 1.0% agarose-formaldehyde gel and blotted to a BioTrans









nylon membrane by capillary action. The RNA was cross-linked to the membrane by UV

irradiation and baked at 800C for 1 h. The RNA filters were hybridized consecutively

with random primer-labeled PGHS-2 and PPAR6 cDNA probes (Balaguer et al., 2005).

After hybridization, RNA filters were washed for 20 min in 50 mL 2X SSC, 0.1% SDS at

room temperature, followed by two 15-min washes in 0.1X SSC, 0.1% SDS at 420C. The

filters were blotted dry and exposed to X-ray films for 6 to 24 h at -800C. Hybridization

signals for each target gene were quantified by densitometric analysis.

Statistical Analyses

Concentrations of PGF2a in cell culture medium and mRNA responses were

analyzed using the General Linear Model procedure of the SAS software package (SAS

Institute Inc., Cary, NC). For PGF2a concentration, the sources of variation included

experiment, treatment, experiment x treatment interaction, and well (experiment x

treatment). One experiment constituted one run in which all treatments (i.e., control,

PDBu, PDBu+EPA) were tested. The well, nested within experiment and treatment, was

considered a random variable and, therefore, the well variance was used as error term to

test the effects of experiment, treatment and experiment x treatment interaction. The

statistical model used was the following:

yijk = |t + Ei + Tj + ETij + W(ET)ijk + Skl

where t = overall mean; Ei = effect of the ith experiment; Tj = effect of the jth

treatment; ETij = the interaction of the ith experiment and the jth treatment; W(ET)ijk = is

the well nested within the ith experiment and the jth treatment; and ijkl = random error.

For PGHS-2 and PPAR6 mRNA responses, the mathematical models included

experiment, treatment and experiment x treatment interaction. Densitometric values for









each target gene were expressed as ratios over the values for 18S ribosomal RNA. When

treatment effects were detected (P < 0.05), means were separated using orthogonal

contrasts.

Results

Prostaglandin F2a production was negligible in control BEND cells (Figure 3-3).

Treatment with PDBu for 6 h stimulated (P < 0.01) PGF2a secretion 10-fold (Figure 3-3).

Pre-incubation of BEND cells with EPA for 24 h decreased (P < 0.01) PGF2a response to

PDBu by 75% (Figure 3-3). However, concentration of PGF2a remained higher (+ 2.7-

fold) in EPA-treated than control cells (Figure 3-3).

Compared to untreated cells, PDBu increased (P < 0.01) steady-state PGHS-2

mRNA concentrations by 20-fold (Figure 3-4). Eicosapentaenoic acid had no detectable

effect on PGHS-2 mRNA responses to PDBu (Figure 3-4).

PDBu increased (P < 0.05) steady-state PPAR6 mRNA concentrations by 1.3-fold,

as compared to untreated cells (Figure 3-5). Eicosapentaenoic acid had no detectable

effect on PPAR6 mRNA response to PDBu (Figure 3-5).

The inhibitory effect of EPA on PGF2a production decreased (P < 0.01) from 88

to 40% as the n-6/n-3 fatty acid ratio in the culture medium increased from 0 to 19

(Figure 3-6). Addition of LA alone to the culture medium had no detectable effect on

PGF2a response to PDBu (Figure 3-6).

Steady-state PGHS-2 mRNA concentrations increased (P < 0.05) from 18 to 93%,

as the n-6/n-3 fatty acid ratio in the culture medium increased from 0 to 19 (Figure 3-7).

Complete substitution of EPA by LA in the medium further enhanced (P < 0.05) PGHS-2

mRNA response to PDBu (+ 149%; Figure 3-7).









Discussion

Evidence is rapidly accumulating that supplemental fatty acids can have major

effects on eicosanoid synthesis in domestic animals (Baguma-Nibasheka et al., 1999;

Mattos et al., 2003; Cheng et al., 2001; Chartrand et al., 2003). Depending on the amount

and type of particular fatty acids reaching the target tissues, supplemental fatty acids can

either stimulate (Burke et al., 1996; Filley et al., 1999) or inhibit (Baguma-Nibasheka et

al., 1999; Mattos et al., 2003; Cheng et al., 2001) prostanoid synthesis. Results of this

study extend previous observations that n-3 fatty acids are potent inhibitors of PG

secretion in mammalian species (Baguma-Nibasheka et al., 1999; Olsen et al., 1992;

Mattos et al., 2000; Mattos et al., 2002; Mattos et al., 2003; Cheng et al., 2001).

Although the exact mechanism by which supplemental n-3 fatty acids inhibit PGF2a

production is not fully understood, it is conceivable that increased concentration of EPA

in membrane phospholipids as a result of treating BEND cells with EPA could displace

AA, leading to increased synthesis of PGs of the 3 series at the expense of PGs of the 2

series (Mattos et al., 2003). This hypothesis does not rule out the possibility that

supplemental EPA also may inhibit PGHS-2 activity in cultured BEND cells. In fact,

incubation of rat hepatoma cells with AA, EPA, DHA, or heineicosapentaenoic acid

(C21:5 n-3) inhibited the PGHS-2 enzyme activity (Larsen et al., 1997). Eicosapentaenoic

acid inactivated the enzyme almost completely when added 30 sec before addition of AA.

The rate-limiting step in PG synthesis involves the cleavage of sn-2 fatty acyl ester

bond of membrane phospholipids by cytosolic PLA2 (Van den Bosh, 1980; Irvine, 1982).

The AA that is released by phospholipid hydrolysis is acted on by PGHS-2 to form

PGH2, which then is converted to PGF2a. Consistent with a previous observation (Mattos









et al., 2003), the present study provided no evidence for EPA regulation of PGHS-2

mRNA abundance in BEND cells. As discussed above, supplemental EPA may alter

endometrial PGF2a production through competitive displacement of AA from membrane

phospholipids and/or through alteration of the PGHS-2 enzymatic activity.

Polyunsaturated fatty acids elicit several physiological changes through the

alteration of the activity or synthesis of nuclear PPARs (Bocher et al., 2002). In mice,

PPAR6 deficiency leads to placental defects and results in frequent mid gestational

lethalities (Barak et al., 2002), suggesting that this nuclear receptor may play an

important role in the control of reproductive processes in mammalian species. Consistent

with a recent in vivo experiment (Palin et al., 2005), supplemental EPA had no detectable

effects on PPAR6 response to PDBu in cultured BEND cells. Results suggest that

supplemental n-3 fatty acids may alter endometrial PGF2a production through a

mechanism which does not require induction of PPAR6 gene. However, whether and

how these fatty acids may control the activity of this nuclear receptor warrants further

investigation.

Conventional cattle diets contain a mixture of n-6 and n-3 fatty acids. Therefore,

we examined the effect of EPA on endometrial production of PGF2a in the presence of

increasing concentrations of LA. The inhibitory effect of EPA on uterine endometrial

PGF2a production decreased from 88 to 40%, as the n-6/n-3 fatty acid ratio in the culture

medium increased from 0 to 19. These findings are consistent with previous reports

(Trujillo et al., 1995; Achard et al., 1997), and suggest that the net inhibition of uterine

PGF2a synthesis by n-3 fatty acids may depend on the amount of n-6 fatty acids reaching

the target tissue. Increasing concentrations of LA in the cell culture system may increase









the availability of AA in membrane phospholipids and therefore decrease the competition

by n-3 fatty acids for the PGHS-2 enzyme. Whether and how increasing n-6/n-3 fatty

acid ratios alter the PGHS-2 activity is yet to be elucidated.

Summary

Phorbol ester stimulated PGF2a production and up-regulated PGHS-2 gene

expression within 6 h in cultured BEND cells. Pre-incubation of confluent BEND cells

with EPA for 24 h decreased PGF2a response to PDBu, but had no detectable effect on

PGHS-2 mRNA abundance. The inhibitory effect of EPA on PGF2a response to PDBu

was reverted when increasing amounts of LA and decreasing amounts of EPA were

added to the culture medium. These findings indicate that the net inhibition of uterine

endometrial PGF2a synthesis by n-3 fatty acids may depend on the availability of n-6

fatty acids within the target tissue.


















Grown to confluence
and split

U
EPA Incubation
24 h


PUEB'1 hIr'iiocn
Elh


S(-)


Control


IM


PDBu


PDBu + EPA
PDBu + EPA


Figure 3-1. Experimental manipulations to examine the effect of eicospaentaenoic acid
(EPA) on prostaglandin F2a (PGF2a) response to phorbol-12,13-dibutyrate
(PDBu) in bovine endometrial (BEND) cells.


I H I H i M




















BEND cells


Grown to confluence
and split



L .CT,+ t ./ (:; (+ (+ (+



PDBu incubation 7(+)))
6h



PDBu + + + + + + +
LA, [M 0 0 50 80 90 95 100
EPA, iM 0 100 50 20 10 5 0
n-6/n-3 0 0 1 4 9 19 oo

Figure 3-2. Experimental manipulations to study the effect of linoleic acid (LA) on
prostaglandin F2a (PGF2a) response to eicosapentaenoic acid (EPA) in bovine
endometrial (BEND) cells.


























30










00
o 20
-

Cl
U-
(9 10
a-




0


Control PDBu PDBu + EPA

Treatments



Figure 3-3. Effect of eicosapentaenoic acid (EPA) on prostaglandin F2a (PGF2a)
response to phorbol-12,13-dibutyrate (PDBu) in bovine endometrial (BEND)
cells. Data represents least square means SEM of three independent
experiments. When treatment effects were detected (P < 0.05), means were
separated using orthogonal contrasts. Contrast 1: (Control) vs. (PDBu),
(PDBu + EPA), P < 0.0001; Contrast 2: (PDBu) vs. (PDBu + EPA), P <
0.0001.















C PDBu PDBu+ EPA


4.4 kb


18S


Control PDBu PDBu+EPA
Treatments


Figure 3-4. Effect of eicosapentaenoic acid (EPA) on prostaglandin endoperoxide
synthase (PGHS-2) mRNA response to phorbol-12,13-dibutyrate (PDBu) in
bovine endometrial (BEND) cells. Ten micrograms of total cellular RNA
isolated from control and treated BEND cells were subjected to Northern blot
analysis (A), and resulting densitometric values were analyzed by the GLM
procedure of SAS (B). The top panel shows a representative Northern blot,
whereas the bottom panel represents means SEM calculated over two
experiments (n = 4 for each treatment). When treatment effects were detected
(P < 0.05), means were separated using orthogonal contrasts. Contrast 1:
(Control) vs. (PDBu), (PDBu + EPA), P < 0.0001; Contrast 2: (PDBu) vs.
(PDBu + EPA), P = 0.4.


(A)


PGHS-2
mRNA

rRNA


(B)


0.3




L 0.2

0
7a
.N
M 0.1
E
o



0.0


















C PDBu


PDBu + EPA


(B,





z
a1
E
Lr
Q-
EL
CL


0.3



z
, 0.2

0.0
0
c-)


M 0.1
E
o



0.0


Control PDBu
Treatments


PDBu+EPA


Figure 3-5. Effect of eicosapentaenoic acid (EPA) on peroxisome proliferator-activated
receptor 6 (PPAR6) mRNA response to phorbol-12,13-dibutyrate (PDBu) in
bovine endometrial (BEND) cells. Ten micrograms of total cellular RNA
isolated from control and treated BEND cells were subjected to Northern blot
analysis (A), and resulting densitometric values were analyzed by the GLM
procedure of SAS (B). The top panel shows a representative Northern blot,
whereas the bottom panel represents means SEM calculated over two
experiments (n = 4 for each treatment). When treatment effects were detected
(P < 0.05), means were separated using orthogonal contrasts. Contrast 1:
(Control) vs. (PDBu), (PDBu + EPA), P = 0.04; Contrast 2: (PDBu) vs.
(PDBu + EPA), P = 0.3.


(A)


PPAR8
mRNA


rRNA


3.5 kb


18S




























Treatments

PFE + + + + + + + + +

EILA i 11.11 19 9" 1

LA 0 0 50 80 90 95 100


Figure 3-6. Effects of increasing n-6/n-3 fatty acid ratios on prostaglandin F2a (PGF2a)
response to phorbol-12-13-dibutyrate (PDBu). Data represents least square
means + SEM of three independent experiments. When treatment effects
were detected (P < 0.05), means were separated using orthogonal contrasts.
Contrast 1: (PDBu), (oo) vs. (0), (1), (4), (9), (19), P < 0.0001; Contrast 2: (0)
vs. (1), (4), (9), (19), (oo), P < 0.0001; Contrast 3: (1) vs. (4), (9), (19), (oo), P
= 0.0001; Contrast 4: (4) vs. (9), (19), (oo), P = 0.0003; Contrast 5: (9) vs.
(19), (oo), P = 0.3; Contrast 6: (PDBu) vs. (oo), P = 0.6.








(A)
PGHS-2I
mRNA
rRNA


(B)


0.5
0.4
0.3
0.2
0.1


n-6/n-3 fatty acid ratio
UPDBU 1 4 9 19 IQ t_2 c


4.4 kb


'i0 18S


Treatments


Figure 3-7. Effects of increasing n-6/n-3 fatty acid ratios on prostaglandin endoperoxide
synthase (PGHS-2) response to phorbol-12,13-dibutyrate (PDBu). Ten
micrograms of total cellular RNA isolated from control and treated BEND
cells were subjected to Northern blot analysis (A), and resulting densitometric
values were analyzed by the GLM procedure of SAS (B). The top panel shows
a representative Northern blot, whereas the bottom panel represents means +
SEM calculated over two experiments. When treatment effects were detected
(P < 0.05), means were separated using orthogonal contrasts. Contrast 1:
(PDBu), (oo) vs. (0), (1), (4), (9), (19), P = 0.1; Contrast 2: (0) vs. (1), (4), (9),
(19), (oo), P = 0.3; Contrast 3: (1) vs. (4), (9), (19), (oo), P = 0.03; Contrast 4:
(4) vs. (9), (19), (oo), P = 0.2; Contrast 5: (9) vs. (19), (oo), P = 0.3; Contrast 6:
(PDBu) vs. (oo), P = 0.0004.


, I II[














CHAPTER 4
STUDIES ON THE MECHANISMS OF ACTION OF EICOSAPENTAENOIC ACID
ON PGF2a PRODUCTION BY CULTURED BOVINE ENDOMETRIAL CELLS

Introduction

Polyunsaturated fatty acids and their metabolites can act at the level of the

nucleus to affect the transcription of a variety of genes. One potential mechanism by

which PUFAs affect mammalian gene expression is through the interaction with nuclear

receptors commonly referred to as PPARs (Berger et al., 2002). Peroxisome proliferator-

activated receptors are members of the nuclear hormone receptor superfamily of ligand-

dependent transcription factors (Ding et al., 2003). Three PPAR subtypes that have been

recognized include PPARca, PPAR6, and PPARy. Of these three different subtypes,

PPAR6 is the only one that has been associated with reproductive processes (Berger et

al., 2002). Recently Balaguer et al. (2005) showed an inverse relationship between

PPAR6 and uterine expression of estrogen receptor alpha and PGHS-2 genes, suggesting

that this nuclear receptor may play an important role in the control of reproductive

processes in mammalian species.

The MAP kinase signaling cascade is believed to function as an important

regulator of PG biosynthesis (Guan et al., 1998), and components of this pathway have

been implicated as mediators of phosphorylation of intracellular substrates such as

protein kinases and transcription factors (Karin, 1994). In mammalian cells, at least three

different subfamilies of MAP kinases have been identified (ERK, JNK/SAPK, and p38),

each having several isoforms (Guan et al., 1998). Recent investigations have









demonstrated the importance of ERK and p38 MAP kinases as regulators of the PG

biosynthetic pathway. Specifically, ERK has been shown to mediate OT-induced PGF2a

synthesis in ovine endometrium (Burs et al., 2001), while p38 regulates the stability of

PGHS-2 mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et

al., 2003). Whether and how supplemental PUFAs interact with the MAP kinases

signaling pathway has not been fully documented.

The objective of this investigation was to study the mechanisms of action of EPA

on PGF2a production in PDBu-stimulated BEND cells. We hypothesized that if EPA

exerts its PGF2a-inhibitory effect through activation of PPAR6, then activation of this

nuclear receptor would mimic the effect of EPA on PGF2a production. Alternatively, if

EPA acts through the ERK or p38 MAP kinase signal transduction pathways, then

blocking these kinases would block the PGF2a-inhibitory effect of EPA.

Materials and Methods

Cell Culture and Treatment

Immortalized BEND cells (American Type Culture Collection # CRL-2398,

Manassas, VA) were cultured as described by Mattos et al. (2003) with the following

modifications. Cells were suspended (0.5 x 106 cells/mL) in growth medium (40% Ham

F-12, 40% MEM, 1% ABAM, 200 U/L of insulin, 0.343 g/L of D-valine, 10% heat-

inactivated FBS and 10% horse serum) and incubated at 370C in a 95% air-5% CO2

environment. Each culture was replenished with fresh medium every 2 days until cells

reached confluence.

Experiment 1. To determine if EPA-mediated inhibition of PGF2a secretion

involves PPAR6 activation, confluent BEND cells were rinsed with HBSS and then









assigned randomly to one of the following treatments: 1) control (medium alone), 2)

PDBu (100 ng/mL), 3) PPAR6 agonist (L-165,041; 1 [tM), or 4) PDBu + L-165,041

(Figure 4-1).

Experiment 2. In another set of experiments, confluent BEND cells were treated

with 1) medium alone (control), 2) PDBu, 3) PDBu + EPA, or 4) PDBu + EPA + PPAR6

antagonist (sulindac sulfide or sulindac sulfone; Figure 4-2). In each experiment, aliquots

(0.5 mL) of cell conditioned media were collected at the end of the culture and stored at -

200C until analyzed for PGF2a concentration. The remaining cell monolayers were lyzed

with Trizol and stored at -800C until PGHS-2 and PPAR6 mRNA quantifications.

Experiment 3. To determine whether EPA-mediated PGF2a secretion involves p38

or ERK MAP kinase activation, confluent BEND cells were treated with EPA (100

ng/mL), a combination of EPA and p38 MAP kinase inhibitor (SB203580 [1 [tM]), or a

combination of EPA and ERK MAP kinase inhibitor (PD98059 [20 [tM]) for 24 h, and

then challenged with PDBu (100 ng/mL) for an additional 6 h (Figure 4-3). After

incubation, aliquots (0.5 mL) of cell-conditioned media were collected and stored at -

20C for subsequent PGF2a radioimmunoassay. The remaining cell monolayers were

rinsed in ice-cold HBSS, lysed with TriZol, and stored at -800C until PGHS-2 mRNA

quantification.

PGF2a Radioimmunoassay

Prostaglandin F2a concentration in cell-conditioned media was measured in

duplicates as described by Danet-Desnoyers et al. (1994) and modified by Binelli et al.

(2000). Assay sensitivity was 5 ng/mL and intra- and inter-assay coefficients of variation

were 8.2 and 15.8%, respectively. To adjust for between well differences in cell density,









final PGF2a concentrations were expressed as picograms per million cells. Cell numbers

in individual dishes were determined using a hemocytometer (Sigma Chemical Co., St.

Louis, MO).

RNA Isolation and Analysis

Total cellular RNA was isolated from control and treated BEND cell cultures

using TriZol reagent according to the manufacturer's instructions. Ten micrograms of

RNA was fractionated in 1.0% agarose-formaldehyde gel and blotted to a BioTrans nylon

membrane by capillary action. The RNA was cross-linked to the membrane by UV

irradiation and baked at 800C for 1 h. The RNA filters were hybridized consecutively

with random primer-labeled PGHS-2 and PPAR6 cDNA probes (Balaguer et al., 2005).

After hybridization, RNA filters were washed for 20 min in 50 mL 2X SSC, 0.1% SDS at

room temperature, followed by two 15-min washes in 0.1X SSC, 0.1% SDS at 42C. The

filters were blotted dry and exposed to X-ray films for 6 to 24 h at -800C. Hybridization

signals for each target gene were quantified by densitometric analysis.

Statistical Analyses

Concentrations of PGF2a in cell culture medium and mRNA responses were

analyzed using the General Linear Model procedure of the SAS software package (SAS

Institute Inc., Cary, NC). For PGF2a concentration, the sources of variation included

experiment, treatment, experiment x treatment interaction, and well (experiment x

treatment). The well, nested within experiment and treatment, was considered a random

variable and, therefore, the well variance was used as error term to test the effects of

experiment, treatment and experiment x treatment interaction. For PGHS-2 and PPAR6

mRNA responses, the mathematical models included experiment, treatment and









experiment x treatment interaction. Densitometric values for each target gene were

expressed as ratios over the values for 18S ribosomal RNA. When treatment effects were

detected (P < 0.05), means were separated using orthogonal contrasts.

Results

To test the hypothesis that EPA may attenuate endometrial PGF2a secretion through

activation of nuclear PPAR6, we examined endometrial PGF2a response to PDBu in the

absence or presence ofL-165,410, a specific PPAR6 agonist. The PPAR6 agonist

completely abolished (P < 0.01) PGF2a response to PDBu (Figure 4-4). The decrease in

PGF2a secretion coincided with a significant reduction in PGHS-2 mRNA abundance in

PDBu + L-165,410-treated cells (Figure 4-5). Basal and PDBu-induced PPAR6 mRNA

concentrations were unaffected by supplemental PPAR6 agonist (Figure 4-6).

When added in combination with EPA, the PPAR6 inhibitor sulindac sulfide

decreased (P < 0.01) PGF2a response to PDBu to a greater extent (-94%) than did EPA

alone (-78%; Figure 4-7). The decrease in PGF2a secretion due to sulindac sulfide

coincided with a reduction (P < 0.01) in steady-state PGHS-2 mRNA concentration in

sulindac sulfide-treated cells (Figure 4-8). The PPAR6 mRNA response to PDBu was

unaffected by supplemental sulindac sulfide (Figure 4-9).

Because sulindac sulfide decreased PGHS-2 mRNA concentration in PDBU-

treated BEND cells, it was unclear whether this PPAR6 inhibitor decreased endometrial

PGF2a secretion via repression of PGHS-2 gene and or through a specific inhibition of

nuclear PPAR6. To test a PGHS-2-independent mechanism of PGF2a inhibition in

endometrial cells, we examined PGF2a response to PDBu in the presence of EPA and

sulindac sulfone, a PPAR6 inhibitor that has been reported to have no effects on PGHS-2









gene expression (Babbar et al., 2003). Sulindac sulfone decreased (P < 0.01) PGF2a

response to PDBu to a greater extent (-86%) than did EPA alone (-71%; Figure 4-10).

However, unlike sulindac sulfide, treatment of BEND cells with sulindac sulfone had no

detectable effects on steady-state PGHS-2 mRNA concentration (Figure 4-11).

Interestingly, priming of BEND cells with EPA alone or EPA and sulindac sulfone

increased (P < 0.05) PPAR6 response to PDBu (Figure 4-12).

To test whether EPA inhibits endometrial PGF2a production through the p38 or

ERK MAP kinase signal transduction pathways, we examined PGF2a response to PDBu

in the absence or presence of SB203580, a p38 inhibitor, or PD98059, an ERK inhibitor.

Pre-incubation of BEND cells for 24 h with either inhibitor had no detectable effect on

PGF2a production as compared to EPA + PDBu treated BEND cells (Figure 4-13). There

was no difference in PGHS-2 mRNA level among any of the treatments tested (Figure 4-

14).

Discussion

Polyunsaturated fatty acids elicit several metabolic changes through alteration of

the activity or synthesis of nuclear PPARs, all of which appear to have distinct patterns of

expression and functional roles (Brassant and Wahli, 1998). Our laboratory recently

detected an inverse relationship between PPAR6 and uterine estrogen receptor alpha and

PGHS-2 genes (Balaguer et al., 2005), suggesting that this nuclear receptor may play an

important role in the control of reproductive processes in mammalian species.

In the present study, PPAR6 activation greatly reduced PGF2a and PGHS-2 mRNA

responses to PDBu in cultured BEND cells (Figure 4-4 and Figure 4-5). This is in

contrast with a previous study in which PPAR6 activation induced PGHS-2 gene









expression in human hepatocellular carcinoma cells (Glinghammar et al., 2003).

Consistent with our findings, Inouie et al. (2000) reported that PPARy activation reduced

the PGHS-2 promoter activity in lipopolysaccharide-stimulated monocytic cells.

Furthermore, Subbamaiah et al. (2001) showed that PPAR ligands could suppress tissue-

type plasminogen activator (TPA)-driven PGHS-2 transcription in human epithelial cells

via activator protein-1 (AP-1) and cAMP response element binding protein (CREB) at the

cAMP response element (CRE) site in the proximal promoter. These studies collectively

indicate that the net effect of PPAR activation may vary depending on the cell type and

likely depends on the presence and or activation of other transcription co-factors in a

given cell system (Lim et al., 2004).

To further characterize the role of PPAR6 in EPA-induced attenuation of

endometrial PGF2a secretion, we examined the effects of two PPAR6 inhibitors on PGF2a

response to EPA in cultured BEND cells. Sulindac sulfide and sulindac sulfone

decreased PGF2a secretion to greater extents than did EPA alone. The decrease in PGF2a

secretion was associated with a significant reduction in PGHS-2 mRNA abundance in

sulindac sulfide-treated, but not sulindac sulfone-treated cells. Results are consistent

with previous reports that sulindac sulfide is a potent inhibitor of cyclooxygenase gene

expression and PG synthesis in several malignant cells (Marnett, 1992; Meade et al.,

1993; Lim et al., 1999). The observation that sulindac sulfone decreased PGF2a secretion

without altering endometrial PGHS-2 mRNA content also is consistent with the literature

data (Thompson et al., 1995,1997; Babbar et al., 2003), and would suggest that the

sulfone derivative of sulindac regulates uterine PGF2a secretion through a mechanism that

is PGHS-2-independent.









Previous studies have shown that the MAP kinase signaling cascade functions as an

important regulator of prostaglandin biosynthesis (Guan et al., 1998), and in ruminants,

there is evidence that ERK and p38 MAP kinases are involved in activation of the PG

biosynthetic pathway (Guan et al., 1997; Lin et al., 1998; Burs et al., 2001; Guzeloglu et

al., 2004). Inhibition of p38 MAP kinase in several cell types resulted in reduced

expression and/or increased turnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al.,

1998; Jang et al., 2000; Subbaramaiah et al., 2003). Most importantly, a recent

investigation by Ait-Said et al. (2003) showed that EPA suppressed p38 MAP kinase

phosphorylation and down-regulated PGHS-2 gene expression in stimulated human

microvascular endothelial cells, suggesting a novel regulatory point for EPA in PGF2a

inhibition.

The present study indicated that p38 or ERK inhibitor had no detectable effect on

PGF2a or PGHS-2 mRNA responses to EPA in cultured BEND cells. This is in contrast

with previous observations which showed that inhibition of the p38 MAP kinase in

several cell types resulted in reduced expression and/or increased turnover of PGHS-2

mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al.,

2003). It is conceivable that the use of PDBu in BEND cells leads to an activation of

PKC and, subsequently, of PGHS-2, which may override the effect of p38 inhibition on

PGHS-2 gene expression.

Collectively, these findings indicate that EPA and PPAR6 alter uterine endometrial

PGF2a secretion through complex mechanisms which may or may not involve PGHS-2

regulation. Whether and how EPA and PPAR6 affect the activities of several enzymes

involved in endometrial PGF2a cascade remains to be elucidated. Further studies are









necessary to investigate the role of p38 and ERK in EPA-mediated PGF2a inhibition in

cultured BEND cells.

Summary

Phorbol ester stimulated PGF2a production and up-regulated PGHS-2 gene

expression within 6 h in cultured BEND cells. Pre-incubation of confluent BEND cells

with a PPAR6 agonist (L-165,041) for 24 h decreased PGF2a and steady-state PGHS-2

mRNA responses to PDBu, but had no detectable effect on PPAR6 mRNA

concentrations. Pre-incubation of BEND cells with PPAR6 inhibitors (sulindac sulfide

and sulindac sulfone) decreased PGF2a response to PDBu to a greater extent than did

EPA alone. Treatment of BEND cells with sulindac sulfide resulted in a reduction in

steady-state PGHS-2 mRNA levels, while use of sulindac sulfone had no detectable

effects on steady-state PGHS-2 mRNA abundance. Basal and PDBu-induced PPAR6

mRNA concentrations were unaffected by supplemental sulindac sulfide, while sulindac

sulfone increased PPAR6 response to PDBu. Inhibiton of p38 and ERK MAP kinases

had no detectable effects on the PGF2a or PGHS-2 response to EPA.

Results of these studies indicate that EPA and PPAR6 alter uterine endometrial

PGF2a secretion through complex mechanisms which may or may not involve PGHS-2

regulation. Further studies are necessary to elucidate the role of MAP kinases in EPA-

mediated PGF2a inhibition in the bovine uterus.




















BEND cells

Grown to confluence
and split



Ag Incubation
24 h (-) (-) (+) (+)
mmmin


PDBu incubation
6h


mm


Control


(+) i-)


PDBu


(+)


PDBu + Ag


Figure 4-1. Experimental manipulations to determine if eicosapentaenoic acid (EPA)-
mediated inhibition of prostaglandin F2( (PGF2() secretion involves
peroxisome proliferator-activated receptor 6 (PPAR6) activation in bovine
endometrial (BEND) cells.




















Grown to confluence
and split

U
EPA incubation
24h
In irtiaaj~.li Eft ub Ui^'n!i


PDTu + EPA


I(+)


PDBu + EPA + I


Figure 4-2. Experimental manipulations to determine if peroxisome proliferator-
activated receptor 6 (PPAR6) inhibition may abolish eicosapentaenoic acid
(EPA) effect on prostaglandin F2z (PGF2() secretion in bovine endometrial
(BEND) cells.


H(-)


(4+)


PDBu incubation
6h


(+)


PDBu




















BEND cells


Grown to confluer
and split


MAPK Inhbltor incubation
24 h
EP i ikJItL12-.I n 1 1


a


ice



(-) (-) (+) (+)
H- I(+) (+) (+

mm mm T


-- --C
PDBu PDBu + EPA PDBu+ EPA PDBu + EPA
+ SB + PD


Figure 4-3. Experimental manipulations to examine the role of mitogen-activated protein
(MAP) kinases in eicosapentaenoic acid (EPA) regulation of prostaglandin F2(
(PGF2a) production in bovine endometrial (BEND) cells.















16



S12




0
u 8

U-


C. 4




Control PDBu Ag PDBu+Ag
Treatments


Figure 4-4. Effect of a peroxisome proliferator-activated receptor 6 (PPAR6) agonist (L-
165,041) on prostaglandin F2a (PGF2a) secretion in bovine endometrial
(BEND) cells. Data represents least square means SEM of two independent
experiments. When treatment effects were detected (P < 0.05), means were
separated using orthogonal contrasts. Contrast 1: (Control), (Ag) vs. (PDBu),
(PDBu + Ag), P < 0.0001; Contrast 2: (PDBu) vs. (PDBu + Ag), P < 0.0001;
Contrast 3: (Control) vs. (Ag), P = 0.2.












C PDBu Ag


PDBu + Ag


Control PDBu Ag PDBu+Pg
Treatments


Figure 4-5. Effect of a peroxisome proliferator-activated receptor 6 (PPAR6) agonist (L-
165,041) on prostaglandin endoperoxide synthase (PGHS-2) mRNA
abundance in bovine endometrial (BEND) cells. Ten micrograms of total
cellular RNA isolated from control and treated BEND cells were subjected to
Northern blot analysis (A), and resulting densitometric values were analyzed
by the GLM procedure of SAS (B). The top panel shows a representative
Northern blot, whereas the bottom panel represents means SEM calculated
over two experiments (n = 4 for each treatment). When treatment effects were
detected (P < 0.05), means were separated using orthogonal contrasts.
Contrast 1: (Control), (Ag) vs. (PDBu), (PDBu + Ag), P = 0.0001; Contrast 2:
(PDBu) vs. (PDBu + Ag), P = 0.004; Contrast 3: (Control) vs. (Ag), P = 0.9.


(A)


PGHS-2
mRNA


rRNA


(B)


4.4 kb


18S


0.3


z
, 0.2
o
a-
0.1
,0
0..
E


0.0
0.0












C PDBu Ag PDBu + Ag


3.5 kb


18S


(B)


<

z
00
E
a"
13_
1:
Q_ v


Control PDBu Ag PDBu+Ag
Treatments


Figure 4-6. Effect of a peroxisome proliferator-activated receptor 6 (PPAR6) agonist (L-
165,041) on PPAR6 mRNA response to PDBu in bovine endometrial (BEND)
cells. Ten micrograms of total cellular RNA isolated from control and treated
BEND cells were subjected to Northern blot analysis (A), and resulting
densitometric values were analyzed by the GLM procedure of SAS (B). The
top panel shows a representative Northern blot, whereas the bottom panel
represents means SEM calculated over two experiments (n = 4 for each
treatment). When treatment effects were detected (P < 0.05), means were
separated using orthogonal contrasts. Contrast 1: (Control), (Ag) vs. (PDBu),
(PDBu + Ag), P = 0.008; Contrast 2: (PDBu) vs. (PDBu + Ag), P = 0.9;
Contrast 3: (Control) vs. (Ag), P = 0.2.


(A)


PPARa
mRNA


rRNA




















48






32
o(



0u
C 1







n


PDBu PDBu+EPA PDBu +EPA+I

Treatments



Figure 4-7. Effect of sulindac sulfide on prostaglandin F2a (PGF2a) response to
eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Data
represents least square means SEM of five independent experiments. When
treatment effects were detected (P < 0.05), means were separated using
orthogonal contrasts. Contrast 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA +
I) P < 0.0001; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P < 0.0001.


______


4V*qV*.












PDBu PDBu + EPA PDBu + EPA +I


PDBu PDBu+EPA PDBu+EPA+I
Treatments


Figure 4-8. Effect of sulindac sulfide on prostaglandin endoperoxide synthase (PGHS-2)
mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial
(BEND) cells. Ten micrograms of total cellular RNA isolated from control
and treated BEND cells were subjected to Northern blot analysis (A), and
resulting densitometric values were analyzed by the GLM procedure of SAS
(B). The top panel shows a representative Northern blot, whereas the bottom
panel represents means SEM calculated over two experiments (n = 4 for
each treatment). When treatment effects were detected (P < 0.05), means
were separated using orthogonal contrasts. Contrast 1: (PDBu) vs. (PDBu +
EPA), (PDBu + EPA + I) P = 0.2; Contrast 2: (PDBu + EPA) vs. (PDBu +
EPA + I), P = 0.0005.


(A)


PGHS-2
mRNA

rRNA


(B)


4.4 kb


18S


OL3


<'
z
L- 0.2



CD
co

o 0.1
0



0.0











PDBu + EPA PDBu + EPA + I


DBLI DBEPEPA FM&BLEPA


Trearrmts


Figure 4-9. Effect of sulindac sulfide on peroxisome proliferator-activated receptor 6
(PPAR6) mRNA response to eicosapentaenoic acid (EPA) in bovine
endometrial (BEND) cells. Ten micrograms of total cellular RNA isolated
from control and treated BEND cells were subjected to Northern blot analysis
(A), and resulting densitometric values were analyzed by the GLM
procedure of SAS (B). The top panel shows a representative Northern blot,
whereas the bottom panel represents means + SEM calculated over two
experiments (n = 4 for each experiment).


(A)


PDBu


PPAR8
mRNA

rRNA


3.5 kb


18S


(B)


o0- 0.1

!.-


















32




24

CI
o

c 16

C
U-
a
8


0
PDBu PDBu+EPA PDBu+EPA+I

Treatments




Figure 4-10. Effect of sulindac sulfone on prostaglandin F2a (PGF2a) response to
eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Data
represents least square means SEM of three independent experiments.
When treatment effects were detected (P < 0.05), means were separated using
orthogonal contrasts. Contrast 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA +
I) P < 0.0001; Contrast 2: (PDBu + EPA) vs. (PDBu + EPA + I), P = 0.0005.












PDBu + EPA


PDBu + EPA + I


4.4 kb


18S


PDBu PDBu+EPA PDBu+EPA+l
Treatments


Figure 4-11. Effect of sulindac sulfone on prostaglandin endoperoxide synthase (PGHS-
2) mRNA response to eicosapentaenoic acid (EPA) in bovine endometrial
(BEND) cells. Ten micrograms of total cellular RNA isolated from control
and treated BEND cells were subjected to Northern blot analysis (A), and
resulting densitometric values were analyzed by the GLM procedure of
SAS (B). The top panel shows a representative Northern blot, whereas the
bottom panel represents means SEM calculated over two experiments (n
= 4 for each treatment). When treatment effects were detected (P < 0.05),
means were separated using orthogonal contrasts. Contrast 1: (PDBu) vs.
(PDBu + EPA), (PDBu + EPA + I) P = 0.07; Contrast 2: (PDBu + EPA) vs.
(PDBu + EPA + I), P = 0.4.


PDBu


(A)


PGHS-2
mRNA

rRNA


(B)


z
< 0.2
Z0cn
E O
E--




C 0.0












PDBu


PDBu+EPA PDBu+EPA+I


3.5 kb


18S


PDBu PDBu+EPA PDBu+EPA+I
Treatments


Figure 4-12. Effect of sulindac sulfone on peroxisome proliferator-activated receptor 6
(PPAR6) mRNA response to eicosapentaenoic acid (EPA) in bovine
endometrial (BEND) cells. Ten micrograms of total cellular RNA isolated
from control and treated BEND cells were subjected to Northern blot analysis
(A), and resulting densitometric values were analyzed by the GLM procedure
of SAS (B). The top panel shows a representative Northern blot, whereas the
bottom panel represents means SEM calculated over two experiments (n
= 4 for each treatment). When treatment effects were detected (P < 0.05),
means were separated using orthogonal contrasts. Contrast 1: (PDBu) vs.
(PDBu + EPA), (PDBu + EPA + I) P = 0.01; Contrast 2: (PDBu + EPA) vs.
(PDBu + EPA + I), P = 0.8.


(A)


PPAR6
mRNA

rRNA


(B)


z
L 0.2
cm
E
al

< N

0



0.0







79








32



24
(5o
e-2
CD



C 8

0




PDBu PDBu+EPA PDBu+E PA+SB PDBu+EPA+PD

Treatments



Figure 4-13. Effect of SB203580 and PD98059 on prostaglandin F2a (PGF2,) response to
eicosapentaenoic acid (EPA) in bovine endometrial (BEND) cells. Data
represents least square means SEM of three independent experiments.
When treatment effects were detected (P < 0.05), means were separated using
orthogonal contrasts. Contrast 1: (PDBu) vs. (PDBu + EPA), (PDBu + EPA +
SB), (PDBu + EPA + PD) P < 0.0001; Contrast 2: (PDBu + EPA) vs. (PDBu
+ EPA + SB), (PDBu + EPA + PD), P < 0.0001, Contrast 3: (PDBu + EPA +
SB) vs. (PDBu + EPA + PD), P = 0.9.











PDBu PDBu+EPA PDBu+EPA+SB PDBu+EPA+PD


4.4 kb


18S


(B)


POBu POBEj+EPA POBu+EPAI-SB POBu4EPA4PO


Treatments


Figure 4-14. Effect of SB203580 and PD98059 on prostaglandin endoperoxide synthase
(PGHS-2) mRNA response to eicosapentaenoic acid (EPA) in bovine
endometrial (BEND) cells. Ten micrograms of total cellular RNA isolated
from control and treated BEND cells were subjected to Northern blot analysis
(A), and resulting densitometric values were analyzed by the GLM procedure
of SAS (B).


(A)


PGHS-2
mRNA

rRNA














CHAPTER 5
GENERAL DISCUSSION

Recent studies have implicated n-3 polyunsaturated fatty acids (PUFAs) in the

reduction of prostaglandin F2a (PGF2a) synthesis in the bovine uterus. Although cattle

diets contain a mixture of n-3 and n-6 fatty acids, currently there is a lack of information

as to how these fatty acids may interact to alter PGF2a biosynthesis in the uterus. Results

of this study extend previous observations that n-3 fatty acids are potent inhibitors of PG

secretion in mammalian species (Baguma-Nibasheka et al., 1999; Olsen et al., 1992;

Mattos et al., 2000; Mattos et al., 2002; Mattos et al., 2003; Cheng et al., 2001).

Although the exact mechanism by which supplemental n-3 fatty acids inhibit PGF2a

production is not fully understood, it is conceivable that increased availability of EPA in

membrane phospholipids as a result of treating BEND cells with EPA could displace AA,

leading to increased synthesis of PGs of the 3 series at the expense of PGs of the 2 series

(Mattos et al., 2003). This hypothesis does not rule out the possibility that supplemental

EPA also may inhibit PGHS-2 activity in cultured BEND cells. In fact, incubation of rat

hepatoma cells with AA, EPA, DHA, or heineicosapentaenoic acid (C21:5 n-3) inhibited

the PGHS-2 enzyme activity (Larsen et al., 1997). Eicosapentaenoic acid inactivated the

enzyme almost completely when added 30 sec before addition of AA.

The rate-limiting step in PG synthesis involves the cleavage of sn-2 fatty acyl ester

bond of membrane phospholipids by cytosolic PLA2 (Van den Bosh, 1980; Irvine, 1982).

The AA that is released by phospholipid hydrolysis is acted on by PGHS-2 to form









PGH2, which then is converted to PGF2a. Consistent with a previous observation (Mattos

et al., 2003), the present study provided no evidence for EPA regulation of PGHS-2

mRNA abundance in BEND cells. As discussed above, supplemental EPA may alter

endometrial PGF2a production through competitive displacement of AA from membrane

phospholipids and/or through alteration of the PGHS-2 enzymatic activity.

Polyunsaturated fatty acids elicit several physiological changes through the

alteration of the activity or synthesis of nuclear PPARs (Bocher et al., 2002). In mice,

PPAR6 deficiency leads to placental defects and results in frequent mid gestational

lethalities (Barak et al., 2002), suggesting that this nuclear receptor may play an

important role in the control of reproductive processes in mammalian species. Consistent

with a recent in vivo experiment (Palin et al., 2005), supplemental EPA had no detectable

effects on PPAR6 response to PDBu in cultured BEND cells. Results suggest that

supplemental n-3 fatty acids may alter endometrial PGF2a production through a

mechanism which does not require induction of PPAR6 gene. However, whether and

how these fatty acids may control the activity of this nuclear receptor warrants further

investigation.

Conventional cattle diets contain a mixture of n-6 and n-3 fatty acids. Therefore,

we examined the effect of EPA on endometrial production of PGF2a in the presence of

increasing concentrations of LA. The inhibitory effect of EPA on uterine endometrial

PGF2a production decreased from 88 to 40%, as the n-6/n-3 fatty acid ratio in the culture

medium increased from 0 to 19. These findings are consistent with previous reports

(Trujillo et al., 1995; Achard et al., 1997), and suggest that the net inhibition of uterine

PGF2a synthesis by n-3 fatty acids may depend on the amount of n-6 fatty acids reaching









the target tissue. Increasing concentrations of LA in the cell culture system may increase

the availability of AA in membrane phospholipids and therefore decrease the competition

by n-3 fatty acids for the PGHS-2 enzyme. Whether and how increasing n-6/n-3 fatty

acid ratios alter the PGHS-2 activity is yet to be elucidated.

For the experiments that focused on studying the mechanisms of action of EPA on

PGF2a production in BEND cells, we hypothesized that if EPA affects PGF2a production

through the activation of PPAR6 or of p38 or ERK, then activation of this nuclear

receptor, or of these MAP kinases, would mimic the effect of EPA on PGF2a production

in cultured BEND cells. In the present study, PPAR6 activation greatly reduced PGF2a

and PGHS-2 mRNA responses to PDBu in cultured BEND cells. This is in contrast with

a previous study in which PPAR6 activation induced PGHS-2 gene expression in human

hepatocellular carcinoma cells (Glinghammar et al., 2003). Consistent with our findings,

Inouie et al. (2000) reported that PPARy activation significantly reduced the PGHS-2

promoter activity in lipopolysaccharide-stimulated monocytic cells. Furthermore,

Subbamaiah et al. (2001) showed that PPAR ligands could suppress TPA-driven PGHS-2

transcription in human epithelial cells via AP-1 and CREB binding proteins at the CRE

site in the proximal promoter. These studies collectively indicate that the net effect of

PPAR activation may vary depending on the cell type and likely depends on the presence

and or activation of other transcription co-factors in a given cell system (Lim et al.,

2004).

To further characterize the role of PPAR6 in EPA-induced attenuation of

endometrial PGF2a secretion, we examined the effects of two PPAR6 inhibitors on PGF2a

response to EPA in cultured BEND cells. Sulindac sulfide and sulindac sulfone









decreased PGF2a secretion to greater extents than did EPA alone. The decrease in PGF2a

secretion was associated with a significant reduction in PGHS-2 mRNA abundance in

sulindac sulfide-treated, but not sulindac sulfone-treated cells. Results are consistent

with previous reports that sulindac sulfide is a potent inhibitor of cyclooxygenase gene

expression and PG synthesis in several malignant cells (Mamett, 1992; Meade et al.,

1993; Lim et al., 1999). The observation that sulindac sulfone decreased PGF2a secretion

without altering endometrial PGHS-2 mRNA content also is consistent with the literature

data (Thompson et al., 1995,1997; Babbar et al., 2003), and would suggest that the

sulfone derivative of sulindac regulates uterine PGF2a secretion through a mechanism that

is PGHS-2-independent.

Previous studies have shown that the MAP kinase signaling cascade functions as an

important regulator of prostaglandin biosynthesis (Guan et al., 1998), and in ruminants,

there is evidence that ERK and p38 MAP kinases are involved in activation of the PG

biosynthetic pathway (Guan et al., 1997; Lin et al., 1998; Bums et al., 2001, Guzeloglu et

al., 2004). Inhibition of p38 MAP kinase in several cell types results in reduced

expression and/or increased turnover of PGHS-2 mRNA (Dean et al., 1999; Guan et al.,

1998; Jang et al., 2000; Subbaramaiah et al., 2003). Most importantly, a recent

investigation by Ait-Said et al. (2003) showed that EPA suppressed p38 MAP kinase

phosphorylation and down-regulated PGHS-2 gene expression in stimulated human

microvascular endothelial cells, suggesting a novel regulatory point for EPA in PGF2a

inhibition.

The present study indicated that p38 or ERK inhibitor had no detectable effect on

PGF2a or PGHS-2 mRNA responses to EPA in cultured BEND cells. This is in contrast









with previous observations which showed that inhibition of the p38 MAP kinase in

several cell types resulted in reduced expression and/or increased turnover of PGHS-2

mRNA (Dean et al., 1999; Guan et al., 1998; Jang et al., 2000; Subbaramaiah et al.,

2003). It is conceivable that the use of PDBu in BEND cells leads to an activation of

PKC and, subsequently, of PGHS-2, which may override the effect of p38 inhibition on

PGHS-2 gene expression.

In summary, this study presents direct evidence that the inhibition of uterine

endometrial PGF2a biosynthesis by n-3 fatty acids depends on the amount of n-6 fatty

acids in the uterus, and that EPA and PPAR6 affect uterine PGF2a synthesis through

complex mechanisms which may or may not involve PGHS-2 gene regulation. Further

studies are needed to fully document the role of p38 and ERK MAP kinases in

endometrial PGF2a biosynthesis.
















h Cj CA(


-4-

CCA


C;~j cn
-m I

C -e
C A j

SO
~ -e ~ 04









61 ij~d b h
> -d ~ 0















aCVj CI
0G ~
0*- a













a) C
C- 0 -
So -e a,



-e e ~
0 09 p,- -







142 > a) C
- Wr,
~ a 0 .e




0, a




CA C41 v



CI) 14 I
0 o) ~0
c~~~, .> 0 ~~P"


















Ccj
o~Y






U, u) 0, P



E_

~C0
N .
1-e rq
to 0c-

y4 *- d
P 1. Z .1.1 ). C


U) jf I m h






IAC un ~ -
09


00 c
ur -e
ow d C ) cP
LfloU C
r* F














R ut 1- ol
0.o
E -Eu
-0 O
L-- w o i