Pregnancy recognition in cattle


Material Information

Pregnancy recognition in cattle effects of conceptus products on uterine prostaglandin production
Physical Description:
xi, 273 leaves : ill. ; 28 cm.
Knickerbocker, Jeffrey John, 1955-
Publication Date:


Subjects / Keywords:
Cattle -- Fertility   ( lcsh )
Cattle -- Reproduction   ( lcsh )
Animal Science thesis Ph. D
Dissertations, Academic -- Animal Science -- UF
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph. D.)--University of Florida, 1985.
Includes bibliographical references (leaves 229-271).
Statement of Responsibility:
by Jeffrey John Knickerbocker.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000505078
oclc - 22770836
notis - ACS5231
sobekcm - AA00004878_00001
System ID:

Full Text








During the last several years, I have had the good

fortune of interacting with a host of generous and talented

individuals. The many contributions, insights and

friendships offered me by these individuals are acknowledged


I wish to extend my heartfelt thanks and appreciation

to Dr. William W. Thatcher, chairman of my supervisory

committee, for giving generously of his time, experience and

knowledge throughout my Ph.D. training. Dr. Thatcher's

insight and contributions with regard to my education and

research have made lasting impressions on my perception of

reproductive function and philosophy of basic reproductive

research. His friendship and interest in my professional

growth and personal well-being are deeply appreciated. Dr.

Thatcher will always be held in my highest esteem as a

scientist, teacher and friend.

Special thanks are given to Dr. Fuller W. Bazer,

cochairman, for his constant interest, availability and

input during the course of my research endeavors. His

substantial contributions to my research program and

understanding of reproductive physiology have been

invaluable. Dr. Bazer's friendship throughout the years and

"support" during the final stages of this degree will always

be appreciated greatly.

Thanks are expressed to Dr. Michael J. Fields for our

numerous stimulating conversations, and for his guidance and

support during early aspects of my research in his

laboratory. My association and friendship with Dr. Fields

over the years have been a pleasure. His contributions to

my research endeavors and education are sincerely


Dr. R. Michael Roberts is acknowledged for his

important role in the development of technical and

conceptual insight relevant to concepts protein

biochemistry and physiology. In this respect, Dr. Roberts

has been a tremendous asset in my research endeavors

relative to concepts secretary protein function in

cattle. I thank Dr. Roberts for his interest and input

while serving on my supervisory committee.

Dr. Donald Caton is acknowledged greatfully for his

input, expert surgical training and assistance in numerous

surgeries at the 34th Street laboratory during our

characterization of uterine responses to estradiol. I am

indebted to Dr. Caton for his contributions while serving on

my supervisory committee and for his friendship.

Thanks are extended to Dr. Robert J. Collier for his

comments and critical evaluation of my dissertation. His

friendship and willingness to participate in my dissertation


defense in the absence of Dr. Thatcher are appreciated


I wish to extend my sincere appreciation and thanks to

Dr. Donald H. Barron and Dr. Maarten Drost for their

gracious and generous support, and valuable interactions and

insight throughout the course of my studies at the

University of Florida. It has been tremendously gratifying

for me to have worked with these gentlemen. Thank you.

Special thanks are given to fellow graduate students,

Dallas Foster, Dr. Louis Guilbault, Dr. Frank 'SkiD' Bartol

and Kimberly Fincher for their valued friendships,

interactions and inestimable assistance over the years.

Friendships and support offered me by Dr. Charlie Wilcox,

Dr. Herb Head, Dr. Dave Beede, Dr. Dan Sharp, III, Dr. Ray

Roberts, Dr. David Wolfenson, Dr. Lee Fleeger, Dr. Wilfrid

Dubois, Dr. Rod Geisert, Dr. Randy Renegar, Dr. Jeff

Moffatt, Dr. Charlie Ducsay, Dr. Juaquim o!oya, Harold

Fischer, John McDermott, Dr. Helton Saturino, Karen

McDowell, Marlin Dehoff, Lokenga Badinga, Deanne Morse, Fran

Romero, Sue Chiachimonsour, Jeff Valet, Kathy Hart, -.

George Baumbach and Leslie Smith are deeply appreciated.

Thank you.

I am greatful especially to Dr. Paul Schneider and son,

Jordan, for their generosity, patience, hospitality and

encouragement during these last 3 months.

I thank Carol Underwood, Candy Stoner, Jesse Jonnson

and Warren Clark for their tolerance, technical assistance

and efficiency in the laboratory, and Cindy Zimmernan for

typing this manuscript so skillfully.

To special friends, Chuck and June Wallace, and Gail

and Bob Knight, most sincere thanks and love are extended

from my family and I. Keep in touch.

I wish to thank my parents, John and Jan, for their

continued love, support and confidence. This proud day

would not have been possible without the principles and

motivation instilled in me by them both.

Lastly, the major contributions and sacrifices made by

my wife, Holly, throughout the course of my studies are more

than greatfully acknowledged. Her constant love, support,

patience and understanding have given me the strength to

endure and reasons to succeed. My deepest love and

affections are hers always.



ACKNOWLEDGMENTS............ .......... ......... ............ ii

ABSTRACT..................................... ........... viii


1 REVIEW OF LITERATURE ............................

Introduction................. .....................1
The Bovine Estrous Cycle....................... 2
The Corpus Luteum Life Cycle...................17
Uterine Control of CL Function.................49
Conceptus-Associated Events During Early
Pregnancy ................................... 59

EXPLANTS ............. ...................... 88

Introduction ... .............. ............... 88
Materials and Methods............ .............90
Results .......................................102
Discussion ........................................129


Introduction .................................139
Materials and Methods.........................141
Results ....... .......... ........ ...... .. .. 146
Discussion........ .............................159

CATTLE ........... ..... ....................... 166

Introduction ................. ... ........... ... 166
Materials and Methods .......................168
Results .......................................176
Discussion .... ...... ... ..... ..................185










Introduction .......... ... ....... ........
Materials and Methods......................
Results ....................................
Discussion ...............................

GENERAL DISCUSSION .........................


CHROMATOGRAPHY ........... .............


54:46) ............................. ...... ...


REFERENCES.... .......... .. .. .......... .... ............

BIOGRAPHICAL SKETCH...................................


. .193

... 201









Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy




May, 1985

Chairman: William W. Thatcher
Cochairman: Fuller W. Bazer
Major Department: Animal Science

In cattle, continued progesterone (P4) production by

the corpus luteum (CL) is required if pregnancy is to per-

sist. Identification of putative conceptus-derived signals

and evaluation of their biological roles relative to CL

maintenance during early pregnancy were goals of this

research endeavor.

Using various chromatography systems, bovine con-

ceptuses (days 19 to 23) exhibited extensive metabolism of

tritiated P4 (90-98%), in vitro. A majority of concepts

metabolites were 56-reduced pregnanes. A major concepts

metabolite was 58-pregnan-3a-ol-20-one (5B-P). Conversely,

endometrial explant cultures metabolized 40 to 50% of P4

substrate to primarily 5a-reduced steroid products.


An in vivo test system to evaluate uterine PGF2, pro-

duction capacity was characterized in experiment two. Exo-

genous estradiol-178 (E2; 3 mg I.V.) stimulated uterine

blood flow, and PGF2a production and metabolism. Concen-

trations of the primary metabolite of PGF2., 15-keto-13,14-

dihydro-PGF2. (PGFM) were significantly correlated (r=.66)

with the E2-induced uterine PGF2. production response. Use

of peripheral PGFM concentrations as an index of uterine

PGF2a production was supported.

In experiment three, CL function, interestrous inter-

val, and spontaneous uterine PGF2a production were evaluated

in cyclic cows following intrauterine administration of

58-P, concepts secretary proteins (CSP) or homologous serum

proteins (Control). Extensions in CL lifespan and

interestrous interval were detected in cows administered CSP

compared to 58-P and Control group responses. Lifespan of

CL and estrous cycle lengths were not different (P>.25)

between 58-P and Control groups. Spontaneous PGF2a episodes

were depressed in CSP-treated cows but not in cows admini-

stered 58-P or serum proteins. In experiment four, E2-

induced uterine PGF2a production (peripheral PGFM response)

was evaluated in cyclic cattle following interuterine

administration of bovine CSP or day 18 pregnant serum

proteins (Control). Mean concentrations of PGFM in CSP-

treated cows were depressed (P<.01) compared to Control

responses. Estradiol injection failed to elicit any PGFM

response in three of five CSP-treated cows. In contrast,

all Control animals exhibited PGFM responses during the

period of E2-induced PGF2a production. Results support a

role for CSP in suppression of uterine PGF2a production

during early pregnancy in cattle.



The recurrent nature of estrous cycles in cattle

centers around development and regression of the ovarian

corpus luteum (CL). Progesterone (P4), the primary product

of CL, directs uterine development and secretary activity

such that an embryotropic uterine environment is established

during each estrous cycle. Luteal maintenance and P4

production are essential if successful pregnancy is to be

established. The process of CL maintenance during early

pregnancy involves numerous and complex biochemical inter-

actions between the concepts and its maternal host. Our

evaluation of mechanisms by which the concepts and maternal

units interact during early pregnancy depend upon identifi-

cation of putative conceptus-derived 'signals' and develop-

ment of biological test systems through which concepts

product function may be assessed. It was in this light that

research described herein was conducted.

The Bovine Estrous Cycle

General Considerations

The nature of the estrous cycle has been characterized

extensively in cattle. Early observations and documentation

of the recurrence of sexual behavior in animals provided the

first clue as to the cyclical nature of reproductive

processes in the female (Heape, 1900; Marshall, 1922;

Hammond, 1927; Asdell, 1969). Wallace in 1876 and

Ellenberger in 1892 (Hammond, 1927; as cited in Marshall,

1922) were among the earliest to report accurately the

interval between "sexual periods" in domestic female cattle.

The cyclical recurrence of estrous behavior follows a

periodicity of approximately 21 days with a range of 17 to

25 days considered normal (Hammond, 1922; Asdell et al.,

1949; Olds and Seath, 1951; Cupps et al., 1969).

Nulliparous heifers tend to exhibit slightly shorter

(approximately 1 day) interestrous intervals than

multiparous cows (Hammond, 1927; Olds and Seath, 1951).

Unlike sheep (Goodman and Karsch, 1981) and horses

(Sharp, 1980), photoperiod does not influence consistently

annual reproductive patterns in mature cattle (Rzepkowski et

al., 1982). Thus, reproductive activities in domestic

cattle are not restricted to specific seasons of the year

(Heape, 1900; Hammond, 1927). However, attainment of

puberty in cattle is enhanced by increased daily exposure to

light (Hansen et al., 1983).

Loss of body condition due to decreased nutrient

availability, lactational stress and parasitism has

detrimental effects on cyclicity and reproductive efficiency

in cattle (Lamond and Bindon, 1969; Dunn et al., 1969; Rakha

and Igboeli, 1971). Thus, wild cattle are known to exhibit

limited breeding and calving seasons under natural

conditions (Marshall, 1922; Heape, 1900). When such

stresses are reduced via management schemes in the confines

of a zoo, wild cattle are capable of breeding at all times

of the year (Heape, 1900).

High ambient temperature and humidity are other

important environmental influences which reduce expression

of estrus, blood flow to the reproductive tract, and may

alter the normal endocrine milieu in cattle (Thatcher and

Collier, 1985). Consequently, heat stress results in

decreased reproductive efficiency.

Historically, observations regarding cyclical

alterations in behavior and ovarian morphology led Walter

Heape (1900) to propose terminology which established the

first classification of these cyclical changes into four

recurrent phases. Thus, the estrous cycle diestrusus cycle"

by Heape's terminology) in cattle consists of four phases:

proestrus (day 19 to estrus), estrus (day 0), metestrus (day

1-3) and diestrus (day 4-18).

Physiology and Endocrinology

Sexual receptivity of the female, or estrus, is a

convenient behavioral landmark which is commonly employed to

mark the start of an estrous cycle (day 0). Estrus is often

characterized by increased restlessness (Kiddy, 1976),

vocalizations, mounting activity and, most notably, standing

to be mounted by other cattle (Howes et al., 1960). Other

variables related to estrus are discussed by Lewis and

Newman (1984). Average length of behavioral estrus

approaches 18 to 20 hours in mature cattle (Trimberger and

Hansel, 1955; Schams et al., 1977); however, cattle in

subtropical regions generally experience shorter (10 to 13

hour) phases of estrus (Branton et al., 1957; Chenault et

al., 1975).

The temporal sequence of events which culminate in

behavioral estrus, and later in the preovulatory gonado-

tropin surge, involve a synchronized and interdependent

cascade of physiological changes in the central nervous

system, hypothalamus, pituitary, ovaries and uterus. Char-

acterization of endocrine profiles and study of mechanisms

controlling their changes have, at least partially,

unravelled the dynamics of various reproductive processes.

Approximately 2 to 4 days before behavioral estrus and

the preovulatory surge of gonadotropins, regression of the

corpus luteum (CL) is initiated, as determined by a

precipitous decline in plasma progesterone (P4)

concentrations (Chenault et al., 1975; Peterson et al.,

1975). In association with this decrease in P4, pulse

frequencies of LH (Rahe et al., 1980; Schallenberger et al.,

1984) and FSH (Schallenberger et al., 1984) increase, as do

basal concentrations of LH, but not FSH (Chenault et al.,

1975; Roche and Ireland, 1981; Milvae and Hansel, 1983b;

Schallenberger et al., 1984).

Plasma concentrations of estradiol-178 (E2) begin to

rise during this period (Chenault et al., 1975; Walters and

Schallenberger, 1984; Schallenberger et al., 1984) as a

result of the initiation and accelerated growth of the

preovulatory follicle (Dufour et al., 1972; Merz et al.,

1981; Staigmiller et al., 1982; Ireland and Roche, 1982;

McNatty et al., 1984a,b). In fact, approximately ten times

more E2 was measured in ovarian vein plasma draining the

ovary containing the preovulatory follicle than the contra-

lateral ovary at estrus in cattle (McNatty et al., 1984a).

A significant correlation exists between LH and E2

pulse frequencies during all phases of the bovine estrous

cycle (Schallenberger et al., 1984; Walters and

Schallenberger, 1984; Walters et al., 1984). Prior to the

gonadotropin surge, LH pulsatile episodes occur every 30

minutes (high frequency) with E2 pulses being generated

responsively (Walters and Schallenberger, 1984). Addi-

tionally, E2 pulse amplitudes gradually increase as estrus

approaches. However, these alterations in E2 pulse

amplitude are not reflected by similar amplitude elevations

in LH episodes (Walters and Schallenberger, 1984). Inis may

suggest that LH pulse frequency and follicle sensitivity to

LH are major regulatory factors for follicular E2 production

in the cow.

Before reviewing data in support of such a claim, a

brief description of follicular anatomy and the two-cell

mechanism of follicular estrogen biosynthesis is in order.

The ovarian follicle consists of two steroidogenically

active cellular compartments: the vascularized theca

internal and the avascular granulosa, which is separated from

the theca internal by an acellular basal laminae (see:

McNatty et al., 1984b, for references). According to the

two-cell mechanism of follicular estrogen production

proposed by Falck (1959), the theca compartment metabolizes

C-21 steroids to androgens which are then utilized by the

granulosa compartment to synthesize estrogens. More recent

evaluation of steroidogenic pathways utilized by bovine

follicular compartments (Lacroix et al., 1974) support this

concept. The theca cell layer utilizes, almost exclusively,

the A5-pathway (pregnenolone and 17a-hydroxy-pregnenolone)

to synthesize androgens, of which androstenedione (A4) is

the major product. In vivo evaluation of A4 concentrations

in ovarian vein plasma and A4 production by the ovary

throughout the estrous cycle in cattle (Wise et al., 1982)

agree with in vitro appraisals of thecal androgen

biosynthesis by Lacroix and coworkers (1974). Metabolism of

A4 to estrogens by the granulosa is very efficient.

Conversely, only small amounts of estrogen are synthesized

by the theca compartment. Lacroix and coworkers (1974)

observed greater yields of estrogen with combined

incubations of theca and granulosa cells than yields with

incubations of either cell type separately, indicative of a

synergism between the two follicle compartments (Falck,


The capacity of follicles to synthesize estrogens

depends upon their ability to respond to gonadotropins. In

the cow, 85% of large (> 8 mm diameter) follicles on both

ovaries bound FSH to granulosa cells and hCG (LH) to theca

internal cells, while only 38% of these follicles also bound

hCG to granulosa. Follicles with granulosa hCG binding

sites were associated with enhanced E2 concentrations in

follicular fluid (England et al., 1981; Merz et al.,

1981). Sheep also exhibit a high positive correlation

between hCG binding to granulosa and follicular E2 content

and production (Webb and England, 1979, 1982). McNatty et

al. (1984a) suggested that there was at least one follicle

greater than 5 mm (range 1 to 3 follicles per cow) with

granulosa cells possessing aromatase activity on day -5

through estrus. However, most, if not all, healthy bovine

follicles greater than 2 mm and many atretic follicles were

capable of secreting A4 in response to LH. In no cases were

follicles with granulosa aromatase activity found in the

absence of an LH-responsive thecal compartment. Lixewise,

Bartol et al. (1981) demonstrated that large follicles

capable of significant E2 production were present in ovaries

of cattle throughout the estrous cycle.

In the rat (Nimrod et al., 1977), pig (Channing, 1975)

and sheep (Weiss et al., 1978) ovarian follicle, granulosa

binding sites for hCG/LH are induced by FSH, thereby

enhancing granulosa cell responsiveness- to LH.

Additionally, P4, dihydroxytestosterone (DHT) and E2 plus

FSH synergistically stimulated granulosa cell responsiveness

to LH above levels found with FSH alone (Rani et al.,

1981). Low concentrations of E2 enhance P4 production by

isolated bovine theca cells (Fortune and Hansel, 1979),

suggesting that E2 may also regulate theca sensitivity to


Thus, follicular development of granulosa aromatase

activity appears to occur secondarily to the thecal capacity

to synthesize androgens. Androgen biosynthesis by the theca

internal compartment is enhanced by exposure to LH, while FSH

may be important in sensitization of granulosa cells to

LH. Steroids, such as P4, DHT and E2 may also potentiate

this FSH effect.

One apparent consequence of granulosa-binding of LH is

activation of the aromatase enzyme complex. Mechanisms

involved in the selection of follicles which become

LH-responsive probably depend upon factors other tnan FSH

alone, perhaps intraovarian factors (Alexander et al., 1978;

Darga and Reichert, 1978; Schomberg, 1979; Hsueh et al.,

1983; Hansel and Convey, 1983), since granulosa cells in a

majority of bovine follicles appear to bind FSH, yet in only

one to three large follicles do granulosa cells develop the

ability to bind LH and acquire aromatase activity (Merz et

al., 1981; Bartol et al., 1981; McNatty et al., 1984).

As mentioned, steroidogenic capacity of ovarian folli-

cular compartments depend upon their ability to respond to

gonadotropins. Additionally, periodicity and magnitude of

gonadotropin release from the anterior pituitary play

important roles in ovarian function. Gonadotropin patterns

are, in turn, regulated by gonadal steroids via long loop

feedback mechanisms. In ovariectomized sheep (Goodman and

Karsch, 1980), frequency of tonic LH episodes is decreased

by P4 with no apparent effect on pulse amplitude, while E2

reduces LH pulse amplitude and does not influence frequency

of LH release. Similar observations were made by Rahe et

al. (1980) following characterization of LH episodes during

mid-luteal and follicular phases of the estrous cycle in

intact cows. High amplitude, low frequency LH episodes

which occur during P4-dominated phases of the cycle are

thought to result from P4 negative feedback on the central

nervous system's (CNS) oscillator regulating episodic

secretion of gonadotropin releasing hormone (GnRH) (Knobil,

1980) and pituitary sensitivity to GnRH (Padmanabhan et al.,

1982). During estrogen-dominated phases, prior to -he

preovulatory surge of gonadotropins, E2 initially decreases

pituitary sensitivity to GnRH resulting in less LH release

per GnRH pulse (Kesner et al., 1981; Kesner and Convey,

1982). However, by approximately 8 hours before the

gonadotropin surge, when highest preovulatory plasma

concentrations of E2 occur (Walters and Schallenberger,

1984), pituitary sensitivity to GnRH reaches its maximum

(Kesner et al., 1981; Kesner and Convey, 1982; Padmanabhan

et al., 1982). Yet during this period, amplitude (but not

frequency) of gonadotropin episodes remains low (Walters and

Schallenberger, 1984). Walters and Schallenberger (1984)

suggested that a negative feedback effect of the elevated E2

on the hypothalamus may explain these endocrine patterns

during the preovulatory period.

Regulation of FSH secretion by gonadal steroids appears

to be more subtle than observed for LH. Concentrations of

FSH are considerably higher than LH during all phases of the

estrous cycle in cattle (Schallenberger et al., 1984;

Walters and Schallenberger, 1984; Walters et al., 1984).

Interpulse frequencies for FSH range from approximately 25

minutes during the preovulatory phase to 50-50 minutes

during the mid-luteal phase of the estrous cycle (Walters

and Schallenberger, 1984; Walters et al., 1984) with pulse

amplitudes declining during the preovulatory period. Thus,

P4 appears to exert only slight reductions in FSH pulse
frequency in cattle. Evidence suggests that E2 exerts an

inhibitory effect on FSH release via the pituitary just

prior to the preovulatory gonadotropin surge (Kesner and

Convey, 1982; Walters and Schallenberger, 1984).

Additionally, exogenous E2 administered to ovariectomized

heifers reduced FSH concentrations to precastration levels

(Kesner and Convey, 1982). Follicular production of inhibin

may also regulate plasma FSH concentrations during periods

of follicular growth (De Jong, 1979; De Paolo et al., 1979;

Henderson and Franchimont, 1982; Padmanabhan et al., 1984),

possibly through decreasing pituitary release of FSH.

Low basal concentrations of P4 and peak follicular

production of E2 potentiate the onset of behavioral estrus

and trigger the preovulatory surge of LH and FSH from the

anterior pituitary (Schams et al., 1977). Although estrous

behavior in cattle may be induced by estrogen treatment

alone (Melampy et al., 1957), a period of P4 pre-exposure

enhances the sensitivity of brain centers responsible for

estrous behavior to E2 (Melampy et al., 1958). Therefore,

P4 priming, during the luteal phase of the cycle, may be an

important aspect of estrous expression (see also: McEwen et

al., 1982; Pfaff and McEwen, 1985). During physiological

states in which P4 priming is absent, as occurs in heifers

approaching puberty (Gonzalez-Padilla et al., 1975; Schams

et al., 1981) or in cows following postpartum anestrous

(Schams et al., 1978; Peters, 1984), behavioral estrus does

not accompany the first gonadotropin surge and ovulation.

Additionally, duration and magnitude of P4 exposure may

influence subsequent luteal function by modulating preovula-

tory follicle development directly as was suggested by the

data of Snook et al. (1969) in heifers treated with bovine

LH antisera and McLeod and Haresign (1984) in the seasonally

anestrous ewe. Similarly, ovulations occurring without the

benefit of P4 priming in pubertal and postpartum cattle

often result in short-lived, less functional CL (Gonzalez-

Padilla et al., 1975; Schams et al., 1978; 1981; Peters,


Elevated P4 concentrations completely eliminate E2

induction of estrous behavior and the gonadotropin surge

(Hobson and Hansel, 1972; Short et al., 1979; Roche and

Ireland, 1981; Ireland and Roche, 1982; Padmanabhan et al.,

1982). With demise of the CL during the preovulatory

period, plasma P4 concentrations fall to below 1 ng/ml thus

removing the negative feedback on GnRH and gonadotropin

release rates. Follicular production of E2 provides

increasingly higher plasma E2 concentrations, the magnitude

of which is proportional to the size of the preovulatory

gonadotropin surge (Walters and Schallenberger, 1984).

Approximately 1 hour prior to the preovulatory surge of

gonadotropins, Walters and Schallenberger (1984) observed a

consistent decline in E2 concentrations and suggested that

this event removed E2 negative feedback on the hypothalamus

and increased GnRH pulse amplitude which triggered trne

gonadotropin surge from the maximally responsive pituitary

(Convey, 1973; Zolman et al., 1973; Kesner et al., 1981;

Kesner and Convey, 1982). The surge of LH and FSH occur

concomitantly near the onset of an 18 to 20 hour behavioral

estrus (Trimberger and Hansel, 1955; Schams et al., 1977)

and lasts 8 to 10 hours (Rahe et al., 1980; Kesner et al.,

1981; Kesner and Convey, 1982; Walters and Schallenberger,

1984). The gonadotropin surge is terminated due to the

refractoriness of the pituitary to GnRH (Kesner et al.,

1981; Kesner and Convey, 1982).

During and immediately following the preovulatory

gonadotropin surge, follicular (granulosa cell) production

of E2 and circulating levels of E2 decline rapidly as

follicular luteinization occurs (Chenault et al., 1975;

Ireland and Roche, 1982). Plasma concentrations of E2, P4

and LH remain low throughout a majority of the metestrous

phase. Pulses of LH are absent for 6 to 12 hours post-

gonadotropin surge. Similarly, FSH pulse amplitudes are

low; however, pulse frequencies remain at a rate comparable

to that observed during the preovulatory gonadotropin surge

(Walters and Schallenberger, 1984). Prior to ovulation and

4 to 12 hours following the gonadotropin surge, FSH

concentrations become elevated as a result of an increase in

FSH pulse amplitude (Kazmer et al., 1981; Hansel and Convey,

1983; Walters and Schallenberger, 1984). The second rise in
FSH is of lower magnitude than the FSH surge and may bz a

result of reduced follicular inhibin production during the

ovulatory process (De Jong, 1979; De Paolo et al., 1979;

Henderson and Franchimont, 1981; Padmanabhan et al.,

1984). The function of this secondary rise in FSH is

currently unknown, however, FSH may play a role in oocyte

maturation (Plachot and Mandelbaum, 1978) or in the

recruitment of preantral follicles (Hansel and Convey,


Ovulation occurs on day 1 of the estrous cycle,

approximately 24 to 30 hours after the preovulatory surge of

LH and FSH (Chenault et al., 1975; Schams et al., 1977;

Hansel and Convey, 1983). Following ovulation, development

of CL function is reflected by the gradual increase in CL

weight (Shemesh et al., 1976; Bartol et al., 1981) and P4

content (Shemesh et al., 1976), plasma P4 concentrations

(Chenault et al., 1975; Peterson et al., 1976; Schams et

al., 1977; Wise et al., 1981; Hansel and Convey, 1983) and

P4 production (Wise et al., 1981), and ovarian blood flow

(Ford and Chenault, 1981; Wise et al., 1981). Follicular

growth and atresia occur continuously throughout the estrous

cycle (Rajakoski, 1960; Marion et al., 1968; Ireland and

Roche, 1983) with moderate rises in plasma E2 concentrations

occurring responsively (Hansel et al., 1973; Hansel and

Convey, 1983; Ireland and Roche, 1983).

During the early luteal phase, both LH and FSH secre-

tion patterns may be characterized as high frequency-low

amplitude (Rahe et al., 1980; Walters et al., 1934). ihe

majority of LH and FSH pulses occur simultaneously during

this period. Similarly, pulsatile episodes of E2 and P4 are

high frequency. High amplitude pulses of E2 follow LH

pulses in 90 to 96% of the cases observed, while occurrence

of low amplitude P4 episodes were more similar to FSH

pulsatile patterns (Walters et al., 1984). These

gonadotropin-steroid pulsatile relationships become more

obvious during the mid-luteal phase. By day 10 to 12 of the

estrous cycle, frequency of LH episodes are significantly

decreased (Rahe et al., 1980; Walters et al., 1984). Rahe

and coworkers (1980) also described increased LH pulse

amplitude during the mid-luteal phase, however, elevations

in LH pulse amplitude were not evident in the data set of

Walters et al. (1984). Pulse amplitude of FSH was similar

during the early and mid-luteal phases, although FSH pulse

frequency decreased during the mid-luteal phase.

Nevertheless, 41% more FSH than LH episodes were observed

during the mid-luteal phase. Low amplitude E2 episodes were

consistently preceded by pulses of LH and 92 to 100% of

concurrent FSH and LH pulses and 97% of separate FSH pulses

were followed by high amplitude episodes of P4. In no case

were separate FSH pulses associated with episodes of E2

(Walters et al., 1984).

The dynamics of endocrine changes during the luteal

phase support previously described roles for FSH and Lr in

follicular steroid production. Additionally, it would

appear that FSH, in conjunction with LH, regulates P4

secretion patterns. The major site of P4 production is the

CL (Shemesh et al., 1976). The bovine CL possesses both FSH

(Mann and Niswender, 1983) and LH (Rao et al., 1979, 1983)

receptors and responds to these gonadotropins, in vitro,

with elevated P4 production (Romanoff, 1966; Hixon and

Hansel, 1979; Milvae et al., 1983).

In cyclic cattle, lifespan of the CL is regulated, to a

large extent, by interactions between the ovary and

uterus. During late diestrus-early proestrus phases of the

cycle, estrogens produced by the developing, large antral

follicle(s) are thought to initiate the process of luteal

regression. Exogenous estrogens are known to shorten the

lifespan of CL in cattle (Greenstein et al., 1958; Wiltbank,

1966; Eley et al., 1979). Luteolytic activity of estrogen

is thought to be mediated through stimulation of uterine

prostaglandin (PG)-F2. synthesis and release (Thatcher et

al., 1984b). The luteolytic activity of exogenous PGF2. in

cattle has been documented by several groups (Hansel et al.,

1973; Hafs et al., 1974; Lauderdale, 1974; Thatcher and
Chenault, 1976). Furthermore, episodic pulses of uterine

venous PGF2a (Nancarrow et al., 1973), and peripheral

measurements of its primary metabolite,

15-keto-13,14-dihydro-PGF2a (PGFM; Granstrom and Kindahl,

1982), are always associated with spontaneous luteal

regression in cattle (Peterson et al., 1975; Kindahl et ai.,

1976; Betteridge et al., 1984). Uterine PGF2a production

requires a period of P4-priming which increases the tissue's

potential to synthesize PGF2, while suppressing copious

secretion of the luteolysin (Hansel et al., 1973; Horton and

Poyser, 1976; Rothchild, 1981).

An indepth discussion of CL characteristics, function

and mechanisms of regression will be reserved for a

subsequent section of this chapter.

The Corpus Luteum Life Cycle

Origin and Development of Corpus Luteum Cells

Cells which make up the corpus luteum (CL) are derived

from the preovulatory follicle. Two distinct luteal cell

types have been described in CL of cattle (Donaldson and

Hansel, 1965b; Ursely and Leymarie, 1979a,b; Koos and

Hansel, 1981; Weber et al., 1984; Alila and Hansel, 1984),

sheep (O'Shea et al., 1979, 1980; Fitz et al., 1982; Rogers

et al., 1983, 1984), swine (Lemon and Loir, 1977; Lemon and

Mauleon, 1982), rat (Kenny and Robinson, 1982), rabbit (Yuh

et al., 1982) and human (Gaukroger et al., 1979). These

cell types are described generally as small and large luteal

cells. Current dogma suggests that small luteal cells are

derived from the theca internal and large luteal cells from

the granulosa of the preovulatory follicle.

In the ewe, large luteal cells of the mature CL are

approximately six-times larger than small luteal cells and

occupy 25% or more of the CL volume (Niswender et al., 1976;

Rodgers et al., 1984) as compared to approximately 18% by

small luteal cells (Rodgers et al., 1984). Small luteal

cells outnumber large luteal cells by nearly five to one.

By far, the most numerous cell type within the mature ovine

CL is endothelium (Rodgers et al., 1984), which attests to

the high vascular density of this tissue.

Cellular mitotic activity in the CL is thought

generally to be at a minimum (Donaldson and Hansel, 1965;

Rodgers et al., 1984). Histological examination and DNA

determinations of the bovine preovulatory follicle (6 hours

following the onset of estrus) and CL throughout the estrous

cycle (Donaldson and Hansel, 1965b) indicated that

granulosa-luteal cell mitotic activity was reduced following

the preovulatory surge of LH and ceased altogether by

approximately day 4. The theca-derived luteal cells,

however, mitosed freely until approximately day 7.

Neovascularization continues in CL until a maximum metabolic

potential is reached around days 10 to 12. In vivo exposure

of developing bovine CL to hCG or LH (Donaldson et al.,

1965a) stimulated mitotic activity in small luteal cells,

vascular endothelium and stromal cells, whereas mitotic

figures were rarely observed in large luteal cells

(Donaldson and Hansel, 1965). Thus, during CL development,

LH appears to limit granulosa cell division and enhance

theca-luteal cell mitosis (Donaldson and Hansel, 1965;

Friedrich et al., 1975; McNatty and Sawers, 1975).

In addition, estimates of granulosa cell numbers in

mature preovulatory follicles compare favorably with numbers

of granulosa-luteal cells in the CL (McNatty, 1979; McNatty

et al., 1984a; Rogers et al., 1984), suggesting that large

luteal cells are derived solely from follicular granulosa

cells. Furthermore, increases in follicular size are

correlated positively with follicular cell numbers (McNatty,

1979; McNatty et al., 1982; Ireland and Roche, 1985). Thus,

it may be assumed logically that events which regulate

viability, proliferation and biosynthetic potential of

preovulatory follicular cells markedly influence the

functionality of the CL subsequently formed post-ovulation

(see: McNatty, 1979; Richards, 1979, 1980; di Zerega and

Hodgen, 1981; Murdoch et al., 1983).

Observations in the cow (Donaldson and Hansel, 1965;

Alila and Hansel, 1984) and ewe (Fitz et al., 1981) support

the hypothesis that CL growth occurs initially via

hypertrophy and hyperplasia by both theca- and granulosa-

derived luteal cells. However, these authors suggest that

as the CL matures, theca-derived lutein cells enlarge to

form large luteal cells. It has further been suggested that

the large luteal cells (granulosa- or theca-derived), whicn

possess receptors for PGF2., play a key role in CL

regression (Fitz et al., 1982; Alila and Hansel, 1984; Hoyer

et al., 1984).

Factors Regulating CL Steroidogenesis

The formation of luteal tissue is begun prior to

ovulation as the preovulatory rise and surge release of LH

promote several biochemical (Savard, 1973) and cytological

(Donaldson and Hansel, 1965b; Blanchette, 1966a,b;

Priedkalns and Weber, 1968; Enders, 1973; Van Blerkom and

Motta, 1978; 1979) changes in both theca and granulosa cells

of the preovulatory follicle.

Unlike the follicle, cells of the bovine CL lack the

endoplasmic reticulum-associated 17-hydroxylase and 19-

hydroxylase-aromatase enzyme systems (Savard and Telegdy,

1965; Savard, 1973). The obvious physiological

manifestations of these biochemical alterations are the

sudden loss of androgen and estrogen biosynthetic capacity

by luteal cells (Savard and Telegdy, 1965; Ireland and

Roche, 1982; Walters and Schallenberger, 1984). Thus,

bovine CL exhibit a very abbreviated steroidogenic pathway

from which P4 is the primary end-product, with lesser

amounts of 208-hydroxy-4-pregnen-3-one (20-P4),

pregnenolone (P5), and 5a-reduced progestins (5a-P) being

generated (Wise and Fields, 1978; Albert et al., 1982).

Corpus luteum development and steroidogenic capacity may be

grossly assessed by monitoring plasma P4 concentrations.

Bovine CL mass increases rapidly from day 3 (< 1 gramn) to

day 7 (4 gram) and is maintained at a weight of 5 to 6 grlans

from days 10 through 17. Likewise, plasma concentrations of

P4 are low (< 1 ng/ml) on day 3, but rise rapidly to levels

of approximately 5 ng/ml by day 7 and are maintained between

6 and 10 ng/ml from days 10 through 17 (Donaldson and

Hansel, 1965b; Erb et al., 1971; Rao et al., 1979; Milvae

and Hansel, 1983a). During regression of the CL, a loss in

function (reduced plasma P4 concentrations) consistently

precedes any decline in CL weight.

Luteal steroidogenic potential is regulated by several

factors. Luteinizing hormone is considered the primary

luteotropin in cattle. During the past 20 years, a

tremendous number of studies have been performed which

support this concept. Administration of exogenous LH

(Donaldson and Hansel, 1965a), hCG (Wiltbank et al., 1961),

and GnRH-agonists, to induce elevated endogenous LH

concentrations (Milvae et al., 1984), increased plasma P4

concentrations and prolonged CL function in cattle. Several

in vitro studies demonstrated increased P4 production by

dispersed luteal cells or luteal slices when incubated with

LH or hCG (Hansel and Seifart, 1967; Hansel et al., 1973;

Hixson and Hansel, 1979; Milvae and Hansel, 1983a; Tan and

Biggs, 1984; Milvae et al., 1985). Antiserum to LH

prevented this luteotropic effect of LH on luteal slices in

vitro (Hansel and Seifart, 1967) and induced a decline in CL

weight and P4 content of CL in intact heifers (Snook et al.,

1969). In hysterectomized heifers, LH treatment increased

P4 content of CL (Brunner et al., 1969) while hysterectomy

plus hypophysectomy or hypophysectomy alone caused a decline

in P4 content of CL, lower plasma P4 concentrations, and a

reduction in CL weight in cattle (Henricks et al., 1969).

Lastly, oxytocin and PGF2a induced CL regression was

circumvented by administration of LH or GnRH to cyclic

cattle (Hansel and Seifart, 1967; Thatcher and Chenault,

1976). Therefore, maximal function of the bovine CL is
contingent upon LH stimulation.

Direct binding studies, using radiolabelled ligands,

revealed the presence of specific, high affinity-low

capacity receptors for LH/hCG in bovine luteal cell

membranes (Rao et al., 1979, 1983) and intracellular

organelles (Rao et al., 1983). Furthermore, luteal

concentrations of LH receptor sites are correlated with P4

secretion during the estrous cycle. Other ligands (to be

discussed later) which bind specifically to luteal tissue

include FSH (Mann and Niswender, 1983), PGF2a (Kimball and

Lauderdale, 1975; Rao et al.,1979; Bartol et al., 1981; Fitz

et al., 1982), PGE1 and PGE2 (Kimball and Lauderdale, 1975;

Fitz et al., 1982), and E2 (Glass et al., 1984).

Recent application of cell separation techniques have

enabled researchers to better characterize and evaluate the

function of specific luteal cell populations with regard to

their roles in P4 production and luteolysis. As described

previously, small luteal cell numbers exceed large luteai

cells by approximately five-fold although large luteal cells

comprise more of the CL volume. In cattle (Ursely and

Leymarie, 1979a,b; Koos and Hansel, 1981; Weber et al.,

1984), sheep (Fitz et al., 1982; Rodgers and O'Shea, 1982;

Rodgers et al., 1983), and swine (Lemon and Loir, 1977;

Lemon and Mauleon, 1982) incubations enriched in large

luteal cells produce consistently more P4 than equivalent

numbers of small luteal cells. However, only small luteal

cells were capable of responding to LH/hCG stimulation, as

determined by a significant increase in P4 production over

nonstimulated cells. This observation suggested, as was

demonstrated by Fitz and coworkers (1982), that the

luteotropic effect of LH/hCG was due to the interaction of

LH/hCG with specific receptor sites on small luteal cells.

Small cell steroidogenesis was also stimulated by cyclic AMP

and PGE2. The deficiency of LH/hCG receptors in the large

luteal cell population and a constant secretion of P4 in the

presence or absence of secretagogues (Fitz et al., 1982)

indicate that peak steroidogenic activity in large luteal

cells does not require receptor-mediated cyclic AMP

generation as it does in small luteal cells (Hoyer et al.,

1984). Instead, maximum potential for P4 production by

large lutein cells appears to be dependent upon precursor

(primarily cholesterol) availability.

Lemon and Mauleon (1982) demonstrated an interaction

between small and large luteal cell types in the porcine CL

with regard to P4 production. These authors utilized a

superfusion system in which medium was pumped from one

chamber, containing small luteal cells, through a second

chamber, containing large luteal cells, and vise-versa.

Progesterone production by the small cell to large cell 'in

series' superfusion exceeded (20%) the sum total of P4

produced by each cell type alone. No such increase in P4

production was detected when the superfusion was conducted

with cell populations in a reversed order. Thus, products

derived from small luteal cells stimulated P4 production by

large luteal cells. When steroid precursors were added to

superfusions of either small or large luteal cells, P5 was

metabolized to P4 equally well by both cell types. In

contrast, only large luteal cells responded to exogenous

cholesterol with increased P4 production compared to control

cells superfused without cholesterol. Metabolism of

cholesterol by large cells was not stimulated by exogenous

LH. Pulses of exogenous LH to small luteal cell

superfusions stimulated P4 production responsively.

However, P4 production in response to LH was not affected by

the addition of cholesterol to small luteal cells. Lemon

and Mauleon (1982) suggested that local transfer of

cholesterol from small to large luteal cells was responsible

for enhanced P4 production by large luteal cells.

Furthermore, LH-stimulated P4 production and cholesterol

mobilization in small luteal cells would provide additional

substrate to large luteal cells. The end result would be a

coordinated increase in P4 production by both cell types.

The following discussion will summarize some of the

important processes involved in cellular cholesterol

regulation. Pertinent references may be found in reviews by

Savard, 1973; Brown et al., 1981; Kaplan, 1981; Savion et

al., 1982; Strauss et al., 1982.

Steroidogenic cells, and extrahepatic cells in general,

synthesize very little cholesterol. Instead, the bulk of

cholesterol required for steroidogenesis and membrane

synthesis is derived from plasma and stored as cholesterol

esters within the cell.

Approximately 70% of the total plasma cholesterol pool

resides within low density lipoprotein (LDL) particles in an

esterified form. These LDL particles are approximately 220
A in diameter and contain a core of nearly pure cholesterol

ester surrounded by two copies of the 250,000 Dalton

glycoprotein, apoprotein B. Cellular uptake of cholesterol

involves binding of LDL to specific, high affinity receptors

which become localized on specialized regions of the plasma

membrane called coated pits (Pastan and Willingham,

1981a,b). Receptor recognition of specific regions of

apoprotein B ensures the selective uptake of LDL. Receptor

binding affinity to LDL is Ca+2-dependent, the importance of

which will become obvious. Receptor-LDL complexes are

rapidly internalized enmass via receptor mediated

endocytosis (Pastan and Willingham, 1981b). Coated pits,

containing numerous receptor-LDL complexes, invaginate and

pinch off to form endocytotic coated vesicles which

ultimately fuse with lysosomal membranes in the cytoplasm.

Receptor-ligand complexes dissociate allowing the LDL

receptors to be recycled for further use. Dissociation is

thought to occur as a result of low Ca+2 concentrations

within the endocytotic vesicle and lysosome. Hydrolysis of

LDL within the lysosome frees cholesterol for use or storage

within the cell.

Intracellular cholesterol concentrations are highly

regulated by the cell. As intracellular cholesterol

concentrations rise following hydrolysis of LDL in the

lysosomes, cellular cholesterol biosynthesis is suppressed

via a decrease in the activity of a pivitol enzyme in the

cholesterol biosynthetic pathway, HMG-CoA reductase.

Secondly, acyl-CoA:cholesterol acyltransferase activity is

enhanced, thus catalyzing the re-esterification of

cholesterol for storage. Most importantly, LDL receptor

biosynthesis is suppressed, preventing further uptake of

cholesterol ester-rich LDL from the plasma.

These data suggest that small and large luteal cells

also adjust to intracellular cholesterol concentrations in

accordance with changing cellular requirements. Thus, large

lutein cells, which synthesize approximately 80% of the

total P4 in unstimulated CL (Hoyer et al., 1984), would be

expected to maintain a higher cholesterol requirement than

small luteal cells. Furthermore, it may be suggested that

large luteal cells in the pig (Lemon and Mauleon, 1982) were

in a cholesterol-deficient state, with regard to their

maximum steroidogenic capacity, since P4 production was

elevated following exposure to exogenous cholesterol.

Conversely, small luteal cells, which utilize cholesterol

less rapidly, may possess larger stores of intracellular

cholesterol and would not be expected to exhibit elevations

in P4 production in response to exogenous cholesterol (Lemon

and Mauleon, 1982). Demand for cholesterol by LH-sensitive

small luteal cells increases following stimulation with LH

or cyclic AMP. However, intracellular stores are apparently

sufficient to accommodate this demand (Lemon and Mauleon,


Luteinizing hormone and cyclic AMP increase cholesterol

esterase and cholesterol side chain cleavage (SCC) enzyme

activities, lipoprotein uptake and P4 production (Savard,

1973; Bisgaier et al., 1979; Caffrey et al., 1979a; Strauss

et al., 1982). However, specific mechanisms of LH action on

luteal cells are not defined clearly.

Intracellular Ca+2 and cyclic AMP concentrations, and

cyclic AMP-dependent protein kinase activation are known to

play essential roles in transduction of receptor-mediated

signals (see: Rasmussen, 1981). General concepts regarding

the mechanism of action of peptide hormones on target cells

have centered around intracellular accumulation of Ca+2 and

cyclic AMP, and activation of cyclic AMP-dependent protein

kinase(s) (Catt and Dufau, 1976; Marsh, 1976). This enzyme

mediates the effects of cyclic AMP via phosphorylation of

specific substrates which in turn regulate cell function.

Cyclic AMP-dependent protein kinase(s) has been localized in

the cytoplasm and numerous cellular oganelles (plasma

membrane, mitochondria, ribosomes, nucleus and endoplasmic

reticulum) suggesting that diversity of cellular responses

to cyclic AMP may depend on availability of specific

substrates within compartmentalized regions of the cell

(Langan, 1973).

Recent evaluation of membrane structure and function

has expanded our understanding of mechanisms involved in

receptor-mediated information flow into the cell. Over the

past decade, it has become apparent that the organization of

phospholipids and proteins within membranes is more complex

than described by the "fluid mosaic" membrane model of

Singer and Nicolson (1972). Several components of

biological membranes (i.e., receptors, enzymes,

phospholipids) are asymmetrically distributed between the

inner and outer bilayer (Rothman and Lanard, 1977; Lodisa

and Rothman, 1979). Furthermore, membrane components may

aggregate preferentially in specific domains of the oilayer

structure (Chapman et al., 1979; Karnovsky et al., 1982).

Assembly of phospholipid and protein domains facilitate

events such as intercellular communications (Loewenstein,

1970; Garfield et al., 1979; Hertzberg et al., 1981),

endocytosis (Pearse, 1980; Schlessinger, 1980; Pastan and

Willingham, 1981a,b), ion movement and enzymatic activity

(Savard, 1973; Strauss et al., 1982; Lentz et al., 1983).

Membrane components are in a state of constant

reorganization, synthesis and breakdown in response to

external stimuli. Current data suggest that, in addition to

Ca+2 and cyclic AMP messenger systems, rapid and transient

alterations in membrane phospholipid composition,

biosynthesis.and metabolism are essential in stimulus-

response coupling in cells (Hirata and Axelrod, 1980; Davis

et al., 1981; Crews, 1982; Farese, 1983; Milvae et al.,

1983). Phospholipid methylation (Hirata and Axelrod, 1980;

Crews, 1982; Milvae et al., 1983) and phosphotidylinositol

(PI) metabolism (Davis et al., 1981; Crews, 1982; Farese,

1985; Nishizuka, 1984b) are two mechanisms by which membrane

phospholipids are transiently altered in response to

receptor-ligand binding.

Methylation of phospholipids is dependent upon two

phospholipid methyltransferase (PMT) enzymes which are

distributed on the inner (PMT-1) and outer (PMT-2) aspects

of the cell membrane bilayer. Likewise, the phospholipid

substrates for PMT-1 (phosphatidylethanolamine, PEF) and PAT-

2 (phosphatidyl-N-monomethylethanolamine, PME) are

asymmetrically distributed within the membrane. Formation

of PME requires the PMT-1 catalyzed transfer of a single

methyl group to PE. Conversion of PME to PC requires two

successive methyl transfers by PMT-2. Successive

methylations of PE by this system result in translocation of

methylated phospholipids (PME and phosphatidylcholine, PC)

toward the outer aspects of the membrane bilayer.

As a result of these chemical alterations in

phospholipid composition, the membrane becomes less

viscous. Enhanced membrane fluidity is thought to result

from the accumulation of deeply embedded PME within the

phospholipid bilayer. With the increase in membrane

fluidity, lateral mobility of membrane proteins is

potentiated thereby enhancing the coupling of surface

receptors with adenylate cyclase and cyclic AMP formation

(Rimon et al., 1978; Axelrod, 1985). Additionally,

formation and deposition of PC in the outer phospholipid

layer of the membrane promotes the unmasking of cryptic

membrane receptors. Thus, receptor availability is enhanced

(Hirata and Axelrod, 1980; Crews, 1982).

Milvae et al. (1983) demonstrated the importance of

phospholipid methylation in expression of LH-stimulated P4

production by dispersed bovine luteal cells. Incorporation

of [3H]-methyl-groups into PME, PC and PI was stimulated by

LH, in vitro. An endogenous methyl-donor, S-aaenosyl-L-

methionine (SAM), enhanced LH-induced P4 production while

methylation inhibitors, 3-deazaadenosine (DZA) and S-

adenosyl-L-homocysteine (SAH), prevented the luteotropic

action of LH on dispersed luteal cells. Cyclic AMP effects

on P4 production were not influenced by DZA or SAH,

supporting the view that inhibition of phospholipid

methylation prevents coupling of ligand-receptor complexes

to adenylate cyclase, cyclase activation and cyclic AMP

generation in the bovine CL.

The degree of membrane phospholipid methylation also

appears to influence cellular Ca+2 fluxes, possibly via

regulation of calcium ion channels and Ca+2-dependent

adenosinetriphosphatase (ATPase) activity (Hirata and

Axelrod, 1980; Crews, 1982). Regulation of cell function by

calcium ion concentrations and intracellular Ca+2 binding

proteins, such as calmodulin, have been reviewed by several

authors (Cheung, 1979; Means, 1981; Rasmussen, 1981).

Receptor mediated Ca+2 influx occurs secondarily to

methylation of phospholipids and, like cyclic AMP

generation, is prevented by methylation inhibitors.

Activation of Ca+2-dependent phospholipase A2 parallels the

influx of Ca+2, as demonstrated by elevated free arachidonic

acid and lysolecithin (lysophosphatidylcholine)

concentrations following receptor activation (Crews,

1982). These observations suggest that hydrolysis of newly

synthesized methylated phospholipids by Ca+2-dependent

phospholipases serve to regulate methylated phospholipid-

induced membrane effects, viz., increased fluidity, Ca+2

influx, receptor availability. It is interesting to note

that Ca+2 is not required for membrane phospholipid

methylation or cyclic AMP generation in several cell

systems, although many of the physiological responses to

cyclic AMP are Ca+2-dependent (Rasmussen, 1970).

In addition to phospholipid methylation, considerable

interest has been directed toward receptor-mediated

alterations of the phosphatidate-inositide cycle (Crews,

1982; Farese, 1983; Nishizuka, 1984b). There are two known

mechanisms by which the phosphatidate-inositide cycle may be

affected by receptor activation. The first involves the

hydrolysis of phosphatidylinositides to diacylglyceride (DG)

and inositol phosphates via phospholipase C catalyzed

removal of the phosphorylated inositide head group. This

mechanism is thought to regulate plasma membrane Ca+2 fluxes

and intracellular Ca+2 distribution. The phosphatidylino-

sitides are localized primarily in the inner phospholipid

layer of the plasma membrane bilayer, and membranes of

endoplasmic reticulum and mitochondria as mono-, di- and

triphosphorylated inositol phospholipids (PI, PIP and PIP2,

respectively). These phospholipids bind Ca+2 avidly and


provide a releasable Ca+2 pool upon hydrolysis. Recent

evidence, reviewed by Nishizuka (1984b), suggests that PIP2,

rather than PI or PIP, is rapidly degraded following

receptor activation. Once hydrolysis of PIP2 is initiated

in the plasma membrane the process is thought to self-

propagate as a result of elevated free Ca+2 concentrations,

increased Ca+2-dependent phospholipase activity, and

extended hydrolysis of phosphatidylinositides within the

plasma membrane and intracellular organelles. One

metabolite of PIP2 hydrolysis, inositol triphosphate, may

serve as a mediator for Ca+2 release from intracellular

stores (Nishizuka, 1984b). A second metabolite, DG, arises

following phospholipase C catalyzed hydrolysis of PIP2, PIP

or PI. Diacylglycerol may serve as a precursor for the

regeneration of phosphatidylinositides as well as PE and

PC. The phosphatidylinositides and DG contain primarily

arachidonate at fatty acid position -2. Thus, hydrolysis by

Ca+2-dependent phospholipase A2 provides free arachidonic

acid for use in prostaglandin and hydroxyperoxylipid

biosynthetic pathways (Ramwell et al., 1977), and various

lysolecithins, many of which are required for membrane

fusion reactions during uptake and secretion of products

(Crews, 1982). Additionally, DG activates the Ca+2- and

phospholipid-dependent protein kinase C (Nishizuka,

1984a,b). Protein kinase C activation has been implicated

in the elicitation of cell proliferation (Nishizuka, 1984a)

and inhibition of gonadal steroidogenesis (Welsh et al.,

1984). It is interesting to note that protein kinase C

serves as a receptor for tumor promoting phorbol esters

(Nishizuka, 1984a,b) which are thought to irreversibly

activate the enzyme. Phorbol esters possess a similar

chemical structure to DG, the endogenous activator of

protein kinase C. Phorbol esters have been shown to innibit

gonadotropin-induced steroidogenesis, in vitro (Welsh et

al., 1984), supporting a similar role for endogenous DG in

the regulation of steroid production. Therefore,

steroidogenesis may be antagonized as a consequence of

phosphatidylinositide hydrolysis.

Conversely, de novo phosphatidate-inositide

biosynthesis may be important in the mediation of

gonadotropin effects on steroidogenesis (Davis et al., 1981;

Crews, 1982; Farese, 1984). Several studies have

demonstrated that LH and ACTH stimulate biosynthesis of

phosphatidylinositides and phosphatidic acid (PA) in gonadal

and adrenal cortex cells, respectively (Davis et al., 1981;

Strauss et al., 1982; Farese, 1984). Data reviewed by

Farese (1984) indicated that identical responses were

triggered by cyclic AMP, suggesting that phospholipid

biosynthesis occurred after cyclic AMP generation. Both

hormone- and cyclic AMP-induced phosphatidate-inositide

biosynthesis, and steroidogenesis are blocked by protein

synthesis inhibitors, puromycin and cycloheximide.

Apparently, the two enzymes responsible for PA biosynthesis,

viz., glycerol-3-phosphate acyltransferase and diglycerile

kinase, are substrate activated and require protein

synthesis. Hormone activation of these enzyme systems is

indirect and thought to occur by increasing glycerol

phosphate, glucose, fatty acid and diglyceride availability

through stimulation of glycogenolysis, glycolysis, glucose

uptake and lipolysis. Phosphatidylinositol kinase catalyzes

the conversion of PA plus ATP to PI, and subsequent

phosphorylations to PIP and PIP2.

Phosphatidate-inositol biosynthesis parallels

steroidogenic activation in adrenal and luteal tissues

following exposure to ACTH and LH, respectively.

Furthermore, addition of certain polyphosphorylated

phospholipids, viz., the phosphatidylinositides and

cardiolipin, to adrenal cortex and luteal cell cultures

stimulated steroid biosynthesis in a dose dependent manner

(Strauss et al., 1982; Farese, 1984). Effects of these

"steroidogenic" phospholipids are not prevented by

cycloheximide or puromycin.

The site of phospholipid action in both adrenal cortex

and luteal cells appears to be the mitochondria. Specifi-

cally, steroidogenic phospholipids stimulate cholesterol

side chain cleavage (SCC) by intact cells and acetone

powders of luteal mitochondria (Strauss et al., 1982;

Farese, 1984), although the mechanism through which this is

accomplished remains unknown.

At present, it is known that trophic hormones, sucii as

LH and ACTH, stimulate steroidogenesis by influencing

several biochemical steps. Free intracellular cholesterol

concentrations are elevated via enhanced cellular uptake of

cholesterol from plasma and mobilization of cholesterol from

intracellular stores following cholesterol esterase

activation. Apparently, hormone induced elevations in free

intracellular cholesterol are not affected by cycloheximide,

but are blocked by cytochalasin B suggesting that microfila-

ments, but not protein synthetic events, are crucial in the

acquisition of substrate for steroid biosynthesis (Gemmel

and Stacy, 1977; Sawyer et al., 1979; Silavin et al., 1980;

Rasmussen, 1981; Gwynne and Condon, 1982). Microtubule

involvement in steroidogenesis is, however, less clear with

evidence reported both pro (Gemmel and Stacy, 1977; Sawyer

et al., 1979) and con (Gwynne and Condon, 1982).

In addition to elevating intracellular substrate

concentrations, LH and ACTH increase mitochondrial content

of cholesterol, presumably as a result of enhanced

cholesterol transfer from cytosol to mitochondria (see:

Rasmussen, 1981, for references). This event requires

protein synthesis and microfilament action. Hormone- and

cyclic AMP-induced biosynthesis of the "steroidogenic"

phospholipids are also dependent on protein synthetic


events, although their stimulatory effects on mitochondrial

SCC activity are not (Strauss et al., 1982; Farese, 1964).

Furthermore, mitochondrial SCC activity is substrate- and

Ca+2-dependent. Thus, it appears as though hormone-induced

steroidogenic activity involves primarily processes which

increase intracellular cholesterol content and transfer of

cholesterol from the cytosol to the SCC enzyme complex,

viz., sterol carrier proteins (see: Savard, 1975, for

references). Translocation of cholesterol across the

mitochondrial membranes may be regulated by the phospha-

tidylinositides and other "steroidogenic" phospholipids

(Strauss et al., 1982; Farese, 1984), the synthesis of which

is dependent on some, as yet undetermined, protein synthetic

event. Increased SCC activity may also involve structural

changes in the enzyme-phospholipid domain (Savard, 1973;

Strauss et al., 1982), and altered mitochondrial Ca+2

concentrations (Crews, 1982; Farese, 1984).

It is becoming more apparent that membrane phospholipid

dynamics, in addition to cyclic AMP and Ca+2 second

messenger effects, are of paramount importance in cellular

information flow and stimulus-response coupling. The reader

is referred to Rasmussen (1981) for further discussions on

this broad and rapidly developing area of cell physiology.

Following side chain cleavage of cholesterol, newly

synthesized P5 is transferred out of the mitochondria to the

endoplasmic reticulum. Here, the membrane bound

38-hydroxy-A5-steroid dehydrogenase/A5-A4 isomlerase enzyme

complex (HSD) catalyzes the rapid conversion of P5 to P ,

the primary steroid end product of the CL. These latter

steps in the biosynthesis of P4 do not appear to be

regulated by trophic hormones. However, luteal P4

production may be modified by local and systemic mechanisms

involving steroids, arachidonic acid metabolites, and

ovarian peptide hormones.

Intraovarian control of CL function

The dependency of CL P4 production on LH has been

extensively documented in the literature, as has the

importance of uterine-derived PGF2a in luteolytic events.

However, the level and duration of luteal P4 production may

also be determined by the local ovarian environment in which

the CL resides.

A large body of evidence supports a role for steroids

in the regulation of steroidogenic enzyme activities in the

ovary, testes, adrenal and placenta (see: Rothchild, 1981;

Gower and Cooke, 1983). Caffrey et al. (1979b) demonstrated

that the affinity of HSD for its product, P4, was

substantially greater than for P5 substrate and that P4 (10-

50 yg) inhibited HSD activity in ovine CL preparations.

Thus, suggesting that P4 may modify its own production via

substrate inhibition of HSD activity in the CL.

Testosterone and E2 (10 Ug) were also found to be potent

inhibitors of HSD activity, in vitro (Caffrey et al., 1979b;

Gower and Cooke, 1983), but not in vivo (Caffrey et al.,

1979b). Similarly, microgram concentrations of testosterone

and E2 were found to inhibit LH-stimulatea P4 production but

not cyclic AMP accumulation in dispersed luteal cells

collected from 2 to 6 months of gestation in cattle.

(Williams and Marsh, 1978). In vivo luteolytic interactions

were reported between E2 and PGF2. in hysterectomized ewes

(Gengenbach et al., 1977), and between E2 (0.5 ug) and

oxytocin (200 mIU) in dispersed luteal cells collected from

cyclic cattle (Tan and Biggs, 1984). However, Hixon and

Hansel (1979) were unable to demonstrate similar luteolytic

effects with subnanogram doses of E2 or E2 plus PGF2a in

dispersed bovine luteal cells from days 12 and 13 of the

cycle. Further support for a direct E2-CL interaction was

provided in a recent study (Glass et al., 1984) in which E2

receptors were characterized in large and small luteal cells

throughout the estrous cycle of the ewe. Concentrations of

E2 receptor rose gradually during the estrous cycle and were

localized primarily in large lutein cells. Additionally,

Sairam and Berman (1979) provided evidence that synthetic

estrogens prevented normal coupling of the gonadotropin-

receptor complex to its catalytic subunit, thus preventing

the initiation of intracellular gonadotropic-induced


Discrepancies with regard to the inhibitory role(s) of

E2 on CL P4 biosynthesis may be due to differences in dosage

of steroid employed, specie and status (cyclic or pregnant)

of the animal from which luteal tissue was collected, and in

vitro versus in vivo experimental conditions. Aitnough

doses of E2 which inhibit CL P4 biosynthesis seem high and

nonphysiological, they do compare favorably with

concentrations reported in follicular fluid of preovulatory

follicles (see: Fortune and Hansel, 1979). Therefore, the

CL may be exposed to high intraovarian concentrations of

various steroids, viz., testosterone and E2, as would be

expected during preovulatory follicle development in the

late diestrum. The potential inhibitory nature on luteal P4

production exhibited by high concentrations of these

steroids supports a functional role in the initiation of CL

regression. In this regard, it is noteworthy that P4 plasma

concentrations begin to fall prior to any detectable rise in

peripheral plasma E2 concentrations or initiation of uterine

PGF2a pulsatile activity (Chenault et al., 1975; Peterson et

al., 1975).

Beyond the luteotrophic effects initiated by LH, P4

production may be "protected" from steroid negative feedback

effects by several intraovarian processes.

For example, binding of steroids to proteins for short-

term storage, viz., steroid carrier proteins (Goddard et

al., 1980; Willcox, 1983) or long-term storage in

cytoplasmic granules (Gemmell et al., 1974; Gemmell and

Stacy, 1979; Sawyer et al., 1979) may reduce intracellular

concentrations of free steroid and reduce feedbacK

inhibition, in vivo. Furthermore, steroid sulfation or

conjugation to fatty acids, viz., lipoidal steroids

(Hochberg et al., 1979; Albert et al., 1980; Schatz and

Hochberg, 1981; Albert et al., 1982), within the CL would

accomplish a similar function. Lastly, ovarian blood flow

parallels luteal P4 production in cattle (Ford and Chenault,

1981; Wise et al., 1982) thus providing an additional means

of reducing CL and ovarian steroid concentrations through

elevated clearance rates.

In recent years, several ovarian peptides, including

steroid-binding proteins (Willcox, 1983), LH-binding

inhibitors (Kumari et al., 1982; Ward et al., 1982),

gonadotropin releasing hormone (GnRH)-like peptides (Ying et

al., 1981; Fraser, 1982; Sharpe, 1982), angiogenic factor

(Gospodarowicz and Thakral, 1978), oxytocin (Fields et al.,

1983; Rodgers et al., 1983; Wathes et al., 1983),

vasopressin (Wathes et al., 1983) and relaxin-like peptides

(Fields et al., 1982) have been discovered. Although

biochemical evidence suggests that these peptides are

produced by the CL and/or follicle, little information is

available concerning their physiological roles in ovarian


Of the peptides mentioned above, oxytocin has been

evaluated most thoroughly. It is known that CL of the cow

and ewe contain high concentrations (ug/g) of oxytocin

(Fields et al., 1983; Flint and Sheldrick, 1985; Flint et

al., 1983; Wathes et al., 1983) which is synthesized and

stored in the large lutein cells (Rodgers et al., 1983).

The bovine CL contains neurophysin, lending further support

for the local synthesis of oxytocin within the CL (Wathes et

al., 1983). Additionally, circulating oxytocin

concentrations are rapidly reduced following ovariectomy

(Schams et al., 1982) and concentrations of immunoreactive

oxytocin were greater in ovarian venous plasma versus

uterine arterial or jugular venous plasma (Flint and

Sheldrick, 1982; Walters et al., 1984). Concentrations of

oxytocin and neurophysin in luteal tissue (Wathes et al.,

1984) and plasma (Sheldrick and Flint, 1981; Flint and

Sheldrick, 1983; Schams, 1983; Walters et al., 1984)

parallel CL P4 production during the luteal phase, although

the decline in oxytocin concentrations occur prior to P4

during CL regression. Oxytocin and P4 are secreted from the

CL in a pulsatile fashion. Walters et al. (1984) reported

that only 29% of P4 episodes were associated with pulses of

oxytocin during the early luteal phase, due largely to the

occurrence of fewer oxytocin than P4 pulses at this time.

During the mid-luteal phase 86% of P4 and oxytocin episodes

occurred simultaneously. Remarkably, 97 to 100% of oxytocin

pulses, during all periods of the luteal phase, were

associated with P4 pulses. These observations suggest that

similar mechanisms may regulate the secretion of P4 and

oxytocin from the CL.

Several lines of evidence support a systemic role for

oxytocin in the process of CL regression (reviewed in a

later section). Additional data suggests that oxytocin may

have an action within the ovary as well. Evidence reviewed

by Wathes (1984) indicates that oxytocin (and vasopressin)

may be involved in P4 production during CL development since

oxytocin (and vasopressin) can promote P4 biosynthesis in

testicular cells through the inhibition of LH-stimulated

androgen biosynthesis. Oxytocin probably acts by selective

suppression of 17a-hydroxylase and C17-20 desmolase

activities. Follicular androgen and estrogen biosynthesis

is suppressed in favor of P4 biosynthesis prior to and

following the preovulatory LH surge. Furthermore, oxytocin

has been measured in preovulatory follicles, ovarian tissue

(including all follicles) devoid of CL, and newly formed CL

(Wathes, 1984; Wathes et al., 1984). Therefore,

preovulatory follicle and luteal oxytocin biosynthesis,

possibly induced by LH, may prevent C-19 and C-18 steroid

biosynthesis and promote P4 formation in CL. High levels of

oxytocin in CL during the mid-luteal phase may reduce

follicular E2 production in a similar manner (Wathes,

1984). Results of Tan et al. (1982) also demonstrated

stimulatory effects of low levels of oxytocin (2 to 20 mIU)

in CL collected from pregnant cattle. Likewise, exogenous

oxytocin, administered to cyclic cattle on days 12 and 1?,

increased P4 content of CL on day 14 (Mares and Casida,

1963). Conversely, high doses of oxytocin (200 to 4u0 mid)

were inhibitory to in vitro P4 biosynthesis by CL of cyclic

(Tan and Biggs, 1984) and pregnant (Tan et al., 1982)


Luteolytic prostaglandins appear to mediate oxytocin

secretion from the CL (Flint and Sheldrick, 1982, 1983;

Schallenberger et al., 1984). Similarly, PGF2a or analogues

of PGF2a stimulate rapid degranulation of large lutein cells

(Heath et al., 1983) which are known to be the synthetic and

storage site of oxytocin in the CL (Rodgers et al., 1983).

Consistent with the hypothesis of oxytocin storage in

granules are observations which demonstrate parallel

patterns of luteal oxytocin content (Wathes et al., 1984)

and secretion (Sheldrick and Flint, 1981; Flint and

Sheldrick, 1983; Schams, 1983; Walters et al., 1984) with

large luteal cell granule content (Gemmell et al., 1974).

Based on the nearly absolute synchrony with which tonic

oxytocin and P4 pulses occur, in vivo (Walters et al.,

1984), and the ability of PGF2a to induce oxytocin

secretion, it may be suggested that luteal prostaglandins

are involved in tonic oxytocin and P4 secretion during the

luteal phase of the estrous cycle. Evidence of luteal

prostaglandin biosynthesis (Shemesh and Hansel, 1975a;

Lukaszewska and Hansel, 1979; Ailvae and Hansel, 1968a)

strengthens this supposition.

Bovine luteal tissue contains massive quantities (2 to

3 mg/g) of arachidonic acid, the bulk of which is esterified
to phospholipids (Lukaszewska and Hansel, 1979, 1980).

Thus, the potential for arachidonic acid metabolism in

either the cyclooxygenase or lipoxygenase pathway is

great. Metabolic pathways utilizing arachidonic acid as

precursor have been reviewed previously (Ramwell et al.,

1977). Characterization of local biosynthetic patterns and

functions of luteal prostaglandins and other arachidonic

acid metabolites have provided additional insight into

mechanisms controlling luteal P4 production and regression.

Shemesh and Hansel (1975b) demonstrated an acute

increase in ovarian venous PGF2a concentrations following

injection of arachidonic acid into CL on days 12 or 13 in

cyclic cattle. Peripheral plasma P4 concentrations fell

rapidly from approximately 6 ng/ml to 2.5 ng/ml, and

remained constant until day 15 or 16 when the experiment

concluded. Additionally, CL weights on day 15 or 16 were

less in arachidonic acid-treated (4.3 1.4 g) than vehicle-

treated (6.3 0.5 g) cattle. Several studies have

demonstrated an inverse relationship between luteal PGF2a

and P4 content or production (Patek and Watson, 1976;

Lukaszewska and Hansel, 1979; Rothchild, 1981). However,

the means by which elevated luteal PGF2a contents reduce P4

production remain to be elucidated. Exogenous PCGF2

administration to cattle (Lukaszewska and Hansel, 1979),

sheep (Diekman et al., 1978) and rats (Hichens et al., 1i74;

Grinwich et al., 1976) initiates a reduction in P4

production and the loss of LH receptors or luteal

sensitivity to LH. The authors suggested that loss of LH

receptor numbers or reduced sensitivity to LH were the

consequence rather than cause of luteolysis since reduced P4

production occurred considerably earlier than either of the

LH related phenomenon.

The PGF2,-induced decline in luteal P4 production

described above may be related more closely to changes in

ovarian vascular resistance. Blood flow to the CL-bearing

ovary parallels closely luteal P4 production (Niswender et

al., 1975; Ford and Chenault, 1981) and PGF2a is known to

reduce ovarian blood flow and luteal P4 production (Nett et

al., 1976; Niswender et al., 1976). Therefore, PGF2

effects may be mediated through increasing ovarian vascular

resistance. However, metabolite requirements are generally

thought to dictate the degree of blood flow to a tissue. In

this light, reduced luteal metabolic activity may initiate

the accumulation of PGF2. and subsequent decline in luteal

blood flow.

Progesterone biosynthetic activity of bovine CL

collected during the estrous cycle is highly correlated

(r2=.93) with luteal prostacyclin (PGI2) biosynthesis

(Milvae and Hansel, 1983a). Sun at al. (1977) demonstr-saed

that PGI2 was the predominant prostaglandin product

following bovine CL membrane metabolism of PGH2.

Furthermore, PGI2 stimulated luteal P4 biosynthesis in vitro

and in vivo (Milvae and Hansel, 1980b). A recent report by

Milvae and Hansel (1985) demonstrated that indomethacin

inhibition of prostaglandin formation early in the estrous

cycle shortened luteal lifespan in cattle. Indomethacin was

administered into the uterine lumen twice daily from days 4

through 6. In conjunction with previous reports, these

authors proposed that luteotropic products of the

cyclooxygenase pathway, viz., PGI2 and PGE2, may be required

for normal CL development. The origin of these

prostaglandins may include both the CL and uterus.

Additionally, synthesis and content of PGI2 and PGF2a in

bovine CL are greatest early in the cycle (Milvae and

Hansel, 1983a). Luteal prostacyclin content was

approximately four times greater than PGF2a early in the

cycle although ratios of PGI2 and PGF2a gradually decline in

CL age increased (Milvae and Hansel, 1983a). Milvae et al.

(1985) provided several lines of evidence suggesting that

lipoxygenase products, viz., 5-hydroperoxy-eicosatetraenoic

acid (5-HPETE) and 5-hydroxyeicosatetraenoic acid (5-HETE),

regulate P4 production by the selective inhibition of

prostacyclin synthetase activity and PGI2 production in

bovine CL. Large quantities of 5-HETE are present in bovine

CL on days 10, 15 and 18 of the cycle and 5-HETE inhibited

in vitro PGI2 and P4 biosynthesis in a dose dependent

fashion without influencing PGF2a production. Simultaneous

addition of LH was unable to prevent the 5-HETE-induced

reduction in luteal cell PGI2 and P4 production, in vitro.

In related experiments (Milvae et al., 1985), arachidonic

acid metabolism through the lipoxygenase pathway was

inhibited by nordihydro-guaiaretic acid (NDGA). When NDGA

was added to dispersed luteal cells, PGI2 production was

increased after a 2 hour incubation. Progesterone and PGF2a

synthesis had increased slightly, although not significantly

(P>.05). Arachidonic acid plus NDGA increased luteal

production of both PGI2 and PGF2a but not P4, suggesting an

antagonism between PGI2 and PGF2. in the regulation of

luteal P4 synthesis (Milvae et al., 1985). This possibility

is also suggested by the declining ratio of PGI2 and PGF2a

contents in bovine CL as luteolysis approaches (Milvae and

Hansel, 1983a). Collectively, these data suggest that HETE,

and perhaps other lipoxygenase products, are involved in

luteolysis by blocking synthesis of the luteotropin, PGI2,

and allowing additional arachidonic acid to be shunted

toward pathways involved in synthesis of luteolytic


Similar mechanisms controlling biosynthesis of

luteotropic and luteolytic prostaglandins may also occur

within the uterus (Milvae et al., 1985; Milvae and Mansel,


Uterine Control of CL Function

There is little doubt that the uterus plays a key role

in regulation of CL lifespan during the estrous cycle. Loeb

(1923) first demonstrated that extirpation of the guinea pig

uterus resulted in extended CL function. Subsequent reports

in cattle (Wiltbank and Casida, 1956; Anderson et al., 1961;

Malven and Hansel, 1964; Anderson et al., 1965), sheep

(Wiltbank and Casida, 1956; Anderson et al., 1969), pigs

(Spies et al., 1958; Du Mesnil Du Buisson, 1966) and horses

(Ginther and First, 1971) provided similar results following

surgical removal of the uterus. Thus, a basis for uterine

involvement in the luteolytic process was developed. The

nature of this utero-ovarian relationship was further

defined following observations that cattle (Bland, 1970) and

sheep (McCrackin and Caldwell, 1960) with congenital absence

of the uterine horn (uterus unicornis) adjacent to the CL

bearing ovary exhibited prolonged luteal lifespans and cycle

lengths which suggested local regulation of CL function by

the ipsilateral uterine horn. Additionally, normal luteal

regression occurred following hemihysterectomy of the

contralateral, but not the ipsilateral uterine horn in

cyclic cattle and sheep (Inskeep and Butcher, 1966; Mioor and

Rowson, 1966; Ginther, 1974, 1981).

As a consequence of these data, Wollmprhaus (1964) and

Ginther (review, 1974) examined the anatomy and histology of

the utero-ovarian vascular architecture as a possible route

of exchange between the uterine horns and their adjacent

ovaries. In cattle and sheep, species which exhibit local

regulation of CL function by the uterus, exchange of

luteolytic substances from the uterine venous effluent

(ovarian vein, International Committee of Veterinary

Anatomical Nomenclature, 1968) into the ovarian arterial

supply is thought to occur just below the ovarian vascular

pedicle. At this point, the ovarian artery follows an

extremely tortuous and convoluted path over the surface of

the ovarian vein (containing both uterine and ovarian venous

blood) thus maximizing the area of contact between the two

vessels. Histologically, the adventitial layers of each

vessel were substantially thinner in the regions of vein-

artery apposition (Wollmerhaus, 1964; Del Campo and Ginther,

1974). From an anatomical and histological perspective,

these observations favored the existence of a functional

venoarterial exchange system between uterine horns and their

adjacent ovaries.

A series of physiological experiments, utilizing

surgical anastomoses of ovarian veins or ovarian arteries in

hemihysterectomized cattle and sheep, provided conclusive

support for the functional ovarian vein and artery

components of the local uteroovarian pathway (Ginzner, 1974;

1981). In such experiments, luteal regression occurred when

either ovarian vein or ovarian artery blood from the

contralateral, uterine-intact side was diverted to the

appropriate vessel on the ipsilateral, hemihysterectomized

side. Conversely, CL were maintained if the surgical

anastomosis became occluded. Thus, it became clear that the

uterus was producing a blood borne product which could move

from the uterine venous drainage into the ovarian arterial

supply and initiate the process of luteolysis in a local


As early as 1940, Hechter and coworkers suggested that

some uterine-derived product may be responsible for

regulation of CL lifespan. However, it was not until the

resurgence of prostaglandin research in the 1950's that the

identity of the uterine luteolytic factor was elucidated.

The chemical structures of several prostaglandins, including

PGF2a, were confirmed in 1962 (Nelson et al., 1982) and by

1969, Pharris and Wyngarden provided evidence that the

vasoactive prostaglandin, PGF2a, may be the uterine

luteolytic substance in the rat. Subsequently, the

luteolytic activity of exogenous PGF2a in cattle has been

documented extensively (Hansel et al., 1973; Hafs et al.,

1974; Lauderdale, 1974; Thatcher and Chenault, 1976).

Several studies have provided evidence indicating that PGF2a

is an endogenous luteolysin in cattle. For example,

elevated levels of PGF2. in uterine venous drainage

(Nancarrow et al., 1975; Shemesh and Hansel, 1975), uterine

tissue (Shemesh and Hansel, 1975) and uterine flushings

(Lamothe et al., 1977; Bartol et al., 1981a) coincide

closely with the period of expected luteal regression. In

addition, luteolysis is prevented in cows immunized

passively against PGF2a (Fairclough et al., 1981) or

following intrauterine administration of the prostaglandin

synthetase inhibitor, indomethacin, late in the estrous

cycle (Lewis and Warner, 1977).

The biological half-life of PGF2a is very short owing

to its rapid metabolism by the lungs (Piper et al., 1970)

and uterine endometrium (Thatcher et al., 1984; Chapter

5). Thus, measurements of its primary metabolite, 15-keto-

13,14-dihydro-PGF2. (PGFM; Granstrom and Kindahl, 1982),

provide an index of uterine PGF2a production. Peripheral

concentrations of PGFM are significantly correlated with the

uterine production of PGF2. (Thatcher et al., 1984; Chapter

3) and concentrations of PGF2a in uterine flushings (Bartol

et al., 1981a). Furthermore, three to five pulsatile

episodes of PGFM were always temporally associated with the

decline in plasma P4 concentrations during luteolysis in

cattle (Peterson et al., 1975; Kindahl et al., 1976;

Betteridge et al., 1984). Collectively, these data indicate

that the endogenous uterine luteolysis in cattle is likely

to be PGF2a.

Finally, in keeping with the concept of a local

venoarterial pathway, transfusion of PGF2a from the uterine

venous drainage into the adjacent ovarian arterial supply

was demonstrated. In an early study, Goding et al. (1971)

induced luteal regression, as demonstrated by the

precipitous decline in plasma P4 concentrations, following

the infusion of physiological concentrations of PGF2a into

the uterine venous drainage of diestrous cattle. Hixon and

Hansel (1974) reported elevated PGF2a concentrations in the

uterine venous drainage (ovarian vein) and ovarian artery,

ipsilateral to the CL bearing ovary, as a result of the

intrauterine deposition of PGF2a. Similarly, Thatcher et

al. (1984b) verified the functionality of the countercurrent

exchange system (PGF2a) during the period of normal luteal

regression in cyclic cattle.

Regulation of Uterine PGF9,Production

Demise of the cyclic CL is uterine and PGF2,-dependent,

as reviewed above. However, interplay between ovarian

estrogen, progesterone and oxytocin with the uterine

endometrium regulate the timing and magnitude of luteolysin

production during the estrous cycle. Estrogens, of

follicular origin, appear to constitute an important

stimulatory component regulating uterine PGF2a production

and CL regression. Destruction of ovarian follicles in

cattle (Villa-Godoy et al., 1981) and sheep (Karsch et al.,

1970; Ginther, 1971) result in prolonged luteal lifespans
and heavier CL compared to follicle-intact controls.

Furthermore, administration of estrogens to late-luteal

phase cows and ewes resulted in production of PGF2a by the

uterus (Barcikowski et al., 1974; Thatcher et al., 1984b;

Chapter 3) and premature CL regression (Greenstein et al.,

1958; Wiltbank, 1966; Stormshak et al., 1960; Caldwell et
al., 1972; Eley et al., 1979). As anticipated, luteolytic

doses of E2 were ineffective in hysterectomized (Brunner et

al., 1969; Akbar et al., 1971; Caldwell et al., 1972) and

indomethacin-treated (Warren et al., 1979; Barcikowski et

al., 1974) cattle and sheep. In another study (Loy et al.,

1960), E2 administration to cattle on day 1 or 4 of the

estrous cycle had no effect on CL examined on day 14.

However, treatment with P4 caused a reduction in CL weight,

P4 content, and number of functional luteal cells.

Administration of exogenous P4 for 1 to 5 days post-estrus

in cattle and sheep results consistently in shorter

interestrous intervals (Woody et al., 1967; Ginther, 1968,

1969; Thwaites, 1971; Lawson and Cahill, 1983), presumably

as a result of premature development and activation of the

uterine PGF2a secretary system (Baird et al., 1976; Ottobre

et al., 1980). These data suggest that the uterus requires

a period of P4 exposure in order to develop the capacity to

secrete luteolytic levels of PGF2a in response to

estrogens. For example, in the cow (SKjerven, 195t;

Martinov and Lovell, 1968) and ewe (Brinsfield anrd Haw,

1973) the degree of endometrial lipid droplet accumulation

varies cyclically, being more prevalent during P4 tnan E2

dominated phases of the estrous cycle. Additionally,

exogenous P4 stimulated accumulation of endometrial lipid

droplets while E2 administration decreased lipid stores

(Brinsfield and Hawk, 1973). Finally, Hansel et al. (1975)

extracted large quantities of arachidonic acid, the

principle substrate for PGF2a synthesis, from bovine

endometrium between days 10 and 14 of the luteal phase.

Collectively, these results support the concept that P4

directs endometrial accumulation of lipid and prostaglandin

precursor concentrations during the estrous cycle.

Estrogens induce the luteolytic response during late


Timing and magnitude of the uterine luteolytic response

are dependent on circulating patterns and concentrations of

P4 and E2 which, in turn, regulate the degree of uterine

sensitivity to ovarian steroids, viz., endometrial P4 and E2

receptor populations. Estrogen is known to stimulate

synthesis of endometrial estrogen and P4 cytoplasmic

receptors (Rc) (Clark et al., 1977; Koligian and Stormshak,

1977a). Conversely, P4 inhibits synthesis of E-Rc and

reduces nuclear retention of the estrogen-receptor complex

in endometrial tissue (Clark et al., 1977; Koligian and

Stormshak, 1977a; Katzenellenbogan et al., 1980).

Progesterone also reduces endometrial synthesis of its own

receptor (Schrader and O'Malley, 1978; Walters and Clark,

1980). With this in mind, direct measurements of steroid

receptors in bovine and ovine endometrium revealed elevated

concentrations of E-Rc during proestrus and metestrus, and

low E-Rc concentrations during diestrus (Senior, 1975;

Koligian and Stormshak, 1977b; Henricks and Harris, 1978).

Additionally, Zelinski et al. (1982) demonstrated a decline

in bovine endometrial P4-Rc concentrations from the

proestrus to diestrus phases of the estrous cycle. Thus, as

the CL develops and plasma E2:P4 ratios decline,

concentrations of endometrial E-Rc fall precipitously.

Plasma P4 thereby inhibits the uterus from responding in an

estrogen-directed fashion. However, elevated plasma P4

concentrations progressively suppress endometrial production

of its own receptor such that estrogen-induced events (i.e.,

prostaglandin biosynthesis) are gradually removed from P4


Induction of uterine PGF2a production by E2 may be

attributed partially to E2-dependent increases in activities

of cyclooxygenase (Huslig et al., 1979) and phospholipase A2

(Dey et al., 1982). Calcium is probably involved in

activating phospholipase A2 since this enzyme is Ca+2

dependent (Brockerhoff and Jensen, 1974). Estrogen was also

shown to increase uterine Ca+2 availability in swine

(Geisert et al., 1982).

Estrogen-induced luteolysis is prevented by inhibition

of uterine DNA dependent RNA synthesis with actinomycin 0

(French and Casida, 1973) suggesting that the process of

PGF2. production requires protein synthesis. Furthermore,

Roberts et al. (1976) and McCracken et al. (19b1, 1984)

provided evidence for estrogen induction of endometrial

oxytocin receptors following a period of P4 priming in

sheep. Exogenous oxytocin stimulates uterine PGF2a

production in cyclic cattle (Newcomb et al., 1977; Milvae

and Hansel, 1980) and sheep (Roberts et al., 1975, 1976;

McCracken, 1981, 1984) and causes premature CL regression in

uterine-intact (Armstrong and Hansel, 1959; Hansel and

Wagner, 1960; Milne, 1963), but not hysterectomized (Ginther

et al., 1967) or indomethacin-treated (Cooke and Homeida,

1983) females. In addition, luteal function was extended in

ewes immunized against oxytocin (Sheldrick et al., 1980;

Schams et al., 1983). Based on these results and reports

concerning secretion of oxytocin and P4 by the CL (reviewed

earlier), McCracken and coworkers (1981, 1984) proposed a

working hypothesis for CL regression in cyclic ewes.

Briefly, their proposed sequence of events are: 1)

declining endometrial P4 receptor population reduces P4

dominance on the uterus during late diestrus and permits

follicular E2 to stimulate endometrial estrogen and oxytocin

receptor formation and activate enzymes essential for P'F2a

biosynthesis; 2) circulating luteal (or pituitary) oxytocin

binds to newly synthesized endometrial receptors and

initiates PGF2. secretion; 3) plasma P4 concentrations

decline slowly, further reducing its influence on the

uterus; 4) uterine-derived PGF2a induces a rapid dumping of

luteal oxytocin which reinforces uterine PGF2a production

and 5) PGF2. initiates a rapid decline in plasma P4

concentrations associated with CL demise.

Available data indicate that PGF2a is the primary

uterine luteolysin in numerous species (Horton and Poyser,

1976). However, recent data implicate additional

arachidonic acid metabolites in the luteolytic process. For

example, the 13,14-dihydro metabolite of PGF2a was as

effective as equivalent amounts of PGF2. in causing CL

regression in heifers (Milvae and Hansel, 1983).

Arachidonic acid metabolites of the lipoxygenase pathway may

also be physiological regulators of luteal lifespan (Milvae

et al., 1985). Lipoxygenase products, such as 5-hydroxy-

eicosatetraenoic acid (5-HETE), specifically inhibit luteal

synthesis of prostacyclin, a potent luteotropin (Ailvae and

Hansel, 1980, 1983), thereby reducing luteal P4 production,

in vitro. When in vitro synthesis of 5-HETE and other

lipoxygenase products was blocked by nordihydroguaiaretic

acid NDGA, luteal cell prostacyclin and P4 production were

increased without influencing PGF2a production. Further,

twice daily intrauterine administration of NDGA to cyclic

heifers on days 14 to 18 extended luteal function

approximately 5 days beyond control heifers. i-iese data

suggest that lipoxygenase products, of uterine and/or

ovarian origin, are involved in the luteolytic process. In

a final study, Milvae and Hansel (1985) provided evidence

that uterine and/or ovarian prostaglandins are required for

normal CL development and function in cattle. Intrauterian

administration of indomethacin to cyclic heifers on days 4,

5 and 6 caused decreased P4 production and early CL

regression. The authors implied that luteotropic

prostaglandins, such as PGE2 and prostacyclin, may be

involved in luteal development.

Conceptus-Associated Events During
Early Pregnancy

An opportunity for establishment of pregnancy is

provided with each estrous cycle. As a consequence of

luteal P4 production, glandular epithelium of the uterine

endometrium develops morphological/ultrastructural

characteristics, and functions of an active secretary tissue

(Skjerven, 1956b; Marinov and Lovell, 1968; Wathes and

Wooding, 1980). Several studies have demonstrated increased

endometrial tissue content or uterine secretion of amino

acids, proteins, fatty acids, lipids, carbohydrates, and

certain ions during the luteal phase of the estrous cycle or

in response to exogenous P4 (cow: SKjerven, 1956a; Fahning

et al., 1967; Carlson et al., 1970; Roberts and ParKer,

1974; Hansel et al., 1975; Bartol et al., 1981; sneep:
Brinsfield and Hawk, 1973; Roberts and Parker, 1976; pig:

Murray et al., 1972; Knight et al., 1974; Bazer, 1975; Bazer

et al., 1977; human: Shapiro et al., 1980). In addition,

elevated P4 concentrations reduce uterine tone and

myometrial contractility (Hays and Van Demark, 1953; Lehrer

and Schindler, 1974). Thus, during the first 15 to 16 days

post-estrus, a complex embryotrophic environment is

established within the uterine lumen.

During early pregnancy there is a clear requirement for

developmental synchrony between the concepts and

endometrium. Embryo transfer studies in cattle and sheep

indicate that a small degree of asynchrony ( 1 or 2 days)

is tolerated with respect to stage of transferred concepts

and recipient uterine development (Rowson and Moor, 1966;

Rowson et al., 1969; Sreeman, 1978; Seidel, 1981). However,

pregnancy success is highest when synchronization is precise

(Cook and Hunter, 1978).

As described earlier, P4 administered early in the

estrous cycle is thought to initiate premature development

and activation of the uterine PGF2. secretary system (Baird

et al., 1976; Ottobre et al., 1980). This system was

utilized recently to evaluate the role of P4 pretreatment,

early in the estrous cycle, on the embryonic state of the

ovine uterus (Lawson and Cahill, 1983). Progesterone

administered during the first 4 days of the ovine estrous

cycle accelerated development of the uterine luminal

environment such that the day 6 P4-treated uterus provided

acceptable support for normal development of transferred 13

day old embryos. Luteolysis had not occurred in 8 of 12

ewes slaughtered on day 25. No day 10 conceptuses survived

following transfer to control (no P4 pretreatment) uteri on

day 6. Estrous cycles in these ewes were of normal

length. These data suggest that uterine development and

complexity of its luminal environment progress sequentially

under the influence of P4. Further, the concepts must

develop in synchrony with this environment to survive.

The inability of embryos to develop beyond the

blastocyst stage, when cultured in vitro (Wright and

Bondioli, 1981) or confined within the oviduct

(Wintenberger-Torres, 1956; Murray et al., 1971; Pope and

Day, 1972; Heap et al., 1979), illustrate that unique

components of the uterine environment become essential for

further concepts development. The essentiality of this

relationship may be due, in part, to increased complexity of

concepts nutrient and hormonal requirements following

blastulation (Daniel, 1971; Biggers and Borland, 1976).

Additionally, Lawson et al. (1983) demonstrated an active

interaction between the uterine environment and rate of

concepts development in sheep. In two experiments,

sixteen-cell embryos (day 4) were transferred to uteri of

ewes on days 4, 6 or 7 of the estrous cycle. The embryos

were then recovered and evaluated 4 or 6 days later.

Embryos transferred to older, more advanced uteri were

viable when recovered (days 8 to 12 of the estrous cycle)

and exhibited accelerated growth rates during the 4 to 6

days in utero as compared to synchronously transferred

controls. A third experiment examined conceptuses on days

12, 14 or 15 of the estrous cycle following synchronous

versus asynchronous embryo transfers, as described above.

Results indicated that viability and accelerated growth

rates of asynchronously transferred embryos were halted by

day 11 or 12 post-estrus. Further, these conceptuses were

unable to maintain CL function in recipient ewes.

Similarly, P4 has been implicated in regulating ovine

concepts growth during normal pregnancy (Bindon, 1971) and

in superovulated recipients of embryo transfers

(Wintenberger-Torres, 1968; Wintenberger-Torres and

Rombauts, 1968). Presumably, P4 effects on embryonic growth

are mediated by alterations in the uterine luminal milieu.

Collectively, these data demonstrate that the concepts is

capable of responding to physiochemical cues in its uterine

environment or conversely, the uterus exerts a regulatory

influence on some aspects of concepts development during

early pregnancy. Either of these viewpoints may explain how

small degrees of developmental asynchrony between the uterus

and concepts are tolerated in large domestic species

(Rowson and Moor, 1966; Rowson et al., 1969; Sreenan, 197d;

Seidel, 1981). However, there appears to be a developmental

limit, beyond which the concepts may not "catch up" to Lts

uterine environment and prevent luteolysis (Lawson et al.,


Regardless of this physiological flexibility, a

considerable percentage of embryos do not survive beyond day

30 of pregnancy in large domestic species. In cattle, the

majority of embryonic mortality occurs between days 16 and

26 of pregnancy (Hawk et al., 1955; Boyd et al., 1969;

Ayalon, 1978; Ball, 1978; Hawk, 1979). Dramatic changes in

bovine concepts growth, differentiation (Melton et al.,

1951; Chang, 1952; Greenstein and Foley, 1958a, b;

Greenstein et al., 1958; King et al., 1982) and endocrine

activity (Shemesh et al., 1979; Chenault, 1980; Lewis et

al., 1982; Eley et al., 1983; Bartol et al., 1985) occur

during this relatively short period. Pregnancy success

between days 16 to 30 of gestation depends upon 1)

appropriate physiochemical communication between concepts

and endometrium such that CL maintenance is achieved, and

2) the initiation of placental (cotyledon and caruncle)

development and function. "Maternal recognition of

pregnancy" (Short, 1969), occurring by days 16 to 17 in

cattle (Betteridge et al., 1980; Northey and French, 1980;

Dalla Porta and Humblot, 1983), and day 12 in sheep (Moor

and Rowson, 1966; Rowson and Moor, 1967) and swine (Ohinasa

and Dziuk, 1968), represents the first period during early

gestation when production of concepts signals becomes

essential for luteal maintenance and continued endometrial

secretary support. The concepts does not appear to be

essential for CL maintenance or initiation of an embryonic

uterine environment prior to maternal recognition of

pregnancy. This was clearly demonstrated by Betteridge et

al. (1980) when pregnancies were established following

synchronous embryo transfers as late as day 16 in cyclic

cattle. Additionally, noticable histological alterations in

endometrial development between pregnant and cyclic cattle

do not occur prior to day 17 (King et al., 1981, 1982).

Thus, pregnancy associated events occurring prior to

maternal recognition of pregnancy may be thought of as being

mediated primarily by the CL and endometrium, whereas events

following this period are under direction of the concepts.

Conceptus Development (Days 15 to 30)

As noted earlier, developmental phases including

maternal recognition of pregnancy through definitive

placentome formation occur between days 16 and 30 of

gestation in cattle. The large degree of embryonic

mortality during this period attests to the critical nature

of these developmental processes with regard to

establishment and maintenance of pregnancy.

On day 16 of gestation, the bovine concepts is

characteristically bilaminar (trophectoderm and enaoder.),

with more advanced conceptuses exhibiting varying degrees of

mesodermal germ layer development (Chang, 1952; Greenstein

and Foley, 1958 a, b). Trophectodermal and endodermal germ

layers initiate a phase of rapid elongation by days 15 to 16

of pregnancy. Lengths of extraembryonic membranes vary

tremendously during this period (1.5 to 225 mm; Hawk et al.,

1955; Greenstein and Foley, 1958b; Betteridge et al., 1980)

with mean diameters calculated between 50 and 90 mm.

Histological and ultrastructural evaluation of rapid

and progressive transition of porcine concepts forms, viz.

spherical, tubular and elongated filamentous, between days

10 and 12 of gestation (Geisert et al., 1982) provided

evidence that rapid concepts elongation occurs as a result

of trophectodermal and'endodermal reorganization and not

hyperplasia. Whether or not this phenomenon occurs during

concepts elongation in the cow or sheep is not known.

By day 17, mean concepts extraembryonic membrane

length reaches approximately 150 mm (Chang, 1952; Greenstein

and Foley, 1958b). Histochemical evaluation of

trophectodermal cell types (days 16 to 33; Greenstein et

al., 1958) based on staining characteristics with Sudan

Black B or Oil Red O (demonstration of lipids), Periodic-

Acid-Schiff (PAS; demonstration of glycogen, glycoproteins),

and phloxine-methylene blue (demonstration of cytoplasmic

basophilia or acidophilia) delineated three distinct types

of cells by day 16 and 17. These included: 1) PAS (+),

trophoblastic giant cells (GC) which were occasionally

binucleated at this stage; 2) undifferentiated trophoblast

"stem cells"; and 3) normal columnar trophoblast cells which

consistently contained basally located lipid inclusions.

The embryo proper, viz. germinal or embryonic disc, is

notably oval in shape by day 17 and demonstrates signs of

initiating organogenesis, as indicated by the appearance of

the primitive groove and node. Additionally,

ultrastructural appearance of day 17 endometrial epithelium

was indistinguishable in pregnant versus cyclic cattle (King

et al., 1981), suggesting that no physical interaction

between concepts and endometrium had occurred at this stage

of pregnancy.

Conceptus adhesion to maternal endometrium was noted to

occur by days 18 to 20 (Wathes and Wooding, 1980; King et

al., 1980, 1981), although no microvillous interdigitation

was observed. This physical interaction between the

concepts and endometrium is consistent with the observed

reduction in uterine epithelial cell height by day 18 of

pregnancy (Wathes and Wooding, 1980; King et al., 1981) and

the apparent migration of binucleated trophoblastic GC into

the uterine epithelial cell layers (Wathes and Wooding,

1980). Extraembryonic membranes continue to elongate,

extending throughout the gravid and nongravid uterine horns

by days 19 and 20. In addition, numbers of binucleated GC

continue to increase, constituting as much as 2C; of the

total trophectodermal cell population on day 23 (vathes and

Wooding, 1980).

Between days 18 and 20, amnionic formation is completed

and the allantoic diverticulum develops via an evagination

of embryonic hindgut splanchnopleure (Greenstein and Foley,

1958a; Greenstein et al., 1958). Conversely, the highly

vascularized yolk sac reaches its developmental peak by day

20 and begins regressing thereafter as the allantois emerges

(Greenstein et al., 1958). This period of development marks

a coordinated transition from conceptus-derived (chorio-

vitelline placenta) to maternal-derived chorioallantoicc

placenta) nutritional support.

Mutual microvillous interdigitation between

trophoblastic and endometrial epithelium is initiated by day

24 (King et al., 1980, 1981; Wathes and Wooding, 1980) while

definitive attachment occurs by day 27 (King et al.,

1980). Additionally, immature fetal cotyledonary villi are

apparent adjacent to maternal caruncles from approximately

days 27 to 30 of gestation (Greenstein et al., 1958; King et

al., 1981). Considerably more lipid accumulation was noted

in trophectodermal epithelium from day 27 onward (King et

al., 1980) suggesting that functional exchange of nutrients

from maternal to concepts tissues was accelerated following

definitive attachment. There is a continued increase in the

number of bi- and multinucleated GC reported in

trophectoderm and endometrial epithelial cell layers

throughout this period (Greenstein et al., 1958; King at

al., 1980; Wathes and Wooding, 1980). Multinucleated GC

were reported to comprise 25% of total endometrial

epithelial cell population and 50% of cell area in the

gravid uterine horn on day 24 (Wathes and Wooding, 1980).

Regression of the yolk sac is complete by day 24.

However, allantois expansion and vascularization are not

completed until approximately days 30 to 33 (Greenstein et

al., 1958). Diameter of the allantois ranges from 0.8 to

5.5 cm on day 24 to over 35 cm on day 33. Dramatic

expansion and vascularization of the allantois is initiated

between days 28 to 30 (Greenstein et al., 1958). Continued

expansion of allantoic fluid volume and membrane diameter

during the first two months of gestation forces

extraembryonic membranes into apposition with the

endometrium, thus facilitating development of firm

cotyledon-caruncle attachment (Eley et al., 1978).

Establishment of a vascularized chorioallantoic placenta

proceeds and is essential for promotion of substantial fetal

weight gains during later gestation (Eley et al., 1978).

Conceptus Signals and Maternal Recognition
of Pregnancy

A functional CL is required if pregnancy in cattle is

to persist. Maintenance of the CL during pregnancy depends

on the presence of a viable concepts within the uterine

lumen and its ability to appropriately "signal" a receptive

maternal system, such that processes mediating cyclic

regression of the CL are circumvented. In cattle, maternal

recognition of pregnancy (Short, 1969) must be initiated by

days 16 to 17 post-estrus if pregnancy is to become

established. For example, Betteridge et al. (1980)

demonstrated that synchronous embryo transfer to recipients

on or prior to day 16, but not day 17 post-estrus, resulted

in normal pregnancies. Additionally, nonsurgical embryo

removal from pregnant cattle on days 16 through 19 resulted

in extended CL maintenance and interestrous intervals (range

4 to 8 days) compared to nonmated controls (Northey and

French, 1980; Dalla Porta and Humblot, 1985). Embryo

removal prior to day 16 of pregnancy, viz. days 9, 13, 14,

and 15, had no effect on cycle length or occurrence of

luteolysis. These data provided clear evidence that CL

maintenance required the presence of a viable concepts

within the uterine lumen of cattle by days 16 to 17 post-

estrus. Further data indicated that twice daily

intrauterine injections of homogenized day 17 or 18

conceptuses (Northey and French, 1980) or freeze-killed day

16 conceptuses (Dalla Porta and Humblot, 1980) on days 14

through 18 or days 15 through 19, respectively, also

extended luteal function. Northey and French (1980) noted

that intrauterine injections of concepts homogenates did

not stimulate luteal P4 production, but extended the

interval to CL regression. Intrauterine administration of

two freeze-killed, day 12 embryos/injection had no effect on

CL lifespan (Dalla Porta and Humblot, 1983). Thus,

potential for bovine concepts production of potent,

biologically active substances by day 16 of pregnancy was

-supported, as was their involvement in prolongation of CL


Subsequent studies in cattle suggested that there are

probably several interdependent strategies by which

concepts signals may initiate maintenance of the CL during

early pregnancy. Potential effectors, or biologically

active signals, which are synthesized and secreted by bovine

conceptuses (days 13 to 27) include steroids (Shemesh et

al., 1979; Chenault, 1980; Gadsby et al., 1980; Eley et al.,

1985; Chapter 2), prostaglandins (Shemesh et al., 1979;

Lewis et al., 1982; Lewis, 1984; Lewis and Waterman, 1984),

and proteins (Masters et al., 1982; Bartol et al., 1985;

Chapter 4). This subject has been reviewed recently by

Thatcher et al. (1984a, b; 1985) with respect to maternal

recognition of pregnancy in cattle.

In several studies, higher plasma P4 concentrations

were detected in pregnant versus bred-nonpregnant or cyclic

cattle from as early as days 9 to 10 post-estrus (Henricks

et al., 1971; Ford et al., 1979; Lukaszewska and Hansel,

1980), suggesting a luteotropic role of the concepts during

early pregnancy. However, pregnancy-related elevations in

P4 concentrations were not supported by others (Batson et
al., 1972; Folman et al., 1973; Hasler et al., 1980).

Progesterone production by dispersed luteal cells was

stimulated by homogenates and aqueous extracts of day 18

bovine conceptuses (Beal et al., 1981). Progesterone

stimulatory activity of concepts extracts was lost

following heating or removal by dialysis of components less

than a relative molecular weight (Mr) of 12,000 Daltons.

Similarly, the elongating ovine concepts contains products

which stimulate luteal P4 production, in vitro (Godkin et

al., 1978). The products) responsible for this luteotropic

activity in ewes is not proteinaceous since 100 pg of

concepts protein had no effect on synthesis of P4 or cyclic

AMP by dispersed ovine luteal cells (Ellinwood et al.,

1979). Furthermore, no LH-like activity was detected in

bovine (day 16 to 20; Henricks and Poffenbarger, 1984) or

ovine (day 14 to 15; Ellinwood et al., 1979) conceptuses as

demonstrated under bioassay and RIA conditions,

respectively. Thus the presumptive luteotropin(s) may be a

steroid or prostaglandin. Marsh (1970) demonstrated that

PGE2 stimulated bovine luteal production of P4 in vitro.

However, evaluation of in vivo PGE2 effects on bovine luteal

function and maintenance are not clearly indicated.

Prostaglandin-E2 administered into the uterine lumen alone

(Gimenez and Henricks, 1983; Chenault, 1983) or in

combination with E2 (Reynolds et al., 1983) extends luteal

function only slightly beyond cessation of intrauterine PGE2

treatments. No stimulation of luteal P4 production was

noted in vivo. Systemic P4 concentrations declined within

12 h following cessation of PGE2 treatment (Gimenez and

Henricks, 1983) or during the treatment period (Reynolds et

al., 1983) suggesting that PGE2 (plus E2) will not prevent

production and transfer of uterine luteolytic substances,

but affects luteal function directly (Marsh, 1970; Henderson

et al., 1977; Reynolds et al., 1981). Others have reported

no effect of intrauterine PGE2 administration on CL

maintenance in cattle (Dalla Porta and Humblot, 1983;

Chenault et al., 1984). As in cattle, reports in sheep

indicate that luteal function may be prolonged with PGE2 for

only short periods beyond the time of normal luteolysis

(Pratt et al., 1979; Huie et al., 1981; Magnuss et al.,

1981). In sheep, numerous studies have demonstrated that

the CL becomes refractory to exogenous PGF2a during early

pregnancy (Inskeep et al., 1975; Mapletoft et al., 1976b;

Pratt et al., 1977; Silvia and Niswender, 1984) or following

PGE2 administration (Pratt et al., 1977; Henderson et al.,

1977; Reynolds et al., 1981). Furthermore, PGE2 is
synthesized by bovine (Shemesh et al., 1979; Lewis et al.,

1982) and ovine (Hyland et al., 1982; Lacroix and Kann,

1982) conceptuses and endometrium. Luteal protective

substances, originating from the gravid uterine horn in

cattle, were demonstrated by Del Campo et al. (1980). In

this study, uterine horns were isolated surgically and

embryos transferred to uterine horns contralateral or

ipsilateral to the CL containing ovary. In three or four

cases, pregnancies in the ipsilateral horn maintained CL

viability, whereas all pregnancies on the contralateral side

failed to maintain CL function. However, when the vein

draining the uterus of the contralateral, pregnant horn was

anastomosed surgically with the ipsilateral, nonpregnant

uterine venous drainage, luteal regression was prevented in

three of three cows. Thus, it was apparent that some blood-

borne products) of pregnancy was able to circumvent the

luteolytic effects of the nonpregnant uterine horn. Similar

findings were reported in the ewe (Mapletoft et al.,

1976b). These results supported previous data demonstrating

existence of a unilateral uteroovarian relationship (veno-

arterial pathway) during early pregnancy (Del Campo et al.,

1977) and luteolysis (Ginther, 1974; Ginther et al., 1974)

in cattle. Collectively, data generated in cattle and sheep

suggest that concepts (and/or uterine)-derived luteal

protective substances provide the CL with some resiliency

against PGF2a-induced luteolysis during early pregnancy.

Evidence supports PGE2 as a possible luteal protective

product of conception.

Movement of substances to and from the pregnant uterus

and ipsilateral ovary may be regulated, to some degree, by

blood flow rates and intraovarian vascular hemodynamics

(Lamond and Drost, 1974; Ford and Chenault, 1980). During

early pregnancy (days 14 to 18) in cattle, blood flow to the

gravid uterine horn is transiently increased (Ford et al.,

1979; Ford and Chenault, 1980; Ford, 1982). Estrogens and

PGE2 are known to stimulate uterine blood flow in cattle

(Roman-Ponce et al., 1978; Thatcher et al., 1984b; Chapter

3). Elevated luminal concentrations of estrogen and PGE2 in

the gravid uterine horn (Bartol et al., 1981; Ford, 1982;

Lewis et al., 1982) and evidence supporting synthesis of

estrogens (Shemesh et al., 1979; Gadsby et al., 1980; Eley

et al., 1983; Chapter 2) and PGE2 (Lewis et al., 1982) by

bovine conceptuses during early pregnancy support a role for

these concepts products in pregnancy-associated elevations

in uterine blood flow. Recent reports by Ford (1982) and

Ford and Reynolds (1983) suggested that hydroxylated

metabolites of estrogens viz., catecholestrogens, may

mediate uterine blood flow responses to follicular-(estrus)

and conceptus-derived (early pregnancy) E2. They propose

that an antagonistic interaction of catecholestrogens with

uterine vascular a-adrenergic receptors promote reduced

vascular contractility and facilitate elevated rates of

uterine blood flow. Additionally, increased endometrium

vascular permeability was suggested in pregnant ovine uteri

as evidenced by positive Pontamine Blue dye uptake (Boshier,

1970). Local induction of uterine blood flow and vascular

permeability by the concepts may increase transfer

efficiency of luteotropic/luteal protective substances from

the pregnant uterus to the CL (Reynolds et al., 1935;

Thatcher et al., 1984b, 1985). However, evaluation of

tritiated PGF2a uptake in bovine endometrium and

uteroovarian vascular components suggested that PGF2a

permeability may be decreased during early pregnancy

(Thatcher et al., 1984b). Tissue samples, viz.,

endometrium, ovarian vein and ovarian artery, were collected

on day 17 of pregnancy or the estrous cycle from the side

ipsilateral to the CL-bearing ovary. Data demonstrated

reduced PGF2a permeability in pregnant versus cyclic tissues

on day 17. Furthermore, ovarian vein permeability to PGF2a

was less than ovarian artery in pregnant cows, however, this

relationship was reversed in vascular samples collected on

day 17 of the estrous cycle. In pregnant tissues, responses

suggest that concepts induced reductions in vessel PGF2c

permeability were greater with regard to vessel proximity to

the pregnant uterus. It may be hypothesized that concepts

products responsible for reduced PGF2a permeability would be

more concentrated in the uterine venous drainage (ovarian

vein) versus ovarian artery, thus resulting in the responses

observed (Thatcher et al., 1984b). Further studies are

required to determine if this pregnancy-associated reduction

in tissue permeability is specific for PGF2a. For example,

elevated PGF2a and PGE2 contents of uterine flushings from

pregnant versus cyclic cattle (Bartol et al., 1981; Lewis et

al., 1982) may suggest sequestration within the pregnant

uterine lumen.

Another strategy by which the concepts maintains CL

function involves attenuation of PGF2a production by the

uterine endometrium. In one of several studies, reviewed by

Thatcher et al. (1984a, b; 1985), in vitro production of

PGF2a by day 17 pregnant endometrial explants was reduced by

more than 50% of PGF2a produced by day 17 cyclic endometrial

explant cultures. Production of PGE2 was similar for

pregnant and cyclic endometrium. Endometrial prostaglandin

synthesis was stimulated following addition of exogenous

arachidonic acid to cultures, however, relative proportion

of prostaglandins (PGF2a, PGFM and PGE2) within and between

pregnancy status were unchanged. These data support an

antiluteolytic effect of the bovine concepts on endometrial

PGF2a production. Levels of arachidonic acid cascade which

appear to be modified during early pregnancy involve phos-

pholipase A2 (i.e., exogenous arachidonic acid stimulated

prostaglandin synthesis) and cyclooxygenase (i.e., PGF2a

production was attenuated in pregnant endometrium with and

without the addition of arachidonic acid). Although PGE2

production did not differ between pregnancy status, it is

clear that the ratio of PGE2:PGF2a must become elevated with

suppression of PGF2a production during early pregnancy.

Such an increase in the PGE2:PGF2a ratio may be of

physiological importance. In vivo evaluation of temporal

PGF2a episodes in the ovarian arterial supply of cows from

days 17 to 20 of the estrous cycle or pregnancy (Thatcher et

al., 1984b) supported in vitro results demonstrating a

reduction in PGF2a during pregnancy. No differences were

observed in frequency of PGF2a episodes between pregnancy

status. However, plasma concentrations of PGF2a during

episodic release were attenuated dramatically in pregnant


Peripheral patterns of PGFM concentrations were used as

an index of uterine PGF2. episodic production during the

bovine estrous cycle (Peterson et al., 1975; Kindahl et al.,

1976), early pregnancy (Kindahl et al., 1976), and following

embryo transfer on day 16 post-estrus (Betteridge et al.,

1984). Luteal regression, during the estrous cycle and

following embryonic mortality, as determined by declining

plasma P4 concentrations, was always associated temporally

with three to five large episodes of PGFM. Conversely, PGFM

episodes are reduced or completely eliminated in early

pregnancy. Similar observations were also reported during

the estrous cycle and early pregnancy in buffaloes (Batra

and Pandey, 1983). The 11-ketotetranor metabolites of PGF2a

have been evaluated recently in 24 hour, pooled urine

samples of cyclic and pregnant cattle (Harvey et al., 1984;

Plante et al., 1984). Although detection of acute patterns

of uterine prostaglandin production are not possible using

this system, clear elevations of 11-ketotetranor PGF2a are

temporally associated with a decline in plasma P4

concentrations during luteolysis. Peaks of tne urinary

metabolite are absent during early pregnancy, however, basal

concentrations begin a gradual increase as pregnancy

proceeds beyond 18 to 20 days. Similarly, day 17 pregnant

heifers had higher basal concentrations of PGFM than

nonpregnant heifers (Williams et al., 1983). It is probable

that concepts prostaglandin biosynthesis during early

pregnancy contributes to elevated urinary and plasma

metabolites of PGF2a. Conceptus prostaglandin production

increases wizn gestational age in cattle (Lewis et al.,

1982). It should be noted, however, that elevations in

baseline prostaglandin concentrations during early pregnancy

are not characteristic of luteolytic pulses detected during

CL regression (Thatcher et al., 1984a). Additional support

for conceptus-derived anti-luteolytic effects in cattle was

reported following E2-induction of uterine PGF2a production

on day 18 of the estrous cycle, and day 18 and 20 of

pregnancy. The capacity of the uterus to produce PGF2a

(peripheral plasma PGFM) in pregnant cattle was reduced

significantly from day 18 cyclic cattle responses (Thatcher

et al., 1984b). Reduction in E2-induced PGFM response was

more dynamatic in day 20 versus day 18 pregnant cattle

suggesting that as concepts elongation proceeds,

progressively larger regions of endometrium become exposed

to concepts antiluteolytic products, resulting in a graded

decline in total uterine capacity to produce PGF2,.

Additionally, duration of endometrial exposure co concepts

antiluteolytic signals may be important in this regard.

Inhibition or reduction of uterine PGF2a production and

secretary patterns may be the primary mechanism by which the

developing bovine concepts ensures CL maintenance during

early pregnancy. Luteotropic and luteal protective agents,

permeability and blood flow alterations may act as secondary

or support systems to further reduce, dilute, or otherwise

protect CL from luteolytic effects of the uterus. This

argument is based on the inability of putative luteotropic

and luteal protective signals to maintain CL function for

extended periods in cyclic cattle. In contrast, recent

reports in cattle and sheep provide evidence that

antiluteolytic proteins of concepts origin significantly

extend CL lifespan and function.

Data evaluating in vitro protein synthesis and

secretion by elongating conceptuses are available for cattle

(Masters et al., 1982; Bartol et al., 1985), sheep (Godkin

et al., 1982b; 1984a) and swine (Godkin et al., 1982a). It

becomes evident from these studies that qualitative and

quantitative changes in array of proteins produced are

related to stage of concepts development. In all three

species, total concepts protein production increases with

gestational age (Godkin et al., 1982a, b; Bartol et al.,

1985). This trend is similar to data reported for bovine

concepts steroid (Shemesh et al., 1979; Chenault, 1980;

Gadsby et al., 1980; Eley et al., 1983; Chapter 2) and

prostaglandin (Lewis et al., 1982) biosynthetic activity.

Additionally, apparent total protein production per mg of

concepts wet weight, in cattle, becomes elevated as rapid

trophectodermal elongation is initiated between days 16 and

17 of gestation (Knickerbocker et al., 1984; Chapter 4).

During this period of concepts development, a distinct, but

similar family of low Mr, acidic polypeptides constituted a

major portion of proteins secreted into medium (cattle: Mr

22-26,000, pi 5.6-6.5; Bartol et al., 1985; sheep: Mr 17-

21,000, pi 5.5; Godkin et al., 1982b; Martal et al., 1984b;

swine: Mr 20-25,000, pi 5.6-6.2; Godkin et al., 1982a).

Transient concepts production of low Mr polypeptide species

during respective periods of concepts elongation and

maternal recognition of pregnancy in cattle, sheep and swine

suggest that these products may be involved in early

pregnancy maintenance. Experimental evidence in the cow and

ewe support this hypothesis.

Rowson and Moor (1967) demonstrated that intrauterine

administration of day 14 to 15 ovine concepts homogenates

prolonged estrous cycles in ewes. However, no effect on

estrous cycle length was observed following intrauterine

administration of day 25 ovine concepts homogenates or heat

treated, day 12 to 14 ovine conceptuses. Similarly,

introduction of homogenates or extracts of day 14 to 16

ovine conceptuses into the uterine lumen resulted in

prolonged CL function and estrous cycles in eight cf 12 ewes

(Martal et al., 1979). Only one of six ewes ex:hioited

prolonged luteal function and interestrous interval when day

21 to 23 concepts homogenates were administered. Pronase

or heat pretreatment completely eliminate the ability of day

14 to 16 concepts homogenates to extend CL function in the

cyclic ewe. Subsequent characterization of protein

production patterns by elongating bovine (days 16 to 24;

Bartol et al., 1985), ovine (days 13 to 21; Godkin et al.,

1982b), and porcine (days 10.5 to 12; Godkin et al., 1982a)

conceptuses verified transient biosynthesis and secretion of

major, low Mr polypeptides during early pregnancy.

Collectively, these data imply that biologically active

proteins of concepts origin are involved in processes

leading to maternal recognition of pregnancy.

Recent data demonstrated that pooled concepts

secretary proteins (CSP), collected from medium of cultured

bovine (days 16 to 18; Chapter 4) and ovine (days 15 and 16;

Godkin et al., 1984b) conceptuses, substantially extend CL

function and interestrous interval following intrauterine

administration to cyclic cattle (days 15 through 21) and

sheep (days 12 through 21), respectively. In addition,

intrauterine administration of the prominent, low Mr, acidic

polypeptide, termed ovine trophoblast protein-1 (oTP-1;

Godkin et al., 1984b), also extended luteal function in

cyclic ewes. Thus, oTP-1 may be a primary protein signal

responsible for luteal maintenance in sheep. Conceptus

protein signals probably act locally at the level of uterine

endometrium. For example, immunohistochemical data suggest

that oTP-1 is secreted by extraembryonic trophectoderm of

the ovine concepts and binds specifically to receptors

localized on uterine endometrial epithelium (Godkin et al.,

1984a). In the same report, oTP-1 was shown to stimulate

secretion of specific endometrial proteins in culture. In

cattle (Chapters 4 and 5) and sheep (Fincher et al., 1984;

Fincher, 1984), pooled CSP dramatically reduced spontaneous

uterine PGF2a episodes (Chapter 4) and E2-induced production

of uterine PGF2a (peripheral plasma PGFM response; Fincher

et al., 1984; Fincher et al.; Chapter 5). Collectively,

these data demonstrate a role for conceptus-derived

secretary proteins as signals during early pregnancy in

cattle and sheep. It is proposed that proteinaceous

concepts signals interact with the uterine endometrium such

that endometrial capacity to synthesize and secrete PGF2a is

dampened. At this time, it is not known whether CSP

antiluteolytic effects are exerted directly or mediated by

some induced endometrial product (Godkin et al., 1984a).

Direct antiluteolytic effects of CSP may include regulation

of steroid and/or oxytocin receptor populations on

endometrium. In the rat, both E2 and P4 nuclear receptors

are increased at the site of blastocyst implantation on day

5 and 6 of gestation (Logeat et al., 1980). Findley et al.

(1982) reported a decline in caruncular E2 receptors

associated with side of concepts on day 15 of pregnancy in

ewes. Additionally, caruncular and intercaruncular

endometrial production of protein in pregnant ewes was

greater than nonpregnant endometrium. A reduction in

endometrial E2 receptors may prevent E2-induction of

oxytocin receptors and subsequent activation of PGF2a

synthesis (McCracken et al., 1984). Replenishment of

endometrial P4 receptors could also result in decreased

nuclear accumulation of E2 receptor. In this light, Fincher

et al. (1984) demonstrated that oxytocin-induced uterine

PGF2a production in E2 primed ewes was decreased

significantly by intrauterine CSP administration.

Thatcher et al. (1984b) reported that day 17 pregnant

endometrial explants produced less PGF2a than cyclic

endometrium in the presence or absence of exogenous

arachidonic acid. They suggested that both phospholipase A2

and cyclooxygenase activity was reduced during early

pregnancy. Wlodawer et al. (1976) and Shemesh et al. (1981,

1984) reported the presence of prostaglandin inhibitors in

bovine endometrium and placental tissues, respectively. It

is possible that induction of such inhibitors during early

pregnancy may involve concepts protein signals. In such a

scheme, endometrial PGF2a biosynthesis may be selectively

inhibited (possibly at the level of phospholipase A2 and/or

cyclooxygenase) without influencing prostaglandin production

by the concepts (Thatcher et al., 1985).

In cattle, low molecular weight concepts polypeptides,

described by Bartol et al. (1985), may play key roles in

pregnancy maintenance as was demonstrated for oTP-1 in sheep

(Godkin et al., 1984a, b). Similarities in the nature and

function of concepts protein signals in cattle and sheep

have been discussed. Relative to this discussion, Martal et

al. (1984a) provided evidence that concepts signals

(presumably protein in nature) of cattle may be biologically

active in sheep and vice-versa. Two trophoblastic vesicles

(0.5 to 2 mm diameter), composed of extraembryonic

trophectoderm and endoderm (embryonic disc removal), from

day 13 bovine blastocysts were transferred to 11 ewes on day

12 of the estrous cycle. Two ewes exhibited prolonged

luteal maintenance until at least day 38 when ewes were

slaughtered. Interestingly, an elongated (>100 mm)

trophoblastic vesicle was recovered from the uterus in one

of these ewes, although concepts membranes were undergoing

necrosis at this time. Similarly, when two trophoblastic

vesicles from day 11 to 13 ovine blastocysts were

transferred to ten cyclic cattle on day 12, two heifers

exhibited cycle lengths of 31 and 36 days. The authors

suggested that nonspecific concepts signals in the cow and

ewe were sufficient for maintaining CL function and the


biologically active molecules responsible for CL maintenance

in these species may be very similar.

As discussed earlier in this review, concepts

production of biologically active signals is closely

synchronized with concepts elongation during maternal

recognition of pregnancy. Furthermore, precise

synchronization between development of the concepts and

maternal uterine milieu are essential for successful

membrane elongation and signal emission by the concepts.

Immunological rejection of transferred interspecie concepts

membranes undoubtedly discourages this process (Allen, 1979;

Beer and Billingham, 1979; Rossant et al., 1982; Wielen and

King, 1984). In this light, observed extension of CL

function in 20% of cattle and sheep receiving interspecific

trophoblastic vesicle transfers (Martal et al., 1984a) seems

even more remarkable. Recent data (S.D. Helmer, W.W.

Thatcher, F.W. Bazer, P.J. Hansen and R.M. Roberts, 1984,

unpublished observations; see Thatcher et al., 1985) lend

additional support that concepts antiluteolytic protein

signals may be similar in cattle and sheep. Antibodies

developed against oTP-1 (Godkin et al.., 1984a) were used in

an ouchterlony double immuno-diffusion system with

unfractionated bovine concepts secretary proteins. Two

precipitin bands were formed against unfractionated bovine

coneptus secretary proteins, one of which showed partial

identity with either purified oTP-1 or unfractionated ovine

concepts secretary proteins. Subsequently, an

immunoprecipitate of tritiated bovine concepts secretary

protein fraction which crossreacted with oTP-1 antiserum was

solubilized and analyzed by polyacrylamide gel

electrophoresis and fluorography. Immunoprecipitated bovine

concepts secretary proteins were detected in the 20,000 and

22,000 Mr range. Collectively, these data suggest that

analogous mechanism of "Maternal Recognition of Pregnancy,

possibly involving the low Mr, acidic polypeptide species,

may exist in cattle and sheep.

The concepts may also exert an indirect antiluteolytic

effect as a result of subtle suppression and acceleration in

folliculogenesis and atresia, respectively (Dufour et al.,

1984). Such a concepts interaction may be important since

exogenous estrogens have been shown to stimulate uterine

PGF2a production in cattle (Chapter 3) resulting in luteal

regression (Greenstein et al., 1958; Wiltbank, 1966; Eley et

al., 1979). Estrogens may also act directly on the CL to

antagonize LH stimulated P4 production (Williams and Marsh,

1978). Conversely, CL function was extended following

destruction of ovarian follicles during the luteal phase of

the bovine estrous cycle (Villa-Godoy et al., 1981). Thus,

concepts regulation of follicular development and hence,

estrogen production may complement the intrauterine

antiluteolytic efforts of the concepts to suppress uterine

PGF2a synthesis during early gestation.

In conclusion, events associated with "Maternal

Recognition of Pregnancy" reflect dynamic as well as subtle

alterations in maternal and concepts physiology. Numerous

putative concepts signals have been identified, although

their precise roles in "Maternal Recognition of Pregnancy"

are just beginning to be understood. Maintenance of CL

function and an embryotrophic uterine milieu during early

pregnancy is accomplished by several interdependent,

conceptus-mediated strategies. Evaluation and clarification

of the complex interactions between concepts and its

maternal host presents a considerable and exciting task for

the future.



Maternal recognition of pregnancy (Short, 1969) in

cattle occurs by day 16 (Betteridge et al., 1980; Northey

and French, 1980; Dalla Porta and Humblot, 1983). During

this period the bovine concepts initiates a phase of rapid

elongation (Hawk et al., 1955; Greenstein and Foley, 1958b;

Betteridge et al., 1980) and elevated endocrine activity

(Shemesh et al., 1979; Chenault, 1980; Lewis et al., 1982;

Eley et al., 1983; Bartol et al., 1985). Conceptus

endocrine products, in turn, direct events which are

essential for the establishment and maintenance of

pregnancy. Biosynthesis and possible mechanisms of action

for conceptus-derived prostaglandin and protein signals have

been discussed earlier (see: Literature Review). Recent

reports demonstrate the capacity for synthesis and

metabolism of various steroids by elongating bovine

conceptuses (Shemesh et al., 1979; Chenault, 1980; Gadsby et

al., 1980; Eley et al., 1983). Based on these studies it is

clear that small amounts of estrogen biosynthesis results

following concepts incubation with androgen precursors

(androstenedione, A4; testosterone, T; dehydroepiandros-

terone, DHA). In addition, Shemesh et al (1979) reported

detectable amounts of immunoreactive P4, T and estradiol-178

(E2) in some bovine concepts extracts on days 15 and 16 of

gestation. Total content of these steroids were elevated in

medium following 48 hour concepts cultures in the absence

of exogenous precursors. Thus the bovine concepts may

utilize C21 steroids, viz., P4 and pregnenolone, as

precursors for androgen and estrogen production.

Conversely, Eley et al. (1983) were unable to demonstrate

tritiated-P4 (10 ng) conversion to estrogens when shorter

culture periods (3 to 6 hours) and less sensitive procedures

of identification (liquid column chromatography and

recrystallization) were utilized. Currently, no data are

available with regard to relative ratios of estrone (E1), E2

and estriol (E3) production by the early bovine concepts.

Indirect physiological data support concepts estrogen

biosynthesis during maternal recognition of pregnancy.

Blood flow to the gravid uterine horn is transiently

elevated between days 14 and 18 of gestation in cattle (Ford

et al., 1979; Ford and Chenault, 1981). Uterine blood flow

during the bovine estrous cycle and early pregnancy are

highest during periods of low P4:E2 ratios in plasma.

Furthermore, exogenous estrogens stimulate uterine blood

flow in cyclic cattle (Roman-Ponce et al., 1978; Chapter 3).

The bovine concepts possesses an extremely active 58-

reductase enzyme system (Chenault, 1980; Eley et al., 1983)

whereas the endometrium produces primarily 5a-reduced

steroids (Eley et al., 1983). Metabolites identified were

hydroxylated in an a-configuration at the 3 and/or 17 (A4

substrate) and 20 (P4 substrate) positions. Roles of 58-

reduced steroids during early pregnancy in cattle have not

been studied.

It was the objective of this study to characterize

patterns of P4 metabolism to 58/a-reduced products and

estrogens by bovine concepts and endometrial tissues during

early pregnancy and to identify major concepts metabolites

for future evaluation of physiological roles during early

pregnancy (Chapter 4).

Materials and Methods


Minimum essential medium (MEM; GIBCO, Grand Island, NY)

supplemented with non-essential amino acids (GIBCO, Grand

Island, NY), antibiotic/antimycotic (GIBCO, Grand Island,

NY), 200 units of insulin/L (Sigma Chemical Company, St.

Louis, MO) and 1 g glucose/L (Fisher Scientific, Orlando,

FL) was used for rinsing and incubating concepts and

endometrial tissues. All radiolabelled steroids were

purchased from New England Nuclear, Boston, MA. Tritiated