expenrnental and clinical endocrinology &diabetes · a. early work on the endocrinology and...

Exp Clin Endocrinol Diabetes 104 (1996) 183-204

The history of assisted human conception with especial reference toendocrinology

R. G. Edwards

London Women's Clinic, London; Churchill College, Cambridge, UK

Key words: Fertilization, history, human, in vitro, endo-crinology

Summary: This review lecture is primarily concernedwith the study of assisted human conception and es-pecially in-vitro fertilization (IVF). It also places in per-spective the role of endocrinology in history of IVF. Aknowledge of the hypothalamic-pituitary control of ovu-lation, and of ovarian follicle dynamics is assumed. Adetailed consideration of these topics, together withextensive references, are available in a recent textbook(Edwards and Brody, 1995).

A. Early work on the endocrinology andembryology of assisted human conception

Introduction of controlled ovarian hyperstimulationin female mammals and in 'omen

Initial concepts on the introduction of IVF into as-sisted human conception began in Edinburgh, dur-ing studies on experimental embryology in mice(Edwards, 1957). During these studies the occur-rence of ovulation during the natural oestrous

"Berthold Memorial Lecture" of the german Society ofEndocrinology, Leipzig, 1995.

Expenrnental and Clinical

Endocrinology&Diabetes

© 1996 Johann Ambrosius Barth

Many of the early pioneers studying animal repro-duction combined reproductive physiology and endo-crinology, especially Marshall, who analysed oestrousand menstrual cycles in many mammalian species. Theclarification of the roles of pituitary gland and hypo-thalamus in the menstrual cycle and ovulation, and theirregulation by steroidal feedbacks from the gonads gavean immense stimulus to studies on human reproduction(Smith and Engle, 1927; Lewis and Gregory, 1933;Harris, 1970). Three periods of research into assisted hu-man conception are covered in this lecture including theinitial work on the introduction of the endocrinologyand embryology of human IVF, the rapid advances intechnique as it expanded worldwide, and finally some ofthe recent remarkable advances in the field.

cycle was assessed from the examination of serialdaily vaginal smears. Studies on embryology weremade difficult by the need to use this cumbrous andinefficient method of obtaining gametes and achiev-ing fertilization. It was also inadequate to yield suf-ficient freshly-ovulated oocytes and embryos forstudy on topics suc te experimental induction ofgynogenesis, androgenesis and heteroploidy inmouse embryos (Beatty, 1957; Edwards, 1957). Es-sentially, some form of ovarian stimulation wasneeded to stimulate the controlled growth of severalGraafian follicles and induce a timed ovulation.

The work of Ascheim, Zondek and many othersin the 1920's and 1930's had led to the availability

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.

184 Exp Clin Endocrinol Diabetes 104 (1996) 3

Fig. 1 The upper picture shows a pregnant mousewhich was stimulated by Gestyl (pregnant mares' serum,Organon, Oss) and Pregnyl (human chorionic gonado-trophin, Organon, Oss); a non-pregnant female is shownin the lower part of the figure. Many multiple pregnan-cies resulted after this method of inducing controlled su-perovulation in adult mice, with up to 50 or more im-plants in some females (Fowler and Edwards, 1957).

of partially purified gonadotrophin preparations,prepared from extracts of various biologicalsources such as pregnant mare's serum (PMS) andhuman chorionic gonadotrophin (HCG). Severalauthors had used various methods of inducingsuperovulation and oestrus in immature animals,e.g. in mice (Evans and Simpson, 1940; Pincus,1940; Parkes, 1942; Runner and Gates, 1954). Theoverall conclusion was that immature animalscould be induced to superovulate and enteroestrus, but not mature females. Indeed, in manystudies, oocytes or embryos were collected fromimmature females after ovarian stimulation, andtransferred to adult female recipients to continuetheir growth to full term (Gates and Beatty, 1954).

The first successful application of any methodof controlled ovarian hyperstimulation in adultmammals was carried out in mice. This resultedfrom the frustrations of using the natural oestrouscycle to obtain mouse oocytes and embryos, asoutlined above. Controlled ovarian stimulation wasincredibly successful (Fowler and Edwards, 1957).An injection of PMS to stimulate follicle growth,followed 40 hr later by HCG induced multiple fol-licular growth, oestrus in the great majority of fe-males, superovulation, mating and fertilization.Many full-term multiple pregnancies and deliverieswere obtained in the adult mice (Figure 1). Theseinjected hormones imposed a strict uniformity onfolliculogenesis, the time of oestrus, oocyte matu-ration and ovulation (Figure 2) (Edwards & Gates,1959). All stages of oocyte maturation occurred atprecise times with hardly any variability betweenfemales. Oestrus began at 8 hours approximately,

and virtually all stimulated mouse females ovu-lated at 12 hours post-HCG. In some high multi-pregnancies, the combined weight of 40 or morefetuses exceeded the mother's body weight. Theseresults led to the first concepts of inducing timedsuperovulation in women and animal species.

A year or so later, Gemzell began his work onsuperovulation in adult amenorrhoeic womenusing human pituitary extracts to stimulate folliclegrowth and HCG to invoke ovulation (Gemzell,1967). Lunenfeld then introduced the clinical useof preparations of human menopausal gonado-trophins (HMG), prepared by Donini, as follicularstimulants for amenorrhoeic women (Lunenfeld,1969), and an alternative approach utilizing clomi-phene was introduced a few years later (Kistner,1972). Similar results to those in mice were ob-served, with many high-order multiple pregnan-cies.

During those years, the methods of ovarianstimulation pioneered by Gemzell and his col-leagues offered no chance of obtaining humanoocytes for studies on maturation and fertilization.There was only limited information on the numberof follicles that had been stimulated to grow, andnone on the timing of human ovulation after theinjection of HCG. Nor was it known if cyclicwomen would respond to these forms of ovarianstimulation. The necessary surgical methods topermit a simple approach to the ovary had notbeen pioneered, so that laparotomy would beneeded for this purpose. These early advances wereastonishing at the time, but each of the associatedproblems had to be solved before any attemptcould be made to introduce human IVF.

Knowledge and oocytes and embryos in the 1960s

Many early pioneers had established a funda-mental basis of animal embryology. In 1890, Heapewas sufficiently familiar with the collection andshort-term culture of rabbit ova and with theirtransplantation to adult recipients as to obtainfull-term offspring from the transferred ova. Pre-implantation embryos of rabbit and monkeys hadbeen cultured, and had developed through cleavageand blastulation in vitro (Lewis and Gregory, 1933;Lewis and Hartman, 1933). Many investigatorshad studied human oocytes and embryos, as shownespecially by the detailed manuscripts in the Car-negie Institute Contributions to Embryology andother journals (e.g. Hertig et al., 1954; Menkin andRock, 1948; Shettles, 1955). Methods of tissue cul-ture were being steadily improved, and many of themodern methods had been introduced by the1950's. Defined media for the culture of mam-


Fig. 1 The upper picture shows a pregnant mousewhich was stimulated by Gestyl (pregnant mares' serum,Organon, Oss) and Pregnyl (human chorionic gonado-trophin, Organon, Oss); a non-pregnant female is shownin the lower part of the figure. Many multiple pregnan-cies resulted after this method of inducing controlled su-perovulation in adult mice, with up to 50 or more im-plants in some females (Fowler and Edwards, 1957).

of partially purified gonadotrophin preparations,prepared from extracts of various biologicalsources such as pregnant mare's serum (PMS) andhuman chorionic gonadotrophin (HCG). Severalauthors had used various methods of inducingsuperovulation and oestrus in immature animals,e.g. in mice (Evans and Simpson, 1940; Pincus,1940; Parkes, 1942; Runner and Gates, 1954). Theoverall conclusion was that immature animalscould be induced to superovulate and enteroestrus, but not mature females. Indeed, in manystudies, oocytes or embryos were collected fromimmature females after ovarian stimulation, andtransferred to adult female recipients to continuetheir growth to full term (Gates and Beatty, 1954).

The first successful application of any methodof controlled ovarian hyperstimulation in adultmammals was carried out in mice. This resultedfrom the frustrations of using the natural oestrouscycle to obtain mouse oocytes and embryos, asoutlined above. Controlled ovarian stimulation wasincredibly successful (Fowler and Edwards, 1957).An injection of PMS to stimulate follicle growth,followed 40 hr later by HCG induced multiple fol-licular growth, oestrus in the great majority of fe-males, superovulation, mating and fertilization.Many full-term multiple pregnancies and deliverieswere obtained in the adult mice (Figure 1). Theseinjected hormones imposed a strict uniformity onfolliculogenesis, the time of oestrus, oocyte matu-ration and ovulation (Figure 2) (Edwards & Gates,1959). All stages of oocyte maturation occurred atprecise times with hardly any variability betweenfemales. Oestrus began at 8 hours approximately,

and virtually all stimulated mouse females ovu-lated at 12 hours post-HCG. In some high multi-pregnancies, the combined weight of 40 or morefetuses exceeded the mother's body weight. Theseresults led to the first concepts of inducing timedsuperovulation in women and animal species.

A year or so later, Gemzell began his work onsuperovulation in adult amenorrhoeic womenusing human pituitary extracts to stimulate folliclegrowth and HCG to invoke ovulation (Gemzell,1967). Lunenfeld then introduced the clinical useof preparations of human menopausal gonado-trophins (HMG), prepared by Donini, as follicularstimulants for amenorrhoeic women (Lunenfeld,1969), and an alternative approach utilizing clomi-phene was introduced a few years later (Kistner,1972). Similar results to those in mice were ob-served, with many high-order multiple pregnan-cies.

During those years, the methods of ovarianstimulation pioneered by Gemzell and his col-leagues offered no chance of obtaining humanoocytes for studies on maturation and fertilization.There was only limited information on the numberof follicles that had been stimulated to grow, andnone on the timing of human ovulation after theinjection of HCG. Nor was it known if cyclicwomen would respond to these forms of ovarianstimulation. The necessary surgical methods topermit a simple approach to the ovary had notbeen pioneered, so that laparotomy would beneeded for this purpose. These early advances wereastonishing at the time, but each of the associatedproblems had to be solved before any attemptcould be made to introduce human IVE

Knowledge and oocytes and embryos in the 1960s

Many early pioneers had established a funda-mental basis of animal embryology. In 1890, Heapewas sufficiently familiar with the collection andshort-term culture of rabbit ova and with theirtransplantation to adult recipients as to obtainfull-term offspring from the transferred ova. Pre-implantation embryos of rabbit and monkeys hadbeen cultured, and had developed through cleavageand blastulation in vitro (Lewis and Gregory, 1933;Lewis and Hartman, 1933). Many investigatorshad studied human oocytes and embryos, as shownespecially by the detailed manuscripts in the Car-negie Institute Contributions to Embryology andother journals (e.g. Hertig et al., 1954; Menkin andRock, 1948; Shettles, 1955). Methods of tissue cul-ture were being steadily improved, and many of themodern methods had been introduced by the1950's. Defined media for the culture of mam-

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.

Before, 1955

Gemzell, 1967

Lunenfeld, 1996

Various sources, 1990's

Various sources, 1990's

Table O Human gonadotrophic preparations

Human chorionic gonadotro-phin available commercially

Anterior pituitary extracts tostimulate follicle growth

Urinary extracts to stimulatefollicle growth

Purified FSHRecombinant FSH/LH

malian embryos were steadily improved duringthose years (Hammond, 1949; Paul, 1961; Whittenand Biggers, 1968).

Initial work on the introduction of IVF was ledby studies on the maturation of human oocytes re-leased from their follicles into culture medium. IfIVF was to be successful, mature human oocyteswould be needed to achieve fertilization and nor-mal embryonic growth. Pincus and his colleaguesinitiated the study of oocyte maturation in vitro inthe 1930's. Rabbit oocytes required 11-12 hoursto mature, and the maturation of human oocytesin vitro was thought to require 12 hours (Pincusand Saunders, 1939). As it happened, this figurewas incorrect, and it probably led to later investi-gators adding spermatozoa to human oocyteswhich had been matured in vitro for 12 hours(Menkin and Rock, 1948; Shettles, 1955). This er-ror, and possibly the general lack of understandingof the composition and handling of culture media,must have hampered these investigators.

Studies on oocyte maturation on several mam-malian species were carried out in the late 1950'sand early 1960's. They showed beyond doubt howeach species had its own unique duration of oocytematuration from the germinal vesicle stage untilthe formation of metaphase of the second meiotic

Fig. 3 Estimates of the time required for oocyte matu-ration from diakinesis to metaphase 2 in various species.The timing of maturation in vivo was subsequentlyshown to be similar in each species (Edwards 1965a,b).

division (metaphase 2) (Edwards, 1965a). More-over, in all species examined, the interval requiredto complete oocyte maturation in vitro was exactlythe same time as that needed for maturation invivo. The interval between the LH surge or an in-jection of HCG and the onset of ovulation couldthus be predicted exactly. Examples were 12 hoursin mice and 37 hours in pigs.

Several years elapsed before the correct intervalof 37 hours between the onset of culture and theformation of metaphase 2 and the first polar bodyin human oocytes was clarified (Edwards,1965a,b). This discovery was made during studieson the few human ovarian oocytes available inLondon and Glasgow, and confirmed in larger-scale studies in Baltimore in collaboration withGeorgeanna and Howard Jones in the Johns Hop-kins Medical School, and it proved to be decisivefor the introduction of human IVF (Figures 3, 4).Human oocytes could now be obtained from ex-cised pieces of ovarian tissue and cultured to timedstages of maturation. They could be used to assesschromosomal and other aspects of maturation, ega study of diakinesis and metaphase 1 using

R. G. Edwards, History of assisted human conception 185

Ovulation 4.3.3. 7.12.11.91.

First polarbody

extrudedo

o Telophase

1.4.1,3.

2. 3.

8. 3.6. 1. 3,

2.

Anaphae

(n

2, 1.2.4.2,3.3,2, 2, 3.1. 1.

Metaphase 2.3 4,23. 7.1. 3. 3, 8. 4,6. 6,1. 3.2,4. 9,1,1.

Fig. 2 Timing of the meiotic stagesPropha5e 2.4.6,2.5.4. 3, 1. 2. during oocyte maturation to meta-

t I 2 in (Ed-phase and ovulation miceo 2 3 4 5 6 7 8 9 10 11 12 13 14

wards and Gates, 1959)Hours after injection of HCG

Rabbit

MouseCow

Rhesus

Pig

Human

O 10 20 30 40

Hours in cufture

Table O Human gonadotrophic preparations

Before, 1955 Human chorionic gonadotro-phin available commercially

Gemzell, 1967 Anterior pituitary extracts tostimulate follicle growth

Lunenfeld, 1996 Urinary extracts to stimulatefollicle growth

Various sources, 1990's Purified FSH

Various sources, 1990's Recombinant FSH/LH

malian embryos were steadily improved duringthose years (Hammond, 1949; Paul, 1961; Whittenand Biggers, 1968).

Initial work on the introduction of IVF was ledby studies on the maturation of human oocytes re-leased from their follicles into culture medium. IfIVF was to be successful, mature human oocyteswould be needed to achieve fertilization and nor-mal embryonic growth. Pincus and his colleaguesinitiated the study of oocyte maturation in vitro inthe 1930's. Rabbit oocytes required 11-12 hoursto mature, and the maturation of human oocytesin vitro was thought to require 12 hours (Pincusand Saunders, 1939). As it happened, this figurewas incorrect, and it probably led to later investi-gators adding spermatozoa to human oocyteswhich had been matured in vitro for 12 hours(Menkin and Rock, 1948; Shettles, 1955). This er-ror, and possibly the general lack of understandingof the composition and handling of culture media,must have hampered these investigators.

Studies on oocyte maturation on several mam-malian species were carried out in the late 1950'sand early 1960's. They showed beyond doubt howeach species had its own unique duration of oocytematuration from the germinal vesicle stage untilthe formation of metaphase of the second meiotic

Fig. 3 Estimates of the time required for oocyte matu-ration from diakinesis to metaphase 2 in various species.The timing of maturation in vivo was subsequentlyshown to be similar in each species (Edwards 1965a,b).

division (metaphase 2) (Edwards, l965a). More-over, in all species examined, the interval requiredto complete oocyte maturation in vitro was exactlythe same time as that needed for maturation invivo. The interval between the LH surge or an in-jection of HCG and the onset of ovulation couldthus be predicted exactly. Examples were 12 hoursin mice and 37 hours in pigs.

Several years elapsed before the correct intervalof 37 hours between the onset of culture and theformation of metaphase 2 and the first polar bodyin human oocytes was clarified (Edwards,1965a,b). This discovery was made during studieson the few human ovarian oocytes available inLondon and Glasgow, and confirmed in larger-scale studies in Baltimore in collaboration withGeorgeanna and Howard Jones in the Johns Hop-kins Medical School, and it proved to be decisivefor the introduction of human IVF (Figures 3, 4).Human oocytes could now be obtained from ex-cised pieces of ovarian tissue and cultured to timedstages of maturation. They could be used to assesschromosomal and other aspects of maturation, ega study of diakinesis and metaphase 1 using


Ovulation 4.3,3, 7,12,11,91,

First polarbody

extruded1,4,1,3, 8. 3.6. 1. 3

V

o Telophase

Anaphase

(n

2, 3,

2, 1,2.4,2,3,3,2, 2, 3.1, 1.

2.

Metaphase 2.3. 4,2,3, 7,1, 3, 3. 8, 4.6. 6,1. 3,2,4, 9,1,1.

Fig. 2 Timing of the meiotic stagesProphase 2,4.6,2.5,4.

I

3, I, 2,

I I I I

during oocyte maturation to meta-2 in (Ed-phase and ovulation mice

0 2 3 4 5 6 7 8 9 10 11 12 13 14 wards and Gates, 1959)Hours after injection of HCG

RabbitMouse

CowRhesus

Pig

Human

O 10 20 30 40

Hours in cufture

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.

Exp Clin Endocrinol Diabetes 104 (1996) 3

Hours in culture

Fig. 4 Timing of the meiotic stages during maturationof human oocytes in vitro (Edwards 1965b and unpub-lished).

oocytes matured for 24-30 hours. Oocytes ma-tured completely for 37 hours in vitro to meta-phase 2 were invaluable for studies on later meioticstages and to begin detailed analyses on humanfertilization in vitro (Edwards et al., 1969). The ex-act timing of human ovulation in vivo could beestimated at 37 hours after the onset of the LHsurge or an injection of HCG, since it would beidentical to the timing of complete oocyte matu-ration. This prediction was made entirely on esti-mates gained from oocyte maturation in vitro, andbefore the onset of ovulation has been examined inany woman. Oocytes that had been matured in vi-tro for 37 hours could also be used for the firstattempts at human fertilization in vitro.

Many previous studies had identified a need foranimal spermatozoa to undergo capacitation, thena poorly understood process, before they could fer-tilise eggs in vivo or in vitro (Austin, 1951; Chang,1951). The earliest studies on the fertilization ofanimal eggs in vitro depended on the use of capaci-tated spermatozoa collected from the uterus or ovi-duct (Dauzier & Thibault, 1956). Only in one spe-cies, the golden hamster, could fertilization beachieved in vitro using spermatozoa taken fromthe epididymis (Yanagimachi and Chang, 1964).

Nevertheless, in our studies, human fertilizationin vitro was achieved using ejaculated spermatozoathat had been washed free of most seminal plasmaby gentle centrifugation (Edwards et al., 1969).Oocytes matured in vitro were inseminated withthese washed spermatozoa, and incubated at37.5 °C. All the eari and iate st.ages of human fert-ilization were later identified in vitro, and the inci-dence of fertilization was quite high. Studies onhuman embryos in vitro were now becoming pos-sible. However, there were many problems to solve,because rabbit oocytes that had matured entirelyin vitro were unable to develop into normal em-bryos after fertilization, since they degenerated be-fore the blastocyst stage (Chang, 1955). Humanoocytes that had been matured and fertilised in vi-tro also displayed some anomalies. This meant that

186

Germinal vesicle 31 3 1 1 16Stages Diakinesis 6 1

ofmeiosis

Metaphase 1Telophase 1

2 13 5 3

Metaphase 2 1 1 46

0 25 28 32 >3624 27 31 36

1 2 3 4 5 6 7

Number of large foflictes

Fig. 5 Numbers of large GraalTian follicles in relationto urinary oestrogens in women given HMG and HCG,with the correlation coefficient (Steptoe and Edwanis,1970).

successful fertilization and the culture of embryosin vitro, and their development to late stages offetal life, would demand the aspiration of fully-ma-ture oocytes from their ovarian follicles. It also me-ant that mild superovulation and oocyte maurationwould have to be introduced for cyclic women, andtheir oocytes would have to be aspirated just beforeovulation occurred.

At this time, I began a wonderful collaborationwith Patrick Steptoe, perhaps the world's foremostlaparoscopist at the time. Laparoscopy had come ofage in the mid-twentieth century, led primarily byPalmer (1946), Fragenheim (1964) and Steptoe(1967). It now made a full contribution to the devel-opment of human IVF, by permitting a much sim-plified method for approaching the human ovary forthe aspiration of oocytes from their follicles.

The routine of human IVF was established in thelate 1960's. Infertile patients attending hospital fortubal repair were asked to take three injections of225 iu of human menopausal gonadotrophins(HMG) on days 3, 5 and 7 of their menstrua' cyc'e.This was followed by an ovulatory dose of 5000 lUHCG on day 10 or 11. The HCG injection was timedto pre-empt the chances of an endogenous LH surgeand so avoid a spontaneous ovulation out of clinicalcontrol. Urinary oestrogens were measured once ortwice during the stimulation cycle. Laparoscopy wasused to inspect the ovary for follicles as they wetebeing aspirated to collect their oocytes. Ultrasoundwas not available in those days!

Exp Clin Endocrinol Diabetes 104 (1996) 3

Hours in culture

Fig. 4 Timing of the meiotic stages during maturationof human oocytes in vitro (Edwards 1965b and unpub-lished).

oocytes matured for 24-30 hours. Oocytes ma-tured completely for 37 hours in vitro to meta-phase 2 were invaluable for studies on later meioticstages and to begin detailed analyses on humanfertilization in vitro (Edwards et al., 1969). The ex-act timing of human ovulation in vivo could beestimated at 37 hours after the onset of the LHsurge or an injection of HCG, since it would beidentical to the timing of complete oocyte matu-ration. This prediction was made entirely on esti-mates gained from oocyte maturation in vitro, andbefore the onset of ovulation has been examined inany woman. Oocytes that had been matured in vi-tro for 37 hours could also be used for the firstattempts at human fertilization in vitro.

Many previous studies had identified a need foranimal spermatozoa to undergo capacitation, thena poorly understood process, before they could fer-tilise eggs in vivo or in vitro (Austin, I95l Chang,1951). The earliest studies on the fertilization ofanimal eggs in vitro depended on the use of capaci-tated spermatozoa collected from the uterus or ovi-duct (Dauzier & Thibault, 1956). Only in one spe-cies, the golden hamster, could fertilization beachieved in vitro using spermatozoa taken fromthe epididymis (Yanagimachi and Chang, 1964).

Nevertheless, in our studies, human fertilizationin vitro was achieved using ejaculated spermatozoathat had been washed free of most seminal plasmaby gentle centrifugation (Edwards et al., 1969).Oocytes matured in vitro were inseminated withthese washed spermatozoa, and incubated at37.5 °C. All the ear1 and iate stages of human fert-ilization were later identified in vitro, and the inci-dence of fertilization was quite high. Studies onhuman embryos in vitro were now becoming pos-sible. However, there were many problems to solve,because rabbit oocytes that had matured entirelyin vitro were unable to develop into normal em-bryos after fertilization, since they degenerated be-fore the blastocyst stage (Chang, 1955). Humanoocytes that had been matured and fertilised in vi-tro also displayed some anomalies. This meant that

186

Germinal vesicle 31 3 1 1 16Stages Diakinesis 6 1

ofmeiosis

Metaphase 1Telophase 1

2 13 3

Metaphase 2 1 1 46

0 25 28 32 >3624 27 31 36

1 2 3 4 5 6 7

Number of large follicles

Fig. 5 Numbers of large Graafian follicles in relationto urinary oestrogens in women given HMG and HCG,with the correlation coefficient (Steptoe and Edwards,1970).

successful fertilization and the culture of embryosin vitro, and their development to late stages offetal life, would demand the aspiration of fully-ma-ture oocytes from their ovarian follicles. It also me-ant that mild superovulation and oocyte maurationwould have to be introduced for cyclic women, andtheir oocytes would have to be aspirated just beforeovulation occurred.

At this time, I began a wonderful collaborationwith Patrick Steptoe, perhaps the world's foremostlaparoscopist at the time. Laparoscopy had come ofage in the mid-twentieth century, led primarily byPalmer (1946), Fragenheim (1964) and Steptoe(1967). It now made a full contribution to the devel-opment of human IVF, by permitting a much sim-plified method for approaching the human ovary forthe aspiration of oocytes from their follicles.

The routine of human IVF was established in thelate 1960's. Infertile patients attending hospital fortubal repair were asked to take three injections of225 iu of human menopausal gonadotrophins(HMG) on days 3, 5 and 7 of their menstrua' cyc'e.This was followed by an ovulatory dose of 5000 lUHCG on day 10 or il. The HCG injection was timedto pre-empt the chances of an endogenous LH surgeand so avoid a spontaneous ovulation out of clinicalcontrol. Urinary oestrogens were measured once ortwice during the stimulation cycle. Laparoscopy wasused to inspect the ovary for follicles as they werebeing aspirated to collect their oocytes. Ultrasoundwas not available in those days!

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Table 1 Number of Graafian follicles in the human ov-ary after priming patients with HMG and HCG (Ed-wards, 1980)

Dose of HMG Mean number of follicles(iu) (± SE)

Table 2 Success in aspirating human oocytes from theirfollicles after ovarian stimulation with HMG and HCG(Edwards, 1973)

Oocytes classified from their appearance

<33

33_34

35-36Hours after

HCG -37-38

38-39

40-43

0 20 40 60 80

% of women ovulating

Figure 6 Timing of the onset of follicular rupture afteran injection of HCG in women given controlled ovarianstimulation with HMG and HCG (Edwards, 1980).

Many treated patients had four or more follicleswhen examined by laparoscopy at various intervalsafter the HCG injection (Figure 5; Table 1). Fol-licle rupture was observed to begin at 37 hourspost-HCG (Steptoe and Edwards, 1970); this inter-val was exactly as predicted from the timing ofoocyte maturation in vitro (Figure 6; Table 2). La-paroscopy for the aspiration of preovulatory fol-licles was accordingly timed at 36 hours post-HCG. Follicle numbers and diameters were re-corded (Table 1) and Steptoe's laparoscopy deliv-ered a steady number of preovulatory and otheroocytes (Figure 7). The superovulation of cyclicwomen using HMG and HCG had begun, andmarked another step along the road to humanbirths after in vitro fertilization (IVF) (Steptoe andEdwards, 1970).

Preovulatory human oocytes were enclosed in amass of viscous follicular fluid, and an average oftwo per patient recovered in the initial work roseto four or more later. Many other oocytes had only

Fig. 7 Steptoe was always filming his laparoscopic ope-rations as shown in this picture taken in his operatingtheatre in the mid-1960's.

partly matured, as assessed by the state of theircumulus oophorus. Wide variations betweenoocytes collected from each patient were revealedby assays of up to nine steroids in their ac-companying follicular fluids, using cluster analyses(Fowler et al., 1978a,b). Several cohorts of follicleshad apparently been stimulated by the HMG, andthis had seemingly led to the varying steroid ratiosin the follicular fluids (Figure 8a). In contrast, onlytwo classes of follicles were identified during thenatural menstrual cycle, i.e. non-ovulatory andpreovulatory respectively.

To achieve fertilization in vitro, ejaculated sper-matozoa washed in culture medium and used toinseminate the oocytes one hour after collection.More than 60% of the oocytes were fertilised invitro (Figure 9), and many of the resulting humanembryos cleaved in vitro to morulae and blasto-cysts (Figure 10; Table 3). The embryos had ap-proximately a diploid chromosome number. Notriploid embryos typical of dispermy or digynywere found. Cleavage rate, blastomere structureand nuclear number, and the hatching of theblastocysts from their zonae pellucidae in vitrowere each assessed. Some hatched embryos grewfor several days in vitro, and one developed to day9 post-fertilization before being fixed for electrommicroscopy (Figure 11). Its embryonic disc wasprominent, and presumably contained many pri-mary stem cells for various organ rudiments. At-tempts to obtain these stem cells (ES cells, as hadbeen achieved in rabbits during the years in Glas-gow (Cole et al., 1965, 1966) by disaggregatingblastocysts or by culturing monolayers of cell out-growths from human blastocysts) did not succeed,

<10 115 53+2?(46%) 7+4?10-17.5 26 18+l?(69%) 4

>19 33 24 (73%) 9 + 2?

300-375 9.2675 8.8 ± 0.8900 10.0 ± 0.4

Follicle No of Oocytes Preovulatorydiameter follicles collected oocytes(mm) aspirated

59

65

23


Table 1 Number of Graafian follicles in the human ov-ary after priming patients with HMG and HCG (Ed-wards, 1980)

Dose of HMG Mean number of follicles(iu) (± SE)

Table 2 Success in aspirating human oocytes from theirfollicles after ovarian stimulation with HMG and HCG(Edwards, 1973)

Oocytes classified from their appearance

<33

33_34

35-36Hours after

36-37HCG

37-38

38-39

40-43

0 20 40 60 80

% of women ovulating

Figure 6 Timing of the onset of follicular rupture afteran injection of HCG in women given controlled ovarianstimulation with HMG and HCG (Edwards, 1980).

Many treated patients had four or more follicleswhen examined by laparoscopy at various intervalsafter the HCG injection (Figure 5; Table 1). Fol-licle rupture was observed to begin at 37 hourspost-HCG (Steptoe and Edwards, 1970); this inter-val was exactly as predicted from the timing ofoocyte maturation in vitro (Figure 6; Table 2). La-paroscopy for the aspiration of preovulatory fol-licles was accordingly timed at 36 hours post-HCG. Follicle numbers and diameters were re-corded (Table 1) and Steptoe's laparoscopy deliv-ered a steady number of preovulatory and otheroocytes (Figure 7). The superovulation of cyclicwomen using HMG and HCG had begun, andmarked another step along the road to humanbirths after in vitro fertilization (IVF) (Steptoe andEdwards, 1970).

Preovulatory human oocytes were enclosed in amass of viscous follicular fluid, and an average oftwo per patient recovered in the initial work roseto four or more later. Many other oocytes had only

Fig. 7 Steptoe was always filming his laparoscopie ope-rations as shown in this picture taken in his operatingtheatre in the mid-1960's.

partly matured, as assessed by the state of theircumulus oophorus. Wide variations betweenoocytes collected from each patient were revealedby assays of up to nine steroids in their ac-companying follicular fluids, using cluster analyses(Fowler et al., 1978a,b). Several cohorts of follicleshad apparently been stimulated by the HMG, andthis had seemingly led to the varying steroid ratiosin the follicular fluids (Figure 8a). In contrast, onlytwo classes of follicles were identified during thenatural menstrual cycle, i.e. non-ovulatory andpreovulatory respectively.

To achieve fertilization in vitro, ejaculated sper-matozoa washed in culture medium and used toinseminate the oocytes one hour after collection.More than 60% of the oocytes were fertilised invitro (Figure 9), and many of the resulting humanembryos cleaved in vitro to morulae and blasto-cysts (Figure 10; Table 3). The embryos had ap-proximately a diploid chromosome number. Notriploid embryos typical of dispermy or digynywere found. Cleavage rate, blastomere structureand nuclear number, and the hatching of theblastocysts from their zonae pellucidae in vitrowere each assessed. Some hatched embryos grewfor several days in vitro, and one developed to day9 post-fertilization before being fixed for electrommicroscopy (Figure 11). Its embryonic disc wasprominent, and presumably contained many pri-mary stem cells for various organ rudiments. At-tempts to obtain these stem cells (ES cells, as hadbeen achieved in rabbits during the years in Glas-gow (Cole et al., 1965, 1966) by disaggregatingblastocysts or by culturing monolayers of cell out-growths from human blastocysts) did not succeed,

<10 115 53+2? (46%) 7+4?10-17.5 26 l8+l?(69%) 4

>19 33 24 (73(y0) 9 + 2?

300-375 9.2675 8.8 ± 0.8900 10.0 ± 0.4

Follicle No of Oocytes Preovulatorydiameter follicles collected oocytes(mm) aspirated

59

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Eo,

40wzo

u,w

o30

LLo

zo

20Ui

zo

o

1-0

-

FOLLICLES PRIMEOWITH EXOGENOUSGONA 001RO PH IN S.

o FOLLICLES ON DAY13 OF A PATIENTWITH 28-DAYMENSTRUAL CYCLES.

FOLL)CLES ON DAY12 OF A PATIENTWITH 35 DAYMENSTRUAL CYCLES.

2-0 3-0

Iog CONCENTRATtON OF OESTRAOIOL17(ngnt.)

Fig. 8a Correlation between the log concentrations ofoestradiol-17b and progesterone in fluids taken from hu-man prevovulatory follicles.b Dendograms showing the clustering and classifica-tion of the stage of growth of individual follicles in wo-men when laparoscopy for oocyte aspiration was per-formed just before the time of ovulation. The height ofthe dendogram indicates increasing variability betweenfollicles. Upper (a). Natural cycles, showing how the in-dividual follicles are easily combined into two groups,ovulatory and non-ovulatory, with a huge variance be-tween the two classifications. Lower (b). HMG/HCGcycles, showing the considerable variation betweenindividual numbered follicles, they have fallen into fourstatistical groups with various levels of difference be-tween the groups. Groups 1, 2 and 3 could be groupedwith modest increase in variance, but not group 4. (Fow-1er et al., 1978 a,b).

and have not succeeded even to the present day. By1971, studies on human embryology had com-pletely surpassed animal studies, since the com-plete growth of oocytes from before fertilization toblastocysts had not been achieved in any mam-malian species, and the embryos of most speciesdid not develope at all in vitro.

The replacement of human embryos into theirmothers with the intention of establishing preg-nancies began in 1971 (Edwards, 1973). The IVFroutine had became established: patients weregiven HMG and HCG, and laparoscopy for oocyteaspiration was performed just before ovulation wasexpected. Fertilization and cleavage in vitro hadbeen achieved, and embryos were replaced in theirmothers at various times between 2.5 and 5 daysafter aspiration.

The endocrine dfficulties in establishing IVFpregnancies

Endocrine problems in the luteal phase soon em-erged. Stimulation with HMG and HCG short-ened the luteal phase to as little as 7-8 days in

110.9

99.3

C

87.6

> -- 29.6

oQ,

180--C

6.3

Group

18.798 b

16.832-

14866-

12.900o

. 10.935

>8.969

7003

5.038-

3 .072-

06

Follicle no. 12 10fl253I 8 175 822212324,7201214169Group 3 4

some patients, and its duration was directly pro-portional to their output of urinary oestrogensduring the follicular phase of their cycle. This earlymenstruation severely restricted the chances of im-plantation (Figure 12) (Edwards and Brody, 1995),and luteal support was essential to overcome it. Itwas possible that progesterone and oestradiol sup-plements would impair the activity of the corpusluteum even more. Moreover, these supplementalsteroids would have to be given daily until week 9or 10 when the placenta became fully active endo-crinologically (Csapo et al., 1974). This form oflong-term treatment using daily injections of pro-gesterone in oil would cause scabbing and pain tothe patients.

An ethical decision to avoid scabbing was there-fore taken to give the patients depot Primulot (17-hydroxyprogesterone acetate) at 5-day intervalssince it was believed to save threatened miscar-riages later in pregnancy. In fact, Primulot was lut-eolytic since progesterone levels fell disastrouslywithin a day of its administration. Moreover, al-though we did know it until ten years later, our

oo,C

-c

ii tiTi I 11H! [TI

Ovulatory Non -ovu1aory


Eo,

4-0wzowu,wIDo

3-O

LL

o

zo

2-0wt-,zo

o,o

l-0

FOLLICLES PRIMEOWITH EXOGENOUSGONADOTROPHINS.

o FOLLICLES ON DAY13 OF A PATIENTWITH 28-DAYMENSTRUAL CYCLES.

FOLLICLES ON DAY12 OF A PATIENTWITH 35-DAYMENSTRUAL CYCLES.

2-0 3-0 4-O

log 10 CONCENTRATION OF OESTRAOIOL 17 ( ngln,t.)

Fig. 8a Correlation between the log concentrations ofoestradiol-17b and progesterone in fluids taken from hu-man prevovulatory follicles.b Dendograms showing the clustering and classifica-tion of the stage of growth of individual follicles in wo-men when laparoscopy for oocyte aspiration was per-formed just before the time of ovulation. The height ofthe dendogram indicates increasing variability betweenfollicles. Upper (a). Natural cycles, showing how the in-dividual follicles are easily combined into two groups,ovulatory and non-ovulatory, with a huge variance be-tween the two classifications. Lower (b). HMG/HCGcycles, showing the considerable variation betweenindividual numbered follicles; they have fallen into fourstatistical groups with various levels of difference be-tween the groups. Groups 1, 2 and 3 could be groupedwith modest increase in variance, but not group 4. (Fow-1er et al., 1978 a,b).

and have not succeeded even to the present day. By1971, studies on human embryology had com-pletely surpassed animal studies, since the com-plete growth of oocytes from before fertilization toblastocysts had not been achieved in any mam-malian species, and the embryos of most speciesdid not develope at all in vitro.

The replacement of human embryos into theirmothers with the intention of establishing preg-nancies began in 1971 (Edwards, 1973). The IVFroutine had became established: patients weregiven HMG and HCG, and laparoscopy for oocyteaspiration was performed just before ovulation wasexpected. Fertilization and cleavage in vitro hadbeen achieved, and embryos were replaced in theirmothers at various times between 2.5 and 5 daysafter aspiration.

The endocrine difficulties in establishing IVFpregnancies

Endocrine problems in the luteal phase soon em-erged. Stimulation with HMG and HCG short-ened the luteal phase to as little as 7-8 days in

110.9

99.3

C

87.6

ID> -o. 29.6oo,E 18.0-

-C

6.3

Group

18.798- b

16.832-

14 866-

12.900-o

.G 10.935-t,>

8.969-oo,C

-c

7 003

5.038-

3 .072-

O

some patients, and its duration was directly pro-portional to their output of urinary oestrogensduring the follicular phase of their cycle. This earlymenstruation severely restricted the chances of im-plantation (Figure 12) (Edwards and Brody, 1995),and luteal support was essential to overcome it. Itwas possible that progesterone and oestradiol sup-plements would impair the activity of the corpusluteum even more. Moreover, these supplementalsteroids would have to be given daily until week 9or 10 when the placenta became fully active endo-crinologically (Csapo et al., 1974). This form oflong-term treatment using daily injections of pro-gesterone in oil would cause scabbing and pain tothe patients.

An ethical decision to avoid scabbing was there-fore taken to give the patients depot Primulot (17-hydroxyprogesterone acetate) at 5-day intervalssince it was believed to save threatened miscar-riages later in pregnancy. In fact, Primulot was lut-eolytic since progesterone levels fell disastrouslywithin a day of its administration. Moreover, al-though we did know it until ten years later, our

1106

Follicle no.Group

M12 l011254 8 175 j822212324 21169

1 2 3 4

OYulotory Non -ovuloory

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Fig. 9 Stages of fertilization in vitro, using preovulatory oocytes aspirated from the ovary. The successive stagesshow (upper) sperm attachment to the zona pellucida, their migration through it to the oolemma, and (lower) thespermatozoon inside the ooplasm, the expanding male pronucleus and a fertilised egg with two pronuclei and polarbodies. (Edwards, 1980).

belief that Primulot sustained the endometriumwas totally mistaken for it actually offered nothingto the luteal endometrium. This ethical decisionled to three years of frustration after we began totransfer embryos to their mothers. There were briefrises in HCG3 levels after embryo transfer in somepatients indicating that short-lived pregnancies hadbeen established. Several years later, we discoveredthat Primulot caused the early abortion of im-plantated human embryos between days 18-25after fertilization!

Meanwhile, after thoroughly and repeatedly test-ing all the systems involved in IVF, suspicionsabout the cause of the failure of implantation fellupon Primulot. It was phased out, and at last preg-nancies were established. The first clinical preg-nancy arose as more natural steroids were used forluteal phase support. A blastocyst implanted afterbeing replaced in its mother after 5 days in culture,and the living fetus was diagnosed on ultrasound.It later proved to be a tubal ectopic pregnancy, andhad to be removed at 11 - 12 weeks of gestation(Figure 13). Another brief pregnancy identified by

rising levels of HCG3 indicated that the endocrin-ology of early pregnancy was being brought undercontrol and that full-term births were near (Ed-wards et al., 1980a).

Several new endocrine approaches were accord-ingly tested (Table 4). Some oocytes and embryoswere cryopreserved with the intention of thawingthem during a later natural cycle for replacementin the mother. Many were in good condition afterthawing (Edwards et al., 1980a, Edwards andSteptoe, 1980, Steptoe and Edwards, 1976; Steptoeet al., 1980). Bromocryptine was introduced intoovarian stimulation to reduce high prolactin levels.Clomiphene and HMG were combined for ovarianstimulation to overcome the short luteal phasewith HMG alone (Table 4). Many oocytes wereharvested and embryo transfers were achieved withall these approaches.

IVF was also tested during the natural cycle.Measurements were made of urinary LH, oestrog-ens and pregnanediol in samples collected eighttimes daily, at 11 pm, 3, 7, 9, 11 am and at 2, 5and 8 pm. LH was measured using Higonavis,


Fig. 9 Stages of fertilization in vitro, using preovulatory oocytes aspirated from the ovary. The successive stagesshow (upper) sperm attachment to the zona pellucida, their migration through it to the oolemma, and (lower) thespermatozoon inside the ooplasm, the expanding male pronucleus and a fertilised egg with two pronuclei and polarbodies. (Edwards, 1980).

belief that Primulot sustained the endometriumwas totally mistaken for it actually offered nothingto the luteal endometrium. This ethical decisionled to three years of frustration after we began totransfer embryos to their mothers. There were briefrises in HCG3 levels after embryo transfer in somepatients indicating that short-lived pregnancies hadbeen established. Several years later, we discoveredthat Primulot caused the early abortion of im-plantated human embryos between days 18-25after fertilization!

Meanwhile, after thoroughly and repeatedly test-ing all the systems involved in IVF, suspicionsabout the cause of the failure of implantation fellupon Primulot. It was phased out, and at last preg-nancies were established. The first clinical preg-nancy arose as more natural steroids were used forluteal phase support. A blastocyst implanted afterbeing replaced in its mother after 5 days in culture,and the living fetus was diagnosed on ultrasound.It later proved to be a tubal ectopic pregnancy, andhad to be removed at li 12 weeks of gestation(Figure 13). Another brief pregnancy identified by

rising levels of HCG3 indicated that the endocrin-ology of early pregnancy was being brought undercontrol and that full-term births were near (Ed-wards et al., 1980a).

Several new endocrine approaches were accord-ingly tested (Table 4). Some oocytes and embryoswere cryopreserved with the intention of thawingthem during a later natural cycle for replacementin the mother. Many were in good condition afterthawing (Edwards et al., 1980a; Edwards andSteptoe, 1980; Steptoe and Edwards, 1976; Steptoeet al., 1980). Bromocryptine was introduced intoovarian stimulation to reduce high prolactin levels.Clomiphene and HMG were combined for ovarianstimulation to overcome the short luteal phasewith HMG alone (Table 4). Many oocytes wereharvested and embryo transfers were achieved withall these approaches.

IVF was also tested during the natural cycle.Measurements were made of urinary LH, oestrog-ens and pregnanediol in samples collected eighttimes daily, at 11 pm, 3, 7, 9, 11 am and at 2, 5and 8 pm. LH was measured using Higonavis,

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Fig. lo Stages of embryonic growth in vitro after the fertilization in vitro of human oocytes aspirated from theirfollicles (Edwards et al., 1980). The stages shown include along the top row 4 cell, 8 cell, a compacting morula, andon the bottom row two living blastocysts and a flattened stained blastocyst with more than 120 cells and manymitoses (Edwards, 1980).

Table 3 Interval in hours between the insemination ofhuman oocytes in vitro and development to variousembryonic stages (Edwards, 1980)

Stage ofdevelopment

2-cell4-cell8-cell16-cellMo rulaBlastocyst

Observations oninterval (hr)betweeninsemination andembryo cleavage

Estimate of(mid-)point ofeach cleavagestage (hr)

From Edwards RG. Conception in the Human Female.London, Academic Press, 1980

which was actually a urinary 'pregnancy test' forHCG with a 2-hour turn-round time (Figure 14a)(Edwards et al., 1980a,b). LH surges appeared inurine 4 hours after their onset in plasma, most ofthem beginning at 7 am, so the plasma LH surgesmust have begun at 3-7 am. The success of oocyterecovery during the natural menstrual cycle, and

Fig. 11 A hatched human blastocyst grown in vitro un-til day 9 post-fertilization. Notice its greater diameterthan that of the discarded zona pellucida. Notice alsothe prominent embryonic disc.

the changing steroidogenic activity of preovulatoryfollicles as they approched rupture, were assessed.The optimal time for follicular aspiration wasidentified.

26 34.9± 1.938 51.2 ± 1.946 67.9 ± 2.568 84.6 ± 3.4

100 100.2 ± 3.0120 112.7 ± 3,8


Fig. lo Stages of embryonic growth in vitro after the fertilization in vitro of human oocytes aspirated from theirfollicles (Edwards et al., 1980). The stages shown include along the top row 4 cell, 8 cell, a compacting morula, andon the bottom row two living blastocysts and a flattened stained blastocyst with more than 120 cells and manymitoses (Edwards, 1980).

Table 3 Interval in hours between the insemination ofhuman oocytes in vitro and development to variousembryonic stages (Edwards, 1980)

Stage ofdevelopment

Observations oninterval (hr)betweeninsemination andembryo cleavage

Estimate of(mid-)point ofeach cleavagestage (hr)

From Edwards RG. Conception in the Human Female.London, Academic Press, 1980

which was actually a urinary 'pregnancy test' forHCG with a 2-hour turn-round time (Figure 14a)(Edwards et al., 1980a,b). LH surges appeared inurine 4 hours after their onset in plasma, most ofthem beginning at 7 am, so the plasma LH surgesmust have begun at 3-7 am. The success of oocyterecovery during the natural menstrual cycle, and

Fig. 11 A hatched human blastocyst grown in vitro un-til day 9 post-fertilization. Notice its greater diameterthan that of the discarded zona pellucida. Notice alsothe prominent embryonic disc.

the changing steroidogenic activity of preovulatoryfollicles as they approched rupture, were assessed.The optimal time for follicular aspiration wasidentified.

2-cell 26 34.9 ± 1.94-cell 38 51.2 ± 1.98-cell 46 67.9 ± 2.516-cell 68 84.6 ± 3.4M orula 100 100.2 ± 3.0Blastocyst 120 112.7 ± 3,8

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


50

Averagefalliclarant ra gen s(q/day)40

30

20

\ Average linear decrease= 7.53 t 2.47 qJ'li,tel dayP < 0.01

Doratian af lt.al phau (days)

Fig. 12 Regression coefficient of the relationship be-tween oestrogen excretion and the duration of the lutealphase in patients given HMG and HCG (Edwards etal., 1980). All urine samples were collected during thefollicular phase from the patients and used to estimatetotal daily output of oestrogens.

This was also the first time such a diurnalrhythm in the onset of the LH surge had been re-ported. It showed that most women ovulated at 4o'clock in the afternoon, i.e. 37 hours after the on-set of the plasma LH surge (Figure 14b). This find-ing led us then and later to question previous workon the timing and control of the endocrinology ofthe human menstrual cycle, and also of the primatemenstrual cycle; the diurnal rhythm of the LHsurge had obviously been overlooked by several in-vestigators who reported LH surges in the middleof the day. It has also led us to question Knobil'shypothesis that the medial basal hypothalamus(MBH) regulated ovulation in primates, since hedid not know of this diurnal rhythm which couldhave been controlled by centres outside the MBHsuch as the suprachiasmatic nucleus.

The second patient treated during her naturalcycle was Lesley Brown. Her ripening oocyte wasaspirated at 11 am, i.e. 26-28 hr after the onsetof the urinary LH surge, immediately inseminated,fertilised and replaced approximately 58 hourslater as an 8 cell embryo (Steptoe et al., 1980).Three more pregnancies were established out of 32replacement cycles. One triploid fetus aborted(Figure 15), a normal boy was delivered, and a se-cond normal boy died 3 days after a prematuredelivery at 19 weeks of gestation, at one week afteran amniocentesis (Steptoe et al., 1980). Three nor-mal children were therefore delivered in this firstseries of IVF pregnancies, but one of them haddied after a most unfortunate premature birth.

Fim br ja IocclusionL

Cannulato fundus

Distorted andocc'uded tubeR

Implantation 5tewith slight bleedinginto occluded tube

Perforation sitewith bleeding intopelvis and tube

Fig. 13 Steptoe's drawing of the ectopic pregnancyshowing the state of the uterus at the time of implan-tation (upper), the probable state of the oviduct at 8weeks after embryo transfer (middle), and conditions inthe right oviduct and pelvis during laparoscopy 10 weeksafter transfer (lower) (Steptoe and Edwards, 1976).

ScientJic, clinical and ethical advances in humanreproduction during this period

Many other attempts to advance knowledge of hu-man reproduction were made during these years(Tables 5, 6). These included the first attempts atGIFT, in an attempt to simplify IVF (Figure 16),and the first studies on preimplantation diagnosisusing whole-mount analyses or the excision ofsmall pieces of trophectoderm from rabbit blasto-cysts. The presence of sex chromatin in the tro-phectodermal nuclei was used to identify femaleembryos, and its absence indicated male embryos.Initially, phase-contrast and fluorescent mi-

croscopy was applied to intact blastocysts pressedgently between a microscopic slide and a cover slip(Edwards & Gardner, 1967), but diagnosis and em-bryo survival were jeopardised by this approach.The excision of small pieces of trophectodermusing micromanipulation, followed by embryo

159 rn 11 12 13 14

70

60


50

Averagefolliculoro.strogens(qg/doy)40

30

20

\ Average linear decrease= 7.53 t 2.47 qg/lvteoi doyP < 0.01

9 10 11 12 13 14 15

Donation of lotool phac, (doya)

Fig. 12 Regression coefficient of the relationship be-tween oestrogen excretion and the duration of the lutealphase in patients given HMG and HCG (Edwards etal., 1980). All urine samples were collected during thefollicular phase from the patients and used to estimatetotal daily output of oestrogens.

This was also the first time such a diurnalrhythm in the onset of the LH surge had been re-ported. It showed that most women ovulated at 4o'clock in the afternoon, i.e. 37 hours after the on-set of the plasma LH surge (Figure 14b). This find-ing led us then and later to question previous workon the timing and control of the endocrinology ofthe human menstrual cycle, and also of the primatemenstrual cycle; the diurnal rhythm of the LHsurge had obviously been overlooked by several in-vestigators who reported LH surges in the middleof the day. It has also led us to question Knobïl'shypothesis that the medial basal hypothalamus(MBH) regulated ovulation in primates, since hedid not know of this diurnal rhythm which couldhave been controlled by centres outside the MBHsuch as the suprachiasmatic nucleus.

The second patient treated during her naturalcycle was Lesley Brown. Her ripening oocyte wasaspirated at 11 am, i.e. 26-28 hr after the onsetof the urinary LH surge, immediately inseminated,fertilised and replaced approximately 58 hourslater as an 8 cell embryo (Steptoe et al., 1980).Three more pregnancies were established out of 32replacement cycles. One triploid fetus aborted(Figure 15), a normal boy was delivered, and a se-cond normal boy died 3 days after a prematuredelivery at 19 weeks of gestation, at one week afteran amniocentesis (Steptoe et al., 1980). Three nor-mal children were therefore delivered in this firstseries of IVF pregnancies, but one of them haddied after a most unfortunate premature birth.

FimbrialOcclusiont.

C an n u lato fundus

Distorted andOccluded tubeR

Implantation Sitewith slight bleedinginto Occluded tube

Perforation sitewith bleeding intopelvis and tube

Fig. 13 Steptoe's drawing of the ectopic pregnancyshowing the state of the uterus at the time of implan-tation (upper), the probable state of the oviduct at 8weeks after embryo transfer (middle), and conditions inthe right oviduct and pelvis during laparoscopy 10 weeksafter transfer (lower) (Steptoe and Edwards, 1976).

Scientific, clinical and ethical advances in humanreproduction during this period

Many other attempts to advance knowledge of hu-man reproduction were made during these years(Tables 5, 6). These included the first attempts atGIFT, in an attempt to simplify IVF (Figure 16),and the first studies on preimplantation diagnosisusing whole-mount analyses or the excision ofsmall pieces of trophectoderm from rabbit blasto-cysts. The presence of sex chromatin in the tro-phectodermal nuclei was used to identify femaleembryos, and its absence indicated male embryos.Initially, phase-contrast and fluorescent mi-

croscopy was applied to intact blastocysts pressedgently between a microscopic slide and a cover slip(Edwards & Gardner, 1967), but diagnosis and em-bryo survival were jeopardised by this approach.The excision of small pieces of trophectodermusing micromanipulation, followed by embryo

70

60

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Table 4 Details of the incidence of pregnancy inpatients given various treatments in Oldham (Edwardset al., 1980a)

Pregnanc-ies

0/15

0/7

0/8

2/16

1/10

0/12

0/2

4/32

0/7

transfer of the sexed blastocysts, led to the success-ful determination of sex in all 18 offspring(Gardner and Edwards, 1968). Similar attempts toidentify sex chromatin and the Y body in humanblastocysts were unsuccessful. Micromanipulationwas also used to place a single donor cell in mouseblastocysts in order to produce chimaeric offspring(Figure 17a) (Gardner, 1968). This work was theforerunner of studies on transgenic mice, using agenetically-engineered donor cell for injection intoa blastocyst (Figure 17b).

The first papers on the ethics of human TVFwere also published in the 1970's (Edwards andSharpe, 1970; Edwards, 1974). These were to setthe scene for many far-reaching debates in the1970's and 1980's, a debate that continues today. Itis worth noting that these debates were initiated bya scientist in the field, accompanied by a lawyer(Edwards and Sharpe, 1970).

B. Forms of ovarian stimulation as IVF expandedworldwide after 1980

The years between 1980 and 1990 included rapidadvances in many aspects of IVF (Table 7). Thesewill be described only briefly, since many of themare thoroughly familiar to this audience.

Introduction of successive forms of ovarianstirn u/a tion

We had no laboratory or clinical facilities to con-tinue our work for two years after Louise Brownwas delivered, as a rapid expansion in IVF around

the world began two years after her birth. Majorattention was now directed at the optimal meansof achieving follicular stimulation. The use ofclo-miphene and/or HMG during stimulated cycleswas reintroduced (Trounson et al., 1981; Jones etal., 1982). The use of clomiphene alone or ofHMG used with or without clomiphene now be-came essential features of follicular stimulation forTVE Less HMG was needed when it was combinedwith clomiphene. This treatment overcame theshort luteal phase typical of HMG alone (Edwardset al., 1980a), and it may have involved some formof compensation by the HMG for an FSH de-ficiency in the mid-follicular phase (Fleming andCoutts, 1990). Some investigators believed thatstimulation with clomiphene and HMG impairedthe endometrium, just as with clomiphene alone.Models for stimulation were developed for poor re-sponders, e.g. giving HMG in pulses, but disadvan-tages with the use of HMG alone included the oc-currence of elevated tonic levels of LH, prematureor attenuated LH surges in 15% of cycles, andweak responses in many patients. These unwantedLH surges were often associated with rising pro-gesterone levels and low pregnancy rates after em-bryo transfer (Howles et al., 1986). A transient hy-perprolactinemia during oocyte collection hadlittle effect on the success of the treatment.

The LHRH agonists became available commer-cially, followed more recently by the LHRH antag-onists. These were combined with HMG and HCGfor stimulation (Fleming and Coutts, 1990). Pulsa-tile treatment with the agonists stimulated follicu-lar growth in men and women suffering from sev-ere hypothalamic deficiencies, and was effective foridiopathic hypogonadotrophic hypogonadism(Leyendecker et al., 1990). Almost 90% of patientsovulated, and pregnancy occurred in >25% ofcycles.

There was much ingenuity in the combinationof the LHRH agonists, contraceptive steroids andHMG or FSH in IVF programmes. The agonistswere given as a single depot injection, in severalinjections or as nasal sprays. LH pulses remainedsuppressed with long-term use whereas FSH levelsrose gradually to early follicular levels. LH surgeswere abolished in many women if agonists weregiven in the preceding luteal phase or on the firstday of menstrual bleeding, and many patients pro-duced large numbers of oocytes and achieved preg-nancies after IVF.

The LHRH agonists were sometimes given as anultra-short flare-up protocol, i.e. for 3-4 days atthe beginning of the cycle, followed by HMG (Fig-ure 18). This protocol was originally introduced forpoor responders since the agonists initially stimu-lated a short-term FSH release (Howles et al.,

H M GIH CG NoneH M GIH CG HCG and progesteroneH M G/H CG Clomiphene

HM G/H CG Hydroxyprogesteronehexanoate and HCG

H M GIH CG H ydroxy progeste ro nehexanoate, progesteroneand HCG

HMG/HCG Bromocryptine

Clomiphene/ NoneHMG/HCGNatural cycle None

Nat. cycle/HCG HCG and progesterone

Stimulation Luteal phase treatment

192 Exp Clin Endocrino! Diabetes 104 (1996) 3

Table 4 Details of the incidence of pregnancy inpatients given various treatments in Oldham (Edwardset al., 1980a)

transfer of the sexed blastocysts, led to the success-ful determination of sex in all 18 offspring(Gardner and Edwards, 1968). Similar attempts toidentify sex chromatin and the Y body in humanblastocysts were unsuccessful. Micromanipulationwas also used to place a single donor cell in mouseblastocysts in order to produce chimaeric offspring(Figure 17a) (Gardner, 1968). This work was theforerunner of studies on transgenic mice, using agenetically-engineered donor cell for injection intoa blastocyst (Figure 17b).

The first papers on the ethics of human IVFwere also published in the 1970's (Edwards andSharpe, 1970; Edwards, 1974). These were to setthe scene for many far-reaching debates in the1970's and 1980's, a debate that continues today. Itis worth noting that these debates were initiated bya scientist in the field, accompanied by a lawyer(Edwards and Sharpe, 1970).

B. Forms of ovarian stimulation as IVF expandedworldwide after 1980

The years between 1980 and 1990 included rapidadvances in many aspects of IVF (Table 7). Thesewill be described only briefly, since many of themare thoroughly familiar to this audience.

Introduction of successive forms of ovarianstimulation

We had no laboratory or clinical facilities to con-tinue our work for two years after Louise Brownwas delivered, as a rapid expansion in IVF around

the world began two years after her birth. Majorattention was now directed at the optimal meansof achieving follicular stimulation. The use ofclo-miphene and/or HMG during stimulated cycleswas reintroduced (Trounson et al., 1981; Jones etal., 1982). The use of clomiphene alone or ofHMG used with or without clomiphene now be-came essential features of follicular stimulation forTVE Less HMG was needed when it was combinedwith clomiphene. This treatment overcame theshort luteal phase typical of HMG alone (Edwardset al., 1980a), and it may have involved some formof compensation by the HMG for an FSH de-ficiency in the mid-follicular phase (Fleming andCoutts, 1990). Some investigators believed thatstimulation with clomiphene and HMG impairedthe endometrium, just as with clomiphene alone.Models for stimulation were developed for poor re-sponders, e.g. giving HMG in pulses, but disadvan-tages with the use of HMG alone included the oc-currence of elevated tonic levels of LH, prematureor attenuated LH surges in l5% of cycles, andweak responses in many patients. These unwantedLH surges were often associated with rising pro-gesterone levels and low pregnancy rates after em-bryo transfer (Howles et al., 1986). A transient hy-perprolactinemia during oocyte collection hadlittle effect on the success of the treatment.

The LHRH agonists became available commer-cially, followed more recently by the LHRH antag-onists. These were combined with HMG and HCGfor stimulation (Fleming and Coutts, 1990). Pulsa-tile treatment with the agonists stimulated follicu-lar growth in men and women suffering from sev-ere hypothalamic deficiencies, and was effective foridiopathic hypogonadotrophic hypogonadism(Leyendecker et al., 1990). Almost 90% of patientsovulated, and pregnancy occurred in >25% ofcycles.

There was much ingenuity in the combinationof the LHRH agonists, contraceptive steroids andHMG or FSH in IVF programmes. The agonistswere given as a single depot injection, in severalinjections or as nasal sprays. LH pulses remainedsuppressed with long-term use whereas FSH levelsrose gradually to early follicular levels. LH surgeswere abolished in many women if agonists weregiven in the preceding luteal phase or on the firstday of menstrual bleeding, and many patients pro-duced large numbers of oocytes and achieved preg-nancies after IVF.

The LHRH agonists were sometimes given as anultra-short flare-up protocol, i.e. for 3-4 days atthe beginning of the cycle, followed by HMG (Fig-ure 18). This protocol was originally introduced forpoor responders since the agonists initially stimu-lated a short-term FSH release (Howles et al.,

Stimulation Luteal phase treatment Pregnanc-ies

HMG/HCG None 0/15

HMG/HCG HCG and progesterone 0/7

HMG/HCG Clomiphene 0/8

HMG/HCG Hydroxyprogesterone 2/16hexanoate and HCG

HMG/HCG Hydroxyprogesterone 1 / 10

hexanoate, progesteroneand HCG

HMG/HCG Bromocryptine 0/12

Clomiphene/ None 0/2HMG/HCGNatural cycle None 4/32

Nat. cycle/HCG HCG and progesterone 0/7

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


b

Totaloestrogens

pg day

a

Pregnanediolmg day

Totaloestrogens

pg day

Pregnanediolmg day

LH byHiGonavisiUh1

100LH byradioimmunoassay 60-

lu day 20:

LHsurge Lap

st

20-

o42

o

64

10 12 14

LHsurge Lap

st

-I-

IIIIf101214 121416

I 11$12 14 16

Day of cyc'e

Day of cycle

LHsurge Lap

st

1986; Matthews et al., 1991). This form of stimula-tion is still suspected even today of providing insuf-ficient flare-up release of FSH, and failing todown-regulate the pituitary gland sufficiently.Consequently, poor responses to HMG and en-dogenous LH surges were reported, jeopardizingthe treatment cycle. Its protagonists stress the ef-fectiveness of the protocol, and its low embryotox-icity and low teratogenicity since the agonist waswithdrawn several days before ovulation.

A "short protocol" consisted of LHRHa ad-ministration from day 2 or from the mid-lutealphase of the preceeding cycle. HMG was givenfrom day 3-4, with the LHRH agonist continu-ing until HCG was given to induce follicle matu-mtion. Treatments beginning during the late lu-teal phase were introduced to avoid activating

lin

LHsurge Lap

st

111315 111315

111315 111315

1

n LinIIIIJf

Fig. 14a Assays of urinaryoestrogens, pregnanediol and LHin the follicular phase of the firstpatients being treated for naturalcycle IVF. Eight samples of urinewere collected at regular intervalsthroughout the day (Edwards etal., 1980a).b Diurnal rhythm in the onset ofthe human LH surge in plasma(histograms), adjusted for its ex-cretion in urine, and in the timingof human ovulation (line). Noticethe stronger LH surge rhythm inspring than in autumn. The ex-pected timing of ovulation wascalculated by adding 37 hours tothe onset of the LH surge. An in-terval of 37 hours is needed foroocyte maturation and folliclerupture in women, as shown ear-lier in this article. Approximately75% of women ovulate before16.00 hr in spring and 60% inautumn.

corpora lutea, but this protocol required moreHMG than the ultra-short protocol. "Long pro-tocols" involved LHRHa alone over many days,to completely down-regulate the pituitary gland.Treatment began during the previous cycle, fromthe follicular or luteal phase, and down-regu-lation could be continued for several weeks("blocking protocols"). These treatments enabledthe physician to choose the day when ovarianstimulation began, and so plan the timing ofoocyte maturation well in advance. Oocyte aspi-ration could be planned to occur on a weekday("Never on Sunday"!). Contraceptive steroidswere also used for blocking and programming,and offered an equivalent and much less expen-sive treatment.

6R4 ...

L

_1 ....L e.

100:60-

20: r-1 rflLH

110LH

surge Lapst

LH LHsurge Lap surge Lap

st st

surgest=

Lap

60:

40

110 j,60 -

40:20-

o42o

linm

LH byHiGonavislU h1

LH byradioimmuno assay

lu day


a

b

Pregnanediolmg day1

LH byHiGonavislU h1

o42o

64

100-LHby -radloimmunoassay 60-

lu day 20-

III,,, 1111111101214 121416

°J

-I-

Day of cycle

Day of cycle

1986; Matthews et al., 1991). This form of stimula-tion is still suspected even today of providing insuf-ficient flare-up release of FSH, and failing todown-regulate the pituitary gland sufficiently.Consequently, poor responses to HMG and en-dogenous LH surges were reported, jeopardizingthe treatment cycle. Its protagonists stress the ef-fectiveness of the protocol, and its low embryotox-icity and low teratogenicity since the agonist waswithdrawn several days before ovulation.

A "short protocol" consisted of LHRHa ad-ministration from day 2 or from the mid-lutealphase of the preceeding cycle. HMG was givenfrom day 3-4, with the LHRH agonist continu-ing until HCG was given to induce follicle matu-ration. Treatments beginning during the late lu-teal phase were introduced to avoid activating

LHsurge Lap

tt

n Lin1111411 1111411111315 111315

LHsurge Lap

ft

lin

LHsurge Lap

ft

LHsurge Lap

5f

n Lin1111111 111141111 13 15 11 13 15

Fig. 14a Assays of urinaryoestrogens, pregnanediol and LHin the follicular phase of the firstpatients being treated for naturalcycle IVF. Eight samples of urinewere collected at regular intervalsthroughout the day (Edwards etal., 1980a).b Diurnal rhythm in the onset ofthe human LH surge in plasma(histograms), adjusted for its ex-cretion in urine, and in the timingof human ovulation (line). Noticethe stronger LH surge rhythm inspring than in autumn. The ex-pected timing of ovulation wascalculated by adding 37 hours tothe onset of the LH surge. An in-terval of 37 hours is needed foroocyte maturation and folliclerupture in women, as shown ear-lier in this article. Approximately75% of women ovulate before16.00 hr in spring and 60% inautumn.

corpora lutea, but this protocol required moreHMG than the ultra-short protocol. "Long pro-tocols" involved LHRHa alone over many days,to completely down-regulate the pituitary gland.Treatment began during the previous cycle, fromthe follicular or luteal phase, and down-regu-lation could be continued for several weeks("blocking protocols"). These treatments enabledthe physician to choose the day when ovarianstimulation began, and so plan the timing ofoocyte maturation well in advance. Oocyte aspi-ration could be planned to occur on a weekday("Never on Sunday"!). Contraceptive steroidswere also used for blocking and programming,and offered an equivalent and much less expen-sive treatment.

Pregnanediolmg day1

LH byHiGonavis

lu h1LH by

radioimmunoassaylU day1

42

o

64

100

60

20

....L

I II10 12 14 12 14

I $

16

LHLH surge Lap

110Total 60 -

surge Lapft ftoestrogens

40 -pg day1

20-

LHLH surge Lap

110Total 60

surge Lapft stoestrogens

40pg day120

linm

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Fig. 15 Illustration of a triploid fetus which was thethird clinical pregnancy established by IVF. It abortedlate in the first trimester after embryo transfer (Steptoeet al., 1980).

Table 5 Scientific advances in assisted conception be-tween 1968 and 1978

Ovarian follicular modelsSuccessive follicle cohorts after ovarian stimulation inwomenES cells in vitroMouse chimaeras from single-cell injections into blasto-cystsPreimplantation diagnosis in rabbits

Table 6 Clinical advances in assisted conception be-tween 1968 and 1978

Human birthsFirst ectopic pregnancy after IVF5 forms of controlled ovarian stimulationFirst GIFT cases doneHuman oculation timedHuman oocytes/embryos cryopresvdOoyte donationFirst ethical debates on IVFHuman preimplantation diagnosis begun

Knowledge gained from controlled 1w/nan ovarianstimula t ion

Much knowledge on human ovarian function hasarisen from the use of controlled ovarian stimula-tion in women. Simple endocrine assays were intro-duced including dual-analyte enzyme immuno-assays for measuring urinary oestrone-glucuronide

Mrs M PODGH 1211/73

150

looo)zwCo

o)wo

>-

z

50

OR TI THE FIRST GIFT

ÇLi proovu atory Oocyte *1.6 X 10 s perm In t

00

HCG (I U.)1500 1500 1500 1500lIN

8 1 2 1 6 20 24 28 32 36

DAYS OF CYCLE

lo

5

o

I

oE

o

wzzCw

Q-

>-

z

Fig. 16 The first GIFT, then called OR ± TI (oocyterecovery with tubal insemination); a single oocyte wasreplaced with spermatozoa in most patients; occasion-ally two oocytes were replaced (Edwards and Steptoe,1980).

and pregnanediol-glucuronide. Home monitorswere introduced to measure urinary glucuronides,and home kits or dipsticks for urinary LH.

Most protocols of ovarian stimulation resultedin large numbers of successive follicle populationsbeing recruited, in multifolliculation and in con-siderable developmental variations between neigh-bouring preovulatory follicles (Figure 8b). A domi-nant follicle produced 400 pg oestradiol severaldays after stimulation began. The sonography offollicular development using first transabdominaland then transvaginal probes became increasinglyimportant for monitoring follicles, and replacedendocrine assays in many clinics. The time of HCGadministration could be delayed for 1 2 days afterfollicles reached 18 mm, without any loss of preg-nancy rates. HCG could be replaced by LHRH orLHRHa to induce ovulation, and progesteronesupplements given in the late follicular phase couldalso induce an LH surge in stimulated patients. Lu-teal phase support with progesterone or intermit-tent doses of HCG were usually given after em-bryo transfer.

Several sequelae of ovarian stimulation becamewell recognised. The consequences of elevatedtonic levels of FSH or LH, of high androgen levelsor luteal phase deficiencies were recognised. Thenature of treatments for patients with polycysticovarian disease (PCOD) was clarified. They dis-played inappropriate negative feedback effects,

150

100-

50-

o0 4 B

I

Ji;

I12 LJO

I ¡24 28 32 36

-10

-5

o


Fig. 15 Illustration of a triploid fetus which was thethird clinical pregnancy established by IVF. It abortedlate in the first trimester after embryo transfer (Steptoeet al., 1980).

Table 5 Scientific advances in assisted conception be-tween 1968 and 1978

Ovarian follicular modelsSuccessive follicle cohorts after ovarian stimulation inwomenES cells in vitroMouse chimaeras from single-cell injections into blasto-cystsPreimplantation diagnosis in rabbits

Table 6 Clinical advances in assisted conception be-tween 1968 and 1978

Human birthsFirst ectopic pregnancy after IVF5 forms of controlled ovarian stimulationFirst GIFT cases doneHuman oculation timedHuman oocytes/embryos cryopresvdOoyte donationFirst ethical debates on IVFHuman preimplantation diagnosis begun

Knowledge gained from controlled human ovarianstimulation

Much knowledge on human ovarian function hasarisen from the use of controlled ovarian stimula-tion in women. Simple endocrine assays were intro-duced including dual-analyte enzyme immuno-assays for measuring urinary oestrone-glucuronide

Mr& M PODGH 1211/73

100

50

OR TI THE FIRST GIFT

Çl_i preovu atorE oocyte *1.6 e 1 6 s perm In t

0(3

8 12 16 20 24 28 32 36

DAYS OF CYCLE

10

S

o

I

oE

Oclwzz(0wzQ-

>-zzz:3

Fig. 16 The first GIFT, then called OR + TI (oocyterecovery with tubal insemination); a single oocyte wasreplaced with spermatozoa in most patients; occasion-ally two oocytes were replaced (Edwards and Steptoe,1980).

and pregnanediol-glucuronide. Home monitorswere introduced to measure urinary glucuronides,and home kits or dipsticks for urinary LH.

Most protocols of ovarian stimulation resultedin large numbers of successive follicle populationsbeing recruited, in multifolliculation and in con-siderable developmental variations between neigh-bouring preovulatory follicles (Figure 8b). A domi-nant follicle produced 400 pg oestradiol severaldays after stimulation began. The sonography offollicular development using first transabdominaland then transvaginal probes became increasinglyimportant for monitoring follicles, and replacedendocrine assays in many clinics. The time of HCGadministration could be delayed for 1 2 days afterfollicles reached 18 mm, without any loss of preg-nancy rates. HCG could be replaced by LHRH orLHRHa to induce ovulation, and progesteronesupplements given in the late follicular phase couldalso induce an LH surge in stimulated patients. Lu-teal phase support with progesterone or intermit-tent doses of HCG were usually given after em-bryo transfer.

Several sequelae of ovarian stimulation becamewell recognised. The consequences of elevatedtonic levels of FSH or LH, of high androgen levelsor luteal phase deficiencies were recognised. Thenature of treatments for patients with polycysticovarian disease (PCOD) was clarified. They dis-played inappropriate negative feedback effects,

5e. Ui

Ï1/MG *1ro) HCG (I U.)3 3 3 3 1500 1500 1500 1500iiiiU i i ii

r

150

100-

50-

o I10

I IB0 4

-10

-5

o24 28 32 36

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


high androgen levels, and changes in gonado-trophin secretion. They were at risk of hyperstimu-lation and premature luteinization, and were usu-ally high responders. New forms of endocrinestimulation for them involved the use of low dosesof FSH and perhaps HMG to avoid hyperstimula-tion, or the stepwise administration of FSH.

Ovarian hyperstimulation syndrome was themost serious complication of ovulation induction.It arose in 1.2% of all patients treated withLHRHa/HMG. Severe hyperstimulation aroseafter HMG treatment in 5% of pregnancy cyclesand l% of non-conception cycles. It was more fre-quent with large cohorts of medium-sized folliclesthan with two or three hyperfunctioning dominantfollicles, high levels of oestradiol, and the occur-rence of pregnany. Moderate and severe forms werea serious health risk to patients (Lunenfeld, 1969).Attempts to avoid it included continuing the ad-ministration of LHRHa until menstruation, re-placing HCG with LHRHa, injecting albumin in-travenously and freezing all embryos after IVF toavoid luteal phase treatments.

b

a

Advances in IVF in the 1980's

The major advance during these years includedthe introduction of ultrasound for the examin-ation of follicles, oocyte aspiration and for stud-ies on blood flow to the ovary and uterus (Table7). It would be hard to imagine IVF today with-out these major inputs. Its diagnostic value isincalculable, for it provides rapid, non-invasivesightings of the number and size of follicles,warnings of PCOD and hyperstimulation, andmay help to witness follicular rupture. Blood flowto the ovarian follicles and the uterus can becalculated. Echogenic patterns in the uterus upto mid-cycle are predictive of implantation. Ul-trasound of the implantation sac at day 16 hasbecome possible, and fetal heartbeat and otherparameters of growth can be scored from day20. Invasive methods relying on ultrasound haveenabled the use of coelocentesis to collectsamples for prenatal diagnosis from the coelomiccavity of human post-implantation embryos aged28 days and older (Jurkovic et al., 1993).

Fig. 17a Micromanipulation of amouse blastocyst to insert a singledonor cell and so form a chimaericembryo. The three needles open ahole in the trophectoderm, allowingthe injection pipette to be insertedinto the blastocoel. b. Chimaericmouse born after the injection of asingle stem cell into the blastocoeliccavity of a recipient blastocyst. Thehost blastocyst was albino, and thedonated cell was pigmented; the areasof pigment shown the sites of colonis-ation of melanocytes derived from thedonor cell. In many chimaeras, thetestis contained spermatogonic cellsderived from the donor cell whichcould achieve fertilization. Back-crosses involving his daughters and achimaeric male led to the birth of off-spring derived originally from the do-nated cell. (Courtesy of RichardGardner).


high androgen levels, and changes in gonado-trophin secretion. They were at risk of hyperstimu-lation and premature luteinization, and were usu-ally high responders. New forms of endocrinestimulation for them involved the use of low dosesof FSH and perhaps HMG to avoid hyperstimula-tion, or the stepwise administration of FSH.

Ovarian hyperstimulation syndrome was themost serious complication of ovulation induction.It arose in 1.2% of all patients treated withLHRHa/H MG. Severe hyperstimulation aroseafter HMG treatment in 5% of pregnancy cyclesand 1% of non-conception cycles. It was more fre-quent with large cohorts of medium-sized folliclesthan with two or three hyperfunctioning dominantfollicles, high levels of oestradiol, and the occur-rence of pregnany. Moderate and severe forms werea serious health risk to patients (Lunenfeld, 1969).Attempts to avoid it included continuing the ad-ministration of LHRHa until menstruation, re-placing HCG with LHRHa, injecting albumin in-travenously and freezing all embryos after IVF toavoid luteal phase treatments.

b

Advances in IVF ¡n the 1980's

The major advance during these years includedthe introduction of ultrasound for the examin-ation of follicles, oocyte aspiration and for stud-ies on blood flow to the ovary and uterus (Table7). It would be hard to imagine IVF today with-out these major inputs. Its diagnostic value isincalculable, for it provides rapid, non-invasivesightings of the number and size of follicles,warnings of PCOD and hyperstimulation, andmay help to witness follicular rupture. Blood flowto the ovarian follicles and the uterus can becalculated. Echogenic patterns in the uterus upto mid-cycle are predictive of implantation. Ul-trasound of the implantation sac at day 16 hasbecome possible, and fetal heartbeat and otherparameters of growth can be scored from day20. Invasive methods relying on ultrasound haveenabled the use of coelocentesis to collectsamples for prenatal diagnosis from the coelomiccavity of human post-implantation embryos aged28 days and older (Jurkovic et al., 1993).

Fig. 17a Micromanipulation of amouse blastocyst to insert a singledonor cell and so form a chimaericembryo. The three needles open ahole in the trophectoderm, allowingthe injection pipette to be insertedinto the blastocoel. b. Chimaericmouse born after the injection of a

a single stem cell into the blastocoeliccavity of a recipient blastocyst. Thehost blastocyst was albino, and thedonated cell was pigmented; the areasof pigment shown the sites of colonis-ation of melanocytes derived from thedonor cell. In many chimaeras, thetestis contained spermatogonic cellsderived from the donor cell whichcould achieve fertilization. Back-crosses involving his daughters and achimaeric male led to the birth of off-spring derived originally from the do-nated cell. (Courtesy of RichardGardner).

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


2Prag. 3

nmoVl 45

>5

o 10 20

Cycles %

Approx. 2000 babies in Bourn HallUltrasound of follicles and blood flowLaws from UK and Spanish ParliamentMany new forms of ovarian stimulationEmbryo cryopreservation introducedMany embryo studiesSperm donation/HRT cyclesDNA probes for preimplantation diagnosisEmbryo implantation rates low

30 40

Fig. 18 Proportion of cycles showing flare-up re-

sponses in LH and progesterone on day 2 after leuprol-ide administration (Courtesy of CD Matthews. Mat-thews et al., 1991).

Table 7 Advances in assisted conception between 1980and 1990

GIFT and ZIFT were also introduced as routineforms of treatment during these years. Their valuein assisted human conception has become some-what controversial. Pregnancy rates during GIFTcould be artificially raised by the fertilization ofthose oocytes remaining in their follicles despite at-tempts at their aspiration. There are suspicionsthat ZIFT involves selection for patients with manyembyros, which could raise the pregnancy rate byselecting the most fertile patients. These two tech-niques became the primary reasons for the reten-tion of laparoscopy for the collection of ocytes,and the need for this operation is perhaps thegreatest handicap with these methods.

Much research was carried out on preimplan-tation human embryos in culture. It became clearthat many human embryos growing in vitro, and

perhaps in vivo, were chromsomally imbalanced.Methods of embryo cryopreservation using pro-panediol, dimethyl sulphoxide and glycerol wereintroduced, primarily for pronucleate eggs, 4 cellembryos and blastocysts respectively (Lasalle etal., 1985). The cryopreservation of pronucleateeggs was the simplest approach, since the eggscould be easily identified, were all in the same stageof growth, and seemed to tolerate cryopreservationsuccessfully (Table 16).

Micromanipulation was introduced to improvehuman fertilization. Zona drilling and the inser-tion of spermatozoa beneath the zona pellucida at-tracted much attention, and some pregnancieswere estabished by this technique. The feasibilityof preimplantation diagnosis attracted the atten-tion of several clinics. Initially, the available meth-ods were too limiting or unsophisticated, but DNAtechnology soon improved sufficiently to produceprobes for various nucleotide sequences (Jones etal., 1987; Handyside et al., 1988). Methods for therecovery of one or more blastomeres for analysis,using micromanipulation on human embryos, de-veloped rapidly. The introduction of techniques tomeasure the uptake of substrates by embryos wasvaluable as a guide to their metabolism, but notfor preimplantation diagnosis since persisting ma-ternal characteristics in the embryos evidentlyswamped the embryonic contribution.

An agonadal patient with the XO syndrome con-ceived after oocyte donation and hormone replace-ment therapy (HRT) and delivered a child (Lutjenet al., 1984). The method was applied to womenwith other forms of agonadism and worked equallyeffectively for them. This study would also trans-form attitudes to pregnancies in post-menopausalwomen in the 1990's.

Ethical arguments about IVF reached parlia-ments in various countries. Debates in Victoriawere among the first and among the most vitriolic,and were dominated by a strict conservative line.Much more liberal attitudes prevailed in Spain andthe UK (only after similar vitriolic debates in theUK parliament), with the passage of legislationthat was sympathetic to research yet insisted on aclose control being maintained on the clinical andresearch aspects of IVF. The majority in favour ofIVF was 3:1 in the UK parliament, and a similarmajority supported the creation of embryos for re-search, provided the studies were completed by 14days (Table 8). Perhaps the efforts made by pro-fessionals in the field of assisted human conceptionto explain the ethics of IVF had convinced parlia-mentarians that research was essential and that itcould be carried out within an acceptable ethicalbasis. A German law based acceptable forms of re-search on human preimplantation embryos on the

<668

1012

LH 14mlU/ml 16

18202224

>24

o 5 10 15 20

Cycles %


2Prag. 3nmoVl 4

5>5

o 10 20

Cycles %

Fig. 18 Proportion of cycles showing flare-up re-

sponses in LH and progesterone on day 2 after leuprol-ide administration (Courtesy of CD Matthews. Mat-thews et al., 1991).

Table 7 Advances in assisted conception between 1980and 1990

Approx. 2000 babies in Bourn HallUltrasound of follicles and blood flowLaws from UK and Spanish ParliamentMany new forms of ovarian stimulationEmbryo cryopreservation introducedMany embryo studiesSperm donation/HRT cyclesDNA probes for preimplantation diagnosisEmbryo implantation rates low

3° 40

GIFT and ZIFT were also introduced as routineforms of treatment during these years. Their valuein assisted human conception has become some-what controversial. Pregnancy rates during GIFTcould be artificially raised by the fertilization ofthose oocytes remaining in their follicles despite at-tempts at their aspiration. There are suspicionsthat ZIFT involves selection for patients with manyembyros, which could raise the pregnancy rate byselecting the most fertile patients. These two tech-niques became the primary reasons for the reten-tion of laparoscopy for the collection of ocytes,and the need for this operation is perhaps thegreatest handicap with these methods.

Much research was carried out on preimplan-tation human embryos in culture. It became clearthat many human embryos growing in vitro, and

perhaps in vivo, were chromsomally imbalanced.Methods of embryo cryopreservation using pro-panediol, dimethyl sulphoxide and glycerol wereintroduced, primarily for pronucleate eggs, 4 cellembryos and blastocysts respectively (Lasalle etal., 1985). The cryopreservation of pronucleateeggs was the simplest approach, since the eggscould be easily identified, were all in the same stageof growth, and seemed to tolerate cryopreservationsuccessfully (Table 16).

Micromanipulation was introduced to improvehuman fertilization. Zona drilling and the inser-tion of spermatozoa beneath the zona pellucida at-tracted much attention, and some pregnancieswere estabished by this technique. The feasibilityof preimplantation diagnosis attracted the atten-tion of several clinics. Initially, the available meth-ods were too limiting or unsophisticated, but DNAtechnology soon improved sufficiently to produceprobes for various nucleotide sequences (Jones etal., 1987; Handyside et al., 1988). Methods for therecovery of one or more blastomeres for analysis,using micromanipulation on human embryos, de-veloped rapidly. The introduction of techniques tomeasure the uptake of substrates by embryos wasvaluable as a guide to their metabolism, but notfor preimplantation diagnosis since persisting ma-ternal characteristics in the embryos evidentlyswamped the embryonic contribution.

An agonadal patient with the XO syndrome con-ceived after oocyte donation and hormone replace-ment therapy (HRT) and delivered a child (Lutjenet al., 1984). The method was applied to womenwith other forms of agonadism and worked equallyeffectively for them. This study would also trans-form attitudes to pregnancies in post-menopausalwomen in the 1990's.

Ethical arguments about IVF reached parlia-ments in various countries. Debates in Victoriawere among the first and among the most vitriolic,and were dominated by a strict conservative line.Much more liberal attitudes prevailed in Spain andthe UK (only after similar vitriolic debates in theUK parliament), with the passage of legislationthat was sympathetic to research yet insisted on aclose control being maintained on the clinical andresearch aspects of IVF. The majority in favour ofIVF was 3:1 in the UK parliament, and a similarmajority supported the creation of embryos for re-search, provided the studies were completed by 14days (Table 8). Perhaps the efforts made by pro-fessionals in the field of assisted human conceptionto explain the ethics of IVF had convinced parlia-mentarians that research was essential and that itcould be carried out within an acceptable ethicalbasis. A German law based acceptable forms of re-search on human preimplantation embryos on the

<668

1012

LH 14mlU/mI 16

18202224

>24

o 5 10 15 20

Cycles %

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Table 8 Voting in the British Houses of Parliament onthe ethics of IVF

House of Lords3:1 in favour of embryo research3: 1 in favour of establishing embryos for research

House of Commons>2: 1 in favour of embryo researchClear majority in favour of research embryos

Table 9 Advances in assisted conception after 1990

ICSI and MESAMolecular approaches to ovary, fertilization, embryoand uterusPreimplantation diagnosis introducedSperm sexingRecombinant and pure FSH and LHPregnancies in women >50 yearsImplantation rates still low!!UK HFEA established

conoept of totipotency (Beier, 1991). This form ofdeciding acceptable legislation is highly unreliableand subject to unexpected changes as scientific opi-nion on the nature of totipotency is modified asnew studies are carried out on preimplantation em-bryos (Gardner, 1995).

C. Recent studies on IVF and especially on humanimplantation

Some modern studies on assisted human conception

The pace of research has accelerated even moreduring the 1990's, and many new advances duringthis time have led to fresh clinical opportunitiesand also ethical problems (Table 9). Pride of placeamong the advances in fertilization in vitro shouldperhaps be given to the introduction of ICSI (Table10) (Palermo et al., 1993; Van Steirteghem et al.,1993). It is very successful, since 60% or more ofthe eggs contain two pronuclei, even using sperma-tozoa collected from the epididymis or testis. Thistechnique now ensures that hardly any forms ofmale fertility remain untreatable, even includingsuch severe conditions as the Sertoli cell only syn-drome (Silber et al., 1995). Embryos are obtainedtoday in couples who produce more eggs thanspermatozoa! The introduction of spermatid injec-tions is likely to increase even more the number ofinfertile men who can conceive.

The introduction of ICSI has shown how noveland simple techniques can lead to completely newforms of treatment, and how these treatments canbe applied across the world in IVF clinics. ICSI is

Table 10 Some recent results with ICSI. Most of thesereports were presented at the ESHRE Annual Meeting,1994. (Edwards and Brody, 1995)

* 1275 ICSI cycles between October 1991 and De-cember 1993 in one clinic included triple sperm defectsin 49.6% men, and the use of testicular spermatozoa in17 cases; 16 109 cumulus eggs were injected, 82% inmetaphase 2. Among 13407 injected metaphase 2

oocytes, 66,4% were fertilised and 11.4% damaged;three-quarters of the embryos were transferrable. Re-placements were made in 91% of cycles, with 2 or 3 em-bryos in 80% of them; positive HGG tests occurred in460 cycles (36.1%), clinical pregnancies in 28.3%, 268prenatal karyotypes were made and 235 children born.* ICSI rescues I day-old unfertilized eggs and gives 48%fertilizations in them.* More human oocytes injected with medium plusa spermatozoon were activated (56%) than if givenmedium alone (14%).* 25/27 GV oocytes and 6/23 metaphase I oocytes de-generated during ICSI, and many of the others had dis-ordered polar body extrusion or pronucleus formation.* Sperm morphology does not effect fertilization or im-plantation rates during ICSI.* Epididymal and testicular spermatozoa have a lowerpotential to give full-term pregnancies with ICSI.* Epididymal spermatozoa gave 42% fertilization inICSI; if spermatozoa cannot be aspirated from the epi-didymis, a testis biopsy provides sufficient to give 42%fertilizations and many pregnancies.* Pressing the tails of spermatozoa against the bottomof the dish during ICSI immobilises and damages thegametes, and PVP probably stabilises this damage.

overall a very safe method, although there aresome unexpected genetic problems. There is appar-ently a higher risk of sex chromosome trisomies inembryos, and the gene causing congenital absenceof the vas has proved to be a variant of cystic fi-brosis. Patients must be carefully counselled anddiagnosed, yet the overall frequency of anomaliesin the children are virtually identical to those aris-ing after IVF or natural conception (Bonduelle etal., 1995). The success of ICSI has also led to a re-thinking of many fundamental aspects of fertili-zation, and has contributed to some fascinatingstudies on the inheritance of Y chromosome de-letions in severely oligozoospermic men (Reijo etal., 1995).

Studies on embryos are also advancing quickly.The frequency of anomalies at fertilization hasbeen well analysed, including a 5% frequency oftripronucleate eggs in vitro, despite the almost to-tal absence of such eggs and of triploid embryos inOldham. The biggest disappointment with modernembryological studies is that methods of embryoculture have not progressed far since the Oldhamdays. Many novel concepts have been applied toimprove embryonic growth, including the use of


Table 8 Voting in the British Houses of Parliament onthe ethics of IVF

House of Lords3: 1 in favour of embryo research3:1 in favour of establishing embryos for research

House of Commons>2:1 in favour of embryo researchClear majority in favour of research embryos

Table 9 Advances in assisted conception after 1990

ICSI and MESAMolecular approaches to ovary, fertilization, embryoand uterusPreimplantation diagnosis introducedSperm sexingRecombinant and pure FSH and LHPregnancies in women >50 yearsImplantation rates still low!!UK HFEA established

concept of totipotency (Beier, 1991). This form ofdeciding acceptable legislation is highly unreliableand subject to unexpected changes as scientific opi-nion on the nature of totipotency is modified asnew studies are carried out on preimplantation em-bryos (Gardner, 1995).

C. Recent studies on IVF and especially on humanimplantation

Some modern studies on assisted human conception

The pace of research has accelerated even moreduring the 1990's, and many new advances duringthis time have led to fresh clinical opportunitiesand also ethical problems (Table 9). Pride of placeamong the advances in fertilization in vitro shouldperhaps be given to the introduction of ICSI (Table10) (Palermo et al., 1993; Van Steirteghem et al.,1993). It is very successful, since 60% or more ofthe eggs contain two pronuclei, even using sperma-tozoa collected from the epididymis or testis. Thistechnique now ensures that hardly any forms ofmale fertility remain untreatable, even includingsuch severe conditions as the Sertoli cell only syn-drome (Silber et al., 1995). Embryos are obtainedtoday in couples who produce more eggs thanspermatozoa! The introduction of spermatid injec-tions is likely to increase even more the number ofinfertile men who can conceive.

The introduction of ICSI has shown how noveland simple techniques can lead to completely newforms of treatment, and how these treatments canbe applied across the world in IVF clinics. ICSI is

Table 10 Some recent results with ICSI. Most of thesereports were presented at the ESHRE Annual Meeting,1994. (Edwards and Brody, 1995)

* 1275 ICSI cycles between October 1991 and De-cember 1993 in one clinic included triple sperm defectsin 49.6% men, and the use of testicular spermatozoa in17 cases; 16 109 cumulus eggs were injected, 82% inmetaphase 2. Among 13407 injected metaphase 2oocytes, 66,4% were fertilised and 11.4% damaged;three-quarters of the embryos were transferrable. Re-placements were made in 91% of cycles, with 2 or 3 em-bryos in 80% of them; positive HGG tests occurred in460 cycles (36.1%), clinical pregnancies in 28.3%, 268prenatal karyotypes were made and 235 children born.* ICSI rescues 1 day-old unfertilized eggs and gives 48%fertilizations in them.* More human oocytes injected with medium plusa spermatozoon were activated (56%) than if givenmedium alone (14%).* 25/27 GV oocytes and 6/23 metaphase 1 oocytes de-generated during ICSI, and many of the others had dis-ordered polar body extrusion or pronucleus formation.* Sperm morphology does not effect fertilization or im-plantation rates during ICSI.* Epididymal and testicular spermatozoa have a lowerpotential to give full-term pregnancies with ICSI.* Epididymal spermatozoa gave 42% fertilization inICSI; if spermatozoa cannot be aspirated from the epi-didymis, a testis biopsy provides sufficient to give 42%fertilizations and many pregnancies.* Pressing the tails of spermatozoa against the bottomof the dish during ICSI immobilises and damages thegametes, and PVP probably stabilises this damage.

overall a very safe method, although there aresome unexpected genetic problems. There is appar-ently a higher risk of sex chromosome trisomies inembryos, and the gene causing congenital absenceof the vas has proved to be a variant of cystic fi-brosis. Patients must be carefully counselled anddiagnosed, yet the overall frequency of anomaliesin the children are virtually identical to those aris-ing after IVF or natural conception (Bonduelle etal., 1995). The success of ICSI has also led to a re-thinking of many fundamental aspects of fertili-zation, and has contributed to some fascinatingstudies on the inheritance of Y chromosome de-letions in severely oligozoospermic men (Reijo etal., 1995).

Studies on embryos are also advancing quickly.The frequency of anomalies at fertilization hasbeen well analysed, including a 5% frequency oftripronucleate eggs in vitro, despite the almost to-tal absence of such eggs and of triploid embryos inOldham. The biggest disappointment with modernembryological studies is that methods of embryoculture have not progressed far since the Oldhamdays. Many novel concepts have been applied toimprove embryonic growth, including the use of

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Table 11 Some recent data on preimplantation diag-nosis for genetic diseases and other traits using the poly-merase chain reaction. Most of these reports were pre-sented at the ESHRE Annual Meeting, 1994. (Edwardsand Brody, 1995)

* Deletion of exon 17 of DMD locus in single blasto-mere to identify non-afflicted male embryos.* Single-copy X- and Y-linked amelogenin sequences(AMX and AMY) to sex cleaving embryos successful inmost embryos.* Rapid multiplexing of conserved amelogenin gene seg-ments from both X and Y chromosomes plus Y-linkedDYZI repetitive elements from biopsies and humanblastocysts with <2% sexing errors and exclusion of ab-errant samples.* Multiplexing of microsatellite, amelogenin and CFprimers gave simultaneous DNA fingerprinting, sexingand detection of CF status.* Amplification of ZFX/ZFY followed by restrictionenzyme digestion to sex single human blastomeres.* Amplification of 1 2 cells from cleaving embryos forcystic fibrosis; affected and normal girl delivered afterdiagnosis.* Amplification of I - or 2-cell biopsies from 8 cell em-bryos of 7 couples at risk of cystic fibrosis, 2 for Lesch-Nyhan, and 25 for sex-linked diseases: 16 clinical preg-nancies, 3 aborted, 4 ongoing and 9 normal babies.* Diagnoses of W1282 and AF508 mutants of cysticfibrosis in single blastomeres identified homozygousnormal, homozygous abnormal and heterotygotes.* Correct diagnoses on 15 normal, 12 carriers and 7homozygous cystic fibrosis patients (AF508) by a newmethod of PCR amplication.

feeder cell layers and more complex media, yet im-plantation rates per embryo remain stubbornlylow. In contrast to this static situation, the intro-duction of preimplantation diagnosis of geneticdisease is moving quickly (Table 11). its molecularbiology is exquisite, since several alleles can beamplified from a single cell using PCR (Li et al.,1989), whole genomic amplification (Zhang et al.,1992; Grifo et al., 1995; Handyside, 1996) and flu-orescent PCR (Findlay et al., 1995). Perhaps 50 ormore genes will soon be amplified from a singlecell. FISH has also been perfected to give a rapid2-hour method of identifying the X and Y chromo-somes, chromosome 18 and chromosome 21 simul-taneously. Some clinics are scoring all embryos intheir patients aged >40, to check for trisomy or formonosomy for XY, 13, 18 and 21. Embryos withthese conditions are not transferred to theirmothers, to avert the birth of children carryingthese chromosomal anomalies.

Other major advances in genetic diagnosis areimminent. Spermatozoa from several species, in-

Sorted for X

RecombinedX&Y

Sorted for Y

Actual

Predicted

I-,

0 20 40 60 80 100

Male rabbits %

Fig. 19 Results of sorting rabbit sperm samples for Xspermatozoa, using Johnson's methods. The upper histo-gram shows the proportion of X spermatozoa in samplessorted for Y spermatozoa (i.e., more than 80% of Xspermatozoa have been removed). The lower histogramshows the results of sorting for X spermatozoa, and themiddle histogram gives the data when sorted sampleswere recombined. The expected frequency is derivedfrom the proportion of X spermatozoa in severalsamples, and the actual frequency gives the sex ratios ofthe offspring. Similar results were gained using humanspermatozoa (Johnson et al., 1993).

cluding human spermatozoa, can now be flow-sorted into two populations containing an over-whelming majority of X or Y spermatozoa respec-tively (Figure 19) (Johnson et al., 1993; Levinsonet al., 1995). The success of X-sorting in 1993 was70-80V0, and better methods are certain to be in-troduced to improve this degree of sorting evenmore. Simple and reliable methods of sex choice,based on successful work with several animal spe-cies, are thus becoming available in clinical prac-tice. These methods will probably be acceptable so-cially if they are used to avert the birth of childrencarrying sex-linked inherited disease. There willprobably be many arguments if they are used forthe choice of boy or girl for social reasons.

Options for stimulation widened further as theLHRH antagonists entered clinical practice.LHRH antagonists such as Nal-G1u6 suppress im-munoreactive and bioactive FSH and LH levels inwomen for 72 hrs or longer. FSH and oestradiollevels rise steadily, while LH levels remain lowwhen Cetrorelix is administered from day 7 ofovarian stimulation with HMG/HCG (Table 12)(Diedrich et al., 1993). Purified and recombinantgonadotrophins also became available and wereused successfully in IVF (Devroey et al., 1992).Doubts remain about the need for pure FSH inovarian stimulation protocols, since it is expensiveand the LH component of HMG may have littleclinical significance. The contributions of manyfine researchers in companies such as Organon,Serono and Schering in making these exciting newproducts available must be applauded. Manyoocytes could be aspirated from the ovaries of


Table 11 Some recent data on preimplantation diag-nosis for genetic diseases and other traits using the poly-merase chain reaction. Most of these reports were pre-sented at the ESHRE Annual Meeting, 1994. (Edwardsand Brody, 1995)

* Deletion of exon 17 of DMD locus in single blasto-mere to identify non-afflicted male embryos.* Single-copy X- and Y-linked amelogenin sequences(AMX and AMY) to sex cleaving embryos successful inmost embryos.* Rapid multiplexing of conserved amelogenin gene seg-ments from both X and Y chromosomes plus Y-linkedDYZ1 repetitive elements from biopsies and humanblastocysts with <2V0 sexing errors and exclusion of ab-errant samples.* Multiplexing of microsatellite, amelogenin and CFprimers gave simultaneous DNA fingerprinting, sexingand detection of CF status.* Amplification of ZFX/ZFY followed by restrictionenzyme digestion to sex single human blastomeres.* Amplification of l-2 cells from cleaving embryos forcystic fibrosis; affected and normal girl delivered afterdiagnosis.* Amplification of I- or 2-cell biopsies from 8 cell em-bryos of 7 couples at risk of cystic fibrosis, 2 for Lesch-Nyhan, and 25 for sex-linked diseases: 16 clinical preg-nancies, 3 aborted, 4 ongoing and 9 normal babies.* Diagnoses of Wl282 and AF508 mutants of cysticfibrosis in single blastomeres identified homozygousnormal, homozygous abnormal and heterotygotes.* Correct diagnoses on 15 normal, 12 carriers and 7homozygous cystic fibrosis patients (AF508) by a newmethod of PCR amplication.

feeder cell layers and more complex media, yet im-plantation rates per embryo remain stubbornlylow. In contrast to this static situation, the intro-duction of preimplantation diagnosis of geneticdisease is moving quickly (Table 11). its molecularbiology is exquisite, since several alleles can beamplified from a single cell using PCR (Li et al.,1989), whole genomic amplification (Zhang et al.,1992; Grifo et al., 1995; Handyside, 1996) and flu-orescent PCR (Findlay et al., 1995). Perhaps 50 ormore genes will soon be amplified from a singlecell. FISH has also been perfected to give a rapid2-hour method of identifying the X and Y chromo-somes, chromosome 18 and chromosome 21 simul-taneously. Some clinics are scoring all embryos intheir patients aged >40, to check for trisomy or formonosomy for XY, 13, 18 and 21. Embryos withthese conditions are not transferred to theirmothers, to avert the birth of children carryingthese chromosomal anomalies.

Other major advances in genetic diagnosis areimminent. Spermatozoa from several species, in-

Sorted for X

RecombinedX&Y

Sorted for Y

Male rabbits %

Fig. 19 Results of sorting rabbit sperm samples for Xspermatozoa, using Johnson's methods. The upper histo-gram shows the proportion of X spermatozoa in samplessorted for Y spermatozoa (i.e., more than 80% of Xspermatozoa have been removed). The lower histogramshows the results of sorting for X spermatozoa, and themiddle histogram gives the data when sorted sampleswere recombined. The expected frequency is derivedfrom the proportion of X spermatozoa in severalsamples, and the actual frequency gives the sex ratios ofthe offspring. Similar results were gained using humanspermatozoa (Johnson et al., 1993).

cluding human spermatozoa, can now be flow-sorted into two populations containing an over-whelming majority of X or Y spermatozoa respec-tively (Figure 19) (Johnson et al., 1993; Levinsonet al., 1995). The success of X-sorting in 1993 was70-80V, and better methods are certain to be in-troduced to improve this degree of sorting evenmore. Simple and reliable methods of sex choice,based on successful work with several animal spe-cies, are thus becoming available in clinical prac-tice. These methods will probably be acceptable so-cially if they are used to avert the birth of childrencarrying sex-linked inherited disease. There willprobably be many arguments if they are used forthe choice of boy or girl for social reasons.

Options for stimulation widened further as theLHRH antagonists entered clinical practice.LI-IR H antagonists such as Nal-Glu6 suppress im-munoreactive and bioactive FSH and LH levels inwomen for 72 hrs or longer. FSH and oestradiollevels rise steadily, while LH levels remain lowwhen Cetrorelix is administered from day 7 ofovarian stimulation with HMG/I-ICG (Table 12)(Diedrich et al., 1993). Purified and recombinantgonadotrophins also became available and wereused successfully in IVF (Devroey et al., 1992).Doubts remain about the need for pure FSH inovarian stimulation protocols, since it is expensiveand the LH component of HMG may have littleclinical significance. The contributions of manyfine researchers in companies such as Organon,Serono and Schering in making these exciting newproducts available must be applauded. Manyoocytes could be aspirated from the ovaries of

Actual

Predicted

I-,

1000 20 40 60 80

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Table 12 Control of mean LH levels with daily injec-tions of the LHRH antagonist Cetrorelix after stimula-tion with HMG/HCG (Diedrich et al., 1993)

* Cetrorelix administered; + Ovulatory HCG admini-stered; progesterone levels remained consistently lowthroughout the follicular phase

Table 13 Oncogenes and oocyte maturation

most patients, and very large numbers from a few,using stimulation protocols consisting of LHRHagonists, HMG or FSH and HCG. Yet, at thistime, concern began to grow about the risks of ov-arian epithelial cancer as a result of prolonged ov-arian stimulation (\Vhittemore et al., 1992).

Molecular analyses became increasingly funda-mental to many aspects of research into humanconception. One example evolved from studies onXenopus and other invertebrates and concerned themle of various oncogenes in oocyte maturation(Table 13). These genes, e.g. pp391110', are presum-ably active in human oocytes. The mos protein en-coded by this gene has been detected in maturehuman oocytes, and as in other species it appar-ently arrests meiosis at metaphase 2 until the sper-matozoon fuses with the oocyte membrane. Theuse of gene knock-out to eliminate pp39hh10S in micehas shown how the oocytes are unable to arrestat metaphase 2, and they immediately proceed toanaphase and beyond if this gene has been elimin-ated from their mother. Molecular studies are alsobeing applied to human implantation, and I shallconclude this lecture with brief analyses of recentmolecular and clinical studies on implantation andpregnancy after IVF.

Molecular basis of human implantation

Preimplantation embryos apparently send varioustypes of signals to the mother. Hatched blastocysts

Table 14 Trophoblast expression: some of the receptorsand other properties displayed by trophoblast of humanand animal embryos as implantation approaches

may signal the mother via embryo-derived plateletactivating factor (EDPAF) within 48 hours of ferti-lization and by early pregnancy factor (EPF). Em-bryos also synthesise oestrogens, e.g. oestradiol issecreted by human blastocysts at 5-8 days post-fertilization (Edgar et al., 1993; Edgar, 1994). Tro-phoblast of the implanting embryo expresses manyreceptors (Table 14), so that the embryo increas-ingly resembles a complex tissue where cytokines,integrins, growth factors and their receptors are ex-pressed.

Modifications in the uterus involve the forma-tion of decidua which is essential to sustain theimplanting embryo. Another uterine activity in-volves the expression of progesterone-dependentpinopods on uterine epithelium between days 19and 21. Their function is seemingly to extract fluidfrom the uterine lumen, so that the embryo andthe epithelium are drawn into a very close physicalcontact. This tight apposition results in an ex-tremely close association between embryo and epi-thelium. The association is so close that inden-tations forced into the trophoblast resemble theoutlines of epithelial pinopodes and microvilli. Itmay also enable reactions to occur between com-plementary binding proteins and other factors onembryos and uterine epithelium during this firststage of implantation (Psychoyos, 1993; Nikas etal., 1995).

Embryo adhesion to the uterine epithelium viapolar trophoblast marks the beginning of the im-plantation process. As trophoblast and epithelialcells are closely apposed, cytoplasmic projectionsfrom the implanting embryo extend long tropho-blastic protusions between epithelial cells whichpenetrate to reach the basement membrane. Fi-bronectin and other substrates may then guide theimplanting embryo until it is downgraded by tro-phoblast. Trophoblast also adheres to laminin,which is distributed around human decidual cells,and it may also secrete this substrate to help withmutual recognition with the uterus (Loke et al.,1989).

Many intermediary factors are expressed in theuterus and embryo during implantation, e.g. EGF,

Day of Mean levels of plasma hormone Integrins: fibronectin receptors (a4b1]cycle FSH LH Oestradiol-1713 lamm receptors

mIU/ml lU/mI pg/mI vitronectin receptors (ab1)UPA receptors

57*

7.07.8

43.5

50100

IL-1 R activated by UFIGF-R, EGF-R, UlF receptor

10 9.0 2.25 250 CSFl4 11.75 2.0 950 Fibronectin

cdc2. cdc2 34 kd kinase cyclins and cdc2 form MPFcdc32. phosphorylates MTOCsc-kit. cell signals, kinase receptorc-fms. Ligand is SL factor (kit)c-myc. nuclear DNA binding proteinc-fos. regulates gene expressionc-abl. protein kinasec-mos. mos protein in M 2 arrest


Table 12 Control of mean LH levels with daily injec-tions of the LHRH antagonist Cetrorelix after stimula-tion with HMG/HCG (Diedrich et al., 1993)

* Cetrorelix administered; + Ovulatory HCG admini-stered; progesterone levels remained consistently lowthroughout the follicular phase

Table 13 Oncogenes and oocyte maturation

most patients, and very large numbers from a few,using stimulation protocols consisting of LHRHagonists, HMG or FSH and HCG. Yet, at thistime, concern began to grow about the risks of ov-arian epithelial cancer as a result of prolonged ov-arian stimulation (Whittemore et al., 1992).

Molecular analyses became increasingly funda-mental to many aspects of research into humanconception. One example evolved from studies onXenopus and other invertebrates and concerned therole of various oncogenes in oocyte maturation(Table 13). These genes, e.g. pp39"'°', are presum-ably active in human oocytes. The mos protein en-coded by this gene has been detected in maturehuman oocytes, and as in other species it appar-ently arrests meiosis at metaphase 2 until the sper-matozoon fuses with the oocyte membrane. Theuse of gene knock-out to eliminate pp3910s in micehas shown how the oocytes are unable to arrestat metaphase 2, and they immediately proceed toanaphase and beyond if this gene has been elimin-ated from their mother. Molecular studies are alsobeing applied to human implantation, and I shallconclude this lecture with brief analyses of recentmolecular and clinical studies on implantation andpregnancy after IVF.

Molecular basis of human implantation

Preimplantation embryos apparently send varioustypes of signals to the mother. Hatched blastocysts

Table 14 Trophoblast expression: some of the receptorsand other properties displayed by trophoblast of humanand animal embryos as implantation approaches

may signal the mother via embryo-derived plateletactivating factor (EDPAF) within 48 hours of ferti-lization and by early pregnancy factor (EPF). Em-bryos also synthesise oestrogens, e.g. oestradiol issecreted by human blastocysts at 5-8 days post-fertilization (Edgar et al., 1993; Edgar, 1994). Tro-phoblast of the implanting embryo expresses manyreceptors (Table 14), so that the embryo increas-ingly resembles a complex tissue where cytokines,integrins, growth factors and their receptors are ex-pressed.

Modifications in the uterus involve the forma-tion of decidua which is essential to sustain theimplanting embryo. Another uterine activity in-volves the expression of progesterone-dependentpinopods on uterine epithelium between days 19and 21. Their function is seemingly to extract fluidfrom the uterine lumen, so that the embryo andthe epithelium are drawn into a very close physicalcontact. This tight apposition results in an ex-tremely close association between embryo and epi-thelium. The association is so close that inden-tations forced into the trophoblast resemble theoutlines of epithelial pinopodes and microvilli. Itmay also enable reactions to occur between com-plementary binding proteins and other factors onembryos and uterine epithelium during this firststage of implantation (Psychoyos, 1993; Nikas etal., 1995).

Embryo adhesion to the uterine epithelium viapolar trophoblast marks the beginning of the im-plantation process. As trophoblast and epithelialcells are closely apposed, cytoplasmic projectionsfrom the implanting embryo extend long tropho-blastic protusions between epithelial cells whichpenetrate to reach the basement membrane. Fi-bronectin and other substrates may then guide theimplanting embryo until it is downgraded by tro-phoblast. Trophoblast also adheres to laminin,which is distributed around human decidual cells,and it may also secrete this substrate to help withmutual recognition with the uterus (Loke et al.,1989).

Many intermediary factors are expressed in theuterus and embryo during implantation, e.g. EGF,

Day of Mean levels of plasma hormone Integrins: libronectin receptors (a4b1]cycle FSH LH Oestradiol-17!3 lamm receptors

mIU/ml IU/ml pg/ml vitronectin receptors (ab1)UPA receptors

57*

7.07.8

43.5

50100

IL-1 R activated by UFIGF-R, EGF-R, LIF receptor

10 9.0 2.25 250 CSF14 11.75 2.0 950 Fibronectin

cdc2. cdc2 34 kd kinase cyclins and cdc2 form MPFcdc32. phosphorylates MTOCsc-kit. cell signals, kinase receptorc-fms. Ligand is SL factor (kit)c-myc. nuclear DNA binding proteinc-fos. regulates gene expressionc-abl. protein kinasec-mos. mos protein in M 2 arrest

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


which may stimulate epithelial cells and gland pro-liferation. Uterine cytokines include interleukins(IL-1 to IL-13), interferons (IFN-a, f3, y), colony-stimulating factors (M-CSF, C-CSF, GM-CSF),tumour necrosis factors (TNF-a, f3), activin, inhi-bin, and nerve- and platelet-derived growth factors(Tabibzadeh et al., 1994) They regulate the uterinemicroenvironment via intercellular signals, andmany of them are influenced by steroid hormonesor locally-released interferons (Bischoff et al.,1995). The four classes of interferons are IFN-al-pha, beta, gamma and omega. T cell interferonsregulate cytokine activity and local cell-cell inter-actions, perhaps via the mediation of PGE2 andHLA-DR. IFN-a, IFN-y, and TNFa may mediateluteal regression. The cytokine cascade is one ofthe many complex interactions in the uterus involv-ing cytokines, and interferons.

Various matrices, basement membranes andblood vessels, including gelatin, collagen, fibronec-tin and laminin, confront the implanting embryo.They may guide and anchor the embryo throughthe uterine epithelium and stroma. Adhesion mol-ecules, e.g. integrins, immunoglobulin superfamily,selectins and cadherins, are expressed on cell sur-faces and interact with matrix membranal sub-strates. Integrins mediate cell binding to fibronec-tin, which is cross-linked with collagens (especiallycollagen I), fibrin, heparin and DNA. Laminin isexpressed, together with collagen IV, in blood ves-sel and gland basement membranes during themid-secretory phase and in stroma and decidualater. Laminin, fibronectin and collagen IV in ex-tracellular matrices are probably regulated by in-terferons.

The three integrin families, B1, B2 and B1, arealso present in the extracellular matrix, and ligandsof B1 include fibronectin, laminin and collagen IV.There is an immense variability of integrin ex-pression, enabling them to bind with laminin, col-lagen, fibronectin, vitronectin and von Wille-brand's factor. Glandular epithelium expressesmainly a2, a1 and a6 (e.g. collagen/laminin a6b1 re-ceptors) and a1 during the secretory phase (Lesseyet al., 1994). Stroma predominantly expresses a5b1(the fibronectin receptor). A defective expressionof ab3 may cause infertility, and an absence of b3(called type I infertility) results in an out-of-phaseendometrium which is correctable by progesterone.Growth factors may regulate ab1 and a4b1 ex-pression on endometrial microvilli as they do withspecific receptors on trophoblast. ab1 (the vi-

tronectin receptor) is expressed cyclically and b1normally appears on day 20. The expression of a1,a4 and b1 on glandular epithelium at days 20-24of the menstrual cycle may define the implantationwindow in the human uterus (Table 15).

Table 15 Uterine integrins in epithelium and stroma,and during the implantation window (Lessey et al., 1994)

Eoitheliuni (glandular + luminal)Constitutive a2, a3, a4, b1, b4Cyclic a1 & a4 luteal only

StroniaConstitutive a5, b1Cyclic a1, ab3 (prolif)

Implantation windoit' days 20-24Coexpression of a1, a4, b3

Invasive trophoblast successively expresses dif-ferent integrin classes. These various classes prob-ably mediate the initial attachment of the embryo,invasion of extracellular matrix 10-15 hours afterhatching, the activation of decidual metalloprote-ases and movement of extravillous trophoblast tothe spiral arteries. The expressed integrins includereceptors for fibronectin, laminin and vitronectin.

Cytotrophoblast secretes the matrix metallopro-teinases (MMPs), i.e. collagenases, gelatinases andstromelysins. Collagenases digest collagen types I,II, III, VII and X, gelatinase digests collagen typeIV and gelatin, and stromelysins digest fibronectin,laminin, collagens IV, V and VII (Matrisian, 1990).Type IV collagenases and serine protease uPA de-grade matrices and remodel extracellular matrixduring implantation. They are regulated by steroidhormones and growth factors, e.g. EGF, IL-1,TGF-f3 which balance the activities of proMMp-1and TIMP-l. Invasive trophoblast also expressescytokeratin, high affinity uPA receptors, HLAclass I framework antigens, receptors for integrinsand LIF, and the integrins a3, a5, b1 and ab3/b5(the vitronectin receptor) which may regulate thepenetration of extravillous trophoblast to the spiralarteries. Extravillous trophoblast expresses thenon-polymorphic histocompatibility antigen HLA-G (Bischoff et al., 1995), and migrates to the uter-ine blood vessels, probably via the expression ofM-CSF and the c-fins and c-kit genes which en-code for specific ligands. Targetted mutations ofthe LIF gene impair implantation in mice, whereasop/op mice (with a congenital absence of CSF)are infertile.

It is obviously essential not to oversimplify theparacrinology of the uterus during implantation.Many complicated interactions occur, each in-

volved with specific uterine components, such asthe limitation of trophoblast invasion, recognitionof the embryonic HLA antigens and the regulationof blood flow to uterus and placenta. The closerelationship between the duration of the implan-tation window, the expression of uterine integrinsand the existence of the pinopods suggests thatthey are controlled by the same regulatory factors


which may stimulate epithelial cells and gland pro-liferation. Uterine cytokines include interleukins(IL-1 to IL-13), interferons (IFN-a, ¡3, y), colony-stimulating factors (M-CSF, C-CSF, GM-CSF),tumour necrosis factors (TNF-a, ¡3), activin, inhi-bin, and nerve- and platelet-derived growth factors(Tabibzadeh et al., 1994) They regulate the uterinemicroenvironment via intercellular signals, andmany of them are influenced by steroid hormonesor locally-released interferons (Bischoff et al.,1995). The four classes of interferons are IFN-al-pha, beta, gamma and omega. T cell interferonsregulate cytokine activity and local cell-cell inter-actions, perhaps via the mediation of PGE2 andHLA-DR. IFN-, IFN-y, and TNFa may mediateluteal regression. The cytokine cascade is one ofthe many complex interactions in the uterus involv-ing cytokines, and interferons.

Various matrices, basement membranes andblood vessels, including gelatin, collagen, fibronec-tin and laminin, confront the implanting embryo.They may guide and anchor the embryo throughthe uterine epithelium and stroma. Adhesion mol-ecules, e.g. integrins, immunoglobulin superfamily,selectins and cadherins, are expressed on cell sur-faces and interact with matrix membranal sub-strates. Integrins mediate cell binding to fibronec-tin, which is cross-linked with collagens (especiallycollagen I), fibrin, heparin and DNA. Laminin isexpressed, together with collagen IV, in blood ves-sel and gland basement membranes during themid-secretory phase and in stroma and decidualater. Laminin, fibronectin and collagen IV in ex-tracellular matrices are probably regulated by in-terferons.

The three integrin families, B1, B2 and B1, arealso present in the extracellular matrix, and ligandsof B1 include fibronectin, laminin and collagen IV.There is an immense variability of integrin ex-pression, enabling them to bind with laminin, col-lagen, fibronectin, vitronectin and von Wille-brand's factor. Glandular epithelium expressesmainly a2, a3 and a6 (e.g. collagen/laminin a6b1 re-ceptors) and a1 during the secretory phase (Lesseyet al., 1994). Stroma predominantly expresses a5b1(the fibronectin receptor). A defective expressionof ab3 may cause infertility, and an absence of b3(called type I infertility) results in an out-of-phaseendometrium which is correctable by progesterone.Growth factors may regulate ab3 and a4b1 ex-pression on endometrial microvilli as they do withspecific receptors on trophoblast. ab3 (the vi-

tronectin receptor) is expressed cyclically and b3normally appears on day 20. The expression of a1,a4 and b3 on glandular epithelium at days 20-24of the menstrual cycle may define the implantationwindow in the human uterus (Table 15).

Table 15 Uterine integrins in epithelium and stroma,and during the implantation window (Lessey et al., 1994)

Eoithelium (glandular + luminal)Constitutive a2, a3, a4, b1, b4Cyclic a1 & a4 luteal only

StromaConstitutive a5, b1Cyclic a1, abs (prolif)

Implantation window days 20-24Coexpression of a1, a4, b3

Invasive trophoblast successively expresses dif-ferent integrin classes. These various classes prob-ably mediate the initial attachment of the embryo,invasion of extracellular matrix 10-15 hours afterhatching, the activation of decidual metalloprote-ases and movement of extravillous trophoblast tothe spiral arteries. The expressed integrins includereceptors for fibronectin, laminin and vitronectin.

Cytotrophoblast secretes the matrix metallopro-teinases (MMPs), i.e. collagenases, gelatinases andstromelysins. Collagenases digest collagen types I,II, III, VII and X, gelatinase digests collagen typeIV and gelatin, and stromelysins digest fibronectin,laminin, collagens IV, V and VII (Matrisian, 1990).Type IV collagenases and serine protease uPA de-grade matrices and remodel extracellular matrixduring implantation. They are regulated by steroidhormones and growth factors, e.g. EGF, IL-1,TGF-13 which balance the activities of proMMp-1and TIMP- 1. Invasive trophoblast also expressescytokeratin, high affinity uPA receptors, HLAclass I framework antigens, receptors for integrinsand UF, and the integrins a3, a5, b1 and ab3/b5(the vitronectin receptor) which may regulate thepenetration of extravillous trophoblast to the spiralarteries. Extravillous trophoblast expresses thenon-polymorphic histocompatibility antigen HLA-G (Bischoff et al., 1995), and migrates to the uter-ine blood vessels, probably via the expression ofM-CSF and the c-fus and c-kit genes which en-code for specific ligands. Targetted mutations ofthe LIF gene impair implantation in mice, whereasop/op mice (with a congenital absence of CSF)are infertile.

It is obviously essential not to oversimplify theparacrinology of the uterus during implantation.Many complicated interactions occur, each in-

volved with specific uterine components, such asthe limitation of trophoblast invasion, recognitionof the embryonic HLA antigens and the regulationof blood flow to uterus and placenta. The closerelationship between the duration of the implan-tation window, the expression of uterine integrinsand the existence of the pinopods suggests thatthey are controlled by the same regulatory factors.

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Table 16 Implantation rates per embryo in acyclicwomen given HRT therapy and in cyclic women usingcryopreserved embryos (Edwards et al., 1991)

No of Implantation ratescycles All patients Pregnant only

Table 17 Pregnancies in women aged between 45 and63, using data collected from various clinics

New clinical concepts on human implantation

Clinical studies attempting to raise pregnancyrates after assisted human conception providemuch information on implantation. One remark-able finding concerns the high fertility of agona-dal and postmenopausal recipients given donatedoocytes plus HRT therapy (Edwards et al., 1991).These patients include women with ovarian dysg-enesis, premature menopause, X-chromosome ab-normalities, and those having an early meno-pause after surgery. They remain highly fertile upto the age of 60 and beyond, and age-for-age aremore fertile than cyclic women of equivalent ages(Edwards and Brody, 1995). Agonadal womenaged <45 are exceptionally fertile, with implan-tation rates per embryo >2O% and even higheras compared with 11% in normal TYF patients.Similar results are gained for women in mid-late40's, some of them post-menopausal (Serhal andCraft, 1989). Women aged 40-43, 44-47 and>47 established 31% clinical pregnancies pertransfer, with no variation due to maternal age(Table 17). Couples aged 50-59 had 38% preg-nancies per attempt and a l9% implantation rate,and pregnancies have been established in womenup to age 63. These exceptional results are notdue to differences in donor age.

Some investigators reason that their high rateof pregnancy after oocyte donation and HRT isdue to a period of amenorrhoea (Marcus andEdwards, 1994; Borini et al., 1995). This con-clusion is based on the high pregnancy rates(>40%) obtained when cyclic women treated forendometriosis are down-regulated with LHRHagonists for 4 months or more, to become highlyfertile after TYF. Their pregnancy rates reach 40%

and more, while controls remain at l0%. Similarvalues are obtained when perimenopausal cyclicwomen of the same ages are given down-regu-lation for several cycles (Borini et al., 1995).

Perhaps amenorrhoea enables uterine steroid-sensitive systems to recover from the effects ofconstant menstrual cycles. The pinopodes arehighly progesterone-sensitive (Nikas et al., 1995).Their malfunction would distort the evacuationof uterine fluids, so that embryos may have diffi-culty in adhering to the uterine epithelium. Newapproaches to implantation could involve the useof progesterone antagonists such as RU 486 toreduce any overexposure of uterine organelles toprogesterone (Beier et al., 1994).

Other approaches to raising pregnancy rates afterIVF

Finally, two other approaches to raising preg-nancy rates after TYF must be considered. Oneinvolves the use of aspirin to improve blood per-fusion in women with an impaired blood flow(Wada et al., 1994). The second approach in-volves the immunology of reproduction. Thetreatment of women with anti-phospholipid anti-bodies who are undergoing TYF with heparin andaspirin raises their pregnancy rates from lO-2O%to almost 50% (Sher et al., 1994). The positiveeffects of the treatment may involve some formof protection to the early syncitiotrophoblast, be-fore or after implantation. Similar results are be-ing gained using anti-gamma globulin injectionsin women undergoing TYF (Coulam et al.,1994).

Clearly, these leads into the improvement ofimplantation rates are of immense importance forTYF and its derivatives. The low rate of implan-tation per embryo (12-15%) has been a problemsince Louise Brown was conceived; perhaps thisperiod is finally coming to an end.

References

Austin CR: Observations on the penetration of thesperm into the mammalian egg. J. Sci. Res. Ser B.,4: 581-596, 1951.

Beatty RA: Parthenogeneis and Polyploidy in Mam-malian Development. Cambridge University Press,1957.

Beier HM, Hegele-Hartung C, Mootz U, Beier-HellwegK: Modification of endometrial cell biology usingprogesterone antagonists to manipulate the implan-tation window. Hum. Reprod., 9: Supplement I,98-115, 1994.

Beier HM: Implications and consequences of the Ger-man Embryo Protection Act. Hum. Reprod., 6:607, 1991.

Details of pregnancies

Clinic 2 3 4Pregnancies 31 50 66 44Clinical pregnancy

rate 28.1 28.5 31.4Implantation rate 14.4 17.4 11.9 12.6Abortions 4 8 7

Cyclic 542 12.1 38.1Acyclic 78 19.4 41.7


Table 16 Implantation rates per embryo in acyclicwomen given HRT therapy and in cyclic women usingcryopreserved embryos (Edwards et al., 1991)

No of Implantation ratescycles All patients Pregnant only

Table 17 Pregnancies in women aged between 45 and63, using data collected from various clinics

New clinical concepts on human implantation

Clinical studies attempting to raise pregnancyrates after assisted human conception providemuch information on implantation. One remark-able finding concerns the high fertility of agona-dal and postmenopausal recipients given donatedoocytes plus HRT therapy (Edwards et al., 1991).These patients include women with ovarian dysg-enesis, premature menopause, X-chromosome ab-normalities, and those having an early meno-pause after surgery. They remain highly fertile upto the age of 60 and beyond, and age-for-age aremore fertile than cyclic women of equivalent ages(Edwards and Brody, 1995). Agonadal womenaged <45 are exceptionally fertile, with implan-tation rates per embryo >20% and even higheras compared with 1 1% in normal IVF patients.Similar results are gained for women in mid-late40's, some of them post-menopausal (Serhal andCraft, 1989). Women aged 40-43, 44-47 and>47 established 31% clinical pregnancies pertransfer, with no variation due to maternal age(Table 17). Couples aged 50-59 had 38% preg-nancies per attempt and a 19% implantation rate,and pregnancies have been established in womenup to age 63. These exceptional results are notdue to differences in donor age.

Some investigators reason that their high rateof pregnancy after oocyte donation and HRT isdue to a period of amenorrhoea (Marcus andEdwards, 1994; Borini et al., 1995). This con-clusion is based on the high pregnancy rates(>40%) obtained when cyclic women treated forendometriosis are down-regulated with LHRHagonists for 4 months or more, to become highlyfertile after IVF. Their pregnancy rates reach 40%

and more, while controls remain at lO%. Similarvalues are obtained when perimenopausal cyclicwomen of the same ages are given down-regu-lation for several cycles (Borini et al., 1995).

Perhaps amenorrhoea enables uterine steroid-sensitive systems to recover from the effects ofconstant menstrual cycles. The pinopodes arehighly progesterone-sensitive (Nikas et al., 1995).Their malfunction would distort the evacuationof uterine fluids, so that embryos may have diffi-culty in adhering to the uterine epithelium. Newapproaches to implantation could involve the useof progesterone antagonists such as RU 486 toreduce any overexposure of uterine organelles toprogesterone (Beier et al., 1994).

Other approaches to raising pregnancy rates afterIVF

Finally, two other approaches to raising preg-nancy rates after IVF must be considered. Oneinvolves the use of aspirin to improve blood per-fusion in women with an impaired blood flow(Wada et al., 1994). The second approach in-volves the immunology of reproduction. Thetreatment of women with anti-phospholipid anti-bodies who are undergoing IVF with heparin andaspirin raises their pregnancy rates from 10-20%to almost 50% (Sher et al., 1994). The positiveeffects of the treatment may involve some formof protection to the early syncitiotrophoblast, be-fore or after implantation. Similar results are be-ing gained using anti-gamma globulin injectionsin women undergoing IVF (Coulam et al.,1994).

Clearly, these leads into the improvement ofimplantation rates are of immense importance forIVF and its derivatives. The low rate of implan-tation per embryo (12 15%) has been a problemsince Louise Brown was conceived; perhaps thisperiod is finally coming to an end.

References

Austin CR: Observations on the penetration of thesperm into the mammalian egg. J. Scï. Res. Ser B.,4: 581-596, 1951.

Beatty RA: Parthenogeneis and Polyploidy in Mam-malian Development. Cambridge University Press,1957.

Beier HM, Hegele-Hartung C, Mootz U, Beier-HeliwegK: Modification of endometrial cell biology usingprogesterone antagonists to manipulate the implan-tation window. Hum. Reprod., 9: Supplement 1,

98-115, 1994.Beier HM: Implications and consequences of the Ger-

man Embryo Protection Act. Hum. Reprod., 6:607, 1991.

Details of pregnancies

Clinic 1 2 3 4Pregnancies 31 50 66 44Clinical pregnancy

rate 28.1 28.5 31.4Implantation rate 14.4 17.4 11.9 12.6Abortions 4 8 7

Cyclic 542 12.1 38.1Acyclic 78 19.4 41.7

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Bischoff P, Haenggeli L. and Campana A: Gelatinaseand oncofetal fibronectin secretion are dependentupon integrin expression in human cytotrophoblasts.Mol. Hum. Reprod. I, see Hum. Reprod., 10:734-742, 1995.

Bonduelle M, Legern J, Derde M-P, Buysse A, Wis-anto A, Devroey P, Van Steirteghem A, Liebars J:Comparative follow-up study of 130 children bornafter intyracytoplasmic sperm injection and 130 chil-dren born after in vitro fertilization. Hum Reprod.,10: 3327-3331, 1995.

Borini A, Violini F, Bianchi L, Bafaro MG, TrevisiMR, Flamigni C: Imoprovement in pregnancy andimplantation rates in cyclic women undergoingoocyte donation after long-term down regulation.Hum Reprod., 10: 3018-3021, 1995.

Chang MC: Fertilizing capacity of spermatozoa de-posited into the fallopian tubes. Nature, 168:697-698, 1951.

Chang MC: The maturation of rabbit oocytes in cul-ture and their maturation, activation, fertilizationand subsequent development in the Fallopian tubes.J. Exp. Zool, 128: 379-405, 1955.

Cole RJ, Edwards RG, Paul, J: Cytodifferentiation incell colonies and cell starins derived from cleavingova and blastocysts of the rabit. Exp. Cell Res., 37:501-504, 1965.

Cole RJ, Edwards RG, Paul, J: Cytodifferentiation andembryogenesis in cell colonies and tissue culturesderived from ova and blastocysts of the rabbit. Dey.Biol., 13: 385-407, 1966.

Coulam C, Krysa LW, Bustillo M: Intravenous immun-oglobulins for in-vitro fertilization Hum Reprod 9:2265-2269, 1994.

Csapo AI, Pulkkinen MO, Kaihala HL: The relation-ship between the timing of luteectomy and the inci-dence of complete abortions. Amer J Obstet Gynecol118: 985-989, 1974.

Dauzier L, Thibault C: Recherche experimentale surla maturation des gametes males chez les mammife-res par l'etude de la fecondation in vitro de l'oeufde la lapine. Third Ann Cong Anim Reprod, Cam-bridge, Section 1, p 58, 1956.

Devroey P, Van Steirteghem AV, Mannaerts-BenninkHC: Lancet 340: 1108-1109, 1992.

Diedrich K, Diedrich C, Santos E, Zoll C, Al-HasaniS, Reissman T, Krebs D, Kingmiller D: Suppressionof the endogenous luteinizing hormone surge by thegonadotrophin-releasing hormone antagonist Cetror-elix during ovarian stimulation. Hum Reprod 9:788-793, 1993.

Edgar DH: Oestrogen and human implantation. HumReprod 10: 2-3, 1994.

Edgar DH, James GB, Mills JA: Seroid secretion byearly human embryos in culture. Hum Reprod 8:277-278, 1993.

Edwards RG: The experimental induction of gynogen-esis in the mouse. I. Irradiation of the sperm by Xrays. Proc Roy Soc Lond, B, 146: 469-487, 1957

Edwards RG: Maturation in vitro of mouse, sheep,cow, pig, rhesus monkey and human ovarianoocytes. Nature 208: 349-351, 1965a.

Edwards RG: Maturation in vitro of human ovarianoocytes Lancet ii: 926-929, 1965b.

Edwards RG: Studies on human conception. Amer JObstet Gynecol 117: 587-601, 1973.

Edwards RG: Conception in the Human Female. Aca-demic Press, London, 1980.

Edwards RG: Fertilization of human eggs in vitro:morals, ethics and the law. Quart Rev Biol 49:3-26, 1974.

Edwards RG, Bavister BD, Steptoe PC: Early stagesof fertilization in vitro of human oocytes maturedin vitro Nature 221: 632-635, 1969.

Edwards RG, Brody S: Principles and Practice of As-sisted Human Reproduction. WB Saunders, Philadel-phia, 1995.

Edwards RG, Gardner RL: Sexing of live rabbit blasto-cysts. Nature, 214: 576-577, 1967.

Edwards RG, Gates AH: Timing of the stages of thematuration divisions, ovulation, fertilization and thefirst cleavage of eggs of adult mice treated withgonadotrophins. J Endocrinol 18: 292-304, 1959.

Edwards RG, Morcos S, Macnamee M, Balmaceda JP,Walters DE, Asch R: High fecundity ofamenorrhoeic women in embryo-transfer pro-grammes. Lancet 338: 292-294, 1991.

Edwards RG, Sharpe DJ: Social values and researchin human embryology. Nature, 231: 87-91, 1971.

Edwards RG, Steptoe PC: A Matter of Life. Hutchin-sons, 1980.

Edwards RG, Steptoe PC, Purdy JM: Establishing full-term human pregnancies using cleaving embryosgrown in vitro. Brit J Obstet Gynecol 87:757-768, 1980.

Evans HM, Simpson ME. Experimental superfecunditywith pituitary gonadotrophins. Endocrinology: 27,305-308, 1940.

Findlay I, Ray P, Quirke P, Rutherford A, Lilford R:Allelic drop-out and preferential amplification in sin-gle cells and human blastomeres: implications forpreimplantation diagnosis of sex and cystic fibrosis.Mol. Hum. Reprod. I, see Hum Reprod, 10,1905-1013, 1995.

Fleming R, Coutts JRT: Induction of multiple follicu-lar development for IVE In: RG Edwards (ed). As-sisted Human Conception. Brit Med Bull 46:596-615, 1990.

Fowler RE, Edwards RG: Induction of superovulationand pregnancy in mature mice by gonadotrophins. JEndocrinol 15: 74-384, 1957.

Fowler RE, Edwards RG, Walters DE, Chan STH,Steptoe PC: Steroidogenesis in preovulatory folliclesof patients given human menopausal and chorionicgonadotrophins as judged by the radioimmunoassayof steroids in follicular fluid. J Endocrinol 77:161-169, 1978a.

Fowler RE, Fox NL,, Edwards RG, Walters DE,Steptoe PC: Steroidogenesis by cultured granulosacells aspirated from human follicles using pregneno-lone and androgens as precursors. J Endocrinol 77:171-183, 1978b.

Frangenheim H: Geburtsh. Frauenhlkd. 24: 470, 1964.Frydman R, Cornel C, de Ziegler D, Taieb J, Spitz

IM, Bouchard P: Spontaneous luteinising hormonesurges can be reliably prevented by the timely ad-ministration of a gonadotrophin releasing hormoneanatgonist (Nal-Glu) during the late follicular phase.Hum Reprod 7: 930-933, 1992.


Bischoff P, Haenggeli L. and Campana A: Gelatinaseand oncofetal fibronectin secretion are dependentupon integrin expression in human cytotrophoblasts.Mol. Hum. Reprod. 1, see Hum. Reprod., 10:734-742, 1995.

Bonduelle M, Legern J, Derde M-P, Buysse A, Wis-anto A, Devroey P, Van Steirteghem A, Liebars J:Comparative follow-up study of 130 children bornafter intyracytoplasmic sperm injection and 130 chil-dren born after in vitro fertilization. Hum Reprod.,10: 3327-3331, 1995.

Borini A, Violini F, Bianchi L, Bafaro MG, TrevisiMR, Flamigni C: Imoprovement in pregnancy andimplantation rates in cyclic women undergoingoocyte donation after long-term down regulation.Hum Reprod., 10: 3018-3021, 1995.

Chang MC: Fertilizing capacity of spermatozoa de-posited into the fallopian tubes. Nature, 168:697-698, 1951.

Chang MC: The maturation of rabbit oocytes in cul-ture and their maturation, activation, fertilizationand subsequent development in the Fallopian tubes.J. Exp. Zoo!, 128: 379-405, 1955.

Cole RJ, Edwards RG, Paul, J: Cytodifferentiation incell colonies and cell starins derived from cleavingova and blastocysts of the rabit. Exp. Cell Res., 37:501-504, 1965.

Cole RJ, Edwards RG, Paul, J: Cytodifferentiation andembryogenesis in cell colonies and tissue culturesderived from ova and blastocysts of the rabbit. Dey.Biol., 13: 385-407, 1966.

Coulam C, Krysa LW, Bustillo M: Intravenous immun-oglobulins for in-vitro fertilization Hum Reprod 9:2265-2269, 1994.

Csapo AI, Pulkkinen MO, Kaihala HL: The relation-ship between the timing of luteectomy and the inci-dence of complete abortions. Amer J Obstet Gynecol118: 985-989, 1974.

Dauzier L, Thibault C: Recherche experimentale surla maturation des gametes males chez les mammife-res par l'etude de la fecondation in vitro de l'oeufde la lapine. Third Ann Cong Anim Reprod, Cam-bridge, Section 1, p 58, 1956.

Devroey P, Van Steirteghem AV. Mannaerts-BenninkHC: Lancet 340: 1108-1109, 1992.

Diedrich K, Diedrich C, Santos E, Zoll C, Al-HasaniS, Reissman T, Krebs D, Kingmiller D: Suppressionof the endogenous luteinizing hormone surge by thegonadotrophin-releasing hormone antagonist Cetror-eux during ovarian stimulation. Hum Reprod 9:788-793, 1993.

Edgar DH: Oestrogen and human implantation. HumReprod 10: 2-3, 1994.

Edgar DH, James GB, Mills JA: Seroid secretion byearly human embryos in culture. Hum Reprod 8:277-278, 1993.

Edwards RG: The experimental induction of gynogen-esis in the mouse. I. Irradiation of the sperm by Xrays. Proc Roy Soc Lond, B, 146: 469-487, 1957

Edwards RG: Maturation in vitro of mouse, sheep,cow, pig, rhesus monkey and human ovarianoocytes. Nature 208: 349-351, l965a.

Edwards RG: Maturation in vitro of human ovarianoocytes Lancet ii: 926-929, 1965b.

Edwards RG: Studies on human conception. Amer JObstet Gynecol 117: 587-601, 1973.

Edwards RG: Conception in the Human Female. Aca-demic Press, London, 1980.

Edwards RG: Fertilization of human eggs in vitro:morals, ethics and the law. Quart Rev Biol 49:3-26, 1974.

Edwards RG, Bavister BD, Steptoe PC: Early stagesof fertilization in vitro of human oocytes maturedin vitro Nature 221: 632-635, 1969.

Edwards RG, Brody S: Principles and Practice of As-sisted Human Reproduction. WB Saunders, Philadel-phia, 1995.

Edwards RG, Gardner RL: Sexing of live rabbit blasto-cysts. Nature, 214: 576-577, 1967.

Edwards RG, Gates AH: Timing of the stages of thematuration divisions, ovulation, fertilization and thefirst cleavage of eggs of adult mice treated withgonadotrophins. J Endocrino! 18: 292-304, 1959.

Edwards RG, Morcos S, Macnamee M, Balmaceda JP,Walters DE, Asch R: High fecundity ofamenorrhoeic women in embryo-transfer pro-grammes. Lancet 338: 292-294, 1991.

Edwards RG, Sharpe DJ: Social values and researchin human embryology. Nature, 231: 87-91, 1971.

Edwards RG, Steptoe PC: A Matter of Life. Hutchin-sons, 1980.

Edwards RG, Steptoe PC, Purdy JM: Establishing full-term human pregnancies using cleaving embryosgrown in vitro. Brit J Obstet Gynecol 87:757-768, 1980.

Evans HM, Simpson ME. Experimental superfecunditywith pituitary gonadotrophins. Endocrinology: 27,305-308, 1940.

Findlay I, Ray P, Quirke P, Rutherford A, Lilford R:Allelic drop-out and preferential amplification in sin-gle cells and human blastomeres: implications forpreimplantation diagnosis of sex and cystic fibrosis.Mol. Hum. Reprod. 1, see Hum Reprod, 10,1905-1013, 1995.

Fleming R, Coutts JRT: Induction of multiple follicu-lar development for IVE In: RG Edwards (ed). As-sisted Human Conception. Brit Med Bull 46:596-615, 1990.

Fowler RE, Edwards RG: Induction of superovulationand pregnancy in mature mice by gonadotrophins. JEndocrinol 15: 74-384, 1957.

Fowler RE, Edwards RG, Walters DE, Chan STH,Steptoe PC: Steroidogenesis in preovulatory folliclesof patients given human menopausal and chorionicgonadotrophins as judged by the radioimmunoassayof steroids in follicular fluid. J Endocrinol 77:161-169, 1978a.

Fowler RE, Fox NL,, Edwards RG, Walters DE,Steptoe PC: Steroidogenesis by cultured granulosacells aspirated from human follicles using pregneno-lone and androgens as precursors. J Endocrinol 77:171-183, 1978b.

Frangenheim H: Geburtsh. Frauenhlkd. 24: 470, 1964.Frydman R, Corne! C, de Ziegler D, Taieb J, Spitz

IM, Bouchard P: Spontaneous luteinising hormonesurges can be reliably prevented by the timely ad-ministration of a gonadotrophin releasing hormoneanatgonist (Nal-Glu) during the late follicular phase.Hum Reprod 7: 930-933, 1992.

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Gardner RL: Mouse chimaeras obtained by the injec-tion of cells into the blastocyst. Nature, 220:596-597, 1968.

Gardner RL: Can developmentally-significant spatialpatterning of the egg be discounted? Hum ReprodUpdate, I: in press, 1995.

Gardner RL and Edwards RG: Control of the sexratio in the rabbit at full term by transferring sexedblastocysts. Nature 218: 346-348, 1968.

Gates AH, Beatty RA: Independence of delayed fertili-zation and spontaneous triploidy in mouse embryos.Nature, 174: 356-357, 1954.

Gemzell CA. Induction of ovulation with menopausalgonadotrophins with special reference to the dosageand administration pattern of FSH. In: Fifth WorldCongress on Obstetrics and Gynaecology, p 240, 1967.

Grifo JA, Tang YX, Munne S, Alikani M, Cohen J, Ro-senwaks Z: Healthy deliveries from biopsied haumanembryos. Hum Reprod, 9: 912-916, 1994.

Hammond J: Nature, 163: 28-29, 1949.Handyside A: Preimplantation diagnosis. Hum Reprod

Il: Supplement, 1996. In press.Handyside: AH, Kontogianni EH, Hardy K, Winston

RM. Pregnancies from biopsied human preimplan-tation embryos sexed by Y-specific DNA amplifi-cation. Nature, 344: 768-770, 1990.

Harris GW: Phil. Trans. Roy. Soc. Lond. B, 259:165-177, 1970.

Heape W: Preliminary note on the transplantation andgrowth of mammalian ova within a uterine fostermother. Proc. Roy. Soc. Lond., 48: 457-458.

Hertig AT, Rock J, Adams EC, Muklligan WJ: On thepreimplantation stages of the human ovum: A de-scription of four normal and four abnormal specimensranging from the second to the fifth day of develop-ment. Contributions to Embryology, 33: 685-695,1954

Howles C, Macnamee M, Edwards RG, Goswamy RK,Steptoe PC: Effect of high tonic levels of luteinisinghormone on incidence of in-vitro fertilization. Lancet2:521-522, 1986.

Johnson LA, Welch GR, Keyvanfar K, Dorfman A,Fugger EF, Schulman JD: Gender preselection in hu-mans? Flow cytometric separation of X and Y sper-matozoa for the prevention of X-linked disease. HumReprod 8: 1733-1739, 1993.

Jones HW Jr, Jones GS, Andrews MC, et al.: Fertil Steril38: 14-21, 1982.

Jones KW, Singh L, Edwards RG: The use of probes forthe Y chromosome in preimplantation embryo cells.Hum Reprod, 2: 439-445, 1987.

Jurkovic D, Jauniaux E, Campbell S, Pandya P, CardyDLN, Nicolaides KH: Coelocentesis: a new techniquefor early prenatal diagnosis. Lancet, 341: 1623-1624,1933.

Kistner RW: Ovulation: clinical aspects. In: H. Balm, S.Glasser (eds) Reproductive Biology, p 477. ExcerptaMecica, Amsterdam, 1972.

Lassalle B, Testart J, Renard J-P: Human embryo fea-tures that influence the success of cryopreservationwith the use of 1,2-propanediol. Fertil Steril, 44:414-426, 1985.

Lessey BA, Castelbaum AJ, Buck CA, et al.: Fertil Steril62: 497-506, 1994.

Levinson G, Keyvanfar K, Wu JC, Fugger EF, et al.:DNA-based X-enriched sperm separation as an ad-

junct to preimplantation genetic testing for the pre-vention of X-linked disease. Mol Hum Reprod I,Supp Hum Reprod 10: 979-982, 1995.

Lewis WH, Gregory PW: Science, 69: 226-229, 1933.Lewis WH, Hartman CG: Carnegie Inst Wash Public,

24: 187-203, 1933.Leyendecker G, Bernart W, Bremen Th, Beck H, Kunz

G, Waibel S, Blum A, Stenger P, Kaplan-Reiterer H:Influence of the duration of the oestradiol rise on thesuccess rate in GnRH analogue/HMG-stimulated IVFcycles. Hum. Reprod. 5: 52-55 1990.

Li A, Byllensten UB, Cui X, Saki RK, Ehrlich HA,Arnheim N: Amplification and analysis of DNA se-quences in single human sperm and diploid cells. Na-ture, 335: 414, 1989.

Loke YW, Gardner RL, Burland K, King A: Lamininin human trophoblast-decidua interaction. Hum. Re-prod. 4, 457-463, 1989

Lunenfeld B. The use of human menopausal gonado-trophins in gonadal pathology. In: W. Inguilla et al.(eds) The Ovary. Chapter 15. C.B. Thomas, Spring-field, Ohio, 1969.

Lutjen PJ, Trounson AO, Leeton FJ, Wood C, Renou P:The establishment and maintenance of pregnancyusing in vitro fertilization and embryo donation in apatient with primary ovarian failure. Nature,307:174-176, 1994.

Marcus S, Edwards RG: High rates of pregnancy afterlong-term down-regulation of women with severe en-dometriosis. Amer J Obstet Gynecol 171: 812-817,1994.

Matthews CD, Warnes GM, Norman RJ, Phillipson G,Kirby CA, Wang X: The leuprolide flare regime forin-vitro fertilization/gamete intra-Fallopian transferand embryo cryopreservation. Hum. Reprod., 6:8 17-820, 1991.

Matrisian L. Metalloproteinases and their inhibitors inmatrix remodelling. Trends in Genet., 6:121-125,1990.

Menkin MF, Rock J: In vitro fertilization and cleavageof human ovraian eggs. Amer J Obstet Gynecol 55:440, 1948.

Nikas G, Drakakis P, Loutardis D, Mara-Skonfari C,Koumantakis E, Michaelis S, Psychoyos A: Uterinepinopods as markers of the 'nidation window' in cy-clic women receiving exogenous ostradiol and pro-gesterone. Hum. Reprod., 10: 1208-1213, 1995.

Palermo G, Jons H, Devroey P, Van Steirteghem AC:Pregnancies after intracytoplasmic injection of singlespermatozoa into an oocyte. Lancet, 340: 17 18, 1992.

Palmer R: La coelioscopie. Acad Chir 77: 363, 1946.Parkes AS: Induction of superovulation and superfe-

cundation in rabbits. J Endocrinol, 3: 268-279, 1942.Paul J: Cell and Tissue Culture. E. and S. Livingstone,

Edinburgh, 1961.Pincus G: Superovulation in rabbits. Anat Rec, 77:

1-8, 1940.Pincus G, Saunders B: The comparitive behaviour of

mammalian eggs in vivo and in vitro. VI. The matu-ration of human ovarian ova. Anat. Rec., 75,537, 1939.

Psychoyos A: Proceedings of the VIIth InternationalIVF Conference, Kyoto, 1993.

Reijo R, Lee T-Y, Salo P, et al.: Diverse spermatogenicdefects in humans caused by Y chromosome deletionsencompassing a novel RNA-binding protein. NatureGenet, 19: 383-393, 1995.


Gardner RL: Mouse chimaeras obtained by the injec-tion of cells into the blastocyst. Nature, 220:596-597, 1968.

Gardner RL: Can developmentally-significant spatialpatterning of the egg be discounted? Hum ReprodUpdate, 1: in press, 1995.

Gardner RL and Edwards RG: Control of the sexratio in the rabbit at full term by transferring sexedblastocysts. Nature 218: 346-348, 1968.

Gates AH, Beatty RA: Independence of delayed fertili-zation and spontaneous triploidy in mouse embryos.Nature, 174: 356-357, 1954.

Gemzell CA. Induction of ovulation with menopausalgonadotrophins with special reference to the dosageand administration pattern of FSH. In: Fifth WorldCongress on Obstetrics and Gynaecology, p 240, 1967.

Grifo JA, Tang YX, Munne S, Alikani M, Cohen J, Ro-senwaks Z: Healthy deliveries from biopsied haumanembryos. Hum Reprod, 9: 912-916, 1994.

Hammond J: Nature, 163: 28-29, 1949.Handyside A: Preimplantation diagnosis. Hum Reprod

11: Supplement, 1996. In press.Handyside: AH, Kontogianni EH, Hardy K, Winston

RM. Pregnancies from biopsied human preimplan-tation embryos sexed by Y-specific DNA amplifi-cation. Nature, 344: 768-770, 1990.

Harris GW: Phil. Trans. Roy. Soc. Lond. B, 259:165-177, 1970.

Heape W: Preliminary note on the transplantation andgrowth of mammalian ova within a uterine fostermother. Proc. Roy. Soc. Lond., 48: 457-458.

Hertig AT, Rock J, Adams EC, Mukiligan Wi: On thepreimplantation stages of the human ovum: A de-scription of four normal and four abnormal specimensranging from the second to the fifth day of develop-ment. Contributions to Embryology, 33: 685-695,1954

Howles C, Macnamee M, Edwards RG, Goswamy RK.Steptoe PC: Effect of high tonic levels of luteinisinghormone on incidence of in-vitro fertilization. Lancet2: 521-522, 1986.

Johnson LA, Welch GR, Keyvanfar K, Dorfman A,Fugger EF, Schulman JD: Gender preselection in hu-mans? Flow cytometric separation of X and Y sper-matozoa for the prevention of X-linked disease. HumReprod 8: 1733-1739, 1993.

Jones HW ir, Jones GS, Andrews MC, et al.: Fertil Steril38: 14-21, 1982.

Jones KW, Singh L, Edwards RG: The use of probes forthe Y chromosome in preimplantation embryo cells.Hum Reprod, 2: 439-445, 1987.

Jurkovic D, Jauniaux E, Campbell S, Pandya P, CardyDLN, Nicolaides KH: Coelocentesis: a new techniquefor early prenatal diagnosis. Lancet, 341: 1623 1624,1933.

Kistner RW: Ovulation: clinical aspects. In: H. Balm, S.Glasser (eds) Reproductive Biology, p 477. ExcerptaMecica, Amsterdam, 1972.

Lassalle B, Testart J, Renard J-P: Human embryo fea-tures that influence the success of cryopreservationwith the use of 1,2-propanediol. Fertil Steril, 44:414-426, 1985.

Lessey BA, Castelbaum AJ, Buck CA, et al.: Fertil Steril62: 497-506, 1994.

Levinson G, Keyvanfar K, Wu JC, Fugger EF, et al.:DNA-based X-enriched sperm separation as an ad-

junct to preimplantation genetic testing for the pre-vention of X-linked disease. Mol Hum Reprod 1,Supp Hum Reprod 10: 979-982, 1995.

Lewis WH, Gregory PW: Science, 69: 226-229, 1933.Lewis WH, Hartman CG: Carnegie Inst Wash Public,

24: 187-203, 1933.Leyendecker G, Bernart W, Bremen Th, Beck H, Kunz

G, Waibel S, Blum A, Stenger P, Kaplan-Reiterer H:Influence of the duration of the oestradiol rise on thesuccess rate in GnRH analogue/HMG-stimulated IVFcycles. Hum. Reprod. 5: 52-55 1990.

Li A, Byllensten UB, Cui X, Saki RK, Ehrlich HA,Arnheim N: Amplification and analysis of DNA se-quences in single human sperm and diploid cells. Na-ture, 335: 414, 1989.

Loke YW, Gardner RL, Burland K, King A: Lamininin human trophoblast-decidua interaction. Hum. Re-prod. 4, 457-463, 1989

Lunenfeld B. The use of human menopausal gonado-trophins in gonadal pathology. In: W. Inguilla et al.(eds) The Ovary. Chapter 15. C.B. Thomas, Spring-field, Ohio, 1969.

Lutjen Pi, Trounson AO, Leeton Fi, Wood C, Renou P:The establishment and maintenance of pregnancyusing in vitro fertilization and embryo donation in apatient with primary ovarian failure. Nature,307:174-176, 1994.

Marcus S, Edwards RG: High rates of pregnancy afterlong-term down-regulation of women with severe en-dometriosis. Amer J Obstet Gynecol 171: 812-817,1994.

Matthews CD, Warnes GM, Norman RJ, Phillipson G,Kirby CA, Wang X: The leuprolide flare regime forin-vitro fertilization/gamete intra-Fallopian transferand embryo cryopreservation. Hum. Reprod., 6:817-820, 1991.

Matrisian L. Metalloproteinases and their inhibitors inmatrix remodelling. Trends in Genet., 6:121-125,1990.

Menkin MF, Rock J: In vitro fertilization and cleavageof human ovraian eggs. Amer J Obstet Gynecol 55:440, 1948.

Nikas G, Drakakis P, Loutardis D, Mara-Skonfari C,Koumantakis E, Michaelis S, Psychoyos A: Uterinepinopods as markers of the 'nidation window' in cy-clic women receiving exogenous ostradiol and pro-gesterone. Hum. Reprod., 10: 1208-1213, 1995.

Palermo G, Jons H, Devroey P, Van Steirteghem AC:Pregnancies after intracytoplasmic injection of singlespermatozoa into an oocyte. Lancet, 340: 17 18, 1992.

Palmer R: La coelioscopie. Acad Chir 77: 363, 1946.Parkes AS: Induction of superovulation and superfe-

cundation in rabbits. J Endocrinol, 3: 268-279, 1942.Paul J: Cell and Tissue Culture. E. and S. Livingstone,

Edinburgh, 1961.Pincus G: Superovulation in rabbits. Anat Rec, 77:

1-8, 1940.Pincus G, Saunders B: The comparitive behaviour of

mammalian eggs in vivo and in vitro. VI. The matu-ration of human ovarian ova. Anat. Rec., 75,537, 1939.

Psychoyos A: Proceedings of the VIIth InternationalIVF Conference, Kyoto, 1993.

Reijo R, Lee T-Y, Salo P, et al.: Diverse spermatogenicdefects in humans caused by Y chromosome deletionsencompassing a novel RNA-binding protein. NatureGenet, 19: 383-393, 1995.

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.


Runner MN, Gates AH: Conception in prepuberal micefollowing artificially induced ovulation and mating.Nature, 174: 222-223, 1954.

Serhal P, Craft IL: Oocyte donation in 61 patients. Lan-cet i: 1185-1187, 1989.

Sher G, Feinman M, Zouves C, Kuttner G, MassaraniG, Salem R, Matzner W, Ching W, Chong P: Highfecundity rates following in-vitro fertilization and em-bryo transfer in antiphospholipid antibody seroposi-tive women treated with heparin and aspirin. Hum.Reprod. 9: 2278-2283, 1994.

Shettles LB: A morula stage of human ova developed invitro. Fertil Steril 6: 287, 1955.

Silber SJ: What forms of male infertility are there left tocure? Hum Reprod. 10: 503-504, 1995.

Smith PE, Engle ET: Experimental evidence regardingrole of anterior pituitary in development and regu-lation of genital system. Amer J Anat, 40: 159-217,1927.

Steptoe PC. Laparoscopy in Gynaecology. E. and S. Liv-ingstone, Edinburgh, 1967.

Steptoe PC, Edwards RG: Laparoscopic recovery of pre-ovulatory human ovarian oocytes after priming ofovaries with gonadotrophins. Lancet i: 683-689,1970.

Steptoe PC, Edwards RG: Reimplantation of a humanembryo with subsequent tubal pregnancy. Lancet, i:880-882, 1976

Steptoe PC, Edwards RG, Purdy JM: Clinical aspects ofpregnancies established with embryos grown in vitro.Brit J Obtset Gynaecol, 87: 757-768, 1980.

Tabibzadeh S: Cytokines and the hypothalamic-pitu-itary-ovarian-endometrial axis. Hum Reprod, 9:947 967.

Trounson AO, Leeton JF, Wood C, Webb J, Wood J:Pregnancies in humans by fertilization in vitro andembryo transfer in the controlled ovulatory cycle. Sci-ence, 212: 681, 1981.

Van Steirteghem AC, Nagy Z, Jons H, Liu J, StaessenC, Smitz J, Wisanto, Devroey P: High fertilizationand implantation rates after intracytoplasmic sperminjection. Hum Reprod, 8, 1961-1068, 1993.

Wada I, Hsu CC, Williams G, Macnammee MC,Brinsden PRI: The benefits of low-dose aspirin ther-apy in women with impaired uterine perfusion. Hum.Reprod9: 1954-1957, 1994.

Whittemore AS, Harris R, Intyre J, and the Collabora-tive Ovarian Cancer group. Characteristics relating toovarian cancer risk: collaborative analysis of 12 UScase-control studies. IV. The pathogenesis of epithelialovarian cancer. Am J Epidemiol, 18: 538-545, 1992.

Whitten WK, Biggers JD: Complete development in vi-tro of the preimplantation stages of the mouse in asimple chemically defined medium. J Reprod Fertil 17:339-401, 1968.

Yanagimachi R, Chang MC: In vitro fertilization ofgolden hamster ova. J Exp Zool, 156: 361-376, 1964.

Zhang L, Cui XF, Schmitt K, Hubert R, Navidi W,Arnheim N: Whole genome amplification from a sin-gle cell: implications for genetic analysis. Proc NatAcad Sci., 89: 5847-5851, 1992.

Correspondence address: Prof. Dr. R. G. Edwards, Chur-chill College, Cambridge CB3 ODS, 113 115 HarleyStreet, London, United Kingdom


Runner MN, Gates AH: Conception in prepuberal micefollowing artificially induced ovulation and mating.Nature, 174: 222-223, 1954.

Serhal P, Craft IL: Oocyte donation in 61 patients. Lan-cet i: 1185-1187, 1989.

Sher G, Feinman M, Zouves C, Kuttner G, MassaraniG, Salem R, Matzner W, Ching W, Chong P: Highfecundity rates following in-vitro fertilization and em-bryo transfer in antiphospholipid antibody seroposi-tive women treated with heparin and aspirin. Hum.Reprod. 9: 2278-2283, 1994.

Shettles LB: A morula stage of human ova developed invitro. Fertil Steril 6: 287, 1955.

Silber SJ: What forms of male infertility are there left tocure? Hum Reprod. 10: 503-504, 1995.

Smith PE, Engle ET: Experimental evidence regardingrole of anterior pituitary in development and regu-lation of genital system. Amer J Anat, 40: 159-217,1927.

Steptoe PC. Laparoscopy in Gynaecology. E. and S. Liv-ingstone, Edinburgh, 1967.

Steptoe PC, Edwards RG: Laparoscopic recovery of pre-ovulatory human ovarian oocytes after priming ofovaries with gonadotrophins. Lancet i: 683-689,1970.

Steptoe PC, Edwards RG: Reimplantation of a humanembryo with subsequent tubal pregnancy. Lancet, i:880-882, 1976

Steptoe PC, Edwards RG, Purdy JM: Clinical aspects ofpregnancies established with embryos grown in vitro.Brit J Obtset Gynaecol, 87: 757-768, 1980.

Tabibzadeh S: Cytokines and the hypothalamic-pitu-itary-ovarian-endometrial axis. Hum Reprod, 9:947 967.

Trounson AO, Leeton JF, Wood C, Webb J, Wood J:Pregnancies in humans by fertilization in vitro andembryo transfer in the controlled ovulatory cycle. Sci-ence, 212: 681, 1981.

Van Steirteghem AC, Nagy Z, Jons H, Liu J, StaessenC, Smitz J, Wisanto, Devroey P: High fertilizationand implantation rates after intracytoplasmic sperminjection. Hum Reprod, 8, 1961-1068, 1993.

Wada I, Hsu CC, Williams G, Macnammee MC,Brinsden PRI: The benefits of low-dose aspirin ther-apy in women with impaired uterine perfusion. Hum.Reprod9: 1954-1957, 1994.

Whittemore AS, Harris R, Intyre J, and the Collabora-tive Ovarian Cancer group. Characteristics relating toovarian cancer risk: collaborative analysis of 12 UScase-control studies. IV. The pathogenesis of epithelialovarian cancer. Am J Epidemiol, 18: 538-545, 1992.

Whitten WK, Biggers JD: Complete development in vi-tro of the preimplantation stages of the mouse in asimple chemically defined medium. J Reprod Fertil 17:339-401, 1968.

Yanagimachi R, Chang MC: In vitro fertilization ofgolden hamster ova. J Exp Zool, 156: 361-376, 1964.

Zhang L, Cui XF, Schmitt K, Hubert R, Navidi W,Arnheim N: Whole genome amplification from a sin-gle cell: implications for genetic analysis. Proc NatAcad Sci., 89: 5847-5851, 1992.

Correspondence address: Prof. Dr. R. G. Edwards, Chur-chill College, Cambridge CB3 ODS, 113 115 HarleyStreet, London, United Kingdom

Thi

s do

cum

ent w

as d

ownl

oade

d fo

r pe

rson

al u

se o

nly.

Una

utho

rized

dis

trib

utio

n is

str

ictly

pro

hibi

ted.

expenrnental and clinical endocrinology &diabetes · a. early work on the endocrinology and...

Documents