The Freezing of. Mammalian Embryos. Ciba Foundation Symposium 52 (new series) Elsevier. Excerpta Medica. North-Holland. New York. Amsterdam.

Transcription

1 The Freezing of Mammalian Embryos Ciba Foundation Symposium 52 (new series) 1977 Elsevier. Excerpta Medica. North-Holland Amsterdam. Oxford + New York

2

3 The Freezing of Mammalian Embryos

4 The Ciba Foundation for the promotion of international cooperation in medical and chemical research is a scientific and educational charity established by CIBA Limited - now CIBA-GEIGY Limited - of Basle. The Foundation operates independently in London under English trust law. Ciba Foundation Symposia are published in collaboration with Elsevier Scientific Publishing Company Excerpta Medica North-Holland Publishing Company in Amsterdam. Elsevier / Excerpta Medica / North-Holland, P.O.Box 21 1, Amsterdam

5 The Freezing of Mammalian Embryos Ciba Foundation Symposium 52 (new series) 1977 Elsevier. Excerpta Medica. North-Holland Amsterdam. Oxford + New York

6 0 Copyright 1977 Ciba Foundation All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage and retrieval system, without permission in writing from the publishers. ISBN Published in October 1977 by Elsevier/Excerpta Medica/North-Holland, P.O. 21 1, Amsterdam and Elsevier North-Holland, Inc., 52 Vanderbilt Avenue, New York, N.Y Suggested series entry for library catalogues : Ciba Foundation Symposia. Suggested publisher s entry for library catalogues: Elsevier/Excerpta MedicaiNorth-Holland Ciba Foundation Symposium 52 (new series) 340 pages, 84 figures, 36 tables Library of Congress Cataloging in Publication Data Symposium on the Freezing of Mammalian Embryos, London, The freezing of mammalian embryos. (Ciba Foundation symposium; 52 (new ser.)) Includes bibliographies and indexes. 1. Embryology-Mammals-Congresses. 2. Cryobiology-Congresses. I. Title. 11. Series : Ciba Foundation. Symposium; new ser., 52. QL959.S ISBN Printed in The Netherlands by Casparie, Alkmaar.

7 Contents D. G. WHITTINCHAM Introduction 1 C. POLGE The freezing of mammalian embryos: perspectives and possibilities 3 Discussion I3 P. MAZUR Slow-freezing injury in mammalian cells 19 Discussion 42 J. FARRANT, H. LEE and c. A. WALTER Effects of interactions between cooling and rewarming conditions on survival of cells 49 Discussion 63 s. P. LEIBO Fundamental cryobiology of mouse ova and embryos 69 Discussion 92 D. G. WHITTINGHAM animals 97 Discussion 108 Some factors affecting embryo storage in laboratory J. D. BIGGERS, R. M. BORLAND and R. D. POWERS Transport mechanisms in the preimplantation mammalian embryo 129 Discussion 146 M. A. EDIDIN and v. A. PETIT The effect of temperature on the lateral diffusion of plasma membrane proteins I55 Discussion 166 V

8 VI CONTENTS S. M. WILLADSEN Factors affecting the survival of sheep embryos during deep-freezing and thawing 175 Appendix Transplantation of sheep and cattle embryos after storage at -196 "c(s. M. WILLADSEN, C. POLGE, A. 0. TROUNSON and L. E. A. ROWSON) 190 Discussion 194 N. w. MOORE and R. J. BILTON Frozen storage of embryos of farm animals: progress and implications 203 Discussion General discussion : Freezing techniques for embryos 221 Ultra structural developmental changes 226 Freezing of advanced blastocysts of the cow 228 Freezing of pancreatic cells 23 1 R. G. EDWARDS and P. c. STEPTOE The relevance of the frozen storage of human embryos in clinicai practice 235 Discussion 243 M. J. ASHWOOD-SMITH and E. GRANT Genetic stability in cellular systems in the frozen state 251 Discussion 268 M. F. LYON, D. G. WHITTINGHAM arid P. GLENISTER Long-term storage of frozen mouse embryos under increased background irradiation 273 Discussion 283 D. W. BAILEY Genetic drift: the problem and its possible solution by frozenembryo storage 291 Discussion 299 J. KLEIN A mouse geneticist's impatient waiting for the arrival of embryofreezing techniques 305 Discussion Final general discussion 3 17 Index of contributors 323 Subject index 325

9 Participants Symposium on The Freezing of Mammalian Embryos, held at The Ciba Foundation, London, 18th-20th January 1977 Chairman : D. G. WHITTINCHAM MRC Mammalian Development Unit, Wolfson House (University College London), 4 Stephenson Way, London NWl 2HE E. ANDERSON Laboratory of Human Reproduction and Reproductive Biology, Harvard Medical School, 4.5 Shattuck Street, Boston, Massachusetts , USA M. J. ASHWOOD-SMITH Department of Biology, University of Victoria, PO Box 1700, Victoria, British Columbia, Canada V8W 2Y2 D. W. BAILEY The Jackson Laboratory, Bar Harbor, Maine 04609, USA H. BANK Department of Pathology, Medical University of South Carolina, 80 Barre Street, Charleston, South Carolina 29401, USA J. D. BIGGERS Laboratory of Human Reproduction and Reproductive Biology, Harvard Medical School, 45 Shattuck Street, Boston, Massachusetts 02115, USA W. F. BODMER Genetics Laboratory, Department of Biochemistry, University of Oxford, South Parks Road, Oxford OX1 3QU M. A. EDIDIN Mergenthaler Laboratory, The Johns Hopkins University, Baltimore, Maryland , USA R. G. EDWARDS Physiological Laboratory, University of Cambridge, Downing Street, Cambridge CB2 3EG VI1

10 VlII PARTICIPANTS J. FARRANT Division of Cryobiology, MRC Clinical Research Centre, Northwick Park Hospital, Watford Road, Harrow, Middlesex HA1 3UJ T. GREVE Institute of Animal Reproduction, The Royal Veterinary and Agricultural University, 13 Bulowsvej, DK-1870 Copenhagen V, Denmark P. HOPPE The Jackson Laboratory, Bar Harbor, Maine 04509, USA J. KLEIN Department of Microbiology, University of Texas, Southwestern Medical School, 5323 Harry Hines Boulevard, Dallas, Texas 75235, USA s. P. LEIBO Biology Division, Oak Ridge National Laboratory, PO Box Y, Oak Ridge, Tennessee 37830, USA MARY F. LYON Genetics Section, MRC Radiobiology Unit, Harwell, near Didcot, Oxfordshire OX1 1 ORD R. R. MAURER US Meat Animal Research Center, US Department of Agriculture, Agricultural Research Service, PO Box 166, Clay Center, Nebraska 68933, USA P. MAZUR Biology Division, Oak Ridge National Laboratory, PO Box Y, Oak Ridge, Tennessee 37830, USA N. w. MOORE Department of Animal Husbandry, University of Sydney, Werombi Road, Camden, NSW 2570, Australia c. POLGE ARC Unit of Reproductive Physiology and Biochemistry, Animal Research Station, 307 Huntington Road, Cambridge CB3 OJQ z. SMORAG Department of Physiological Reproduction and Artificial Insemination of Animals, Zootechnical Institute, Balice K, Krakowa, Poland A. TROUNSON Monash University, Department of Obstetrics and Gynaecology, The Queen Victoria Memorial Hospital, Melbourne, Australia 3000 s. M. WILLADSEN ARC Unit of Reproductive Physiology and Biochemistry, Animal Research Station, 307 Huntingdon Road, Cambridge CB3 OJQ

11 PARTICIPANTS IX I. WILMUT ARC Animal Breeding Research Organization, Field Laboratory, Roslin, Midlothian EH25 9PS MRC Laboratory Animals Centre, Woodmansterne Road, Carshalton, Surrey SM5 4EF MAUREEN WOOD G. H. ZEILMAKER Department of Endocrinology, Growth and Reproduction, Erasmus University, PO Box 1738, Rotterdam, The Netherlands Editors: KATHERINE ELLIOTT (Organizer) and JULIE WHELAN

12

13 Introduction D. G. WHITTLNGHAM MRC Mammalian Developmenr Unit, Univevsily College London It is particularly significant that this meeting on the freezing of mammalian embryos should be held by the Ciba Foundation, because two of its earlier symposia, one published in 1953 on germ cells, where Dr Audrey Smith reported on the attempted freezing of rabbit embryos, and one published in 1970 on the freezing of cells, record work that has now converged and has led to the development of techniques for freezing mammalian embryos. In 1972, two independent reports described the first successful storage of mouse embryos (Whittingham et NI. 1972: Wilmut 1972). I was fortunate to have the opportunity of collaborating with Dr Stanley Leibo and Dr Peter Mazur. We combined our respective research interests of early mammalian development and cryobiology and were able to define the optimal conditions for freezing and thawing mouse embryos as well as establishing their ultimate viability-namely, the birth of live offspring from embryos stored at temperatures as low as -269 "C. In this meeting we shall see how the technique has been developed to include the embryos of several other mammalian species but we shall also see that the embryos of the different species have different sensitivities to cooling depending upon the stage of development. One special feature which I hope will be brought out in this meeting is that the mammalian embryo as such is a very suitable model for studying the behaviour of cells during freezing and thawing. The embryo is a special model because, unlike most other cell types at present used for basic low temperature studies, it consists of a group of undifferentiated cells varying in size during early cleavage and having the potential to give rise to a complete new individual. This leads on to the important practical application of embryo storage in the fields of genetics and animal breeding which we shall discuss in detail later in the meeting. I

14 2 D. G. WHITTINGHAM In organizing this meeting we realized that a fundamental knowledge of the physiology and biochemistry of the mammalian embryo is an essential prerequisite to the understanding of the behaviour of embryonic cells during cooling and thawing. Similarly, for the purposes of embryo collection and embryo transfer after storage, knowledge of mammalian reproductive physiology is required too. Therefore we have attempted to bring together people from these different disciplines to enable us to discuss the problems involved in both the basic and applied aspects of embryo freezing and to ascertain what are the more important aspects for future research in this field. That is how we have tried to set the scene. Let us hope our ambition is achieved at least in part. References CIBA FOUNDATION (1953) Mammalian Germ Cells (Ciba Found. Symp.), Churchill, London CIBA FOUNDATION (1970) The Frozen Cell (Ciba Found. Synzp.), Churchill, London WHITTINGHAM, D. G., LEIBO, S. P. & MAZUR, P. (1972) Survival of mouse embryos frozen to -196" and -269 "C. Science (Wash. D.C.) 178, WILMUT, I. (1972) Effect of cooling rate, warming rate, cryoprotective agent and stage of development on survival of mouse embryos during freezing and thawing. Life Sci. 11, part 2,

15 The freezing of mammalian embryos: perspectives and possibilities C. POLGE A. R.C. Institute of Animal Physiology, Animal Research Station, Cambridge Abstract Since the Ciba Foundation Symposium in 1952 on Mammalian Germ Cells when Dr Audrey Smith reported that exposure to very low temperatures is not incompatible with further development of mammalian eggs, much progress has been made on the problem of freezing mammalian embryos. The significant steps leading to this progress are reviewed and an attempt is made to assess the extent of our current knowledge and to relate this to experience gained in other fields of low temperature biology. There is good evidence that certain basic principles concerning cooling and warming rates are applicable to the preservation of all mammalian embryos so far studied, but differences between species and between stages of development within species exist, particularly in their resistance to cooling in temperature ranges above 0 C. Some of these differences are illustrated by reference to experiments with pig embryos. Clearly there are many problems remaining to be solved, but practical applications of techniques for long-term storage of mammalian embryos are already feasible and offer interesting possibilities for future development. One of the early symposia organized by the Ciba Foundation was that on Mammalian Germ Cells in At that time my old friend and former colleague, Dr Audrey Smith, gave a paper in which she described her pioneering work on freezing rabbit eggs (Smith 1953). She had managed to get just one per cent of the fertilized eggs to continue division in culture after freezing and thawing, but she concluded: these results are su@cient to prove that exposure to very low temperatures is not incompatible with the further development of mammalian eggs. She added, there is little doubt that with appropriate modijications in technique a high survival rate will be obtainable. Evidence of the truth of these words is provided here when we are meeting once again at the Ciba Foundation in order to discuss specifically the freezing of mammalian embryos at a time when high survival rates are indeed obtainable. But 25 years have passed; so this is a Silver Jubilee year in more ways than 3

16 4 C. POLGE one and the Ciba Foundation should be congratulated on marking this occasion in a most appropriate manner. During the 25 years since Audrey Smith gave the paper to which I have referred, quite a lot has happened to extend the dimensions of this subject and to make it an exceedingly interesting one from a number of points of view. In the first place, it is probably true to say that cryobiology is now emerging as a science. Although nobody working in the field of low temperature biology would claim to have answers to all the problems, at least a large number of cells and tissues have now been frozen and thawed successfully. These studies have led to a clearer understanding of some of the principles involved in the interaction between cells and their environment when ice is formed. The recent experiments on embryos have been especially interesting to cryobiologists because they have demonstrated very dramatically the interrelationships between rates of freezing and thawing. Secondly, work on mammalian eggs and embryos has increased tremendously, particularly as regards the development of techniques for studying embryos in vitro. It is now possible, for example, to fertilize the eggs of a number of mammals in vitro and to maintain the development of the resulting embryos for considerable periods of time in tissue culture conditions. Thus knowledge of early embryonic development, although still scanty in many respects, is expanding both from the morphological and functional point of view. Finally, the techniques of transplanting eggs or embryos from donor animals to recipient foster mothers have been improved or developed not only in laboratory mammals but also in the larger farm animals, and embryo transplantation is now being applied both as an experimental tool and in a practical manner. Despite these advances, we are still at a relatively early stage in the development of satisfactory techniques for freezing and thawing embryos of many species. The number of people working in this field is still quite small; probably the majority are present at this meeting. Progress is also bound to be somewhat slow because of the difficulty and expense of obtaining embryos in large numbers from some animals. Nevertheless, it is important that we should work with the embryos of a large variety of species because experience has taught us that similar cells or tissues from different species may vary quite considerably in the conditions required to obtain their survival at very low temperatures. So, this meeting has been arranged at an opportune time. It has brought together a group of people of quite widely divergent interests-cryobiologists, embryologists, cell biologists, reproductive physiologists, geneticists and animal breeders. This emphasizes the breadth of the subject and the fact that we are concerned not only with cryobiology, but

17 THE FREEZING OF MAMMALIAN EMBRYOS 5 also with the equally important aspects of obtaining and manipulating embryos and the application of results. My role here is to look at perspectives and possibilities and I would like to do this by viewing the subject of embryo freezing from three angles. First, we should consider the type of material that we are trying to preserve at very low temperatures. Secondly, we should review how far we have progressed in experiments on freezing and thawing mammalian embryos and what are the major problems still to be overcome. Finally, the practical possibilities opened up by the ability to store eggs or embryos for prolonged periods of time should be assessed, and in this respect applications in animal breeding are of paramount importance. EGGS AND EMBRYOS Probably one of the most important features of the mammalian egg from the point of view of the cryobiologist is its size, because the oocyte at the time of ovulation is generally the largest cell found in most animals. It is a sphere varying in diameter from about pm in mice to about !bm in species such as the sheep or pig. It is surrounded by a transparent, non-cellular membrane, the zona pellucida. The principal component of the cytoplasm is yolk, but the eggs of some mammals also contain quite large amounts of lipid. Differences in the composition of the cytoplasm of oocytes of different species are quite obvious on simple inspection by phase-contrast microscopy. The eggs of mice, for example, are quite transparent and it is possible to discern nuclear elements with the cytoplasm. By contrast, pig eggs appear very dark and opaque; there is a large number of lipid droplets and no features can be distinguished clearly within the cytoplasm unless the eggs are fixed and cleared. The eggs of some carnivores such as the ferret appear to be even blacker. Shortly before ovulation the egg nucleus is activated to resume its meiotic division and in most mammals the eggs are ovulated as secondary oocytes -that is, with a first polar body extruded and the egg chromosomes arranged on a spindle at the second meiotic metaphase. In this condition they might be regarded as relatively unstable since it is fairly characteristic of unfertilized ovulated eggs that with time the spindle tends to break down and the chromosomes form micronuclei. The unfertilized egg may also be activated to undergo some early parthenogenetic development. The ovulated unfertilized oocyte is obviously a candidate for low temperature preservation. So, too, is the fertilized single-cell egg in which male and female pronuclei are developing. A possible important difference between fertilized and unfertilized oocytes is the change which occurs in the zona pellucida after sperm penetration, leading

18 6 C. POLGE to disruption of the cortical granules and the development of the block to polyspermy. Embryos at later stages of development are generally easier to obtain and to transplant and are therefore more suitable candidates for freezing. After syngamy, the egg divides within the zona pellucida by a process of cleavage resulting in a reduction of cell size. In fact the initial stages of development are characterized by negative growth. Cleavage from two to four and four to eight cells is relatively, but not entirely, synchronous between the blastomeres, but beyond the eight-cell stage the division of individual blastomeres becomes more and more asynchronous. At this stage-the morula-the cells become compacted and tight junctions between them are formed. The time at which compaction occurs varies in different species, but it always seems to occur before the blastocyst stage. The blastocoele is formed and the blastocyst expands by the active transport of fluid into this cavity. The expanded blastocyst then consists of a single peripheral layer of cells, which will form the trophoblast, and an inner cell mass, which will form the embryo. During the latter stages of blastocyst expansion there is a marked thinning of the zona pellucida. The time that blastocysts hatch from the zona pellucida also varies somewhat between species but is usually at about 5-7 days after fertilization. In some species (e.g. rabbit, mouse, man) implantation occurs shortly before or just after hatching of the blastocyst. In other species (e.g. pig, cow, sheep, horse) the time of attachment to the endometrium is much later and there is therefore a relatively long time during which the embryo is living free within the uterus. These embryos can still be collected and transplanted, but in order to establish pregnancy in a normal unmated recipient the embryos must be transferred at a time before the presence of an embryo within the uterus becomes necessary in order to maintain ovarian progesterone secretion. Sheep embryos have been transplanted successfully up to 12 days after ovulation (Moor & Rowson 1966) and cow embryos can possibly be transplanted somewhat later (Betteridge et al. 1976). In some species a considerable elongation of the blastocyst occurs at about the 13th day and this is particularly dramatic in the pig. This brief and obviously superficial look at the material with which we are concerned illustrates its great variety. When we are discussing embryo freezing we can be considering anything from a huge single cell to a complex developing organism containing many thousands of differentiated cells. It is a fascinating problem and we shall obviously be concerned with how various changes in the developing embryo affect its reactions to freezing and thawing and the differences which might exist between species.

19 THE FREEZING OF MAMMALIAN EMBRYOS 7 FREEZING AND THAWlNG Early work during the 1950s provided some valuable clues to the preservation of eggs at very low temperatures, but just missed some vital steps so that very little was achieved in the way of actual practical success. Audrey Smith's (1952, 1953) experiments were with unicellular fertilized rabbit eggs and with glycerol as a cryoprotective agent. She noted that the eggs shrank irreversibly when exposed to 15% glycerol at room temperature or 5 "C, but if the concentration of glycerol was increased gradually by stepwise addition at 37 "C, the eggs shrank initially and then re-expanded. Permeation with 15 % glycerol was not harmful because, after stepwise removal of the glycerol at 37 "C, the eggs developed normally in culture. A high proportion (69 %) of eggs survived storage for three days at -15 "C provided they remained in the supercooled state. However, longer storage usually resulted in crystallization of the medium and a drastic reduction in survival rate. Similarly, after slow cooling of eggs in 15 % glycerol to temperatures of -79 "C or below only six out of 600 showed any development after rapid thawing, removal of glycerol and culture. In hindsight it seems reasonable to guess that the low survival was probably due to using a rate of cooling that was too fast. In those days 'slow cooling' usually meant cooling at about 1 "C per minute to a temperature of around -15 "C and then somewhat faster to -79 "C. At this rate it seems probable that a considerable amount of intracellular ice would have been formed and it is doubtful whether slow re-warming would have been of any additional benefit. The interesting experiments of Sherman and his colleagues do not always get the credit that they deserve (Lin et al. 1957; Sherman & Lin 1958a, 6, 1959). They worked with unfertilized ovulated mouse oocytes and evidence of survival after treatment with glycerol or after cooling and re-warming was provided by fertilization and embryonic development after transfer to the oviducts of mated, genetically distinct, recipients. Treatment with glycerol at 5 "C caused shrinkage, as was the case with rabbit eggs, but they re-expanded at 37 "C, suggesting that permeation with glycerol occurred at this temperature. The presence of glycerol within the eggs did not interfere with subsequent fertilization and embryonic development. Neither was development reduced after transfer after abrupt cooling to 0 "C. Some eggs in 5 % glycerol survived in the presence of extracellular ice after 30 minutes exposure to -10 "C, but none survived after six hours. Although Sherman was not able to demonstrate functional survival of oocytes after freezing to very low temperatures, he did make some very interesting observations relating to their morphology. He was one of the first to question seriously the idea that cells were protected by glycerol

20 8 C. POLGE during freezing and thawing only after full permeation. He concluded that morphology was better preserved if glycerol did not permeate the cells or if cooling was very slow ( "C/min). He therefore proposed that the site of protective action of glycerol was extracellular and it was achieved in part by cellular dehydration which occurs in the presence of glycerol at high concentrations or at low concentrations during slow cooling (Sherman 1963). So it can be seen that success was only a hair's breadth away. It wasn't achieved because thawing was always done at a fast rate. The most decisive demonstration in the 1950s of the survival of oocytes after freezing and thawing was provided by experiments on freezing ovarian tissue (Parkes & Smith 1953). During the attempts to freeze recently ovulated eggs it was noted that although the majority of the eggs failed to survive, some of the surrounding cumulus cells grew in tissue culture after slow freezing to -79 or -196 "C. Rat ovaries were therefore sliced or minced and soaked in 15% glycerol, and viability after freezing and thawing was tested by subcutaneous grafting. Endocrinological activity of the grafts was demonstrated, but histological examination showed that all the oocytes within Graafian follicles appeared to be degenerating. Within a few days, however, numerous primordial follicles could be seen. Best results were obtained when the ovarian tissue was frozen in glycerol-serum rather than glycerol-saline (Parkes 1958). It was estimated that between 1 and 5% of oocytes in primordial follicles had survived freezing and thawing in slices of infantile ovaries examined 24 hours after grafting (Deanesly 1957). A clear illustration of the viability of surviving oocytes was later provided in experiments in which frozen and thawed mouse ovarian tissue was successfully grafted into the ovarian capsule of mice previously sterilized by irradiation. The surviving eggs were capable of maturation and fertilization and gave rise to normal young (Parrott 1960). The fact that it was only the very small oocytes which survived and not those which had grown within Graafian follicles suggests that it was the size of the oocytes which governed their survival during the freezing and thawing regimes used at that time. A temporary lull ensued until interest was again revived by the work of our chairman published in the early 1970s (Whittingham 1971). In this he claimed survival of early mouse blastocysts treated with 7.5 % polyvinyl pyrrolidone and cooled rapidly to -79 "C. Unfortunately, these experiments could not be repeated by others, and I am sure David Whittingham will not mind my saying that the approach he adopted was probably leading away from the advances that had been made by others in previous years. It seems to me that what he must have achieved is a remarkable degree of supercooling, for he noted that no embryos survived storage at -79 "C for more than

21 THE FREEZING OF MAMMALIAN EMBRYOS 9 30 minutes. What his paper did do, however, was to rekindle a vital interest in this subject and it is this that has led to the successes achieved in more recent years. In fact the breakthrough came a year later. Whittingham then joined Mazur and Leibo at Oak Ridge. Whittingham's experience and skill with embryos, coupled with the cryobiological wisdom of Mazur and Leibo, constituted a formidable team with which to attack the problem of embryo freezing. They applied to this problem some basic cryobiological principles and success was at hand. Perhaps the most important concept in cryobiology is that for different cells there is an optimal cooling rate which varies according to the type of cell. Experiments of numerous workers have shown this to be true, and it is well described by Mazur (1970). A major cause of freezing injury is intracellular ice formation when cells are cooled too rapidly. Reducing the cooling rate permits extracellular ice formation which leads to gradual dehydration of the cell. Damage to cells at slower rates of cooling may be caused by solution effects which in turn can be reduced by protective media such as glycerol. Mazur and Leibo calculated that, in order to avoid intracellular ice formation, cells the size of mouse eggs should not be cooled faster than about 1 "C per minute. The validity of this concept was borne out by the experiments with mouse embryos in which effects of suspending media, cooling rate and warming rate were studied (Whittingham et al. 1972). The results were dramatic. Glycerol or dimethyl sulphoxide (DMSO) were used as cryoprotective agents. Very few embryos survived after cooling at rates faster than 2 "C per minute, but a high proportion (around 80% in some experiments) survived after slower cooling, the optimal rates being 0.3 to 0.4 "C per minute. Even so, the effect of cooling rate was probably not the most important observation that they made. Rapid rewarming, even with the slowly cooled embryos, was very damaging and the high survival rates were achieved only after slow warming at 4 "C to 25 "C per minute. While this advance was made in the USA, Wilmut in our laboratories in Cambridge had come independently and at the same time to almost analogous conclusions (Wilmut 1972). The chief interest of these observations in cryobiological terms is that the optimal cooling rate established for the mouse embryos is probably the lowest for any animal cell yet examined and the vital importance of slow warming has brought to light a new aspect of cell survival during freezing and thawing. Earlier observations had shown that cells cooled rapidly were generally more sensitive to slow warming than to rapid warming, due possibly to the growth of ice nuclei within the cells. But the damaging effect of rapid warming in embryos that have been cooled very slowly (and which therefore are probably extremely dehydrated) is more difficult to explain.

22 10 C. POLGE Possibly it is related to a necessity for slow rehydration-later papers in this symposium will, I hope, throw more light on this topic. From the practical point of view these experiments have heralded new opportunities for the preservation of embryos of a variety of species. Whittingham et al. (1972) worked with mouse embryos from the single cell to the blastocyst stage of development and viability after freezing and thawing was tested not only by culture techniques in vitro, but also by transfer to foster mothers. Sixty-five per cent of the recipients became pregnant and more than 40% of the embryos in the pregnant mice gave rise to normal living young. Survival of embryos frozen at the one- or two-cell stage was somewhat better than that of embryos frozen as blastocysts, but Wilmut (1972) obtained about 80 % survival of embryos frozen as blastocysts, although viability was only tested in vitro. During recent years similar techniques have been applied to the embryos of a number of other species. In laboratory animals, both rabbit (Bank & Maurer 1974; Whittingham & Adams 1976) and rat (Whittingham 1975) embryos have been successfully frozen and thawed, but the proportions surviving have not been so high as with mouse embryos. Damage caused to the zona pellucida of rabbit embryos has perhaps been responsible for reduced survival of embryos after transplantation. In our laboratory we have concentrated on experiments with the embryos of farm animals, notably sheep, cow and pig. High survival rates have now been obtained with sheep and cow embryos and the details of this work will be presented in this symposium by Willadsen (see pp ). Experiments with cow and pig embryos deserve some special comment because they illustrate some important observations on the sensitivity of embryos to cooling in temperature ranges above 0 "C. The initial experiments of Wilmut on cow embryos were with hatched blastocysts obtained from donors on Day 12 of the oestrous cycle. A proportion survived slow freezing in medium containing I.SM-DMSO followed by rapid thawing and two calves were born as a result of transplantation (Wilmut & Rowson 1973). Embryos at somewhat earlier stages of development are probably more suitable for transplantation, but when attempts were made to freeze eight-cell embryos, survival rate was negligible. It was then noted that the eight-cell embryos did not recover even from the initial stages of cooling to 0 "C (Wilmut et al. 1975). The sensitivity to cooling diminished at later stages of embryonic development and in another series of experiments it was found that expanded blastocysts were quite resistant (Trounson et al. 1976). Thus it is the expanded blastocyst that has been used almost exclusively in further experiments on freezing and thawing.

23 THE FREEZING OF MAMMALIAN EMBRYOS 11 The sensitivity of pig embryos to cooling appears to be even more marked (Polge et al. 1974). In experiments with embryos collected at any stage of development from four cells up to the hatched blastocyst, none has survived cooling to 0 "C. The critical temperature at which damage occurs is around 15 "C. Above this temperature virtually all the embryos survive, but when cooled below 15 "C the best that has been achieved so far is that a few cells in expanding blastocysts have shown evidence of growth in tissue culture. No viable embryos have developed after transplantation to recipients. Attempts to increase the resistance of pig embryos to cooling by addition to the media of phospholipids which are known to protect boar spermatozoa (Butler & Roberts 1975) have also been unsuccessful. It is tempting to speculate that the damage caused to pig embryos during cooling might be associated with lipid phase changes within the membranes, but as yet there is no direct evidence to support this idea. These results illustrate the diversity of the problems associated with the preservation of embryos at very low temperatures and the need for a greater understanding of the nature of some of the differences which exist between embryos of different species and of changes which occur during early development. It is clear that we are concerned with problems additional to those which are caused by the formation of ice. Nevertheless, in species in which the embryos at particular stages of development are not damaged by cooling per se, considerable progress is now being made towards their successful preservation in the frozen state. This has been achieved by using DMSO as a cryoprotectant coupled with techniques of very slow cooling and rewarming. Further progress will undoubtedly be made through experiments designed to study specifically the interactions between cooling and warming rates within different temperature ranges. A systematic approach to these problems may not be too difficult with embryos of some of the laboratory species in which sufficient numbers can be obtained at relatively low cost. By contrast, the sheer problem of obtaining large numbers of embryos from some of the domestic species will always be a limiting factor. PRACTICAL APPLICATIONS Practical applications arising from the ability to store embryos in the frozen state can already be foreseen. Probably one of the most important will be in relation to the breeding of farm animals. The technique of embryo transplantation is already being applied to a limited extent in cattle breeding and it has been used specifically for the introduction of new breeds to different countries and for the more rapid multiplication of genetically superior stock.

24 12 C. POLGE The distance over which embryos can be transported from donors in one area to recipients in another is obviously limited by the time for which embryos can be stored in vitro. Low temperature preservation will abolish such limitations. Thus, one of the first applications will be in relation to the export and import of embryos. Storage at very low temperatures has the advantage that the embryos can be kept in quarantine for sufficient time to establish that the donor animals were free of certain diseases at the time the embryos were collected. The establishment of embryo banks may also be useful in relation to livestock improvement schemes in which it is desired to measure more accurately than has been possible hitherto the actual rate of genetic gain that is being made over a period of years. Applications such as these, however, are dependent on obtaining high survival rates of embryos after freezing and thawing. Techniques of superovulation and transplantation at the present moment are quite expensive, and it would not be very practicable to transplant embryos with a low chance of survival. In many cases only one embryo is transplanted to each recipient. We would probably require, therefore, at least a 50% chance of survival. It is encouraging that the recent results with sheep and cow embryos suggest that survival rates of this order or greater can now be achieved. Practical advantages of storing embryos of laboratory species and also possible applications in human medicine will be described later in this symposium. Many Ciba Foundation symposia have been notable for the fact that they have focused attention on areas of science in which rapid progress has since been made. I am sure that the topic of embryo freezing will be such a one and that our discussions here should stimulate further developments in the very near future. References BANK, H. & MAURER, R. R. (1974) Survival of frozen rabbit embryos. Exp. Cell Res. 89, BETTERIDGE, K. J., MITCHELL, D., EAGLESOME, M. D. & RANDALL, G. C. B. (1976) Embryo transfer in cattle days after estrus. Proc. 8th Internat. Congr. Anim. Reprod. & A.I. (Krakow) 3, BUTLER, W. J. & ROBERTS, T. K. (1975) Effects of some phosphatidyl compounds on boar spermatozoa following cold shock or slow cooling. J. Reprod. Fertil. 43, DEANESLY, R. (1957) Egg survival in immature rat ovaries grafted after freezing and thawing. Proc. R. Soc. Lond. B Biol. Sci. 147, LIN, T. P., SHERMAN, J. K. & WILLErT, E. L. (1957) Survival of unfertilized mouse eggs in media containing glycerol and glycine. J. Exp. Zool. 134,

25 THE FREEZING OF MAMMALIAN EMBRYOS 13 MAZUR, P. (1970) Cryobiology: freezing of biological systems. Science (Wash. D.C.) 168, MOOR, R. M. & ROWSON, L. E. A. (1966) The corpus luteum of the sheep: functional relationship between the embryo and the corpus luteum. J. Endocrinol. 34, PARKES, A. S. (1958) Factors affecting the viability of frozen ovarian tissue. J. Endocrinol. 17, PARKES, A. S. & SMITH, A. U. (1953) Regeneration of rat ovarian tissue grafted after exposure to low temperatures. Proc. R. SOC. Lond. B Bid. Sci. 140, PARROTT, D. M. V. (1960) The fertility of mice with orthotopic ovarian grafts derived from frozen tissue. J. Reprod. Fertil. I, POLGE, C., WILMUT, I. & ROWSON, L. E. A. (1974) The low temperature preservation of cow, sheep and pig embryos. Cryohiology 11, 560 SHERMAN, J. K. (1963) Questionable protection by intracellular glycerol during freezing and thawing. J. Cell Comp. Physiol. 61, SHERMAN, J. K. & LIN, T. P. (1958~) Effect of glycerol and low temperature on survival of unfertilized mouse eggs. Nature (Lond.) 181, SHERMAN, J. K. & LIN, T. P. (1958b) Survival of unfertilized mouse eggs during freezing and thawing. Proc. Soc. Exp. Bid. Med. 98, SHERMAN, J. K. & LIN, T. P. (1959) Temperature shock and cold-storage of unfertilized mouse eggs. Fertil Steril. 10, SMITH, A. U. (1952) Behaviour of fertilized rabbit eggs exposed to glycerol and to low temperatures. Nature (Lond.) 170, SMITH, A. U. (1953) In vitro experiments with rabbit eggs, in Mammalian Germ Cells (Ciha Found. Symp.), pp , Churchill, London TROUNSON, A. O., WILLADSEN, S. M., ROWSON, L. E. A. & NEWCOMB, R. (1976) The storage of cow eggs at room temperature and at low temperatures. J. Reprod. Fertil. 46, WHITTINGHAM, D. G. (1971) Survival of mouse embryos after freezing and thawing. Nature (Lond.) 233, WHITTINGHAM, D. G. (1975) Survival of rat embryos after freezing and thawing. J. Reprod. Fertil. 43, WHITTINGHAM, D. G. & ADAMS, C. E. (1976) Low temperature preservation of rabbit embryos. J. Reprod. Fertil. 47, WHITTINGHAM, D. G., LEIBO, S. P. & MAZUR, P. (1972) Survival of mouse embryos frozen to -I96 "C and -269 "C. Science (Wash. D.C.) 178, WILMUT, I. (1972) Effect of cooling rate, warming rate, cryoprotective agent and stage of development on survival of mouse embryos during cooling and thawing. Life Sci. 11, part 2, WILMUT, I. & ROWSON, L. E. A. (1973) Experiments on the low temperature preservation of cow embryos. Vet. Rec. 92, WILMUT, I., POLGE, C. & ROWSON, L. E. A. (1975) The effect on cow embryos of cooling to 20, 0 and -196 "C. J. Reprod. Fertil. 45, Discussion Whittingham: I think it is important to stress that unlike the freezing of sperm and other tissue cells, the actual freezing of the embryo is only part of the whole procedure. For sperm freezing, the collection of sperm and insemination after thawing are relatively simple non-surgical procedures but in the majority of cases the collection and transfer of embryos involves surgical interference.

26 14 DISCUSSION In addition, we have to determine ways of obtaining maximum numbers of embryos from each female, the most appropriate embryonic stage for freezing, suitable culture media for collection, manipulation and culture in vitro, and finally the correct synchrony between embryo and foster mother at the time of transfer. Biggers: You perhaps didn t indicate sufficiently how variable different species are in these early stages, Dr Polge. You mentioned egg volume but you specified diameters only. The differences are more striking when one compares volumes. The smallest mammalian egg is that of Microtus (the vole), which is about 65 pm in diameter, and one of the largest is the human egg, about 130 pm. That gives a diameter ratio of two but a volume ratio of eight. Also, in several mammalian species the volume is reduced during early cleavage. In the mouse, by the eight-cell stage the total volume is reduced by 25%. The protein content also falls by a similar amount. Up to compaction each blastomere is essentially operating independently but then tight junctions develop, resulting in a much closer relationship between the cells. These changes affect the total permeability properties of the embryo. As you said, there are also big differences in the inclusion bodies of mammalian eggs. The mouse and rabbit egg, which are very different in size, are both translucent, with inconspicuous inclusions. The pig egg and many carnivore eggs have many fatty inclusions which make their cytoplasm very opaque. On the question of the synchrony of early cleavage, it is perfectly normal to see a three-cell or a five-cell stage at the time of early cleavage. There is about half an hour to an hour s lag between some of the divisions. The rabbit rat and mouse all show this near synchrony. Polge: Slightly later, however, asynchrony becomes more marked, and in a late blastocyst with several hundred cells one may find only about 5% of the cells in metaphase at any one time. Whittingham: In mice, there is a variation in the synchrony of division both within and between embryos from a particular female as preimplantation development proceeds. Willadsen: In the species we are working with-the sheep, the cow and the pig-an apparent synchrony completely overshadows any minor asynchrony. Take freshly ovulated oocytes, which are obvious candidates for freezing: one may safely assume that these will all be at the metaphase of the second meiotic division. During cleavage synchronous development is also pronounced. Only from the morula stage (20-30 cells), which is reached 4-5 days after the onset of oestrus, and onwards does there appear to be quite a bit of variation between embryos with respect to cell number, the time of blastulation and the time at which the zona pellucida is shed.

27 THE FREEZING OF MAMMALIAN EMBRYOS 15 Biggers: You mentioned hatching, Dr Polge. Do you attach any particular significance to hatching, in relation to the freezing of embryos? Some people question whether hatching is a normal event in development, but believe that the zona is normally destroyed by a lytic agent. Polge: In culture one can often see blastocysts that appear to be hatching from the zona. There is probably a lytic agent involved, but it does not seem to affect the whole zona because the blastocyst hatches from a crack which develops in part of it. It would be interesting to know the significance of the zona in relation to freezing and thawing and whether there are marked differences in survival between hatched and unhatched embryos. Should we, for instance, expect the embryos within the zona to react as if they were one large cell and the hatched embryos as a tissue? It has been suggested that with embryos which are, say, at the eight-cell stage, either the majority of the cells survive freezing and thawing or they are all damaged. Is this true? Whittingham: On the whole, yes; you get a low percentage with degeneration or lysis of some blastomeres at the eight-cell stage but this percentage is increased when something has gone wrong in the freezing procedure, such as supercooling of samples before ice induction, or suboptimal rates of cooling and warming, or incorrect diluting of the cryoprotective agent on thawing. Polge: What do we know about the permeability of the zona to glycerol or dimethyl sulphoxide (DMSO)? Mazur: It is high. Polge: When you add glycerol or DMSO at low temperature and the embryos shrink, does this mean that the cryoprotectant is not getting through the zona and water is being sucked out through it, or is the cryoprotectant getting through the zona but not entering the cells and is shrinking them within the zona? Leibo: I would say the latter. The evidence for this is the following. Recently, Jackowski (1977) has measured the permeability of fertilized and unfertilized mouse ova to glycerol. She found that removal of the zona, by gentle enzymic digestion, had no effect on the rate of permeation of glycerol into ova, whether fertilized or not. Trounson: When you compare the hatched and unhatched embryo, the results will be confounded by changes in individual cell size, membrane changes and structural changes. Therefore, methods of freezing advanced embryos are likely to be different from those suitable for early stage embryos. For example, when cow embryos aged 12 days (or more) are frozen using the established procedures for freezing Day 7 early blastocysts, none survive. We found it necessary to thaw rapidly and to raise the DMSO concentration to M. Survival rate was also improved by increasing the cooling rate to

28 16 DISCUSSION 'CC/min (see also p. 228). This difference is probably not due to the presence or absence of a zona pellucida or to the presence of a large blastocoelic cavity in advanced embryos, because embryos induced to deflate before freezing appear to be similar to those that are fully inflated. Morphological changes in embryo development are probably more significant in determining the response to different freezing procedures. To establish the role, if any, played by the zona pellucida one has to compare the results of freezing embryos with and without zonae at the same stage of development. Whittinghum: There is one practical difficulty here: if you remove the zona from, say, the eight-cell mouse egg and freeze the embryo you will disperse the blastomeres. It would be very difficult to say what part the zona is playing during freezing and thawing, other than the normal function attributed to it during the early cleavage stages, namely keeping the blastomeres of each individual embryo together (Whittingham 1968). Trounson: You could make a small hole in the zona. Whittingham: There is probably a hole left in the zona pellucida after sperm entry, as suggested by the work of Professor Tarkowski's group in Poland (unpublished). They found that when one-cell eggs are exposed to hypotonic solutions, the contents of the fertilized eggs start to protrude through the zona pellucida at one particular point, but this does not occur with similarly treated unfertilized eggs. This suggests that there is some continuity between the perivitelline space and the external environment outside the zona pellucida after sperm penetration. Willadsen: In the cow and the sheep, we don't know whether the collected embryos are in fact hatching normally: by flushing the reproductive tract one probably gets more to hatch artificially than would normally hatch at that point. But cow and sheep embryos which have compacted may be transferred successfully even if the zona is absent. So with these stages of development one can disregard freezing damage to the zona, which again makes freezing much easier. Secondly, on Alan Trounson's comments: there is now ample evidence that there are many ways in which to freeze embryos, and when you say that you have to increase the DMSO concentration or the freezing rate, you are speaking in the context of your particular approach. Muzur: The zona is so permeable both to water and, as Dr Leibo has pointed out, to the cryoprotective additives that one would not expect it to affect the freezing response of embryos, which is what Dr Willadsen finds. Bunk: The question of whether the zona imposes a permeability barrier to the penetration of the so-called cryoprotective additives presupposes that the mechanism of cryoprotection is intracellular. If the embryos are protected

29 THE FREEZING OF MAMMALIAN EMBRYOS 17 primarily by the presence of extracellular agents, such as occurs after brief exposure to glycerol at low temperature, the basic consideration is the permeability of the plasma membrane and zona to water. Survival after brief exposure to glycerol may be due either to dehydration or to direct extracellular protection of the plasma membrane. In rabbit embryos frozen in the presence of dimethyl sulphoxtde, Dr Maurer and I (Bank & Maurer 1974) found that gradual dilution of the freezing media containing 'normal' embryos, after thawing, is critical for survival. However, embryos in which the zona was damaged during freezing could withstand relatively abrupt dilution indicating, at least in this species, that the zona may act as a permeability barrier. Whittingham: In the rabbit embryo you also have the much coat in addition to the zona pellucida, which makes it different from other mammalian embryos. Edwards: Dr Polge said that in the cow the period before the morula was transformed into the blastocyst was smsitive to cooling and the embryo would die if that stage were frozen. Several factors could be involved here, and I wonder if there are any leads to indicate which of them could be important. Tight junctions are forming between adjacent trophoblast cells and I wonder if this affxts the results of cooling at that time. Of course, these junctions are also present in the blastocyst later on. A second factor could be the secretion of blastocoelic fluid, and I wonder if an enlarging cell loaded with blastocoelic fluid secretions becomes more sensitive. This leads me to speculate whether there are species differences in the formation of the blastocoele : in other words, IS there a different form of expansion in those species where embryos can be frozen successfully as compared with those which are more difficult to freeze? Trounson: I think that species differences are more marked in the developmental stages before blastulation. In the cow, eight-cell embryos are more sensitive to damage by cooling than morulae and these in turn are more sensitive than blastocysts. The early cow blastocyst is the preferred stage of development for freezing. Polge: I agree. But if one thinks of a possible effect of tight junctions on sensitivity to cooling one must distinguish between species. In the mouse, for instance, there does not appear to be any stage which is particularly sensitive to cooling, either before or after the formation of tight junctions. Zeilmaker: I wonder whether the bid survival of pig embryos after cooling is related to the large lipid granules, because usually membranes are thought to be the structures most vulnerable to freezing and these lipid globules must be surrounded by a number of membranes. Are embryos of other species which contain lipid granules, such as the ferret, also difficult to freeze? Polge: In a pig blastocyst that has been cooled to + 15 "C and has survived,

30 18 DISCUSSION we see distinct lipid droplets throughout the cytoplasm. By electron microscopy these droplets are seen to be surrounded by endoplasmic reticulum. By contrast, in a blastocyst that has not survived cooling, the lipid droplets become aggregated into pools. Whether this is the lesion, or whether it is the membranes which are damaged, which then allows the droplets to coalesce, we don t know. Anderson: The lipid droplets within the cell are not membrane-bounded. They are incompletely encompassed by a thin cisterna of endoplasmic reticulum. Whittinghum: Is that rough or smooth endoplasmic reticulum? Anderson: Rough and smooth. Muzur: Is there any evidence that a high degree of lipid granulation is correlated with a high lysosomal content? There is an interesting parallel that highly granular cells are the most difficult to freeze. Human granulocytes, for example, cannot be frozen successfully although human lymphocytes, which do not contain granules, can be frozen easily. At least some of the granules of the granulocyte are lysosomal in nature. We were struck too by the large number of granules in the cow embryo, which I gather are lipids. Is there anything known about the lysosomes of this embryo? Anderson: The rabbit and mouse eggs have a large population of lysosomes, primary and heterolysosomes. it is interesting that you bring up this point in relation to lysosomes in the granulocyte. Heterogeneity of lysosomes may be a major factor in freezing various cell types. Biggers: Species differences are extremely important. One other difference that should be mentioned is in the way the blastocyst forms. i like to consider that there are two main types of mammalian blastocyst, the minimally expanding and the maximally expanding types. The minimally expanding type is found in the mouse, which forms a blastocoele cavity but the increase in its size is less than twofold. Most primate blastocysts seem to be like this, including the human. in the rabbit, which typifies the maximally expanding type of blastocyst, the volume increases by many orders of magnitude. The pig is another extreme example of this type of blastocyst. References BANK, H. & MAURER, R. R. (1974) Survival of frozen rabbit embryos. Exp. Cell Res. 89, JACKOWSKI, S. C. (1977) Physiological differences between fertilized and unfertilized mouse ova; glycerol permeability and freezing sensitivity. Ph.D. Dissertation, The University of Tennessee WHITTINGHAM, D. G. (1968) Development of zygotes in cultured mouse oviducts. 1. The effect of varying oviductal conditions. J. Exp ,