Enigmas and Variations Among Mammalian Embryos

This brief essay, based on my closing presentation at the workshop on ‘Oocyte maturation and Embryo Quality’ held in Ghent, on 7 January 2000, expresses my personal views on the value of comparative studies in mammalian embryology and is by no means a review of the subject. During the workshop much progress was reported towards understanding the physiology of follicle growth and oocyte maturation, as well as towards the improvement of embryo culture and thence, presumably, ‘embryo quality’. An excellent feature of the programme was its comparative nature, with much emphasis on the value of parallel studies of various mammalian species, particularly cattle and humans. An emphasis on comparing species in order to understand reproductive processes is hardly new in the Benelux countries. It was here, more that 300 years ago, that Regnier de Graaf recognized and illustrated rabbit embryos in the reproductive tract within 72 h of mating, and was one of the first to show that the ovary contributes to the formation of embryos (Jocelyn and Setchell 1972; Betteridge 1981). His name, of course, lives on each and every time a Graafian follicle is discussed. Perhaps, though, he deserves at least as much credit as a doyen of the comparative approach. Although de Graaf was correct in his belief that the ovary produces eggs from its follicles, he also concluded that follicular fluid was part of the egg. That conclusion he based on two comparative observations: first, that sow’s ovaries look like those of hens; second, that the follicular contents of boiled human ovaries resembled the white of a hen’s egg in appearance, consistency and taste! Evidently de Graaf’s dedication to his science knew no bounds! The title of this essay is intended to be a reminder that (a) there are still very many puzzles to unravel in studies of mammalian embryos, and (b) understanding the similarities and differences among those of various species should be very instructive for those researchers that are trying to make practical use of embryos. An elementary, but often forgotten, point about the study of reproduction is made in Fig. 1. To confine studies to humans, to humans and mice, or even to humans, mice and the domestic animals, is really to touch just the tip of the iceberg as far as the wonderful range of mammalian reproductive strategies is concerned. The ‘iceberg’ of reproductive strategies, illustrating the danger of a restricted training narrowing one’s outlook on reproductive biology With that in mind, I shall try to illustrate my belief that: • there are extremely important warning signs to be drawn from comparative studies with regard to human clinical use of animal reproduction technologies (ARTs); • there is enormous potential for comparative studies to suggest new avenues of embryo manipulation for animal production; • there is still much room for straightforward descriptive studies of mammalian development to complement the explosion in knowledge based on genomic information. The epidemiology of embryonic loss in farm animals leads one to realize the general principle that a long interval can separate the time at which an embryo experiences ‘abuse’, and the time at which the adverse effects of that abuse become manifest (Betteridge 1983). Nevertheless, the demonstration that that interval can be extended, experimentally, into the postnatal period (Iannaccone 1984) came as an unpleasant surprise to many of us involved in the collection, manipulation and transfer of embryos. There are three important points to make from Iannaccone’s study. The first, and most salient of course, is that the potent, direct-acting mutagen methylnitrosourea (MNU), can alter the developmental program of mouse blastocysts and can manifest effects long after exposure, indeed after birth (Fig. 2). The second point is that, as recently as 1984, the Introduction to Iannaccone’s paper could state that: ‘The preimplantation embryo has been traditionally considered refractory to the effects of exposure to teratogens’. The third point of relevance in the context of this workshop is that this very important study depended on the use of embryo transfer, and nicely illustrates the value of that procedure as a research tool. The postnatal survival of mice born from blastocysts that had been exposed to methylnitrosourea (MNU) or control solvent (CON) before transfer to recipient mice. Note that, during the first year after birth, offspring developed from blastocysts exposed to MNU in vitro were found to have a three-fold greater crude mortality rate than offspring developed from blastocysts exposed to solvent. [Redrawn from Iannaccone 1984; Cancer Res 44, 2785–2789, with permission] As MNU is an established mutagen, and those of us working with embryos are not in the habit of deliberately using such agents, how relevant to the ARTs are these delayed manifestations of insults to the embryo? The answer, unfortunately, is ‘very relevant’, as has been clearly demonstrated by more recent embryo manipulation and transfer work, particularly in farm animals. Commenting on his pioneering work on cloning sheep and cattle by nuclear transfer, Willadsen (1989) noted that: ‘… a large number of nuclear transfer calves has been produced. The vast majority have appeared to be normal in every respect. Among the earliest, were some that were considerably larger than average at birth’. At first, this phenomenon (later to become known as the ‘large offspring syndrome’) was considered to be an effect of nuclear transfer itself. Soon, however, it became clear that the problem could arise from mere culture of embryos. The sporadic nature of the problem has made its experimental investigation difficult and so survey studies and data from large-scale field studies have been exceptionally useful in the now extensive attempts to explain its causes. Recent reviews of the topic include those of Walker et al. (1996), Rieger (1998), Leese et al. (1998), Robinson et al. (1999) and the several papers presented at a symposium in Maastricht in January, 2000 [Theriogenology, 53 (2)]. These form the basis of the following paragraphs. The consensus is that the large offspring syndrome is the result of pre-implantation effects on the oocyte and/or the developing embryo. It also seems clear that useful parallels can be drawn between the effects of maternal nutrition in vivo and culture in vitro. These generalities, however, mask much that remains unknown, controversial, and intriguing about how fetal oversize is brought about. From general principles, it seems likely that excessive growth may be just one extreme of a spectrum of effects on embryonic and fetal development. Consequently, a consideration of ways in which the pre-implantation environment can affect subsequent events in general may cast light on the large offspring syndrome in particular. Even before, or shortly after, ovulation, the oocyte’s potential to produce a blastocyst can be affected by the dam’s nutritional status. This effect is thought to be mediated by changes in the concentrations of steroids and/or insulin-like growth factors (IGF’s) that are available to the oocyte within the follicle. In ewes, a diet that includes an excess of rumen-degradable protein, which results in elevated concentrations of ammonia in the maternal circulation, leads to increased embryo metabolism, extended gestations, and increased average birth-weight of lambs. In vitro, it is thought that inclusion of serum in the culture medium during oocyte maturation (IVM) and fertilization (IVF) may affect either process via the production of ammonia and result in oversized lambs or calves. Although embryos produced in vitro and resulting in large offspring could have been affected during IVM, IVF or subsequent culture (IVC), IVC is likely to be the culprit because embryos produced in vivo and then subjected to culture are also prone to the syndrome. Mysteriously, not all fetuses derived from embryos cultured together under exactly the same conditions are affected. In addition to oversize as a whole, there can be significant relative increases in the size of key organs (most notably the heart) of affected fetuses – effects which can persist postnatally. Culture in vitro is known to be associated with an increase in intracytoplasmic lipids in developing embryos, relative to their in vivo counterparts, but the significance of this in relation to oversize remains unknown. The exposure of early embryos to a more rapid than usual rise in progesterone in their environment, induced either by nutrition, temporary transfer to a more advanced recipient, or by injection of exogenous progesterone, certainly accelerates early growth. Again, though, the relevance of this to IVC-induced oversize is conjectural. At the moment, the most promising avenue of investigation directed at understanding the large offspring syndrome seems to be molecular: understanding how the expression of imprinted genes is controlled, particularly those for proteins affecting fetal growth and development (IGF-2, the corresponding receptor IGF-2r, insulin-2 and H19; Young and Fairburn, 2000; Blondin et al. 2000). In human medicine, there is no direct equivalent of the large offspring syndrome following embryo production in vitro. However, it has been pointed out that this may reflect the more intensive obstetric management accorded individual mothers after human IVF (Barnes 2000). Furthermore, there is evidence that the incidence of babies that are small, or very small, for gestational age is higher than normal following IVF. This, and the evident relationship between intra-uterine growth during normal pregnancy and patterns of disease in adulthood in humans (Barker 2000) make it imperative that the progeny human ART procedures be, at least, most carefully monitored over the long-term (Van der Lende et al. 2000). Personally, like many others (cf Cummins and Jequier 1995; Rieger 1998; Leese et al. 1998; Ménézo et al. 2000), I believe that the increasingly complicated procedures in use in today’s human ART clinics call for thorough comparative investigation in other mammals. Elsewhere (Betteridge 1995) it has been argued that those researchers that are involved in embryo transfer and related techniques have much to learn from a wider range of species than are of direct concern. One of the issues raised was that of polarity of the oocyte which, in mammals, had not been considered to be of much importance to development. Perhaps this disregard of polarity was because, apart from the positioning of the germinal vesicle and meiotic metaphase spindles, it is so inconspicuous in the oocytes of mice and humans compared with, for example, the markedly polar horse oocyte. Mounting evidence now suggests that early polarity is involved in determination of the subsequent anterior–posterior axis of the fetus (Gardner 1999). Furthermore, from a practical standpoint, it has been suggested that ‘a greater understanding of cytoplasmic compartmentalization will …. lead to refinement in manipulative procedures and improvements in the development of manipulated eggs’ (Fulka et al. 1998). Stepping beyond the domestic animals, yet barely penetrating the base of the iceberg depicted in Fig. 1, there are tantalizing features of reproductive strategies that surely have lessons to teach us if we would only take the time to learn them. One such is the differentiation of the embryo proper from cells of unilaminar blastocysts (i.e. with no segregated inner cell mass) in many marsupials and some eutherian mammals; how useful an understanding of this could be to those trying to derive embryonic stem cells from better known species. Holding the blastocyst in a state of suspended animation (embryonic diapause) is an essential component of reproduction in a wide variety of mammals (Flint et al. 1981). The mechanisms underlying this strategy remain poorly understood; how useful it would be, for both applied and research purposes, to be able to turn embryo development on and off with equal efficiency in vitro. As an alternative to cloning by nuclear transfer, blastomere separation offers the advantage of producing animals with identical cytoplasmic as well as nuclear components to their cells (Willadsen 1982, 1989; Johnson et al. 1995; Chan et al. 2000). This, however, is an arduous and inefficient artificial procedure. In contrast, the armadillo naturally and routinely produces identical quadruplets or octuplets (McLaren 1982); would not an understanding of how this is achieved be of enormous practical and fundamental significance? Penman (1991) has discussed the regrettable propensity of humans to believe in what he calls ‘the single, sure road to truth’ and to defend their viewpoint rather too staunchly. Biologists, he argues, have been particularly subject to this failing, dating back to the time of Aristotle. ‘Today’, says Penman, ‘… self styled “reductionists” hold the high ground in biological arguments. They posit that rigorous science is only that which yields a protein sequence, or at the very least, bands on a gel electrophoretogram. All other experimental studies, even if possessed of ingenuity and subtlety, risk being contemptuously labeled “descriptive”.’ As does Penman, I consider that any such attitude ‘should warn us of the dangers of trying to dictate methods’. Without in any way diminishing the value of reductionism, let us list some examples of the value of descriptive studies to our branch of science. It needed descriptive studies of genomic activation and energy metabolism in ruminant embryos to show that the mouse is not necessarily a universal mammalian model (see Betteridge and Rieger 1993). It was observational studies in cattle, as well as mice, that revealed that the sex of embryos influences their early development (Xu et al. 1992). It is censuses in reindeer abattoirs (Godkin 1986) that suggest to us that rates of embryonic death can be extremely low in some ruminant species, and that understanding why could be very useful indeed. In the horse (for references see Betteridge 2000), it was van Niekerk’s observation of many old, unfertilized eggs in the mare’s oviduct that made us think of interactions between embryos and the mother even in the oviduct – a concept since extended to several other species. It was Ginther’s painstaking use of ultrasound that showed us that horse embryos behave very differently from those of most species in the uterus, too. The fact that the horse conceptus is surrounded by a mucin-like capsule suggests many functional questions to investigate with molecular techniques. The hypotonicity of yolk-sac fluid, and the rising concentrations of oxytocin, arginine vasopressin and fructose in the fluid, are enigmas of relevance to blastocyst expansion in general, and probably to the cryopreservation of embryos. Why do pig embryos rely on steroids to establish pregnancy? What is it about them that makes them so difficult to freeze, and how will the answer help other cryobiological studies? What, precisely, allows larger embryos to be in synchrony with their uterine environment and survive better than smaller embryos? These are questions that depend firstly on careful description and to discourage such research would, I believe, be a huge mistake. If we have much to learn from comparative studies of embryos from as wide a range of mammals as possible, how might such studies best be conducted? Molecular biology, of course, has revolutionized the conduct of embryo research over the past two decades. History repeats itself, and so it seems equally likely that other techniques will contribute, in ways that we cannot imagine, to our understanding of how the embryo develops and interacts with its maternal environment. New imaging and microscopic techniques and methods of image analysis are likely to be among these but it is a safe bet that other methods, currently considered remote from those of biomedicine, will also be incorporated into our armamentarium (Dill 1999). Thus, young investigators need to approach our subject with an open and imaginative mind – a need best summed up by an old German proverb, as used with relish by Merkt (cited by Betteridge 1987): ‘Carry a spoon in your pocket, ready for the day that it rains soup’. The author wishes to thank Dr J.I. Raeside, Dr D. Rieger, and Dr R.O. Waelchli for their constructive criticism of this manscript, Allison Fex for graphical assistance, and NSERC, Canada, and the Ontario Ministry of Agriculture and Rural Affairs for financial support.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.