How to Make an Egg

In vitro gametogenesis (IVG)

“A hen is only an egg’s way of making another egg.” Samuel Butler’s readers may have dismissed a wisecrack that seemed to mock nature, even attack the dignity of our organic selves, but later in the nineteenth century the eminent biologist August Weismann made sense of it.

We regard reproduction as a choice, but in nature it is the raison d’être, the only avenue to avoid a genetic lineage going extinct. We might expect physiology would execute the crucial process perfectly, but trade-offs impose compromises that sometimes come with a tragic cost.

Single-celled organisms replicate endlessly, but the evolution of multicellularity introduced a mortal body (the soma) to convey germ cells along a potentially immortal line. Each body hands over its genetic cargo to the next generation, like passing the baton in a relay race, although genes get scrambled by assortment in meiosis. Weismann pictured somatic and germ cells as distinct domains that segregate immediately after fertilization, so they can never share information or injuries.

The Weismann barrier carried immense implications for embryology and evolution. It implied that the history of soma cannot affect the germplasm and, hence, the next generation. It seemed to invalidate in a swoop the Lamarckian theory of evolution by inheritance of acquired characteristics, epitomized by the stock example of a blacksmith’s son who can’t inherit the brawniness of his father from swinging hammers. Weismann’s germplasm theory became a cornerstone of the new science of genetics and bolstered neo-Darwinism in the next century. We shouldn’t blame Weismann when the imperfections of his theory showed up because that’s always a risk for pioneers. Life is more complex than we know, or perhaps we can know. Evolution is not a law; it is contingent. We study a lawless science.

The two lineages don’t separate as strictly as he supposed in all animals. Somatic cells in the ovaries and testes—and possibly further off—provide not just nutrition and governance of germ cell development, but to convey proteins and microRNAs in extracellular vesicles that can modulate gene expression in developing eggs by blocking messenger RNAs. Epigenetic effects carry over for a generation or more. Too much significance was lent to germ cell autonomy and too little to the community of cells needed to make an egg (or sperm).

The embryology of sea squirts, bristle worms, and sea urchins casts shadows over the absolutist doctrine. These creatures can replace germ cells in embryos, larvae, or even at later stages with somatic cells under experimental conditions and perhaps in nature.

Sea urchins still blaze in my memory after pulling spines out of my soles after paddling in the Mediterranean! So they are my example. If the embryonic patch destined for germ cells is eliminated, they are replaced by somatic cells elsewhere, confirmed by the canonical germ line marker Vasa and subsequently by making eggs or sperm. This prickly exception to the germplasm theory shows the tenacity of reproduction. Life always finds a way.

But there is an enigma in warm-blooded vertebrates, including hens and humans. Females don’t have the plasticity of invertebrates that generate new germ cells by multiplication throughout life or by the conversion of somatic cells. A rare exception is the queen of naked mole rats for its eusocial life history (matriarchal). Otherwise, ovarian germline stem cells irreversibly cease activity before or shortly after birth, so females must eke out a limited egg store that steadily flows into the pipeline to ovulation or until emptied at menopause. No such constraint exists in spermatogenesis. The striking difference between sexes tempted some researchers to ask if conventional belief is false—if the ovary behaves more like the testis than we thought—but exhaustive studies have confirmed the dogma, with immense implications in reproductive medicine.

There must be an evolutionary rationale for a perilous strategy in females. An obvious reason is that we don’t scatter masses of eggs in a hostile environment, like invertebrates that leave their survival to chance. As mammals (and birds), we invest in exceptionally large, high-quality eggs, and release a small number at ovulation to match the capacity of the womb (or nest) and postnatal care. Instead of eliminating an excess number of eggs after ovulation, we eliminate them before shedding through hormonal control. Superovulation rescues those that would otherwise degenerate by atresia. It appears to be a random process and no respecter of egg quality. If atresia existed to remove defective eggs instead of a numerical regulator, superovulation would be less safe in clinical practice.

Weismann formulated his “doctrine” before the discovery of DNA, even before genes. He thought segregation of the germline ensured its immortality to guarantee the vitality of posterity compared to somatic cells that are doomed to mortality in every generation. The matter is complicated, although it remains true that germ cells have more privileges than other cell types.

There is a risk of point mutations every time DNA replicates in a cell cycle—about six events per diploid nucleus, and a similar frequency across the animal kingdom for thermodynamic reasons. Most are harmless, but a mutational load accumulates over many cycles, along with the hazards of epigenetic drift.

Germ cell precursors in the testes have undergone hundreds of mitotic cycles by middle age, accounting for 80% of fresh mutations in children. Sperm are shot with mutations in their mitochondria; fortunately, they are not inherited by babies. The minor balance of 20% responsibility for mutations coming from eggs is due to their history of only 20 to 30 cycles happening before birth. The difference between sexes also explains why germ cells cause only 5% of ovarian tumors compared to 95% of testicular tumors in young men and adolescents.

Eggs also benefit from efficient DNA proofreading and repair mechanisms. And the risk of a mitochondrial mutational meltdown is minimized by a genetic bottleneck in fetal germ cells, meaning that a tiny number of organelles will give zygotes a healthy start. Nevertheless, egg cells are not entirely safe from errors. Reactive oxygen species cause damage, especially when they are more metabolically active coming out of a prolonged dormancy, and probably contribute to chromosome errors, especially at advanced maternal ages. Perfection is illusory in nature.

Perhaps no other factors limit fertility treatment more than the scarcity and quality of eggs. How are they managed, and what are the options in the future?

Ovarian stimulation with gonadotropic hormones originated in the late 1950s. In vitro maturation (IVM) launched afterward is getting renewed attention (discussed elsewhere on this platform). Although one is a clinical treatment and the other a laboratory procedure, they have something in common. The hormone FSH is fundamental for superovulation and is sometimes a primer for IVM; both of them work by increasing the number of eggs available by rescuing follicles from the fate of atresia. Safety is assured after thousands of healthy babies have been born over the decades, but they sometimes fall short of the number of eggs needed. In theory, a larger harvest is possible by drawing from the pool of abundant small follicles that are unresponsive to injected hormones.

This goal was launched in the 1990s by Gil Greenwald in Kansas City, John Eppig in Bar Harbor, Maine, and by our Edinburgh group, and followed up by technical refinements elsewhere. It seemed obvious to try culturing small follicles to the stage when they progress to maturity by IVM methods. We called the technology in vitro growth or IVG because it involves the major phase of oocyte growth. The acronym stuck after the original goal extended from growing eggs taken from the ovary to generating them de novo from stem cells by “in vitro gametogenesis.”

It is much harder to grow follicles in vitro than to culture zygotes to the blastocyst stage. It takes longer and requires support from somatic cells (granulosa), so culture conditions must satisfy both cell types. Moreover, there is no increase in cytoplasmic mass up to the blastocyst stage (slightly the opposite), whereas eggs must grow to full size with a corresponding increase in the cloud of granulosa cells. Moreover, IVG is even more challenging for human follicles than animal models because they grow to the size of grapes. In static culture, they exceed the limits for diffusion of gases, metabolites, and nutrients normally met by a blood supply. Can these hurdles be overcome, or were we dreaming?

The Eppig lab deserves credit for being the first to grow oocytes to maturity entirely in vitro from the smallest stage taken from neonatal mouse ovaries. The follicles initiated growth spontaneously, and when they reached a certain stage, they were isolated as granulosa-oocyte complexes using enzymes for further growth on hydrophobic membranes. Oocytes developed on stalks (resembling the fruiting bodies of slime molds) to undergo maturation divisions of meiosis before fertilization. Embryos were transferred to surrogate mothers for the remainder of development. The firstborn, called Eggbert, was obese and short-lived, but a spurious case because subsequent pups were healthy and fertile.

The Edinburgh group adopted a different method, growing whole follicles individually in tubes, more suitable for physiological studies than mass production. Some follicles ovulated into the culture fluid for collection before IVF. Evelyn Telfer and Helen Picton, working in Scotland and Yorkshire respectively, adapted methods for farm animal follicles, setting the stage for a collaboration with Edinburgh clinicians who supply human ovarian biopsies.

Evelyn’s group managed the greater complexity by extending the two stages of Eppig’s method in mice to four, with IVM as the final stage for human follicles. The prospects of growing human eggs in vitro looked doubtful until recently, but they have achieved eggs matured in vitro. This is a cause for celebration, and even if the efficiency is still low, refinements will surely lead them to routine success. The group has applied to the Human Fertilisation and Embryology Authority for a licence to test the fertility of IVG-generated eggs by fertilization with donor sperm.

Clinical embryologists have the skills needed for IVG. Among the clear clinical needs are hormonally resistant ovaries, growing eggs from frozen ovarian tissue for cancer patients, and dwindling follicle numbers in poorly responsive ovaries before premature menopause. The methods won’t help women after menopause or with congenital hypogonadism who want to have a genetically related child—their options are limited to egg/ embryo donation or adoption. But even this hurdle will likely be overcome eventually with somatic cells through stem cell technology, which will overturn Weismann’s doctrine and make him roll in his grave!

In the past two decades, induced pluripotent stem cells (iPSCs) discovered in Kyoto laboratories have largely replaced ethically contentious embryonic stem cells for research. They are now transforming the outlook for regenerative medicine, drug discovery, and disease models. When iPSCs are generated from somatic cells using a few key transcription factors, these pluripotent cells can be canalized for differentiation into virtually any cell type, including fertile gametes of both sexes in rodent models. The technology is conquering the barrier against generating germ cells from the soma, which sea urchins could always do!

The engineering of new human eggs starts with a biopsy of somatic cells (epidermal cells hold promise). They will be converted to iPSCs before passaging through successive stages of egg development to reach maturity for IVF or ICSI. After we learn if IVG eggs in Edinburgh are safe for human reproduction, even more rigorous tests will be needed for eggs derived from stem cells. They have a higher risk of epigenetic abnormalities after genetic manipulation and a bigger mutational load from more cell cycles. These technologies will likely jump one day out of the research laboratory into the clinic unless animal models are misleading.

They are the equivalent of moonshots in reproductive science today. It will be a long journey to landing and to celebrate breaking the Weismann barrier, although that will still stand for something.

Thanks to Professor Evelyn Telfer, C.B.E., F.R.S.E. at the University of Edinburgh for helpful suggestions

Image Credit: Sea urchin. Unsplash (unknown species and photographer)