To err is human: How human embryos rarely get it right

Every human being originates from a single cell, called a zygote, which is formed when an egg is fertilised by a sperm. The zygote immediately starts dividing, with each division producing two daughter cells.

This process repeats throughout embryo development, during and after pregnancy, and as the resulting child grows, until the approximately 30 trillion cells that make up the body of a human adult have formed. In adulthood, some cells continue to divide to support the renewal of tissues such as blood, skin and the gut lining, and to promote wound healing.

During cell division – known as mitosis – the contents of a cell are divided equally between the two resulting daughter cells. These contents include chromosomes, which are pieces of DNA carrying the instructions for how to make a human. Correct cell function requires a complete set of chromosomes, which in humans means 46 chromosomes in the nucleus of each cell.

To maintain this number, chromosomes are replicated and the two identical sets that result are separated during mitosis. Surveillance mechanisms help prevent chromosomes from being lost or misplaced, ensuring that each daughter cell receives a complete set.

Deviation in the number of chromosomes is known as aneuploidy and is associated with abnormalities. For example, aneuploidy is often a feature of cancer. If aneuploidy occurs early in development, it can lead to developmental syndromes, such as Down's syndrome, which is caused by an extra copy of chromosome 21, or to pregnancy loss, because many aneuploid embryos cannot develop normally.

Interestingly, cell divisions in the early human embryo – especially the first few divisions following fertilisation – seem to work in a distinct way. The first division of human embryos is particularly prone to chromosome segregation errors, with more than 30 percent of zygotes dividing with abnormalities. This is surprisingly high, considering that fewer than one percent of the cells of a developed human body divide with abnormalities.

These segregation errors can cause differences in the chromosomal content of daughter cells if chromosomes are lost or end up in the wrong cell. This in turn can result in a 'mosaic' embryo, meaning an embryo whose cells have different chromosomal makeups.

In recent years, it has become clear that human embryos are commonly mosaic. Some evidence suggests that more than 80 percent of day five embryos, known as blastocysts, are mosaic. Remarkably, such embryos can result in successful pregnancies and the birth of healthy babies, despite the presence of aneuploid – that is, chromosomally abnormal – cells.

One clinical trial studying mosaic embryos showed that the chances of establishing a healthy pregnancy are very similar for chromosomally normal ('uniform') embryos and mosaic embryos, provided that the abnormal cells make up a minority of cells within the mosaic embryo (see BioNews 1122, 1312 and 1316). Other evidence suggests that abnormal cells are progressively removed as development progresses.

The precise mechanism behind this removal is unclear. Theories suggest that it might occur through directed cell death, and/or the extrusion of abnormal cells into the outer embryonic layer – the trophectoderm – which subsequently develops into tissues that support the pregnancy. In any case, it seems that human embryos can, at least to some extent, tolerate abnormal cells.

All that being said, we should also recognise that most human embryos fail at some point. In fact, evidence suggests that fewer than a third of conceptions result in a live birth, regardless of whether conception is assisted or natural. This figure is consistent with the current IVF success rate per embryo transfer in the UK, which was 30 percent on average in 2024.

Such low efficiency in human reproduction may be due – at least in part – to the high levels of aneuploid cells in early developing embryos. If so, one way to improve IVF success rates would be to reduce the frequency of chromosome segregation errors during the early stages of embryo development.

This is much easier said than done, since we do not fully understand why embryos are susceptible to aneuploidy. However, my colleagues and I have explored this question using a small molecule called UMK57, which affects the cellular machinery that physically separates chromosomes during mitosis. We found that human embryos treated with UMK57 displayed significantly fewer errors during the first division of the zygote.

These findings are exciting, and UMK57 could be tested in a larger cohort study. However, the safety of such an intervention must also be considered. We must also consider a broader question: could the presence of aneuploid cells in early embryos have a positive effect, potentially explaining why human embryos are so prone to chromosomal abnormalities?

Perhaps it has something to do with the incredible maternal investment involved in a pregnancy, with aneuploidy somehow serving as an indication of overall embryo quality. Alternatively, it may have something to do with implantation potential. As mentioned above, cancer cells are commonly aneuploid, particularly when they can migrate around the body and settle away from their original site (a process called metastasis). Following to a similar logic, aneuploid cells in an embryo might help the embryo to implant in the uterine lining and thereby establish a pregnancy.

All these hypotheses remain to be investigated. In the meantime, establishing a technique to improve the accuracy of cell division is an important step towards a better understanding of early embryonic development and, in the longer term, potentially improving IVF success rates.

Much of the research discussed above has been made possible by fertility patients donating their unused embryos for research. Such donations have played an invaluable role in a number of important discoveries, and continue to enable advances in assisted reproductive technologies. I think I speak for all my fellow researchers when I express my heartfelt gratitude to all patients who have donated embryos for research, or who are considering doing so.

Acknowledgements:

This article is based on work conducted at the McAinsh lab (Warwick Medical School, University of Warwick) in collaboration with the Centre for Reproductive Medicine (CRM) within University Hospitals Coventry and Warwickshire. The CRM facilitates embryo donations by fertility patients to our research study, which we greatly appreciate. More information about human embryo research in the McAinsh lab can be found here.