Base Editing Reveals NANOG Builds All Tissues

A precision genome-editing technique has answered one of the most fundamental open questions in human developmental biology — and in doing so, has exposed a critical blind spot in decades of animal research.

On Thursday, scientists at the University of Cambridge’s Loke Centre for Trophoblast Research published a paper in Nature demonstrating for the first time that a technique called base editing can be used to study gene function in living human embryos with minimal collateral damage to the genome. Their target was NANOG, a transcription factor long recognized as central to stem cell biology.

When they silenced it, the results were unequivocal: human embryos without NANOG cannot form the epiblast, the thin inner layer of cells from which every tissue, organ, limb, and structure of the human body eventually arises. No epiblast, no body.

The finding carries an equally significant corollary. In mice, NANOG is important but not irreplaceable — other molecular pathways can partially compensate for its loss. In human embryos, no such backup system exists.

For the millions of couples worldwide who undergo IVF cycles each year, and for researchers working to reduce the majority of embryos that fail to establish a viable pregnancy, the gap between what mouse experiments suggested and what human embryos actually require has now been precisely mapped for one critical gene.

What Base Editing Does That CRISPR Cannot

The study’s scientific foundation is a gene-editing technique that represents a meaningful departure from the tool that electrified biology a decade ago. CRISPR/Cas9 — the system that earned its developers a Nobel Prize in Chemistry in 2020 — functions by delivering molecular scissors to a precise location in the genome and cutting through both strands of the DNA double helix.

That double-strand break triggers cellular repair machinery that, in delicate cells such as early-stage embryos, frequently causes chromosomal chaos: large deletions, rearrangements, and errors that make it difficult to determine whether observed developmental failures stem from the intended gene knockout or from unintended genomic damage. Niakan’s own lab demonstrated this problem in a 2017 Nature paper, showing that CRISPR used in human embryos reliably caused the loss of entire chromosomes.

Base editing sidesteps the cut entirely. Instead of severing both DNA strands, it chemically converts one nucleotide — a single letter in a three-billion-letter genetic text — into another, without ever breaking the backbone of the helix.

The Cambridge team used a variant called adenine base editing, specifically the ABE8e tool, to introduce a splice-site mutation that disabled NANOG’s function entirely. The resulting Nature paper confirmed that this approach produced no genotoxicity and only limited off-target editing — a significant technical improvement over what was possible before.

There are real limitations. Some edited embryos exhibited mosaicism, meaning not all cells in a given embryo carried the edit uniformly. Bystander mutations — small, unintended changes near the target site — were also detected, consistent with findings from a separate base editing study in human embryos published three weeks earlier by a Columbia University team led by geneticist Dieter Egli. Neither paper claimed clinical readiness; both confirmed that base editing, while substantially less damaging than standard CRISPR, still requires further refinement before any clinical application could be contemplated.

What NANOG Does, and What Happens Without It

NANOG was first functionally characterized in 2003, when two independent research groups identified it as one of a small cluster of master transcription factors required to maintain embryonic stem cells in their pluripotent state — the condition of being able to differentiate into virtually any cell type in the body. It works in concert with two partners, OCT4 and SOX2, to sustain a gene regulatory network that keeps early embryonic cells from committing prematurely to a specialized fate.

Its precise role in the living human embryo had never been definitively established, however. The Cambridge team’s experiment provided that definition with unusual clarity.

When ABE8e was used to knock out NANOG in early-stage human blastocysts — the roughly 200-cell embryo that forms around day six after fertilization — the epiblast, the inner pluripotent cell layer from which all fetal tissues arise, failed to form. The cells that would normally become placenta and yolk sac were largely unaffected. The cells destined to become every organ, limb, and nervous system structure simply did not emerge as a distinct population.

Critically, cells did not remain as unspecified precursors. In NANOG-null human embryos, cells appeared to redirect toward a primitive endoderm — the yolk sac’s transcriptional program — instead. Losing NANOG does not simply halt development; it reroutes it, and the body-forming lineage is the specific casualty.

Why Mouse Studies Could Not Have Revealed This

The most scientifically significant aspect of the finding may not be what NANOG does, but what it reveals about the limits of animal models as a proxy for human biology. Janet Rossant, a developmental biologist at the Hospital for Sick Children in Toronto, has argued that understanding the molecular rules of human embryo development requires studying human embryos directly.

The Cambridge results provide a specific, crisply documented example of why. In mice, the absence of NANOG produces a different and less absolute outcome — other regulatory molecules can substitute for its function during the equivalent developmental window. That functional compensation, well documented in mice, does not exist in humans. Any pre-clinical developmental biology conclusion based on murine NANOG experiments is not wrong about mice. It is simply not applicable to human embryos.

The study’s research design addressed one of base editing’s known limitations: mosaicism. Base-editing components were introduced at the time of intracytoplasmic sperm injection, so the editing took effect in the fertilized egg before the first cell division. This meant edits were present from the start, reducing the proportion of embryos where only some cells carry the change.

A Landmark Result Lands in a Charged Moment

The Cambridge paper arrived on Thursday, three weeks after a separate study by Egli’s team at Columbia set off a wave of scientific and ethical debate. That work — published as a preprint on bioRxiv on June 1 — showed that base editing could be used to correct disease-causing mutations in the human embryonic genome, targeting genes linked to cholesterol dysregulation and blood disorders. Leading gene therapy organizations, including the International Society for Cell and Gene Therapy and the American Society of Gene and Cell Therapy, had called for a ten-year moratorium on heritable human genome editing as recently as May 2025.

Two peer-reviewed studies demonstrating the technical feasibility of base editing in human embryos within the same month sharpened those concerns. The core anxiety is not specific to either paper’s stated purpose — the Cambridge study explicitly used embryos that were never intended for implantation — but rather cumulative. Each successful precision edit raises the technical floor, makes the next experiment easier to defend, and draws the field closer to a threshold that governance frameworks have not yet defined with legal force.

Biomedical ethicist Hank Greely at Stanford University has expressed concern that affluent individuals could use such research as a springboard toward trait-selection editing of embryos. Fyodor Urnov, who studies molecular therapeutics at the University of California, Berkeley, has argued that editing the genomes of human embryos to treat disease amounts to a solution in search of a problem, given that preimplantation genetic testing already allows couples to select unaffected embryos before implantation.

The 2018 case of He Jiankui — the Chinese researcher who implanted CRISPR-edited embryos and produced live-born children with heritable genetic changes — sent shockwaves through the scientific community and produced calls for international consensus-building. Eight years later, no binding global governance framework for germline editing exists. Countries operate under radically different legal regimes, from the United Kingdom’s tightly licensed system to contexts where oversight is minimal or absent.

What Strict Oversight Made This Research Possible

The Cambridge team conducted the research under a license from the Human Fertilisation and Embryology Authority, the UK government’s independent regulator for fertility treatment and embryo research. All embryos used were donated by patients specifically for research purposes. The study also received approval from the Newcastle and North Tyneside Research Ethics Committee. Embryos were cultured only to the blastocyst stage and were not transferred or implanted.

The UK’s Human Fertilisation and Embryology Act, enacted in 1990 and updated in 2008, establishes the 14-day rule as a legal limit on embryo culture in vitro. The HFEA has since recommended to Parliament that this limit be extended, citing advances in developmental biology that make the current restriction scientifically limiting; as of now, the 14-day rule remains in force. The act explicitly prohibits altering the nuclear DNA of embryos intended for reproductive use.

Dr. Helen O’Neill of University College London’s Institute for Women’s Health, commenting on the study, emphasized that the research’s purpose was epistemic rather than clinical — to understand the genetic rules governing the first week of human life — and that its value lay precisely in what it revealed about human-specific biology that animal models cannot provide.

What Comes Next for IVF and Heritable Disease Research

The immediate scientific implications flow in two directions. A more detailed understanding of the molecular signals that govern epiblast formation could improve how embryos are selected and supported during IVF. Failures of very early development — before implantation, before a pregnancy is established — account for a large proportion of IVF cycle failures and recurrent miscarriage. Identifying which genes are non-negotiable in human embryos, as opposed to dispensable as mouse data suggested, gives reproductive medicine a clearer map of where those failures originate.

The study also demonstrates that base editing is a viable, lower-risk tool for interrogating gene function in human embryos — a proof of concept that will encourage further work on other transcription factors and developmental regulators whose human-specific roles remain unknown.

Base editing could in principle be used to correct embryos carrying mutations for conditions such as cystic fibrosis or Huntington’s disease before implantation. That use would not be legally permissible in the United Kingdom under current law, and the Cambridge team was explicit that extensive safety testing, further technical refinement, and broad public deliberation would all be prerequisites for any clinical application. What the study has established is that the precision to attempt such work — safely enough to study rather than harm — now exists.