# Base Editing Tests NANOG in Human Embryos

> Researchers disabled NANOG in donated research embryos and found it essential for forming the pluripotent epiblast, but the experiment does not establish clinical germline editing as safe.

_Source: Nature research article, verified against PubMed, Crossref, OpenAlex and HFEA-regulated study records described by the authors · 2026-09-25 · 8 min read · Verified against primary sources_

Canonical: https://iyu.app/e/nanog-base-editing-human-embryos

## The 60-second version

ABE8e base editing of donated human embryos showed that NANOG is essential for establishing the pluripotent epiblast lineage.

**Key points**

- Twelve of 28 targeted embryos began blastocyst formation; the key analysis used 359 cells from 11 edited blastocysts.
- Loss of NANOG shifted cells toward primitive endoderm or trophectoderm rather than pluripotent epiblast.
- Targeted assays found limited off-target editing and no detected chromosome loss at tested regions, but the safety sample was small and incomplete.
- The work was licensed research on donated embryos and did not test implantation, birth, long-term health or clinical benefit.

**Verdict.** Strong evidence for NANOG's developmental role and for base editing as a research tool, not evidence that heritable embryo editing is clinically safe.

## Full explainer

Disabling **NANOG** with adenine base editing prevented donated human embryos from maintaining the pluripotent epiblast program. Cells instead moved toward primitive-endoderm or trophectoderm identities, revealing a human-specific lineage response that partly differs from mouse embryos.

> **⚑ Caveat:** This was licensed laboratory research on donated embryos, not fertility treatment. It did not test implantation, pregnancy, birth, long-term health or multigenerational effects, and it does not establish clinical germline editing as safe.


### Mechanism — Editing a splice site to switch off NANOG

NANOG is a transcription factor central to pluripotency. The team used ABE8e, which changes an adenine base through a Cas9 nickase and deaminase rather than cutting both DNA strands. Editing the splice donor after NANOG exon one disrupted RNA splicing and caused functional loss of the protein.

- **Material:** Human eggs, sperm and embryos donated after assisted reproduction; embryos were used only for research.
- **Oversight:** UK HFEA research licences R0162 and R0152, plus research ethics review and donor consent.
- **Intervention:** ABE8e targeted the NANOG exon-one splice donor; AAVS1 editing served as an experimental control.
- **Question:** Whether NANOG is required to establish and maintain early human embryonic lineages.


### Experiment — What happened in the edited embryos

- **28** — NANOG-targeted embryos
- **12** — initiated blastocyst formation
- **359** — single cells from 11 edited blastocysts
- **42** — cells from 2 edited control regions

Twelve of 28 targeted embryos initiated blastocyst formation and expansion. The decisive single-cell dataset came from 359 cells across 11 edited blastocysts; one of 12 collected biopsies lysed before sequencing. The analysis also used two AAVS1-edited controls and published reference embryos.

Cells carrying disruptive NANOG edits lost the pluripotent epiblast state and instead expressed primitive-endoderm or trophectoderm programs. Human embryos retained primitive-endoderm differentiation, unlike Nanog-null mouse embryos. The species share a failure to specify epiblast, but not every downstream lineage response.


### Safety evidence — Promising signals within a narrow test

- **1. Chromosomes:** low-pass whole-genome sequencing covered 23 biopsies from 5 blastocysts and found no loss at the target or top five predicted off-target regions.
- **2. Predicted DNA sites:** deep sequencing of 10 predicted off-target sites in 30 biopsies from 10 embryos found editing of 0.15% or less.
- **3. RNA:** an assay in human embryonic stem cells found apparent off-target changes near the estimated technical-error range.
- **4. Scope:** these assays do not exclude unpredicted, rare, structural, mosaic or later developmental effects.

> **⚑ Caveat:** A biopsy samples only part of a potentially mosaic embryo. Selected-site tests and low-pass sequencing cannot prove a genome is unchanged, and no embryo was transferred for reproductive use.


### Interpretation — A mechanism result, not a clinical green light

The study shows that base editing can be used as a functional probe in early human development and provides direct evidence that NANOG is essential for epiblast specification. It does not demonstrate improved fertility, prevention of pregnancy loss or a safe way to alter inherited traits.

The authors state that sample sizes were not predetermined statistically and experiments were not blinded. They call for larger cohorts, more loci, broader genotoxicity and developmental testing, ethical oversight and public debate before any clinical consideration.


#### Disclosed interests

David Liu disclosed co-founding genome-editing companies and co-inventing base- and prime-editing patents. Oliver Bower joined a company studying gene-editing safety after his work on the project. The other authors declared no competing interests.

> The experiment clarifies what NANOG does; it does not answer whether edited embryos should ever initiate a pregnancy.


### Takeaway — Keep discovery and reproductive use separate

The appropriate next step is replication and more comprehensive safety analysis under strict oversight. Readers should treat this as a developmental-biology advance and a research-tool result, not as evidence that clinical embryo editing is ready or justified.


## Primary sources

- [Telegram post 1505](https://t.me/CNSmydream/1505)
- [Nature paper](https://doi.org/10.1038/s41586-026-10792-1)
- [PubMed record](https://pubmed.ncbi.nlm.nih.gov/42350792/)
- [Crossref record](https://api.crossref.org/works/10.1038/s41586-026-10792-1)
- [OpenAlex record](https://openalex.org/W7165921062)
- [Public study data](https://doi.org/10.6084/m9.figshare.32336139.v3)

---
_Published by iyu (https://iyu.app) — the day's AI news, checked against primary sources and rewritten in plain language. Free to quote with attribution and a link to the canonical URL._