Issue Date: September 18, 2017 | Web Date: September 15, 2017
Doubts raised over validity of CRISPR-edited human embryo study
Last month, an international team reported using CRISPR/Cas9 gene-editing techniques to remove a common genetic mutation from human embryos that causes the heart condition hypertrophic cardiomyopathy. Now, experts in embryology and genetics are calling that report into question.
In the initial study, published Aug. 2 in Nature, researchers led by Shoukhrat Mitalipov of Oregon Health & Science University claimed they used CRISPR to correct a mutation in the MYBPC3 gene introduced into embryos by a sperm donor’s DNA (DOI: 10.1038/nature23305). To do so, they simultaneously injected the sperm and CRISPR/Cas9 components into healthy eggs. The CRISPR system snipped out the mutation, and, the researchers contended, natural repair mechanisms used the egg’s maternal gene as a template to rebuild a healthy version.
Lack of direct proof that this repair process occurred is central to the doubts raised by Maria Jasin of Memorial Sloan Kettering Cancer Center, Dieter Egli of Columbia University, and four other scientists. The group uploaded a paper on Aug. 28 to the biology preprint server bioRxiv outlining their concerns (DOI: 10.1101/181255).
One of the issues the authors highlight is that maternal and paternal DNA are too far apart inside a brand-new embryo for the maternal gene to guide repair of CRISPR-made cuts in the paternal gene. They suggest instead that an error-prone repair mechanism may have added or deleted enough nucleotides to render the cut region unrecognizable. Only the mother’s healthy copy of the gene would be detectable in this situation, leading to an incorrect conclusion, the authors argue. The preprint paper is under review with a journal, so the authors did not speak with C&EN for this story other than to confirm its factual accuracy.
The doubts raised in the bioRxiv paper are legitimate concerns, says Tony Perry, a developmental biologist at the University of Bath. After fertilization, maternal and paternal genomes must cross tens of micrometers—“intergalactic distances, in cellular terms,” he says—before they are close enough for the kind of DNA repair suggested in the Nature paper. Determining which repair process actually occurred will take serious bioinformatics effort, Perry says.
Mitalipov indicated in a statement that he and his coauthors stand behind their findings and are working on a point-by-point response to the bioRxiv paper.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society