CRISPR and Down syndrome: removing an extra chromosome from cells

CRISPR and Down syndrome came together in an unusual experiment: instead of correcting a single mutation, the researchers tried to change the number of whole chromosomes. In cultured human cells with trisomy 21, they were able to make a fraction of cells lose one selected copy of chromosome 21, shifting from three copies to two.

Down syndrome is caused by the presence of three copies of chromosome 21 rather than the usual two. That makes an obvious but technically difficult strategy conceivable: remove the extra copy. The challenge is that the three chromosome 21 homologs are highly similar. If CRISPR cuts them indiscriminately, it could damage the wrong copy and create new genomic problems.

How do you choose one chromosome out of three?

Ryotaro Hashizume and colleagues exploited small DNA-sequence differences among the three copies of chromosome 21. These ordinary genetic variants allowed the team to design guide RNAs that preferentially recognized one chosen homolog.

In this system, Cas9 was not used to repair a single DNA letter. Instead, the researchers directed it to make multiple double-strand breaks along one chromosome. The more cuts they introduced, the more often that damaged chromosome was lost from the cell.

The experiments used as many as 13 cut sites. In direct FISH measurements, an optimized 13-cut treatment produced an average chromosome-loss rate of 17.4%. In a separate experiment in which individual cell clones were grown after treatment, 22 of 72 clones — 30.6% — had two copies of chromosome 21 rather than three. Conventional metaphase karyotyping found 46 chromosomes instead of 47 in 15 of 40 metaphases.

Those percentages should not be treated as the same efficiency measure: they came from different assays and different stages of the experiment. Taken together, however, they support the central finding that the selected chromosome could be lost while the other two copies were retained.

The team also temporarily suppressed two DNA-repair pathways. This increased the chance that a chromosome carrying multiple breaks would not simply be repaired and retained. That feature captures both the appeal and the risk of the approach: it deliberately harnesses a type of large genomic event that is normally regarded as a dangerous complication of CRISPR editing.

What changed after the chromosome was lost?

Most experiments were performed in induced pluripotent stem cells derived from skin fibroblasts of a child with full trisomy 21. The same strategy was then tested in the original primary fibroblasts. In those cells, chromosome loss reached 13.9% after treatment, compared with 1.4% in the scramble-guide control.

The authors also asked whether chromosome loss could occur in cells that were not dividing at the time. In that experimental system, chromosome loss was detected in about 3.2% of nondividing cells, compared with 0.4% in the control. This matters for any future therapeutic concept because most mature cells in the body are not continuously proliferating.

The shift from trisomy to two copies of chromosome 21 also altered gene activity. RNA-seq showed that corrected cells clustered separately from the original trisomic cells; they also proliferated somewhat faster and produced less reactive oxygen species. In other words, removing one chromosome changed more than a karyotype image — it altered measurable cellular properties.

That does not mean that features of Down syndrome were reversed in a person. Changes in gene expression in cultured cells cannot be directly translated into effects on brain development, cognition, or the physiology of an entire organism.

Why this is not a treatment yet

The study is limited in scale. It used a single iPS-cell line and two cell types, all derived from one donor. No animals, embryos, or patients were treated, and the problem of delivering such a CRISPR system to the right tissues in a living body remains unsolved.

There is also a more fundamental safety issue. If Cas9 cuts the selected chromosome but the cell does not lose it, the breaks can be repaired with small insertions, deletions, or structural variants. Whole-genome sequencing also detected some cutting of the two nontarget chromosome 21 copies at intended target loci. The authors identify protection of those nontarget copies as a major challenge.

An allele-specific strategy also requires knowing which DNA variants belong to each of the three chromosome 21 copies and then designing suitable guides. In its present form, that makes the approach considerably more complex than a universal drug.

So this paper does not show that Down syndrome can now be treated by genome editing. It shows something narrower but scientifically important: in human trisomy 21 cells, a CRISPR system can be designed to distinguish among three nearly identical copies of one chromosome and increase the probability that one selected copy is eliminated. The harder question is whether that can ever be done safely, reproducibly, and selectively in a living organism.