Novel miCas9 technology increases large size gene knock-in rates,
reduces undesirable off-target events and reduces on-target indel events. 


The 2020 Nobel Prize in Chemistry was awarded to Drs. Emmanuelle Charpentier and Jennifer Doudna for their contributions in the development of CRISPR/Cas9. This transformative technology has revolutionized the gene targeting field. Yet to realize its clinical potential to cure genetic disease, many issues remain to be addressed. For example, large size gene knock-in efficacy rate is still low. With respect to safety, substantial off-target editing events occur, raising concerns of tumorigenicity associated with the technology. Equally alarming, the on-target insertion or deletion (indel) rates are often higher than those of the desired precise correction.

To overcome efficacy and safety issues associated with Cas9 at the DNA level, scientists at the University of Michigan (UM) looked into the DNA repair mechanisms. Gene editing nucleases are efficient in generating DNA double-stranded breaks, which are repaired by either the error-prone non-homologous end joining (NHEJ) pathway or the homology-directed repair (HDR) pathway. The balance between NHEJ and HDR determines the outcome of gene editing applications, for which HDR is preferred in precise gene editing applications.

Based on this knowledge, the group focused on RAD51, the major player
in the HDR pathway. UM scientists modified the spCas9 protein by fusing
a thirty-six amino acid long peptide encoded by BRCA2 Exon 27 (Brex27),
which is known to bind RAD51. This new Cas9 variant is named meticulous
integration  Cas9 (miCas9), to reflect its extraordinary capacity to enable
“maximum integration” yet with “minimal indels”, as well as to recognize
its development at the University of Michigan. 

In comparison to spCas9, miCas9 satisfactorily addresses the aforementioned
efficacy and safety deficiencies: it increases large size gene knock-in rates
by multiple folds; it reduces off-target indel rates; and importantly it reduces
undesirable on-target indel rates, the first nuclease that can achieve this
to the best of our knowledge. 
It is further demonstrated that the fusion motif,
Brex27, can be used as a plug-and-play module to improve other gene
editing nucleases. This may benefit many “new” or “understudied” nucleases,
to which a simple fusion of Brex27 is expected to improve both their efficacy
and safety performances.

Brex27’s small size is advantageous since unlike other Cas9 fusion motifs,
Brex27’s addition only increases spCas9’s size by 2%. When it comes to in vivo
delivery of therapeutic biologics, “size matters” and any “room saving” helps,
making this an important aspect of the Brex27 addition. In this regard, Brex27
is the smallest effective HDR promoting motif to date.

In summary, this rationally designed Cas9 variant, miCas9,
possesses a unique combination of desirable features
including improving knock-in rates, reducing undesirable off-target
events, and reducing undesirable on-target indel events,
providing a “one small stone for three birds” tool in gene editing.

new spcas9micas9 construct.jpg

Design of miCas9. (A) Double-strand DNA splicing spCas9 is modified by the covalent attachment of a 36 amino acid motif (Brex27) which enables enhanced recruitment of endogenous RAD51 oligomer, a DNA repair enzyme. (B) miCas9-expressing plasmid DNA construct, with enhanced formation of RAD51 stabilizes the single-stranded nucleoprotein filaments (C) Shows construction of miCas9 comprising a 1368 amino acid spCas9 nuclease with a nuclear localization signal (NLS) and the 36 amino acid Brex27 modification. (D) The amino acid sequence of the Brex27 motif. Brex27 can be used as a plug and play module that is compatible with other Cas9 variants

Mi-Cas9 Reduces / Abolishes Off-Target Edits


Comparison of spCas9 to miCas9 for percent off-target insertions at highly susceptible gene sites. Shown are three highly susceptible sites in the VEGFA gene where VEGFA (A) has 1, VEGFA (B) has 3, VEGFA (C) has 2 off-target sites.
The FANCF site 2 (D), EMX1 (E), and the RUNX1(F) genes each have 1 off-target site.