NEW YORK – A team led by researchers at the European Molecular Biology Laboratory and GlaxoSmithKline's Cellzome has found that a substantial proportion of genes knocked out via CRISPR-Cas9-induced frameshift mutations still produce proteins.
In a study published today in Nature Methods, the researchers used RNA sequencing and mass spectrometry to measure transcripts and proteins produced by 136 genes with 193 CRISPR-induced deletions and found residual protein expression in roughly one-third of these cases.
Investigating several of the knockout cell lines in detail, the researchers found that the proteins produced by the impaired genes were truncated but retained some functionality.
In a CRISPR-induced frameshift mutation, the Cas9–sgRNA complex breaks the DNA at the target sequence, and the break is then repaired by non-homologous end joining, which the authors noted is "an error-prone pathway introducing insertion or deletion mutations that can lead to frameshifts and a premature termination codon (PTC) in the expressed transcript, resulting in nonsense-mediated decay (NMD) of the mRNA and aberrant peptide products that are degraded."
However, they added, the effectiveness of these deletions is typically assessed at the gene level, leaving open the question of what, exactly, is produced at the mRNA and protein level.
To explore this question, the researchers used RNA sequencing to measure mRNA expression levels in 174 of the 193 knockout cell lines they produced, finding that the reduction in mRNA expression ranged widely across the different cell lines. This observation, they noted, tracked with other recent research that found "high variation in NMD efficiency depending on a complex set of biological factors."
The researchers followed this experiment with a mass spec analysis using TMT isobaric labeling to quantify levels of proteins produced by the deletion-carrying genes, finding that roughly one-third of the knockout lines showed residual protein expression. They also found that there was no correlation between mRNA and protein expression levels.
Looking closely at several of the knockout lines, they found that the residual protein expression was in some cases due to alternative splicing that skipped the CRISPR-induced frameshift mutation while in others it was caused by translation reinitiation at a site not impacted by the mutation.
The findings indicate potential limitations of the CRISPR technology, though the authors noted that "it should be possible to at least partially suppress or circumvent these mechanisms by adapting the experimental design, for instance, by using combinations of sgRNAs and/or modifying splicing sites."
The study also suggests that the effectiveness of CRISPR-induced knockouts should be assessed not only at the gene or transcript level but also at the protein level using mass spectrometry, they wrote, which is more sensitive than immunoblotting approaches.