5 thoughts on “Poll: impact of off-target effects on CRISPR as a tool?”
anonymous stem cell repairman
The data’s basically there already. Highly specific in embryos and ESCs and iPSCs, not so much in transformed cell lines (mostly those beginning with “H” and ending with “EK293T”). From my viewpoint, the missing piece is that the mechanism for this isn’t worked out, particularly why HEK293T seems so much more prone to use an error-prone repair pathway than what primary cells appear to use.
The human embryo offtarget analysis was pointed out as largely flawed by Doug Mortlock I believe, leading to it actually being more on-target than those authors stated due to not controlling for genetic variation within the population the embyros came from.
@Jeanne,
The off-target effects will definitely depend on the specific gene being targeted. They have to be evaluated on a case-by-case basis. So it is hard to make generalizations. However, we can hope over time that data will accumulate to paint general pictures about the incidences, the nature, etc.
I’m not sure that anyone has done a systematic analysis of different variables in CRISPR gene editing human pluripotent stem cells.
Paul
I would just like to know if there are consistent observations. I’m sure that the off-target effects are dependent on the gene being targeted as well as the design of the experiment.
Sorry if I missed it, but has anyone done an investigation of multiple variables in hESCs or iPSCs? Back in the old days when we were doing straightforward homologous recombination to modify mouse ESCs, different labs got different results targeting the same genes. Is that happening in this case?
Jeanne
Doesn’t it depend a lot on the context? If it’s work where the accuracy doesn’t matter too much – in vitro work where good v bad cells can subsequently be sorted or in vivo work with organisms whose health isn’t a major concern (C elegans), it wouldn’t matter. If it were in human embryos intended to give rise to babies, it would matter enormously!
The data’s basically there already. Highly specific in embryos and ESCs and iPSCs, not so much in transformed cell lines (mostly those beginning with “H” and ending with “EK293T”). From my viewpoint, the missing piece is that the mechanism for this isn’t worked out, particularly why HEK293T seems so much more prone to use an error-prone repair pathway than what primary cells appear to use.
The human embryo offtarget analysis was pointed out as largely flawed by Doug Mortlock I believe, leading to it actually being more on-target than those authors stated due to not controlling for genetic variation within the population the embyros came from.
@Jeanne,
The off-target effects will definitely depend on the specific gene being targeted. They have to be evaluated on a case-by-case basis. So it is hard to make generalizations. However, we can hope over time that data will accumulate to paint general pictures about the incidences, the nature, etc.
I’m not sure that anyone has done a systematic analysis of different variables in CRISPR gene editing human pluripotent stem cells.
Paul
Paul and all:
I would just like to know if there are consistent observations. I’m sure that the off-target effects are dependent on the gene being targeted as well as the design of the experiment.
Sorry if I missed it, but has anyone done an investigation of multiple variables in hESCs or iPSCs? Back in the old days when we were doing straightforward homologous recombination to modify mouse ESCs, different labs got different results targeting the same genes. Is that happening in this case?
Jeanne
Doesn’t it depend a lot on the context? If it’s work where the accuracy doesn’t matter too much – in vitro work where good v bad cells can subsequently be sorted or in vivo work with organisms whose health isn’t a major concern (C elegans), it wouldn’t matter. If it were in human embryos intended to give rise to babies, it would matter enormously!
Hi Hank,
You are right. It will depend. In this poll I was just trying to get the overall feeling on how serious this is or not to people.
Paul