I’ve been intending to blog on this for a while now and finally got around to it. This paper comes from a group at the Jackson labs; the first author is Wenning Qin, and the senior author is Haoyi Yang who also holds a primary appointment at an institute in Beijing.
Qin et al. Efficient CRISPR/Cas9-mediated genome editing in mice by zygote electroporation of nuclease. Genetics, 2015 v.200 (2) pp. 423-430.
In this paper they show that CRISPR-Cas9 reagents can be electroporated directly into mouse zygotes to generate gene-edited animals. It’s not quite as efficient as direct injection into zygotes – but it’s not too shabby, considering the ease of the electroporation step.
To remind those who are unfamiliar, the main method to deliver DNAs or RNAs into mouse zygotes is through direct pronuclear or cytoplasmic injection, using ultrafine capillary needles. It requires micromanipulators for the needles, carefully controlled pressure to deliver the injected material, and a quality inverted microscope with high-contrast optics. Plus a steady hand and skill at injections. In mammals, zygotic DNA transgenesis has generally required direct injection of DNA into the pronuclei (unless a retrovirus is used, which has its own drawbacks). This was shown in papers from the early days of transgenic mouse research.
Historically, electroporation has not been used for engineering mouse zygotes. There are some good reasons for this. First, zygotes have a zona pellucida surrounding the zygote itself. This can be dissolved away rather easily with brief acid treatment – but without the zona, the embryos are sticky and much more difficult to handle. Second, electroporation doesn't immediately transfer material into the zygote nucleus, and the chance of DNA integrating into the genome is very low. In fact, even direct injection of DNA into the zygote cytoplasm does not yield transgenic mice efficiently – you’ve got to inject it into the pronuclei.
Now, there are research applications for zygote injection apart than transgenesis. You might just want to transiently express an mRNA in a zygote, for example. Embryologists who work with zebrafish and xenopus will be totally familiar with this idea. It’s not done frequently in mice but can be done. I think the labs that ever do this with mouse embryos are hardcore enough they have access to microinjection equipment, and presumably haven’t bothered to try electroporation much - why would you, if you are all set up to perform the established method.
However…what if you either (1) want to try transgenic manipulation, but don’t have access to a microinjection apparatus, or (2) you just want to really streamline the labor involved? Then electroporation might be useful… Enter CRISPR, in which we actually do want to transiently express the reagents in zygotes. At Jax they fall into the (2) category.
To develop this method, Qin et al. first confirmed previous reports that brief incubation in acid can be carried out to weaken the zona pellucida without completely dissolving it, while not affecting embryo viability. Next, they tested electroporation parameters to optimize both the media/TE mixtures compatible with embryo survival and the maximum voltages the embryos could tolerate and still live. Finally, they mixed acid-treated embryos with Cas9 mRNA plus guide RNAs for known pre-validated targets in the Tet1 or Tet2 genes and did electroporations. Surviving embryos were either genotyped after in vitro culture, or transferred into recipient females and analyzed after birth.
Bottom line: they could generate mutant animals at double digit percentages. Not surprisingly, efficiency increased with higher RNA concentrations. The final standard conditions involve at least 30 to 50 embryos per electroporation, in a total volume of 20 µl buffer/media with final concentrations of 600 ng/µl Cas9 mRNA and 300 ng/µl guide RNA. Note that this requires a total of 12 µg Cas9 mRNA and 6 µg gRNA per batch of 50 embryos. They also showed HDR is possible by co-electroporation with donor oligo DNAs.
So what is the efficiency? Table 2 is a very nice comparison of microinjection vs. electroporation across ten genes. Kudos to them for a nice big data set! Good news: electroporation generated mutants for 5/10 genes tested. However, 8/10 microinjections were successful for the same genes/reagents. The overall rate of mutants was lower in the electroporation set as well; the average efficiency in the 5 successful electroporations was 20%, while with microinjections it was 42%. Interestingly, the overall rate of embryo survival to birth was higher for electroporations – about double that of microinjections. So in the end, the lower efficiency of mutant generation via electroporation might be pretty well balanced by the higher birth rate per embryo.
I’ll still note that zygote manipulation may not be for the faint of heart even if you don’t have a microscope/microinjection setup handy. Remember that to make engineered live mice you have to transfer the manipulated zygotes back into a pseudopregnant recipient female mouse. This requires microsurgery, people.
However – maybe even this last step can be improved upon. Takahashi et al have now reported that embryos can be electroporated – and CRISPR-mutagenized – without taking them out of the oviduct. (Takahashi et al, Scientific Reports, June 22 2015 (5:11406).) This still requires injection of the CRISPR RNA solution into the oviduct of a live mouse, but it’s probably easier than transferring embryos back in to the oviduct. Hmm… maybe one doesn't need to pre-weaken the zone pellucida? Their method requires rather high RNA concentrations, but bears promise.