I was wondering if anyone else has used the Dickinson method to N-terminally tag a lethal gene? In the paper (before heat shock), it says to balance the non-red rolling worms, and the resulting progeny should be: dead (SEC/SEC), a het (SEC/balancer), and balancer/balancer.
I balanced my worms and I am now getting non-rolling, non-balanced worms (along with rolling balanced worms, balanced non-rolling worms and non-balanced rolling worms). Does the presence of non-rolling, non-balanced worms mean that my CRISPR did not work and I just got an EX array?
We tried this for 2 different (lethal) genes, and we get the same results with both.
It shouldn’t be an extrachromosomal array - that was why there was an rfp co-injection marker, and why you had to pick non-red rollers.
It does sound like you’re not getting a lethal phenotype, though. Can you get apparent roller homozygotes, from the non-balancer rolling worms? What does PCR suggest is going on at the targeted locus? On the other hand, can you get non-balancer non-rolling worms, that would suggest the sqt-1 transgene is integrated elsewhere in the genome, or maybe is extrachromosomal despite what I first said?
I’ve used the Dickinson method to N-terminally tag a lethal gene 
Three questions: What did the worms look like before you balanced them? Were you able to get plates with 100% Rollers? If so, then your insertion is not lethal. It’s hard to know what you’d predict without knowing what gene you were targeting, but multiple and/or cryptic promoters can sometimes lead to failure to observe an expected lethal phenotype for an N-terminal insertion. If you were able to see dead eggs or arrested larvae, that would certainly increase the confidence that your insertion is indeed lethal.
Second, can you see fluorescence (from GFP) in your Rollers? And if so, are your non-balanced non-rolling worms also fluorescent? I’m asking because we do sometimes see spontaneous excision of the marker without heat shock, which could give you back non-rolling (but viable) worms. If you’re lucky you might be seeing the same.
Third, what do the ratios of balanced Roller vs. non-balanced Roller vs. Balanced non-roller look like? Is it at least roughly consistent with Mendelian segregation? If not, and if you’re seeing a large majority of non-rolling worms, you probably have a false positive. As Hillel points out, you should have used Christian Frokjaer-Jensen’s mix of array markers which are pretty effective at eliminating arrays. But if you didn’t use all three markers at their recommended concentrations, or you didn’t look carefully, or you just got really, really unlucky, it’s possible that some arrays could still sneak through.
Definitely something weird is happening, because if LOF of your gene is strictly (zygotic) lethal, you also should not see non-balanced Rollers (Roller homozygotes should be dead). So either it’s not really lethal, or you don’t have a knock-in, or you’re getting spontaneous excision which restores the wild-type phenotype. I suppose it’s also possible that your balancer is not a balancer, but I’m assuming you’ve chosen an appropriate balancer and so can rule out that possibility.
We have had a very similar situation when N-terminally tagging an embryonic lethal with GFP using Dan’s vectors.
With some careful PCR, we determined that, as Dan suggested, it was a rare occurrence of the SEC spontaneously excising itself even at 20C.
Yeah, we tagged one essential gene where we got multiple spontaneous excisions of the SEC. None with many other tags. So, context-specific.
We always have the primers at the ready to detect any insertion, whether in an essential or inessential gene.