Important context: I am not trying to express this piece in E.coli, in fact, it shouldn't be expressing in bacteria at all(apparently life uh... finds a way). I am just trying to clone it into a plasmid for another organism, which just requires the E.coli to not die before the miniprep....

I am having issues getting the C-terminal part of split-Cas9, rapamycin inducible cloned into several different plasmids. This region contains 3 out of the 4 nuclease domains and it seems that "sometimes" the domain is not toxic (I can transform the original Zhang vector no problem) but it is toxic in other vectors, presumably depending on the presence of cryptic bacterial promoters/expression which seems to be unavoidable in the plasmids that I have to use.

Here is the rundown of the issue I am experiencing and steps I have taken to try and solve it.

1. The C-terminal piece (it is a codon optimized IDT geneblock) is toxic in the TOPO zero blunt vector, as I am unable to get colonies back (I do get colonies with the N-terminal part geneblock). I have seen cryptic expression of GFP in TOPO before so this is not shocking if the protein is toxic. The C-terminal part from the original Zhang vector (PCR product) is also toxic in TOPO, so it does not seem to be a DNA sequence issue, but an amino acid sequence issue.

2. I was able to get the C-terminal piece and the N-terminal piece into one of my vectors in two sequential gibson assembly reactions (very exciting), but when I tried to add the selection cassette for my parasites it suddenly, magically, became toxic again and only colonies that had excised a chunk of the plasmid came back. The colonies, which were positive by colony PCR "died" after forming a colony and then a few individual bacteria from the colony recovered after streaking and seemingly removed a chunk of the plasmid to save themselves.

3. At this point I made an entirely new vector, finding new promoters/UTRs that would be suitable and making the entire vector as minimal as possible. This made other stuff I am doing easier, but did not help with this toxicity issue.

4. I tried mutating an important residue (H840A) in the HNH nuclease domain of this part and it seems to resolve the toxicity, but that is pretty useless to me.

5. I attempted to clone this piece into my plasmids using the "CopyCutter" E.coli which maintain low plasmid copy numbers until induction. This was unsuccessful, suggesting that even a few copies of the plasmid are enough to cause toxicity from the cryptic expression.

There is nothing I can find in the literature that suggests the C-terminal piece by itself should be cutting anything or be toxic and any way in E.coli since it cannot bind guide RNA at all.

There are other splitCas9 systems (light inducible etc.), but there is no reason to expect the C-terminal domain from those systems to be any less toxic since only the control system is different, so we are at a dead end as far as split-cas9 systems going, not because they don't work in our target parasite, but because we can't clone it into a plasmid! It is so absurd that I feel like I am losing my mind.

Does anyone have any magic ideas to solve this nonsensical problem? I am pretty desperate here.

Edit: I have considered first expressing anti-Cas9 protein in my E. coli, but the literature suggests it only binds to/inhibits Cas9 that has a guide RNA bound to it (not the apo form) so no reason to expect it to work here.