Hi,
I'd like to create a plasmid that a pathogen (like Klebsiella pneumoniae) can carry/replicate into the mammalian cells it infects. Then, that plasmid (assuming a small number of them survive being phagocytosed) has a mammalian promoter and downstream of it contains a fluorescent protein that localizes to the nucleus. As a result, a cell that contains KP would fluoresce with a nuclear fluorescent readout. Is this possible?
I've seen engineered KP variants have pET plasmid vectors but I'm afraid they have bacterial promoters. How can I have a plasmid vector that contains a mammalian promoter and fluorescent protein but that it'll replicate well in the KP?
I've had an experience where I electroporated some mammalian plasmids into KP but it affected their growth massively. Those plasmids contained a CMV promoter.
How would your plasmid get out of the Klebsiella? I think it will stay in the bacterium for as long as it is alive, and it will be destroyed when the Klebsiella is destroyed.
But OK we'll assume some of the bacterium makes it into the cytoplasm, and then dies there, and leaks out intact plasmid. It won't actively localize to the nucleus. The best you can hope for is that the cell wants to divide, disassembles its nuclear membrane, and then encloses the plasmid when it reassembles it. To get it to actively go to the nucleus, you would need to do something harebrained, like express a DNA-binding protein containing a mammalian nuclear localization signal that would drag the whole plasmid along. This is not well established. You could do a PhD thesis or 2 on that part alone.
So, I hate to rain on your party, but I think none of your scheme will work -- deal with it. But the part that you asked about is actually easy. You design a construct containing all the features you need for mammalian expression, like promoter, Kozak, gene of interest, stop, terminator. Clone it into the pET vector. The mammalian cell will ignore any bacterial expression features on the vector. It does not hurt it to have them there. And expression in the bacterium will not work, for lack of a bacterial ribosome binding site. (just don't put one there...)
These are great points you mentioned John Schloendorn . I guess, I was interested by this plasmid approach here: Article Use of a Dual Reporter Plasmid to Demonstrate Bactofection w...
Could you please provide any insight into why theirs may have worked?
I guess it comes down to defining what "it worked" means for you. Are you content finding "some" (how many?) cells, in which the plasmid leaked in just the right way to demonstrate some (how much?) protein expression? Or do you want to build a robust gene delivery platform that does something useful, and requires a certain efficiency?
As always, assume that the "it worked" paper optimized their conditions very carefully, and published only the very "best" result they ever obtained during months or years of trials. Could you live with seeing substantially fewer cells expressing less protein than published? If so, by all means go for it, quick and painless.
- Badges
- Science topic