Hello Dominique,

the effect you see could be due to some contamination of your culture with another yeast strain.

As you say, GAL1 promoter leakage can also be responsible. I would suggest to check for the presence of the plasmid on glucose by PCR (WB detection might be not sensitive enough).

Good luck!

To add to that, depending on if you're working with a few more people in the lab, it's possible that sometimes galactose simply gets into the DEX media somehow, or worst case into the Dextrose powder by contamination. Just a small amount of GAL in the mix can result in expression..

Your background strain may still have your essential gene. What resistance ensures the KO in the chromosome ? Conversly, if the KO is not made properly, you may have had homologous recombination which restored the initial gene in the chromosome.  This becomes more likely has you restreak a lot the colonies.

Thanks for your responses. For Artemis the deletion has been performed by homologous recombination replacing the entire ORF by TRP1 cassette which prevent any recombination between the plamidic copy and any homologous region in the genome. For Christian T, a slight contamination of the glucose stock by galactose wouldn't induce anything because it wouldn't relieve any of the repression exerced by glucose on GAL promoter... And for Christian QS, yes the plasmid is present in cell growing on glucose... That's why I am more and more thinking of the leaky expression effect driven from a high copy number plasmid...

The construct with the GAL1 promoter do you know the sequence? The reason that I ask is because it may be missing the sequence for Mig1 binding.  Also, did you check for the KO by using PCR to check for both the KO fragment and WT. Sometimes there can be chromosomal duplications.

I would make sure, that you don't have a background mutation that suppresses the deletion of your essential gene. The acquisition of suppressor mutations is a common problem for working with essential genes and you won't get rid of it through re-streaking. 

A good sign for this would be that you get very few colonies growing on 5-FOA. Those would be cells that spontaneously have lost the plasmid (or acquired a mutation in URA3, but it's more likely to be plasmid loss), and they would only grow if you have a suppressor mutation. 

Thanks for more comments!

Christoper, I don't have anything growing on FOA which means the presence of the plasmid is essential (ie. there is no suppressor mutation).

Evelyn, the GAL1 promoter comes from a commercial plasmid which I suppose correct and yes I've made the PCR control you mentioned.

Hi Dominique,

GAL promoters are in fact a little leaky, but not enough for detection by WB. On top of this 2µ plasmids can wary in copy number (from ca.5 to more than 200). The more pressure for survival the higher the copy number. So if you only need small amounts of your essential protein for survival this might explain your phenotype. What you can try is using an ARS/CEN based plasmid, which is always low in copy number (1-3 the most), and repeat your experiment.

Thanks Curt, that's what I was thinking of.

You can also try to add more Myg tags (3-5) in total this might make it possible to detect the minute ammounts of protein in your samples. Often people use multible tags for localisation studies, to enhance the signal.