I am trying to construct a Piggybac retrotransposon vector to stably express my target gene in a mammalian cell line. However, I have a big problem: only false positive colonies are obtained after transformation.

To construct the vector, the Piggybac vector and a plasmid containing the target gene were digested by EcoRI and NotI., and ligation was peformed using Mighty Mix (TaKaRa, 6023). Chemical competent DH5a cells was used for transformation, and cells were incubated at 37 °C on LB agar plate containing 20 ug/mL ampicillin. On the next day, a lot of colonies (more than 1,000) were obtained, and 20 colonies were picked up for subsequent mini culture in 3 mL LB containing 20 ug/mL ampicillin. However, proliferation of the cells was not observed at all on the next day.

I considered following possibilities:

1. Piggybac terminal repeats disturb duplication of the plasmid.

The Piggybac plasmid has repeat sequences, as retrovirus and lentivirus vectors have long terminal repeat sequences. When I handled retrovirus vectors previously, I also got a lot of false colonies, but I was able to overcome this problem by incubating DH5a at 30 °C instead of 37 °C. However, this didn’t work in the case of the Piggybac vector. I also tried Stbl2 cells, which are resistant to repeat sequence, but it didn’t work neither.

2. Ampicillin concentration is low to select positive colonies.

I have done cloning with 20 ug/mL ampicillin for years, but I have never had any problem before. As my colleagues pointed out the concentration is low, I tried to select colonies with 100 ug/mL ampicillin. Fewer colonies were obtained, but all colonies didn’t grow in 3 mL LB (100 ug/mL ampicillin).

3. Ligation is not successful.

I haven’t checked this possibility with my plasmid yet. My colleague inserted a target gene into the Piggybac vector using Gibson assembly, and he also got colonies. However, all colonies he got didn’t grow in LB medium. He also performed colony PCR and got no band. Oddly, he got bands when he performed PCR with Gibson assembly mixture containing fragments of the vector and target gene, and presumably his goal vector. (Primers were designed inside and outside the target gene to ensure amplification of the fragments does not occur.) Considering these results, it is unlikely the phenomenon we are currently facing might not be caused by unsuccessful ligation or Gibson assembly.

I predict that DH5a cells once obtain resistance to ampicillin by expressing ampicillin resistant proteins, but somehow lose the plasmid while growing. After repeated proliferation on LB plates, concentration of ampicillin resistant proteins reduces and they lose resistance to ampicillin, and thus no proliferation in LB medium.

Does anyone have an idea why this phenomenon occurs and how to overcome it?

Thank you for reading very long story. I appreciate your cooperation.