Reconstitution of acetosyringone-mediated Agrobacterium tumefaciens virulence gene expression in the heterologous host Escherichia coli.J Bacteriol. 2001 Jun; 183(12):3704-11.JB
The ability to utilize Escherichia coli as a heterologous system in which to study the regulation of Agrobacterium tumefaciens virulence genes and the mechanism of transfer DNA (T-DNA) transfer would provide an important tool to our understanding and manipulation of these processes. We have previously reported that the rpoA gene encoding the alpha subunit of RNA polymerase is required for the expression of lacZ gene under the control of virB promoter (virBp::lacZ) in E. coli containing a constitutively active virG gene [virG(Con)]. Here we show that an RpoA hybrid containing the N-terminal 247 residues from E. coli and the C-terminal 89 residues from A. tumefaciens was able to significantly express virBp::lacZ in E. coli in a VirG(Con)-dependent manner. Utilization of lac promoter-driven virA and virG in combination with the A. tumefaciens rpoA construct resulted in significant inducer-mediated expression of the virBp::lacZ fusion, and the level of virBp::lacZ expression was positively correlated to the copy number of the rpoA construct. This expression was dependent on VirA, VirG, temperature, and, to a lesser extent, pH, which is similar to what is observed in A. tumefaciens. Furthermore, the effect of sugars on vir gene expression was observed only in the presence of the chvE gene, suggesting that the glucose-binding protein of E. coli, a homologue of ChvE, does not interact with the VirA molecule. We also evaluated other phenolic compounds in induction assays and observed significant expression with syringealdehyde, a low level of expression with acetovanillone, and no expression with hydroxyacetophenone, similar to what occurs in A. tumefaciens strain A348 from which the virA clone was derived. These data support the notion that VirA directly senses the phenolic inducer. However, the overall level of expression of the vir genes in E. coli is less than what is observed in A. tumefaciens, suggesting that additional gene(s) from A. tumefaciens may be required for the full expression of virulence genes in E. coli.