Plant-specific promoter sequences carry elements that are recognised by the eubacterial transcription machinery.

During evolution the promoter elements from prokaryotes and eukaryotes have developed differently with regard to their sequence and structure, implying that in general a transfer of eukaryotic promoter sequences into prokaryotes will not cause an efficient gene expression. However, there have been reports on the functionality of the 35S promoter from cauliflower mosaic virus (CaMV) in bacteria. We therefore decided to experimentally investigate the capability of plant promoter sequences to direct gene expression in various bacteria. Accordingly, we tested ten different plant-specific promoters from Solanum tuberosum, Nicotiana tabacum, CaMV, Agrobacterium tumefaciens, and A. rhizogenes for their ability to initiate transcription in five different eubacterial species (Escherichia coli, Yersinia enterocolitica, A. tumefaciens, Pseudomonas putida, and Acinetobacter sp. BD413). To monitor the strength of the plant-specific promoters in bacteria we created fusions between these promoters and the coding region of the luciferase genes from Vibrio harveyi and measured the luminescence in the bacteria. Heterologous gene expression was observed in 50% of the combinations analysed. We then mapped the transcription start site caused by one of the plant-specific promoters, the ST-LS1 promoter from S. tuberosum, in these bacterial species. The location of the mapped transcription start site indicated that the sequences of the plant promoter themselves were recognised by the bacterial transcription apparatus. The recognition of plant-specific promoter sequences by the bacterial RNA polymerase was further confirmed by site-directed mutagenesis of the ST-LS1 promoter and the analysis of the effects of the mutations on the strength of gene expression in E. coli. Using these mutants in our reporter assays we could localise the sequences of the ST-LS1 promoter serving as -10 region in E. coli. The results of our study show that promoter sequences are much less specific than is generally assumed. This is of great importance for our knowledge about the evolution of gene expression systems and for the construction of optimised expression vectors.