Transcriptional analysis was used to examine the effect of a genomically encoded artificial RNA on Escherichia coli in rich and minimal media. Only the expression of a single gene, deoC, was unequivocally affected under both conditions. E. coli marker strains of this type may be useful in monitoring the fate and transport of bacteria in various applications.A previously constructed Escherichia coli tracking strain known as PCPHR expresses a genomic gene encoding an artificial RNA (aRNA) which contains a unique 17-nucleotide identifier sequence (2). The aRNA gene was originally derived from Vibrio proteolyticus 5S rRNA by deleting a key functional region and replacing it with the identifier sequence (21). The aRNA product is expressed at high levels from an rRNA promoter and accumulates in the cell without an obvious effect on growth rate (2). Although it resembles 5S rRNA, the aRNA is inactive and not integrated into the E. coli ribosomes (2,11,14,21). The high accumulation level of the aRNA allows its presence to be readily detected by many routine molecular methods, including RNA size, hybridization, fluorescence, and immunohybridization (15,16,22). Large numbers of alternative but equally unique identifier sequences could be introduced into the aRNA (12), thereby making it possible to simultaneously differentiate between multiple sources of E. coli. These aRNA strains are, however, genetically modified organisms whose use in field applications, such as source tracking or bioremediation, would require regulatory approval.The issue, which was addressed here, is whether or not the presence of the aRNA tracking gene will significantly modify the behavior of the cells carrying it. E. coli has eight genes for 5S rRNA that are found in seven operons. The synthesis of the rRNAs is strongly linked with cell growth and subject to sophisticated regulation (9). For example, it has been shown that if the levels of 16S and 23S rRNA are artificially reduced by deletions in one or more rRNA operons, the cell will upregulate its remaining rRNA operons to compensate for the loss (9). In contrast, the successive deletion of 5S rRNA genes revealed that there is apparently no ability to directly compensate for underexpression of 5S rRNA (3). These results suggest that the addition of an artificial RNA to the genome would not have a significant impact on other cellular processes. Nevertheless, the safety and utility of tracking strains expressing aRNAs that resemble 5S rRNA will depend on whether their presence affects other components of the cellular machinery in a significant or unexpected way.In order to address this issue, arrays of PCR products targeting each known open reading frame in the E. coli genome were used to determine what transcriptional changes are induced in the PCPHR strain compared to the wild-type (WT) E. coli EMG2 strain. Such arrays have been widely employed in the study of E. coli gene expression with considerable success (5-7, 17, 23-25). These experiments have generated extensive lists of genes that respo...