NEW YORK (GenomeWeb News) – The food poisoning bacteria Listeria monocytogenes exhibits a wide variety of transcriptional patterns under different growth conditions, according to a paper appearing online yesterday in Nature.
A team of French and Swedish researchers used custom tiling and gene expression arrays to compare the transcriptional profiles pathogenic and non-pathogenic stages of L. monocytogenes. In so doing, they identified stage-specific changes in coding and non-coding transcript expression, created an operon map for the bug and identified new regulatory RNA patterns.
"We could really see that there was some kind of reshaping of transcription [in one compartment compared to another]," senior author Pascale Cossart, a researcher at the Pasteur Institute, told GenomeWeb Daily News.
In 2001, Cossart and her colleagues sequenced the genomes of both L. monocytogenes and a non-pathogenic Listeria species called L. innocua. For the latest paper, they expanded on this sequencing work, conducting an unbiased transcriptomic study that "depicted the whole genome transcriptional landscape of L. monocytogenes in diverse growth conditions in vivo."
To do this, they extracted RNA from bacteria grown at a stationary phase of 37C or under low oxygen or temperature conditions, using custom Affymetrix tiling and gene expression arrays to assess Listeria transcripts under each condition.
They also did similar experiments on Listeria RNA from bacteria grown in human blood or isolated from the intestine of germ-free mice fed L. monocytogenes a day earlier. The intestinal infection in mice is very similar to the way the bacteria infect humans, Cossart noted, though in humans the bug is likely influenced by natural flora that are missing from germ-free mice.
Using this data, the researchers were able to come up with a fine-resolution operon map for Listeria, containing 517 polycistronic operons coding for 1,719 genes and delineating the 5' and 3' edges of transcripts under each condition.
Their results suggest that most of the L. monocytogenes genome is expressed. After comparing genome-wide transcription patterns, the team found that the bug expresses more than 98 percent of its open reading frames under all of the conditions tested.
Hundreds of genes were up- or down-regulated depending on the temperature or growth conditions evaluated. But the researchers' main focus was on non-coding RNAs, Cossart said. They found 21 previously identified small RNAs and 29 new sRNAs. Of these, just a third of the newly identified sRNAs were also detected in L. innocua.
While some of the potential sRNA transcripts seem to contain previously unrecognized small open reading frames, many appear to be authentic sRNAs targeting other sRNAs or genes involved in processes such as metabolism.
The expression of these sRNAs shifted in L. monocytogenes cells grown in the mouse intestine compared with those in human blood. For example, the expression of sRNA called rli38 was ramped up about 25 times when the bug was grown in blood.
The team's analysis of non-coding sequences revealed other surprises as well, including long overlapping 5' and 3' untranslated regions, RNA expression from the strand opposite coding sequence, and riboswitches acting as terminators for upstream genes.
"[O]ur study provides evidence that bacterial transcription is more complex than previously anticipated and urges investigations at a single bacterium level," the authors concluded. "Such studies will no doubt help in unraveling unclear phenotypes encountered when performing deletion analysis or insertion mutagenesis."
In the future, Cossart said the team plans to look at how other microbes influence Listeria transcription. They are currently doing experiments adding one bacterial species at a time to Listeria. They are also keen to evaluate Listeria transcription using RNA-Seq to see how data from that method compare to results obtained using arrays. Data from the current paper are available online at ArrayExpress.