Metabolism, stress response and gene expression are all controlled by regulatory networks in living systems. The factors involved are diverse and the interactions complex. Well-known regulators are proteins, which function as enzymes, chaperones or transcription factors to regulate numerous processes. Less well-known are RNA-molecules, which also regulate a multitude of processes: small RNAs or sRNAs. Dr. Jens Georg from the Department of Genetics and Experimental Bioinformatics at the University of Freiburg has made it his goal to identify these sRNAs and their interactions and physiological functions in bacteria. In order to do so, along with colleague he has developed a new computer-assisted tool, the online software CopraRNA, which can be used to make extensive predictions about the small RNAs.
“Interestingly, the sRNAs function almost like transcription factors, they have large networks of targets, which they regulate, similar to microRNA in eukaryotes,” says the biologist. The small RNAs represent a large heterogeneous class of bacterial regulatory factors, which bind to proteins and modify their function as well as interact with mRNA targets and regulate gene expression. They function post-transcriptionally, meaning the transcription is regulated by proteins and the small RNAs then prevent or activate translation. “Basically we can say that every mechanism in the cell is also regulated by small RNAs at some point,” Georg adds. Their functions are diverse. Some of them are formed as a result of stress situations such as oxidative stress, iron deficiency or light stress and are part of a complex regulatory network. Others are involved in the regulation of the outer membrane proteins, and in some bacteria virulence genes are regulated by sRNAs, such as in the process of toxin production. The small RNA called RyhB, for example, is only formed during iron deficiency and inhibits the translation of non-essential iron-containing proteins. Many of the sRNAs, their individual functions as well as the networks in which they are involved, have so far mainly been discovered through extensive experimental research in the lab. In the past few years the use of new sequencing techniques, which make it possible to identify all the RNAs in an organism at once, enabled the transcriptome, and hundreds, possibly more than a thousand different RNAs to be found. It is not known how many are yet to be discovered.
Together with the biologist Patrick Wright, Prof. Dr. Rolf Backofen from the Institute for Bioinformatics and Prof. Dr. Wolfgang Hess, Dr. Jens Georg has now developed a new computer programme CopraRNA (Comparative Prediction Algorithm for sRNA Targets), which can easily predict genetic sequences representing interaction sites for sRNA binding. This tool takes advantage of the complementarity of the nucleic acids in certain areas, in which complementary bases can pair up. The binding site of this RNA-RNA interaction can be predicted based solely on complementary sequences and is also referred to as target. The results are the genes that are targeted and regulated by the selected sRNA.
For more reliable results of sRNA targets, George and his colleagues use a comparative approach from the field of bioinformatics in their CopraRNA. They incorporate phylogenetic information as to how strongly RNA targets vary in related bacterial strains, in other words, how much they were conserved during evolution. The scientists postulate that if sRNA is conserved by inheritance then the same will happen with the target. The researchers therefore take advantage of this relationship. “So if a target has newly developed during evolution in just one bacteria, we will not be able to predict this with our programme,” admits Georg . Scientists who work with CopraRNA can access the already sequenced bacterial genomes in the database. From this wealth of data they can pull out the sequences of interest to them, use them for making a prediction and then obtain a result as to how probable the target is as a real domain for sRNA target recognition. The probability values of homologous genes in different bacterial strains are then converted by the programme to a significance value from zero to one. “The list is then sorted by the computer according to these values,” Georg explains, “and the top of the list is where the smallest numbers and most probable values for real targets can be found.” This then have to be verified in the lab, but only using the data from the top of the list.
According to Georg, “it is experimentally impossible to test all of these in the lab, therefore, we have developed the CopraRNA tool for all researchers interested in RNA-based regulation to enable them to pull out useful information from the plethora of transcriptome data.”This prediction software saves Georg and his colleagues a great deal of time and effort. In addition to the information as to which target gene is regulated by the selected sRNA, the computer also provides the exact location on the gene of the interaction sites as well as the physiological function of the sRNA in different parts of the metabolism during events such as iron deficiency or oxidative stress. This is important when trying to combat pathogenic agents in medicine or when using bacteria for biotechnological purposes. Jens Georg is convinced that “how a bacteria reacts to environmental conditions cannot be completely understood without the knowledge of RNA-based regulation.”
Further information:Dr. Jens GeorgPhone.: 0049 761 / 203 - 2708Email: jens.georg(at)biologie.uni-freiburg.de