Elsevier

Methods

Volume 43, Issue 2, October 2007, Pages 131-139
Methods

Computational approaches for the discovery of bacterial small RNAs

https://doi.org/10.1016/j.ymeth.2007.04.001Get rights and content

Abstract

Recent work has uncovered a growing number of bacterial small RNAs (sRNAs), some of which have been shown to regulate critical cellular processes. Computational approaches, in combination with experiments, have played an important role in the discovery of these sRNAs. In this article, we first give an overview of different computational approaches for genome-wide prediction of sRNAs. These approaches have led to the discovery of several novel sRNAs, however the regulatory roles are not yet known for a majority of these sRNAs. By contrast, several recent studies have highlighted the inverse problem where the functional role of the sRNA is already known and the challenge is to identify its genomic location. The focus of this article is on computational tools and strategies for identifying these specific sRNAs which function as key components of known regulatory pathways.

Introduction

Noncoding RNAs perform a variety of critical regulatory functions in both prokaryotes and eukaryotes [1]. In bacteria, these RNAs correspond to small genes (typically ∼50–500 nt in length) that are transcribed but not translated and are generally referred to as small RNAs (sRNAs). Recent work, often combining experimental and computational approaches, has led to a dramatic increase in the discovery of bacterial sRNAs. In Escherichia coli alone, more than 60 sRNAs have been identified experimentally (surveyed in [2]) and several more have been predicted by computational searches. While the functions are not known for a majority of these sRNAs, an emerging trend is that they play crucial regulatory roles in bacterial adaptation to changing environments [3], [4]. Several recent reviews document the regulatory roles of sRNAs in bacteria [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15] and one of the major challenges for future work is both the discovery of novel sRNAs and the elucidation of their regulatory functions.

Bacterial sRNAs discovered so far can be broadly categorized into two major classes based on their mode of action. The majority of known sRNAs participate in post-transcriptional regulation by base-pairing with the target mRNA, which results in translation inhibition and/or degradation of the mRNA–sRNA complex [16], [17]. In many cases, this interaction is mediated by the RNA chaperone protein Hfq, e.g., DsrA regulates rpoS translation in a Hfq-dependent manner [18], [19]. The other class consists of sRNAs that act by interacting with RNA-binding proteins to modify the activity of the protein. For example, the CsrB and CsrC sRNAs in E. coli regulate levels of CsrA by binding to multiple copies of the protein [20], [21], [22]. Another example, illustrating the diversity of sRNA–target interactions, is the recent demonstration that the sRNA Rcd binds to the enzyme tryptophanase thereby increasing its enzymatic activity resulting in increased production of indole which in turn leads to cell division arrest [23], [24]. These examples highlight the varied roles of sRNAs in critical cellular processes and underscore the need for identification of sRNAs functioning in known pathways to better understand their roles in cellular regulation.

A timeline of sRNA discovery in E. coli reveals two distinct phases. Prior to 2001, about a dozen sRNAs were identified in E. coli mostly by accident, in many cases by genetic studies focusing on the phenotypes conferred by multicopy plasmids or through analysis of the regulation of divergently transcribed proteins [7], [25]. An analysis of the common characteristics of these sRNAs served as a guide for systematic, genome-wide searches for sRNAs initiated in 2001 by different groups [26], [27], [28], [29]. Since then, more than 50 sRNAs have been found experimentally in E. coli alone [2], [30]. Building on insights from the initial discoveries, several computational tools were developed for genome-wide prediction of sRNAs leading to many more predictions of candidate sRNAs. These approaches have also led to the discovery of sRNAs in other bacterial genomes [31], [32] indicating that systematic searches of bacterial genomes can uncover many more sRNAs.

While the above global searches have led to the discovery of novel sRNAs, deciphering the functional roles of these sRNAs remains a challenge. On the other hand, recent experiments have highlighted studies where the situation is reversed: the functional role of the sRNA is known and the challenge is to identify its genomic location. This typically occurs when experimental evidence indicates that a ‘missing link’ in the regulatory pathway under study corresponds to a sRNA. Examples of such studies include the discovery of sRNAs involved in iron homeostasis [33] and the discovery of multiple sRNAs in the quorum-sensing pathways in Vibrio cholerae and Vibrio harveyi [34]. In such situations, computational strategies which combine criteria used in whole-genome searches (corresponding to common features of known sRNAs) with specific search criteria based on available information about the sRNA can contribute to identifying sRNAs. The focus of this article will be on providing an overview of the different computational approaches and tools involved in these specific searches.

The rest of this paper is organized as follows. In Section 2, we provide a brief overview of different strategies for sRNA discovery which have been discussed in detail in several reviews recently [25], [35], [36]. Section 3 focuses on strategies for targeted searches which use available experimental information to identify specific sRNAs. In Section 4, we illustrate the application of these methods to a problem where the sRNA has not yet been discovered experimentally and conclude with a brief summary in Section 5.

Section snippets

Genome-wide methods for sRNA discovery

As mentioned in the previous section, an initial set of ∼12 sRNAs were discovered in E. coli mostly by accident as a byproduct of studies not specifically designed to search for sRNAs [7]. However, the discovery of these sRNAs outlined their common characteristics which formed the basis of systematic searches for sRNAs by several groups in 2001 [26], [27], [28], [29]. Three of these studies combined computational screens with experimental validation of selected candidates leading to the

Specific sRNA searches

The approaches outlined in the previous section were, for most part, global genome-wide approaches to sRNA prediction and verification. In several cases, however, the challenge is to identify specific sRNAs functioning in known pathways based on available experimental evidence. Typically such a situation arises when experiments indicate post-transcriptional regulation of a target gene that requires the RNA chaperone protein Hfq. In such cases, previous experience suggests that identifying

Application to a specific sRNA search

So far, the search strategies outlined have been illustrated by examples for which the sRNAs involved have been experimentally verified (in some cases based on the bioinformatic predictions). It would be of interest to apply these approaches to a case where the sRNA involved has not yet been identified experimentally as detailed below.

Recent experiments by Ruiz and Silhavy provide evidence for the existence of another, as yet unidentified, sRNA involved in the regulation of rpoS in E. coli [61]

Summary

In summary, sRNA involvement in bacterial adaptation to changing environments is becoming an increasingly recurring theme in recent research in microbiology. It is likely that future research will uncover many more examples indicating involvement of sRNAs as important components of regulatory pathways. In this article, we have highlighted bioinformatic techniques for discovery of bacterial sRNAs and presented a general framework for locating specific sRNAs based on experimental inputs. It is

Acknowledgments

We thank Mark Mandel for several helpful discussions and Natividad Ruiz and Thomas Silhavy for sharing their data with us. We acknowledge funding support from the Jeffress Memorial Trust and the ASPIRES Award from Virginia Tech.

References (65)

  • J. Vogel et al.

    Curr. Opin. Microbiol.

    (2006)
  • E. Masse et al.

    Curr. Opin. Microbiol.

    (2003)
  • K.M. Wassarman

    Cell

    (2002)
  • S. Gottesman

    Trends Genet.

    (2005)
  • A. Huttenhofer et al.

    Trends Genet.

    (2005)
  • G. Storz et al.

    Curr. Opin. Microbiol.

    (2004)
  • L. Argaman et al.

    Curr. Biol.

    (2001)
  • E. Rivas et al.

    Curr. Biol.

    (2001)
  • D.H. Lenz et al.

    Cell

    (2004)
  • A. Huttenhofer et al.

    Curr. Opin. Chem. Biol.

    (2002)
  • S. Chen et al.

    Biosystems

    (2002)
  • S. Washietl et al.

    J. Mol. Biol.

    (2004)
  • N. Yachie et al.

    Gene

    (2006)
  • K. Robison et al.

    J. Mol. Biol.

    (1998)
  • M. Dsouza et al.

    Trends Genet.

    (1997)
  • C. Valverde et al.

    J. Biol. Chem.

    (2004)
  • A.G. Blanco et al.

    Structure

    (2002)
  • S.R. Eddy

    Nat. Rev. Genet.

    (2001)
  • R. Hershberg et al.

    Nucleic Acids Res.

    (2003)
  • F. Repoila et al.

    Mol. Microbiol.

    (2003)
  • T. Geissmann et al.

    Handb. Exp. Pharmacol.

    (2006)
  • N. Majdalani et al.

    Crit. Rev. Biochem. Mol. Biol.

    (2005)
  • S. Gottesman

    Annu. Rev. Microbiol.

    (2004)
  • K.M. Wassarman et al.

    Trends Microbiol.

    (1999)
  • G. Storz et al.

    Annu. Rev. Biochem.

    (2005)
  • S. Gottesman et al.

    Cold Spring Harb. Symp. Quant. Biol.

    (2001)
  • M. Guillier et al.

    Genes Dev.

    (2006)
  • E. Massé et al.

    Genes Dev.

    (2003)
  • D.D. Sledjeski et al.

    EMBO J.

    (1996)
  • D.D. Sledjeski et al.

    J. Bacteriol.

    (2001)
  • M.Y. Liu, G. Gui, B. Wei, III, J.F.P., L. Oakford, Ü. Yüksel, D.P. Giedroc, T. Romeo, J. Biol. Chem. 272 (1997)...
  • T. Weilbacher et al.

    Mol. Microbiol.

    (2003)
  • Cited by (0)

    View full text