The coming of age of the LeuO regulator
E-mail email@example.com; Tel. (+52) 777 329 1645; Fax (+52) 777 313 8673.
LeuO is a quiescent genetic regulator present in many bacteria, which forms part of the H-NS regulon. LeuO in turn has been proposed to activate a subset of genes of the regulon by antagonizing H-NS. In the paper by Dillon et al., binding of LeuO to the S. Typhimurium genome was observed by ChIP-chip to some of the previously described LeuO-regulated genes, upon growth under stress conditions. However, studies at a higher LeuO concentration from a cloned inducible promoter rendered many more binding sites, pointing towards the importance of the abundance of the regulator in the cell, in a given moment. Binding of LeuO was observed not only to intergenic sequences, but in the majority of cases to intragenic sequences, and co-binding was observed with H-NS in many sites and with RNA polymerase to the majority of sites. The authors define a binding motif that allowed the detection of several other LeuO-regulated genes that were not detected by ChIP-chip, which were possibly missed because LeuO binds and bridges distal sites, in those instances. The observations reported open new questions regarding the mode of action for LeuO.
The LeuO regulator
The fascination with the bacterial LeuO regulator stems from the fact that it is quiescent, being expressed at very low levels when cells are grown under standard laboratory conditions. The leuO gene is repressed by the H-NS nucleoid protein and silencing is relieved by LeuO itself, as part of a promoter relay mechanism (Chen and Wu, 2005). Actually, H-NS silences the transcription of many genes apparently towards enhancing the competitive fitness of bacteria, which have evolved anti-H-NS activity represented at least in part by activator proteins (Dorman, 2007). One such activator protein is LeuO, which thus enables the expression of genes that are usually quiescent.
The LysR-type transcriptional regulators (LTTRs) represent the largest family of transcriptional factors in prokaryotes (Pareja et al., 2006). The structural organization of these proteins consists of an N-terminal DNA-binding helix–turn–helix motif and a C-terminal co-inducer-binding domain (Maddocks and Oyston, 2008). LeuO is a LTTR which was originally described as a member of the leucine operon in Salmonella Typhimurium (Hertzberg et al., 1980), after which it was recognized as a member of the LysR family (Henikoff et al., 1988). The first gene whose expression was affected upon LeuO overexpression was cadC (Shi and Bennett, 1995), and a negative regulatory effect was also reported in Escherichia coli for dsrA, which codes for a small regulatory RNA for the translation of the σS stress factor (Klauck et al., 1997; Repoila and Gottesman, 2001). LeuO was shown to relieve the Bgl− phenotype for the utilization of certain β-glucosides, ascribing it a role as a positive regulator (Ueguchi et al., 1998). LeuO was also reported as a positive regulator for the ompS2/ompN gene in S. Typhi (Fernández-Mora et al., 2004). In addition, LeuO and BglJ were proposed to counteract H-NS repression at the bgl operon (Madhusudan et al., 2005). At present, both positive and negative regulatory effects have been shown for LeuO at many genes involved in diverse biological processes, such as transport, regulation, metabolism, detoxification, virulence and stress conditions, thus suggesting that LeuO has a pivotal role in free-living bacteria and in the pathogenesis of various microorganisms, such as Vibrio cholerae, pathogenic Yersiniae, S. Typhi, S. Typhimurium and E. coli (Fang et al., 2000; Tenor et al., 2004; Moorthy and Watnick, 2005; Lawley et al., 2006; Rodríguez-Morales et al., 2006; Lawrenz and Miller, 2007; Hernández-Lucas et al., 2008; Shimada et al., 2009; 2011).
The binding of LeuO to the S. Typhimurium genome
In this issue, Dillon et al. (2012) report the first investigation of the binding in vivo of LeuO at the genomic level in S. Typhimurium by ChIP-chip. Thus, several relevant observations are reported. One is the interaction of LeuO with the genome under two conditions: one upon growth in a minimal low-phosphate medium (LPM), simulating the Salmonella-containing vacuole in the host cell, to stationary phase, where LeuO expression has been shown to increase (VanBogelen et al., 1996; Fang et al., 2000), and the other by overexpressing the LeuO protein from an arabinose-inducible pBAD promoter system. Interestingly, 178 target genes for binding to LeuO were found under LPM. These initial target genes included the ompS2 quiescent porin gene of the LeuO regulon in S. Typhi (Hernández-Lucas et al., 2008) and some of those described for E. coli (Shimada et al., 2011).
The contribution by Dillon et al. (2012) actually not only extends the members of the LeuO regulon, but now identifies genes previously shown to be involved in pathogenesis. Notable new LeuO targets in LPM include the Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR), and sopA of the Salmonella pathogenicity island I (SPI-1) and sifA of SPI-2. Also included are the gene coding for the RcsA regulator, of the RcsA/RcsB system, where RcsB has recently been shown to regulate the expression of the leuO gene in conjunction with BglJ in E. coli (Stratmann et al., 2012).
Quite remarkably, upon overexpression LeuO was found to bind now to 331 targets. These included the S. Typhi ompS1/ompS quiescent porin gene in accordance with a previous observation that lower concentrations of LeuO are needed for activation of ompS2 than for ompS1 (De la Cruz et al., 2007). Thus, the contribution by Dillon et al. (2012) constitutes an important illustration of the effect of the concentration of a regulator on the number of genes being regulated.
Another novel contribution of this work is the fact that only a third of the LeuO binding sites were to non-coding sequences, and the rest was to intragenic sequences: a finding of much consideration. Actually, this is in agreement with intragenic LeuO binding sites found in the E. coli genome, albeit only 20 genes were reported (Shimada et al., 2011). It could well be that, by doing so LeuO might exert a negative regulatory function impeding the progress of the transcriptional complex or else contribute to the nucleoid structure by bridging separate sites on the genome, or both. As transcriptional units have been located in intragenic regions in the E. coli genome (Mendoza-Vargas et al., 2009), and this could well be occurring in other bacteria, LeuO could possibly be exerting new positive or negative regulatory functions beyond our current knowledge.
Out of a total number of 140 LeuO binding sites detected in E. coli, 95% were found to co-bind H-NS (Shimada et al., 2011), supporting the role of LeuO as an antagonist of H-NS action as postulated by several authors (Chen and Wu, 2005; De la Cruz et al., 2007; Shimada et al., 2011). In this respect, Dillon et al. (2012) found that LeuO co-bound with H-NS at 38% of the sites, supporting the notion that LeuO acts by antagonizing or even in conjunction with H-NS to activate or repress genes.
Dillon et al. (2012) predict a 28-nucleotide LeuO DNA binding site, an AT-rich region containing the LTTR box motif (T-N11-A). Their bioinformatic prediction was validated by the experimental binding of LeuO to the envR and pipA genes, which were not detected by ChIP-chip. Furthermore, this consensus allowed the identification of additional LeuO-dependent genes that were not identified by ChIP-chip under growth in LPM, such as leuO (Chen and Wu, 2005), ompS1 (De la Cruz et al., 2007), assT (Gallego-Hernández et al., 2012), the CRISPR-associated casA and cas3 regions (Westra et al., 2010; Medina-Aparicio et al., 2011) and in the yjjQ-bglJ operon (Stratmann et al., 2008), although binding to ompS1 (as pointed above) and the cas3 region was observed upon LeuO overexpression. Moreover, putative LeuO binding sites within SPI-1 and SPI-2 were predicted, encompassing several of the regulatory genes. These observations have led the authors to propose that many actual binding sites were not detected in their study, given the distinct possibility that LeuO could be binding to a variety of distal sites.
New perspectives for research on LeuO
A particularly exciting idea is the proposed functional consortium between LeuO, H-NS and RNA polymerase with the ensuing novel mechanistic possibilities for gene regulation. This concept is based on the fact that there is co-binding of LeuO and H-NS at many sites and that in the vast majority of LeuO binding sites there was a concomitant binding by RNA polymerase. Thus, the elucidation of possible interactions of LeuO with H-NS and RNA polymerase should be pursued in future studies, as well as with other regulators, specially other LTTRs that might be regulated by LeuO. The determination of the functional oligomeric state and crystal structure for LeuO will also be crucial, among other avenues of research.
It is becoming clear that the levels of LeuO determine the gene content of the regulon and this concept should be extended to the study of other regulators. It is conceivable that bacteria might tightly control varying intracellular concentrations of LeuO in short-time pulses, to contend with particular environmental conditions, including those found in the host cells. Hence, knowledge of the environmental cues that regulate LeuO intracellular concentration constitutes a fundamental challenge in the field.
Research in our laboratory was supported by grants to E. C. from CONACyT, Mexico (82383) and DGAPA/UNAM (IN216310), and to I. H.-L. from CONACYT (89337) and DGAPA/UNAM (IN203312).