1. Top of page
  2. Summary
  3. References

Since their classification as members of the σ70 superfamily, Group IV alternative σ factors have been found to control gene expression in response to diverse environmental or stress signals. Activity of the Streptomyces coelicolor Group IV family member, σR (SigR), is increased by changes in the oxidation-reduction state of cytoplasmic disulphide bonds. Once released by its cognate anti-σ factor RsrA, σR activates expression of gene products that help cells reduce cytoplasmic disulphide bonds. In this issue of Molecular Microbiology, Kim and co-workers provide new insights into positive and negative control of σR activity. The authors show that a transcript derived from the inducible σR-dependent sigRrsrA p2 promoter operon encodes a σR protein of a higher molecular weight (termed σR′) than is found in uninduced cells. One major difference between σR′ and the smaller σR protein found in uninduced cells is the rapid proteolysis of σR′ by the ClpP1/P2 protease system. The genes for the ClpP1/ClpP2 protease subunits are themselves members of the σR regulon. The newly identified positive (σR′ synthesis) and negative control (selective σR′ turnover) aspects of this circuit are either found or predicted to exist in other related Group IV σ factor family members.

Bacterial RNA polymerase containing the primary σ factor, σ70, direct and initiate transcription of a large number of genes commonly expressed in exponential growth phase (Borukhova and Severinov, 2002; Young et al., 2002). Over the past several decades, it has been shown that alternative σ factors in the σ70 superfamily direct the respective pool of RNA polymerase molecules to transcribe suites of genes that allow adaptation to changing environmental conditions (Lonetto et al., 1992; 1994; Gruber and Gross, 2003; Paget and Helmann, 2003).

Given the adaptive role of these alternative σ factors, it is not surprising that their synthesis or activity is regulated in response to specific signals (Gruber and Gross, 2003). The mechanisms known to modulate the activity of different alternative σ factors include phosphorylation (Schurr et al., 1995), binding to another protein to tag the σ factor for degradation (Zhou et al., 2001), activation of an inactive precursor σ factor by removal of an N-terminal extension (Kroos et al., 1999), formation of a complex with a cognate anti-σ factor to prevent RNA polymerase binding or transcription (Helmann, 1999; Chadsey and Hughes, 2001; Li et al., 2002; Anthony et al., 2004), or use of multiple initiation codons to produce σ factors of different size or activity (Viollier et al., 2003, Dona et al., 2008).

The Group IV or extracytoplasmic function σ factors are a set of alternative σ factors that were first recognized as members of the σ70 superfamily in 1994 (Lonetto et al., 1994). At that time, the Group IV proteins were termed extracytoplasmic function (ECF) σ factors because the characterized proteins from Streptomyces coelicolor (Jones et al., 1997) and Escherichia coli (Mecsas et al., 1993; Rhodius et al., 2006) controlled expression of genes whose products functioned beyond the cytoplasm (Lonetto et al., 1992; 1994). A recent analysis of completed bacterial genomes predicts that Group IV σ factors account for ∼60% of the proteins (∼3600/6000) predicted to be members of the σ70 superfamily (Campbell et al., 2007). The picture emerging from analysis of Group IV σ factors is that they direct the transcription of genes whose products allow cells to respond to diverse extracellular, environmental or stress signals (Helmann, 2002; Gruber and Gross, 2003; Paget and Helmann, 2003; Francez-Charlot et al., 2009). Complete genome sequences suggest that this group of alternative σ factors can play a major role in gene regulation, as some cells have as many as 50 open reading frames annotated as members of the Group IV family (Bentley et al., 2002; Campbell et al., 2007).

There are typically positive and negative circuits that control activity of Group IV σ factors (Helmann, 1999; Francez-Charlot et al., 2009). In the absence of an inducing signal, activity of Group IV σ factors is often negatively regulated by a specific anti-σ factor that is often encoded by the gene immediately downstream of the σ factor structural gene (Helmann, 1999; Gruber and Gross, 2003; Campbell et al., 2007). Once the inducing signal is generated, the anti-σ factor releases the Group IV σ, allowing it to bind RNA polymerase and activate target gene expression (Helmann, 1999; Helmann, 2002). It is also common to find that the gene encoding the Group IV σ factor contains a promoter that is recognized by this σ factor, providing a positive autoregulatory loop that amplifies the effect of the signal on expression of target genes (Helmann, 1999; 2002; Gruber and Gross, 2003).

In this issue of Molecular Microbiology, Kim et al. (2009) provide exciting new data on positive and negative control of S. coelicolorσR and its homologues. Previously published analysis of S. coelicolorσR and its cognate anti-σ factor, RsrA, demonstrated that this Group IV σ factor activates gene expression in response to changes in the oxidation-reduction of cytoplasmic thiols (so-called disulphide stress) (Paget et al., 1998; 2001a; Park and Roe, 2008). RsrA is one of a large group of zinc-dependent Group IV anti-σ factors (Paget et al., 2001a; Bae et al., 2004; Zdanowski et al., 2006; Campbell et al., 2007) that, in its reduced state, binds σR, preventing expression of target genes (Fig. 1). Upon a change in the oxidation-reduction state of cytoplasmic thiols, RsrA is inactivated by formation of an intramolecular disulphide bond, releasing zinc, causing dissociation of the σR–RsrA complex, and allowing this Group IV family member to activate transcription of its target genes (Fig. 1). The ∼30 known σR target genes include the sigRrsrA operon (encoding the σR–RsrA proteins), those encoding thioredoxins (to reduce cytoplasmic disulphide bonds in proteins), enzymes in mycothiol biosynthesis (the actinobacterial equivalent of the more common low-molecular-weight thiol, glutathione), proteases, and other gene products that help cells mount a disulphide stress response (Paget et al., 2001b). In many regards, the biochemical, molecular and physiological aspects of this system have provided new paradigms for the process and control of Group IV σ factor function (Paget and Buttner, 2003; Zdanowski et al., 2006; Park and Roe, 2008) as well as the role of zinc as a regulator of gene expression (Ilbert et al., 2006).


Figure 1. Model for positive and negative control of σR activity (adapted from Paget and Buttner, 2003; Ilbert et al., 2006; Kim et al., 2009). The thiol-disulphide status of S. coelicolor is controlled by the Group IV σ factor, σR, and RsrA, its cognate, zinc-containing anti-σ factor. Under reducing conditions, the sigRrsrA p1 promoter is active; RsrA binds σR and prevents it from binding RNA polymerase, thereby inhibiting it from activating target gene transcription. Conditions of thiol stress induce formation of an intramolecular disulphide bond in RsrA, which causes it to release zinc and σR, activating transcription of thioredoxin structural genes (trx), genes encoding the clpP1/clpP2 protease, an upstream sigRrsrA p2 promoter, plus others. The transcript derived from the upstream inducible sigRrsrA operon p2 promoter encodes a higher-molecular-weight σR protein (σR) that is capable of activating target gene expression, thus providing positive control to this regulatory loop. However, the N-terminus of the induced σR protein makes it a preferred substrate for proteolysis by the ClpP1/ClpP2 protease. Since the clpP1/clpP2 protease genes are also members of the σR regulon, selective proteolysis of the induced σR provides a negative feedback loop to this control circuit that is proposed to help restore target gene expression to basal uninduced levels when thiol stress is relieved.

Download figure to PowerPoint

The new results provided by Kim et al. (2009) show that the positive control involves the synthesis of a heretofore unrecognized higher-molecular-weight form of σR (31 kDa, referred to as σR) compared with the 25 kDa protein found in uninduced cells. They further show that this induced σR protein is encoded by a sigRrsrA transcript derived from an upstream p2 promoter, which is transcribed by σR-containing RNA polymerase. The induced σR protein is able to bind RNA polymerase and activate transcription (Kim et al., 2009), so use of the sigRrsrA p2 promoter to direct synthesis of the higher-molecular-weight σR factor is a newly discovered part of the positive autoregulatory circuit for a Group IV σ factor (Fig. 1).

The use of multiple translation initiation codons for a gene or σ factor is not a novel feature (Viollier et al., 2003; Dona et al., 2008). While the induced σR protein activates transcription when it is added to core RNA polymerase, Kim et al. (2009) report that this higher-molecular-weight protein differs functionally from the lower-molecular-weight counterpart that is present in uninduced cells. Specifically, they show that the addition of these N-terminal 55 amino acids renders σR more susceptible to proteolysis in vivo. In addition, their experiments identify the ClpP1/P2 proteases as the enzymes responsible for the rapid intracellular turnover of the induced σR protein (Fig. 1).

These findings are significant for several reasons. First, the short t1/2 of the induced σR protein adds a negative feedback loop into the control circuit, providing cells with a mechanism to shut off the stress response when or as cytoplasmic disulphide bonds are reduced. In addition, the structural genes for the ClpP1/P2 proteases are members of the σR regulon. Thus, it appears that the ClpP1/P2-dependent proteolysis of σR is part of homeostatic pathway that controls both the induced levels of target gene expression and allows rapid adaptation to the non-induced state once the stress response is no longer needed (Fig. 1).

Of equal importance, Kim et al. (2009) provide experimental support for the existence of similar positive and negative control loops in mycobacteria that contain orthologues of S. coelicolorσR. In addition, computational analysis of additional genome sequences of Mycobacterium and Streptomyces species that contain σR orthologues predicts that these bacteria could also synthesize higher-molecular-weight isoforms of these proteins. Thus, it seems very likely that other related bacteria use a similar strategy to control the activity of this family of Group IV σ factors. Time and additional experiments will tell how widely this system is distributed throughout bacteria. However, the discovery of these additional features provide an opportunity to gain experimental, mathematical and mechanistic insight into how these newly identified positive and negative circuits influence increases or decreases in target gene expression as cells encounter and recover from this or other stress signals.

Given the number of Group IV σ factors predicted to exist and the wealth of diverse environmental or stress signals encountered by bacteria in nature, it is not surprising that the analysis of these alternative σ factors have provided new insights into systems used for transcriptional control of gene expression. The results of Kim and co-workers illustrate that, even in well-studied systems, there is still much to be learned about the logic and operation of bacterial control circuits. The ever-increasing number of bacterial genome sequences and the ability to study these organisms by traditional and high-throughput approaches provide microbiology with an exciting opportunity to identify additional features of positive and negative control circuits that govern transcriptional output.


  1. Top of page
  2. Summary
  3. References