Regulation of gene expression: cryptic β-glucoside (bgl) operon of Escherichia coli as a paradigm.

Bacteria have evolved various mechanisms to extract utilizable substrates from available resources and consequently acquire fitness advantage over competitors. One of the strategies is the exploitation of cryptic cellular functions encoded by genetic systems that are silent under laboratory conditions, such as the bgl (β-glucoside) operon of E. coli. The bgl operon of Escherichia coli, involved in the uptake and utilization of aromatic β-glucosides salicin and arbutin, is maintained in a silent state in the wild type organism by the presence of structural elements in the regulatory region. This operon can be activated by mutations that disrupt these negative elements. The fact that the silent bgl operon is retained without accumulating deleterious mutations seems paradoxical from an evolutionary view point. Although this operon appears to be silent, specific physiological conditions might be able to regulate its expression and/or the operon might be carrying out function(s) apart from the utilization of aromatic β-glucosides. This is consistent with the observations that the activated operon confers a Growth Advantage in Stationary Phase (GASP) phenotype to Bgl(+) cells and exerts its regulation on at least twelve downstream target genes.

Introduction

Bacteria are the most successful and the most prevalent creatures on earth. In these natural habitats bacteria are subject to various kinds of stress, such as nutrient scarcity, with occasional availability of food, fluctuations in temperature, pH, osmolarity and severe competition for resources from other organisms. Various elaborate survival strategies are employed by microbes to sense and adjust to the external and internal milieu. According to the neutral theory of evolution, genes that do not contribute towards the fitness of an organism are not subjected to natural selection and are lost by genetic drift (. In that way genomes are in a constant state of flux wherein pre-existing genes are lost by means of mutations and genetic drift and new genes are further acquired by horizontal gene transfer as well as mutations in preexisting genes. Cryptic genes are defined as genes that remain silent in the wild type organism but are capable of being activated and expressed by means of certain genetic changes (Hall ). These genes are different from pseudo genes since unlike pseudo genes they can be activated to a functional state. There are several known examples of cryptic genes in different organisms, such as the gene for citrate utilization in E. coli (Hall, 1982) and alcohol dehydrogenase gene in yeast (Paquin and Williamson 1986). In view of this, maintenance of such genes that do not contribute to the fitness of the organism is enigmatic. One possibility is that such genes are expressed under specific conditions and contribute to the organism’s fitness (Thatcher ). In the present review the cryptic Bgl operon of E coli has been discussed to understand its contribution in conferring fitness advantage to Bgl+ cells under stress physiological conditions.

Escherichia coli and β-glucosides Utilization

Identified in 1885 by Theodor Escherich, E. coli is one of the most well studied species of bacteria. While many strains of E. coli are non pathogenic, there are several strains that cause intestinal and extra intestinal infections. E. coli can utilize several carbohydrates, such as phosphorylated sugars, polyols, carboxylates, amino sugars, pentoses, hexoses, dissacharides, and polysaccharidesas as carbon source. However, wild type E. coli, like many other members of Enterobacteriaceae, is incapable of utilizing the β-glucosides as a sole source of carbon and energy. The β-glucosides are sugars mostly of plant origin that have a molecule of glucose linked through β-1, 4 linkage to an aliphatic or an aromatic side group. Some of the commonly found β-glucosides are salicin, arbutin, and cellobiose. The side groups in these sugars are 2-hydroxymethylphenyl, 4-hydroxyphenyl and glucose, respectively. Salicin is a secondary metabolite in the leaves of plants from the genus Salix; arbutin is found in the leaves of plants belonging to families Saxifragiceae, Rosaceae and Ericaceae, while cellobiose is a breakdown product of cellulose and lichenin and does not exist free in nature. There is heterogeneity among the members of the family Enterobacteriaceae with respect to their ability to utilize the β-glucosides as a carbon source. While members such as E. coli, Shigella, and Salmonella are incapable of fermenting these sugars, there are members such as Klebsiella, Enterobacter, Erwinia and Citrobacter which readily metabolize some or all of these sugars (Schaefler, 1967; Schaefler and Malamy, 1969).

Genetic Diversity of β-glucosides Utilization in E. coli

Wild type E. coli is unable to metabolize β-glucosides in spite of having three genetic systems for their utilization. These three genetic systems of E. coli: bgl, asc and chb, are classified as cryptic. Mutational activation of at least one of these systems is required to enable E. coli to metabolize these sugars. The asc operon, located at 58.7 min of E. coli chromosome (Hall ), upon being activated also enables the organism to utilize β-glucosides (Parker and Hall, 1988). This operon comprises a putative repressor, ascG, a PTS permease, ascF and a phospho-β-glucosidase, ascB (Hall and Xu, 1992). The chb operon of E. coli, located at 39 min on the chromosome, is a normal inducible operon for the uptake and utilization of chitobiose (Keyhani and Roseman, 1997). The chb operon comprises six ORFs, chbBCARFG and a regulatory region, chbOP. chbBCA encode three domains of the PTS permease, chbR encodes an activator that also acts as a repressor, chbF codes for phospho-glucosidase and chbG does not have any known function. ChbR, CAP and NagC have been implicated in the regulation of the chb operon by chitobiose (Plumbridge and Pellegrini, 2004). The bgl operon of E. coli (first studied by Schaefler is positioned at 83.8 min on the E. coli chromosome (Bachmann, 1990). The operon comprises three structural genes, bglG, bglF and bglB and a regulatory region bglR (Figure 1) (Mahadevan ; Schnetz ). The first gene of the operon, bglG, encodes an antiterminator that acts at two rho independent terminators flanking bglG (Mahadevan and Wright, 1987; Schnetz and Rak, 1988). The following gene, bglF, encodes a PTS permease that phosphorylates and transports the β-glucosides, salicin and arbutin. In the absence of the inducer BglF phosphorylates BglG, preventing its antiterminator function, thereby acting as a negative regulator of the bgl operon (Amster-Choder ; Schnetz and Rak, 1990). The last gene of the operon, bglB, encodes a phospho-β-glucosidase that cleaves phosphorylated salicin and arbutin. In addition to these three ORFs, the bgl operon also comprises another gene, bglH, which is not essential for the utilization of the β-glucosides. BglH is as sociated with outer membrane and is a porin, specific for the uptake of carbohydrates (Andersen ). In spite of being intact at the genetic level, the bgl operon is kept silent in the wild type organism due to the presence of certain negative structural elements in the regulatory region, bglR (Lopilato and Wright, 1990; Schnetz, 1995; Singh ; Schnetz and Wang, 1996; Mukerji and Mahadevan, 1997).

Figure 1

Schematics for the bgl operon of E. coli. The operon comprises three structural genes, bglG, bglF and bglB and a regulatory region bglR. The first gene of the operon, bglG encodes an antiterminator that acts at two rho independent terminators. The next gene bglF encodes a PTS permease and a negative regulator of the bgl operon. The last gene of the operon, bglB, encodes a phospho-β-glucosidase. In addition to these three ORFs, the bgl operon also comprises another gene, bglH, which is not essential for the utilization of the β-glucosides.

Mutations that Activate the bgl Operon

A variety of mutations, that act in cis or trans, can activate the silent bgl operon of E. coli, enabling the bacteria to utilize the β-glucosides, salicin and arbutin (Reynolds ; Reynolds ; Di Nardo ; Higgins ; Schnetz and Rak, 1992; Giel ). A single mutational event is sufficient to activate this operon. The most commonly occurring activating mutations for this operon are insertions of IS1 or IS5 in a 223 base pair sequence in the regulatory region of the bgl operon and also in some downstream sequences (Reynolds ; Di Nardo ; Higgins ). The insertion elements do not provide promoter element to the bgl operon (Di Nardo ) but activate the operon by disrupting the negative elements from the bgl promoter (Lopilato and Wright, 1990; Singh ). Point mutations in the binding site for atabolite ctivator rotein (CAP), which brings this site closer to the consensus CAP binding sequence also have been shown to activate the bgl operon (Di Nardo ; Lopilato and Wright, 1990). These point mutations result in a higher affinity binding of CAP and exclusion of H-NS binding, since CAP and H-NS binding sites in the bglR are overlapping (Mukerji and Mahadevan, 1997). Mutations in the hns locus are also known to activate the bgl operon, since H-NS acts as a negative regulator of this operon (Defez and Felice, 1981; Higgins ). Change in the supercoiling status of DNA has also been shown to affect the expression of the bgl operon. Mutations in gyrA (48 min) and gyrB (83 min) loci, that are expected to reduce DNA supercoiling, are known to activate the bgl operon (Di Nardo ). Reduced super coiling is expected to destabilize the cruciform structure in the bgl regulatory region, thereby lifting the negative regulation from the bgl operon and allowing it to be expressed at a higher level. This is consistent with the observation that point mutations within the inverted repeat activate the bgl promoter and inhibition of gyrase fails to enhance the expression further (Mukerji and Mahadevan, 1997). In addition, mutations that lead to the over expression of LeuO or BglJ have been shown to activate the bgl operon (Giel ; Ueguchi ). The bgl operon is subject to induction by the β-glucosides after mutational activation. BglG and BglF encoded by the bgl operon bring about this second level of regulation (Mahadevan, 1997).

Growth Advantage in Stationary Phase (GASP)

It has been shown that bacterial population can be maintained at counts of about 106 colony forming units (CFUs) per ml for several years without the addition of fresh nutrients (Finkel, 2006). This is a highly dynamic phase wherein several population take over occur and the culture becomes highly heterogeneous. If bacteria are starved for prolonged periods of time, 99% of the population dies in a phase commonly known as the Death phase. The remaining 1% of the population not only remains alive but also grows during a phase now known as prolonged stationary phase. It has been demonstrated that the GASP phenotype of the older cultures is due to genetic changes in the population (Zambrano ). The occurrence of GASP is a continuous phenomenon wherein older culture will always take over the younger culture. For example a 10-day-old culture takes over a one day old culture and a twenty day old culture takes over a ten day old culture and so on (Zambrano and Kolter 1996; Finkel ).

The Cryptic Bgl Operon Regulates oppA an Oligopeptide Transporter

Global analysis of intracellular proteins from Bgl+ and Bgl− strains revealed that the operon exerts regulation on at least twelve downstream target genes. Of these, oppA, which encodes an oligo-peptide transporter, was confirmed to be up-regulated in the Bgl+ condition (Harwani ). Since the oligopeptide transporter (oppA) has been shown to be up-regulated in the Bgl+ strain, it is conceivable that the functions encoded by oppA contribute to the GASP phenotype exhibited by Bgl+ strains (Harwani ). Interestingly the ZK819-97TΔoppA cells have lost the strong fitness advantage shown by the parent strain ZK819-199 97T in co-culture experiment. This suggests that a part of the growth advantage in stationary phase of the Bgl+ strain is contributed by OppA. The involvement of the bgl operon in the regulation of OppA expression could be direct or indirect. OppA is regulated negatively by a small regulatory RNA (sRNA) gcvB (Argaman ) which has been shown to inhibit translation initiation by binding to the oppA mRNA (Sharma ). In turn, the transcription of gcvB is positively regulated by the GcvA protein, the major transcription factor of the glycine cleavage system (Urbanowski ). Expression of gcvB is high during early log phase, but its level decreases during cell growth (Argaman ). This reduction in gcvB expression was much more pronounced in Bgl+ cells. Similarly, a significant decrease in gcvA transcription in Bgl+ cells was also registered in the stationary phase (Harwani ).

These observations suggest that the regulation of oppA by the bgl operon is via its regulators gcvA and gcvB (Figure 2). In view of this it has been proposed that the ability to transport oligo-peptides, mediated by the over expression of oppA, is partly responsible for the GASP phenotype exhibited by Bgl+ strains. Down-regulation of oppA in a strain carrying a deletion of bglG may be one of the reasons for the loss of the GASP phenotype of the ΔbglG strain. The complete loss of the GASP phenotype in the ΔbglG mutant and its partial rescue in the ΔbglGΔgcvA double mutant suggest that BglG is a master regulator involved in modulating the expression of downstream genes important in stationary phase survival and oppA is one such locus (Harwani ). It has been shown that the BglG decreases gcvA mRNA stability and suggests a specific role for BglG-mediated post transcriptional regulation at this locus and yet at other unexplored loci (Figure 2). The molecular mechanism by which BglG/gcvARAT duplex is exposed to the degradation machinery is still unknown. The regulatory role exerted by BglG on gcvA to control OppA translation would enhance our understanding of the BglG-mediated signalling process.

Figure 2

Schematics for the genetic regulation mediated by the activated bgl operon of an E. coli on oppA. In a co-culture experiment Bgl+ cells exhibit GASP phenotype over Bgl− cells under stationary phase growth condition. BglG interacts at gcvA mRNA to destabilize it which is known to positively regulates GcvB. In absence of GcvB, its negative regulation on oppA is relieved (GcvB is a translational repressor of oppA). The elevated level of oligopeptide transporter can now facilitate transport of small peptides.

Post-transcriptional regulation mediated by BglG on gcvA mRNA leads to its destabilization that affects GcvA translation negatively. The reduced level of GcvA leads to the reduced transcription of gcvB (translational repressor of OppA), and the increased translation of OppA. Thus, elevated levels of OppA in the Bgl+ strain facilitate transport of oligo-peptides, conferring a GASP phenotype over the wild type Bgl− counterpart (Figure 2) (Harwani ).

Conclusion

The maintenance of genetic systems which are apparently of no selective advantage to the organism is an evolutionary paradox. This applies to ‘cryptic’ genes, since they do not seem to function in the wild type organism and require mutational activation for expression. Does their retention in the wild type organism have any physiological significance? Are these genes truly silent or there are specific physiological conditions that can induce their expression? A possible mechanism has been documented by which oppA confers a competitive advantage to Bgl+ cells relative to Bgl− cells in the stationary phase (as described above). The involvement of the bgl operon in functions unrelated to the catabolism of β-glucosides suggests that selection for elevated expression of the operon can occur even in the absence of β-glucosides. This could be achieved by either by mutations or by overriding its negative regulation under specific growth conditions such as stationary phase. Though such elevated expression may not be sufficient to allow utilization of β-glucosides, it may be sufficient for the regulation of the downstream target genes, providing a selective force for the maintenance of the bgl genes over evolutionary time. Conclusively bgl operon has been highlighted for its involvement in the functions unrelated to the β-glucosides utilization. This could be one of the possible signal transduction mechanisms by which bacteria might modulate gene expression upon starvation stimuli. The present review help strengthen the notion that rather than a “cryptic” genetic element, the bgl operon should be considered as a dynamic component of the E. coli genome.