SgrT, a Small Protein That Packs a Sweet Punch

ABSTRACT

The SgrS small RNA (sRNA) has been shown to protect against elevated levels of glucose phosphate by regulating the stability and translation of mRNAs encoding proteins involved in sugar transport and catabolism. The sRNA also was known to encode a protective 43-amino-acid protein, SgrT, but little was known about its mechanism of action. Lloyd et al. (J Bacteriol 199:e00869-16, 2017, https://doi.org/10.1128/JB.00869-16) use cell biological and genetic approaches to demonstrate that the small protein interacts with the PtsG importer to block glucose transport by this phosphotransferase system and promote utilization of nonpreferred carbon sources to maintain growth during glucose-phosphate stress.

TEXT

To survive rapid fluctuations in environmental conditions such as changes in nutrient availability, pH, temperature, reactive oxygen species, and detrimental compounds, bacteria have evolved elaborate responses to sense, protect against, and recover from these stressful conditions. In recent years, small RNAs (sRNAs) have emerged as important players in the regulation of stress responses. Usually, these sRNA regulators act by modulating the transcription, stability, and translation of mRNAs either by binding to the target mRNA via short base-pairing interactions or by binding to and modulating the activities of RNA binding proteins (reviewed in reference 1). In early genome-wide screens to identify new sRNAs, it was realized that a subset of the short transcripts encoded small proteins (2). Since then, other proteins of less than 50 amino acids have been identified in bacteria and eukaryotes, and some have been shown to have important functions in signaling and cellular defenses, many acting at the membrane (reviewed in reference 3). A few sRNAs have even been found to both base pair with mRNA targets and encode a small protein, and thus have been termed “dual-function sRNAs.”

Although increasing numbers of sRNAs and small proteins are being studied, only five dual-function sRNAs have been characterized in any detail in bacteria, namely, Staphylococcus aureus RNAIII and Psm-mec RNA, Streptococcus pyogenes Pel RNA, Bacillus subtilis SR1, and Escherichia coli SgrS (reviewed in reference 4). All five function as sRNAs by base pairing with their respective mRNA targets. The RNAIII, Psm-mec, and Pel sRNAs code for hemolysin or hemolysin-like proteins, but the actions of the SR1- and sgrS-encoded SR1P and SgrT proteins have been more elusive. Other dual-function sRNAs also have been predicted. However, although mRNA targets for these dual-function base-pairing sRNAs are known or can be predicted, the functions of the encoded small proteins remain a mystery. The gap in knowledge about these proteins is in part due to the formidable challenges associated with studying small proteins since standard biochemical methods often cannot be utilized (reviewed in reference 3). These challenges require innovative alternative approaches as illustrated by the current paper from the Vanderpool group. Combining mutational analysis and cell imaging, Lloyd et al. demonstrate that the 43-amino-acid, hydrophobic SgrT protein controls the glucose-specific phosphotransferase system (PTS) (5).

E. coli utilizes three different PTS transporters for glucose uptake (reviewed in reference 6) (Fig. 1). The primary transporter is the glucose-PTS complex comprised of two subunits: the transmembrane EIICB^Glc protein, which is a glucose permease and is encoded by ptsG, and the cytoplasmic EIIA^Glc protein, which functions as an intermediate phosphotransfer protein and is encoded by crr. The EIIABC^Glc complex can also transport α-methyl glucoside 6-phosphate (αMG), βMG, 1-thio-glucose, and 5-thio-glucose with high affinity. Glucose also can be imported by other PTS complexes, such as the mannose-PTS EIIABCD^Man complex encoded by manXYZ and, to a lesser extent, by the N-acetyl-d-glucosamine (GlcNAc)–PTS EIICBA^GlcNAc protein encoded by nagE.

FIG 1.

The SgrS sRNA and SgrT protein act synergistically to combat glucose-phosphate stress. (A) In E. coli, glucose is transported primarily by the EIIABC^Glc complex comprised of the transmembrane PtsG (EIICB^Glc) subunit and the cytoplasmic Crr (EIIA^Glc) subunit. Glucose also can be imported by other PTS complexes, such as ManXYZ and NagE. Glucose import by PtsG leads to the dephosphorylation of the EIIA^Glc subunit, which blocks the import of nonpreferred carbon sources (inducer exclusion) by transporters such as LacY, ensuring the preferential utilization of glucose. (B) Glucose-phosphate stress induces transcription of SgrS and consequently SgrT synthesis. SgrS sRNA base pairs with the ptsG and manXYZ mRNA to inhibit further translation of the transporters encoded by these mRNAs. Simultaneously SgrT protein binds the EIIC^Glc domain of PtsG to inhibit further glucose transport. This inhibition possibly leads to accumulation of phosphorylated EIIA^Glc, which can no longer bind and block other transport proteins (relieving inducer exclusion). Thus, together, SgrS and SgrT stop further glucose-6-phosphate (G6P) accumulation, diminish intracellular G6P levels, and promote the utilization of alternative carbon sources.

The 227-nucleotide SgrS sRNA was first discovered in a computation screen for sRNAs in E. coli (2) and soon thereafter was found to have a role in mediating the cellular response to glucose-phosphate stress (7). Transcription of SgrS is induced by glucose-6-phosphate accumulation, which can be caused by a mutation in the glycolytic pathway or exposure to the nonmetabolizable analogue α-methyl glucoside 6-phosphate (αMG). The accumulation of these sugar phosphates is toxic, and it was reported that cells respond by destabilizing the ptsG mRNA (8). Vanderpool and Gottesman showed this regulation is mediated by SgrS (7). The lack of recovery from growth inhibition in ΔsgrS cells upon glucose-phosphate stress highlights the importance of this sRNA under these stress conditions. The SgrS RNA was shown to base pair with the ptsG mRNA as well as the manXYZ mRNA to repress synthesis of both of these transporters (7, 9). The sRNA also protects against sugar phosphate stress by repressing the translation of the asd, adiY, folE, and purR mRNAs and by stabilizing the yigL mRNA, which encodes a phosphosugar phosphatase (10–12). As for many other sRNAs in enteric bacteria, the RNA chaperone Hfq is required for base pairing and regulation by SgrS (7).

The SgrS sRNA is long compared to other known Hfq-binding sRNAs in E. coli, with the middle region responsible for base pairing with mRNAs and the 3′ end responsible for binding Hfq. The unusual length and conservation prompted further examination of the SgrS sequence and led to the discovery of an open reading frame encoding SgrT at the 5′ end of the sRNA (13). Mutational experiments to separate the properties of the RNA showed that high levels of the individual activities were capable of protecting the cells against the glucose-phosphate stress, indicating that the sRNA and small protein have redundant function. It was observed that SgrT had no effect on the translation or stability of the ptsG mRNA, leading to the hypothesis that the small protein functioned at a level distinct from SgrS to mediate the same or complementary response. Wadler and Vanderpool hypothesized that SgrT might function by inhibiting the uptake of glucose, but they did not know the mechanism (13).

Lloyd et al. (5) now present creative experiments that document that SgrT specifically blocks the glucose-specific PTS complex by inhibiting PtsG activity. The authors first examined the specificity of the SgrT effect by monitoring the growth of the SgrT-overproducing strain in different carbon sources. The observation that these strains fail to grow with glucose as the sole carbon source but not with other carbon sources, such as mannose, fructose, trehalose, and N-acetylglucosamine, suggested specific inhibition of glucose transport. To further test this hypothesis, the authors took advantage of the differences in substrate specificity for the PtsG and ManXYZ transporters. They assayed the expression of an sgrS-lacZ transcription fusion, which is induced by sugar phosphate stress upon the addition of two toxic analogs of glucose, αMG or 2-deoxyglucose (2DG), transported by PtsG and ManXYZ, respectively. SgrT overexpression blocked sgrS-lacZ induction by αMG but not 2DG, indicating that SgrT inhibited the transport activity of PtsG but not that of ManXYZ.

To further characterize SgrT, the subcellular localization of the protein with a C-terminal 3×FLAG tag was examined by immunofluorescent staining. These experiments revealed the majority of SgrT-3×FLAG is at the cell periphery. The size and hydrophobic properties of small membrane proteins can render standard protein tagging and purification approaches difficult, resulting in barriers to the identification of interacting proteins by copurification. Lloyd et al. circumvented the problems associated with biochemical approaches by utilizing the cell biological approach of examining colocalization as a proxy for assessing protein-protein interactions. In monitoring the localization of SgrT-3×FLAG in wild-type and ΔptsG backgrounds, they observed that SgrT localizes to the membrane in a PtsG-dependent manner, again consistent with SgrT interacting with this protein.

PtsG is comprised of three domains: the cytosolic EIIB^Glc domain, the linker, and the membrane-bound EIIC^Glc domain. To delineate what region of PtsG is affected by binding to SgrT, Lloyd et al. assayed a chimeric protein comprised of the transmembrane C-terminal domain of PtsG (EIIC^Glc) and the cytoplasmic B domain NagE (IIB^GlcNAc), which preferentially transports GlcNAc and d-glucosamine (14). The previous study showed that the chimeric protein could phosphorylate glucose but not GlcNAc, indicating the EIIC^Glc domain dictates the glucose specificity of the EIIC^Glc-IIB^GlcNAc chimera. The observation that SgrT expression still inhibits growth in glucose in this background led the authors to conclude that SgrT targets the EIIC^Glc domain.

To further define the EIIC^Glc residues important in the potential SgrT-PtsG interaction, the authors tested the effects of several point mutations previously shown to broaden substrate specificity (15) or reported to impact the SgrT-PtsG interaction (16). Lloyd et al. examined the activity of the P_sgrS-lacZ fusion in the mutant backgrounds in the presence or absence of SgrT overexpression when the cells were subject to glucose-phosphate stress. As expected, the induction of the P_sgrS-lacZ fusion by αMG addition was attenuated by high levels of SgrT consistent with SgrT-mediated inhibition of wild-type PtsG activity. In a strain expressing a V12F derivative of PtsG, αMG still caused induction of the fusion. However, SgrT no longer blocked this induction. These observations suggest the V12F mutation prevents inhibition by SgrT without influencing transport activity, pointing to the importance of this residue in the SgrT-PtsG interaction.

Finally, to test if SgrT inhibited preexisting PtsG protein or instead impacted only newly synthesized protein, the authors measured [¹⁴C]αMG uptake upon overproduction of either the base-pairing region of SgrS or SgrT. Cells expressing the base-pairing region of SgrS gradually accumulated [¹⁴C]αMG, consistent with the model that SgrS base pairing inhibits new PtsG synthesis. In contrast, cells expressing SgrT showed almost no [¹⁴C]αMG accumulation, indicating that SgrT inhibits preexisting PtsG.

Together the experiments described by Lloyd et al. (5) present a compelling case that the 3′ part of the SgrS RNA blocks PtsG synthesis by pairing with the ptsG mRNA and the 5′-encoded SgrT protein interacts with and inhibits the activity of the C-terminal membrane domain of PtsG, providing a powerful two-level block of glucose-phosphate accumulation (Fig. 1). Interestingly, the authors find that one consequence of SgrT inhibition of PtsG is the relief of inducer exclusion. During normal growth in media with glucose and other carbon sources, the dephosphorylated form of EIIA^Glc (Crr), prominent during glucose transport, inhibits transporters of the alternative carbon sources (reviewed in reference 17). SgrT apparently blocks this repression, possibly through the accumulation of the phosphorylated form of EIIA^Glc, given that the small protein allowed cells to grow better in lactose in the presence of αMG. SgrT thus aids full recovery from glucose-phosphate stress not only by inhibiting further glucose transport but also by indirectly facilitating the utilization of alternative carbon sources.

The current study again raises several questions about dual-function sRNAs. What selective advantage is provided by a transcript that both acts as a base-pairing sRNA and encodes a small regulatory protein? How frequently do both activities affect the same pathways? Is one individual RNA molecule acting both as an sRNA and as an mRNA simultaneously or sequentially, or does one pool of RNA molecules function solely as sRNAs and another pool solely as mRNAs? These questions can only be answered by further characterization of the dual-function sRNAs and the identification of the protein targets using creative approaches as described for SgrT.

The views expressed in this Commentary do not necessarily reflect the views of the journal or of ASM.