Inducing Self-Poisoning: a Feasible Antibacterial Strategy?

ABSTRACT

Development of novel antibacterial strategies is required to tackle the alarming threat for global health due to antimicrobial resistance. In this issue of the Journal of Bacteriology, Boulanger et al. provide evidence supporting that the blocking of metabolic pathways to induce accumulation of toxic intermediates can be a possible approach to combat bacterial infections (E. F. Boulanger, A. Sabag-Daigle, M. Baniasad, K. Kokkinias, et al., J Bacteriol 204:e00344-22, 2022, https://doi.org/10.1128/jb.00344-22).

TEXT

In this issue of the Journal of Bacteriology, Boulanger et al. (1) report that the accumulation of phosphorylated intermediates from different carbohydrate metabolic pathways (sugar-phosphates) has a deleterious effect on the growth and virulence of Salmonella.

Accumulation of sugar-phosphates can have toxic effects in many organisms, including humans and bacteria. For instance, the accumulation of galactose-1-phospate (Gal-1P) in galactosemic infants causes poor weight gain, diarrhea, hepatocellular damage, lethargy, pseudotumor cerebri, and even death (2). Gal-1P accumulation also has toxic effects in other animals and different bacteria (3). Furthermore, the accumulation of fructose-1-P (Fru-1P) is detrimental for the growth of the Escherichia coli bacterium (4), and the accumulation of mannose-6-P (Man-6P) inhibits stalk elongation and causes growth arrest in the Caulobacter crescentus bacterium (5). Accumulation of sugar-phosphate occurs when the enzyme that converts this metabolic intermediate is catalytically inactive or is removed. For example, in the galactosemic infants, the enzyme Gal-1P uridyl transferase (GalT) fails to convert Gal-1P to glucose-1-P, and thus they accumulate Gal-1P when they receive galactose from milk (2). Similarly, bacteria lacking the genes encoding the Fru-1P kinase enzyme (FruK) or the Man-6P isomerase enzyme (ManA) accumulate the Fru-1P and Man-6P sugar-phosphates, respectively, in the presence of the corresponding sugar, fructose or mannose (4, 5). Salmonella, a bacterium that produces intestinal and systemic infection in humans and other animals (6), can utilize over 70 different compounds as carbon sources, including many sugars, with a velocity of up to 10⁵ sugar molecules per second (7–9). Nevertheless, Salmonella has a reduced capacity to use phosphorylated compounds, like sugar-phosphates, as the sole carbon source (7). Therefore, a rapid accumulation of sugar-phosphates in Salmonella strains carrying dysfunctional metabolic enzymes could be expected.

Boulanger and colleagues determined the effect of the accumulation of different sugar-phosphates in Salmonella. This study started with the selection of seven enzymes for different sugar metabolic pathways, which were expected to generate accumulation of sugar-phosphates when inactivated. These enzymes are AraD, RhaD, GlpD, MtlD, and ManA, required for the metabolism of the sugars arabinose, rhamnose, glycerol, mannitol, and mannose, respectively, as well as GalT and GalE, required for the metabolism of the sugar galactose. As a strategy alternative to the utilization of small-molecule inhibitors of these enzymes, the activity of each enzyme was eliminated by deleting or affecting by insertion the gene encoding the respective enzyme in Salmonella. Then, Boulanger et al. analyzed the effect of the sugar of interest (that for which the metabolism was blocked to induce accumulation of sugar-phosphate) on the growth of the respective Salmonella mutant. Each mutant grew normally in medium lacking the sugar of interest but containing another carbon source, i.e., in the nutrient-rich lysogeny broth (LB) or M9 minimal medium with fructose. Additionally, each mutant did not grow if the sugar of interest was the only carbon source available. Interestingly, the mutants lacking GalE, GalT, MtlD, RahD, or AraD enzymes showed a reduced growth in the presence of the sugar of interest, no matter if other carbon sources were present. Consistently with this growth defect, by performing targeted metabolomics using mass spectrometry, the authors demonstrated that selected mutants, lacking MtlD or AraD, indeed accumulated the expected sugar-phosphate, mannitol-1P and ribulose-5P, respectively. In addition, a previous study demonstrated that the inactivation of GalE or GalT led to the accumulation of the sugar-phosphate Gal-1P in E. coli (10). To further support the accumulation of sugar-phosphate as a cause that generates growth defects in Salmonella, the authors showed that the absence of enzymes acting upstream of RhaD, AraD, and MtlD in the corresponding metabolic pathway (RhaB, AraB, and MtlA, respectively) does not induce intoxication in the presence of the sugar of interest. Under the laboratory conditions tested, the lack of GlpD or ManA did not, or only slightly, affect the growth of Salmonella and did not generate accumulation of sugar-phosphate. Then, Boulanger et al. evaluated the virulence of the Salmonella mutants in the mouse model for intestinal infection, through competition assays between the wild-type (WT) strain and the respective mutant. In these experiments, the animals were supplemented or not with the corresponding sugar of interest provided in the drinking water. Notably, the absence of RhaD or ManA reduced Salmonella virulence only in the presence of the sugar of interest, whereas the absence of AraD inhibited Salmonella virulence in the presence or not of the sugar of interest, but mostly under the first condition. The absence of GalE or MtlD reduced Salmonella virulence similarly in the presence or not of the sugar of interest, while the absence of GalT or GlpD did not affect Salmonella virulence under either condition. These results showed that the accumulation of sugar-phosphate, induced by blocking the activity of specific metabolic enzymes in combination with the administration, or presence in the animals, of the sugar of interest, attenuated Salmonella virulence. Finally, Boulanger et al. performed phylogenetic analysis to determine which other bacteria may be susceptible to sugar-phosphate toxicities. Interestingly, this analysis revealed that MtlD, AraD, and RhaD enzymes are uncommon in most phyla but present in genera including Salmonella, Escherichia, Enterobacter, Citrobacter, Klebsiella, Serratia, and Hafnia. Thus, the authors proposed that MtlD, AraD, and/or RhaD could be narrow-spectrum targets to combat infections by pathogens of the Enterobacteriaceae family or even could be targets to edit the intestinal microbiota to reduce the risk of chronic illness associated with the referred bacteria, without affecting beneficial microbes. Data from the work of Boulanger et al. and results from a previous report of the same group (11) support FraB, an enzyme from the fructose-asparagine metabolic pathway, as a good target for therapy against Salmonella; FraB is present in Salmonella but extremely rare in other bacteria, and its inactivation generates toxicity mediated by the sugar-phosphate 6-P-Fru-Asp (11). In contrast, GalE is very common in bacteria and has high structural conservation with the human enzyme GalE (12); thus, it is an unlikely bacterial target for therapy.

The study from Boulanger et al. shows that the induction of different sugar-phosphates can attenuate bacterial virulence (Fig. 1). However, much investigation is needed to determine the viability of this strategy for therapy. A major challenge is the identification of small-molecule inhibitors specific for the targets like those identified by Boulanger and colleagues, which would be used for practical therapy. Inhibitors able to cause complete inactivation of targets may be required to accumulate the amount of sugar-phosphate required for toxicity, which could be solved by using a combination of inhibitors. Even more, these inhibitors could be used to potentiate the activity of antibiotics or other antibacterial molecules. Therefore, it is necessary to determine whether the combination of different molecules inducing sugar-phosphate toxicity, or the combination of these inhibitors with antibiotics or other antibacterial molecules, has a synergistic effect. Likewise, the effectiveness of the strategy based on sugar-phosphate toxicity will also depend on the presence of the sugar of interest and its hierarchy of nutrient utilization by bacteria in the niche of infection. Assays to determine the availability of the respective sugar in the niche of infection and the possibility of the sugar reaching this site when exogenously administered are required. On the other hand, it is necessary to know the mechanism of action for the sugar-phosphates, which will help to understand possible protection responses developed by bacteria against these toxic molecules. Of transcendental importance will be the studies to identify possible mechanisms of resistance against the effect of the sugar-phosphates. In this respect, Boulanger et al. determined that the interruption of the carbohydrate metabolic pathway, in a step prior to that generating the sugar-phosphate, also moderately attenuates Salmonella virulence. Therefore, even with this expected mechanism of resistance bacteria would have a decreased fitness. However, it is necessary to test whether the target enzymes can mutate to become immune to the effect of inhibitors while maintaining their metabolic activity.

FIG 1.

Accumulation of sugar-phosphates affects fitness and virulence of Salmonella. Normally, Salmonella can consume the sugars fructose-asparagine, mannitol, arabinose, and rhamnose as carbon sources. Inactivation (⊗) of the enzymes fructose-asparagine deglycase (FraB), mannitol-1-phosphate dehydrogenase (MtlD), ribulose-5-phosphate epimerase (AraD), and rhamnulose-1-phosphate aldolase (RhaD) leads to the accumulation of the sugar-phosphates 6P-fructose-aspartate, mannitol-1P, ribulose-5P, and rhamnulose-1P, respectively, in the presence of the corresponding sugar. The accumulation of each of these sugar-phosphates causes intoxication to and thus affects the fitness and virulence of Salmonella (1, 11).

The study from Boulanger et al. inspires further investigation on an innovative antibacterial strategy, which could be based not only on the induction of sugar-phosphate toxicities but also on the accumulation of toxic molecules from other metabolic pathways.

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