Indole teaches persistence to bacteria

Indole molecule (C8H7N). Source: wikimedia commons

When a bacterial infection is treated with antibiotics, bacteria that are in a so-called dormant, inactive state may escape death – this because antibiotics only kill growing bacteria. It becomes a serious problem when these sleeping beauties start to grow again, in particular when they do so after the period of antibiotic treatment has ended… Thus, an infection that was apparently cured could be followed by a secondary infection days or weeks later. This problematic phenomenon is called bacterial persistence, and it should not be confused with bacterial resistance, in which growing bacteria are immune to one or several antibiotics.

Now what about indole? (The molecule displayed on top of this post.) Actually indole is present in very common and important biomolecules, such as the amino acid tryptophan, the animal hormone serotonin and the plant growth hormone auxin. We have known for more than a century that E. coli produces indole in stationary phase (Lee, 2010), and it does so thanks to an enzyme called tryptophanase, which cleaves tryptophane into indole, pyruvate and ammonia. 

But E. coli is not the only bacterium capable of that: more than 85 species (both Gram-negative and Gram-positive) can synthesize indole (Lee, 2010).  For a long time the biological functions of indole were overlooked, but now we know that indole can act as an extracellular signal and can for instance increase antibiotic resistance and control biofilm formation in E. coli.

Illustration of how bacterial persisters (orange cells) survive antibiotic treatment while growing cells (blue) die. After the treatment, persister cells can turn into growing cells.

The group of Jim Collins, at Boston University, noted that indole was produced under conditions (stationary phase, limited nutrients) conducive to bacterial persistence. They decided to investigate the possible role of indole signaling in the formation of bacterial peristers in E. coli, and they published their results in May 2012 as a brief communication in Nature Chemical Biology (NicoleVega et al., 2012). 

And what they show clearly is that exposure to indole increases the appearance of persisters by a ten-fold factor in a population of E. coli! In addition, they showed that bacteria that were not producing tryptophanase (thus incapable of producing indole from tryptophane) formed much less persister cells. 

How does the signaling work? Well, it is not fully understood yet, but Vega et al. showed that indole is acting extracellularly, because E. coli cells that are missing a specific indole transporter (Mtr, which imports indole inside the bacterium) are more likely to become persistent than normal E. coli cells. 

Curiously, even though the bacteria are exposed to a fixed concentration of indole in a liquid environment, they will not all react the same way: it is clear that the population’s response to indole is heterogeneous. Vega demonstrated this by using a fluorescent reporter system that was responsive to indole signaling. Thus, they could follow the expression of fluorescence in individual cells and study the differences among the population. After antibiotic treatment, the persisters were also the ones that glowed most in response to indole!

Finally, the group of Jim Collins looked at the transcriptome of E.coli during indole signaling, and they found that two pathways, the so-called phage-shock response and oxidative stress response were stimulated by indole. Now if you disrupt both phage-shock and oxidative stress pathways in E. coli, the induction of persistence by indole disappears… Interestingly, they also found that cells exposed to indole had no higher expression in drug export, which is required for bacterial resistance. Hence, it confirms that bacterial persistence, not resistance, is at play here.

Model for indole signaling and persister formation in E. coli, from Vega et al., 2012. Some cells in the population turn into persisters (orange cells) after exposure to indole. I thank Jim Collins for the permission to reproduce the figure here.

But this phenomenon may not be totally specific to indole, since Vega et al. showed that treatment with hydrogen peroxide could also activate the oxidative stress pathway and lead to persister formation.

The authors conclude as follows:

“The bacterial signaling molecule indole is sensed in a heterogeneous manner by a population of cells, causing induction of OxyR and phage-shock pathways via a periplasmic or membrane component, thereby inducing the creation of a persistent subpopulation. Indole is not toxic at physiological concentrations, but it triggers protective responses, acting to inoculate a subpopulation (persisters) against possible future stress.”

What I would be curious to know is what is responsible for the heterogeneous response? The authors did not address this question (maybe in a future work?) but acknowledged its importance:

“These findings add to an understanding of persister formation as bacterial ‘bet-hedging’ strategy in uncertain environments.”

Indeed, it can be valuable for bacteria to maintain diversity within the population, in particular when their life conditions are susceptible to change often. The more you look at it, the more you find such examples of phenotypic variability in microbial ecology. It definitely plays a major role in the bacterial world.

References:

• Lee J. H. &  Lee J. (2010). Indole as an intercellular signal in microbial communities. FEMS Microbiology Reviews 34(4):426-44.
• Jermy A. (2012). From indolence to persistence. Nature Reviews Microbiology 10(5):310-1.
•  Vega N. M., Allison K. R., Khalil A. S. and J. J. Collins (2012). Signaling-mediated bacterial persister formation. Nature Chemical Biology 8(5):431-3.