Indole Pulse Signalling Regulates the Cytoplasmic pH of E. coli in a Memory-Like Manner

Bacterial cells are critically dependent upon pH regulation. Here we demonstrate that indole plays a critical role in the regulation of the cytoplasmic pH of Escherichia coli. Indole is an aromatic molecule with diverse signalling roles. Two modes of indole signalling have been described: persistent and pulse signalling. The latter is illustrated by the brief but intense elevation of intracellular indole during stationary phase entry. We show that under conditions permitting indole production, cells maintain their cytoplasmic pH at 7.2. In contrast, under conditions where no indole is produced, the cytoplasmic pH is near 7.8. We demonstrate that pH regulation results from pulse, rather than persistent, indole signalling. Furthermore, we illustrate that the relevant property of indole in this context is its ability to conduct protons across the cytoplasmic membrane. Additionally, we show that the effect of the indole pulse that occurs normally during stationary phase entry in rich medium remains as a “memory” to maintain the cytoplasmic pH until entry into the next stationary phase. The indole-mediated reduction in cytoplasmic pH may explain why indole provides E. coli with a degree of protection against stresses, including some bactericidal antibiotics.


Concentrating indole with C18 columns
Indole production in E. coli occurs mostly during the transition from exponential to stationary phase 1 . To measure the low concentration of indole accurately during lag and exponential phase, a pre-concentration step was added before the Kovacs assay, using C18 solid phase extraction (SPE) cartridges (SampliQ C18, Agilent, CA, USA). C18 SPE cartridges concentrate low quantities of non-polar analytes, and their use for concentrating various indole analogues has already been established 2 .
The procedure was performed according to the manufacturer's instructions with minor modifications. Before use, each cartridge (500 mg octadecylsilane -6 ml) was equilibrated by flowing-through 10 ml 1-pentanol, followed by 10 ml deionised water. A sample (50 ml) from a lag-phase or exponential-phase bacterial culture was taken, and cells harvested by centrifugation at 2755 x g for 10 min (Eppendorf 5810 R centrifuge). The supernatant was flowed through an equilibrated cartridge and the cartidge was then washed with 10 ml deionised water. Indole was eluted from the cartridge with 5 ml 1-pentanol, resulting in an eluate containing indole concentrated ten-fold.

Kovacs assay
The Kovacs assay is an established technique providing consistent results from different laboratories 3,4 . The assay was performed as previously described 1 to determine the presence of indole in culture supernatants. Briefly, a sample (1 ml) from a stationary-phase culture was removed, and cells harvested by centrifugation at 11337 x g for 3 min (Eppendorf Minispin microfuge). The supernatant was removed and assayed: 300 μl of Kovacs reagent (10 g of p-dimethylamino-benzaldehyde dissolved in a mixture of 50 ml of HCl and 150 ml of amyl alcohol) was added to the supernatant and incubated for 2 minutes.
The presence of indole was indicated by the formation of red colour following the addition of Kovacs reagent.
The Kovacs assay was also used to determine the concentration of indole in lag and exponential phase culture supernatants, after a pre-concentraion step with C18 columns.
The assay was performed in a 96-well plate as previously described 5  Overnight cultures were diluted to OD600 = 0.05 in fresh LB or M9 and cultures were grown at 37 °C and sampled at intervals between OD600 = 0.1 -0.6 for LB and OD600 = 0.1 -0.4 for M9. The final sample was taken after overnight incubation (~ 18 hrs). The transition between exponential to stationary phase was observed around the same time (~ 2 hrs after sub-culture) in both media. Supplementary Fig. S2: The absence of indole production correlates with the higher cytoplasmic pH. The cytoplasmic pH of the WT E. coli is ~ 7.2 when indole is produced and ~ 7.8 when indole is absent. The mutant strain lacks the ability to convert tryptophan to indole under any conditions so the cytoplasmic pH is always ~ 7.8. The presence of indole is indicated by the formation of red colour following the addition of Kovacs reagent. Samples were taken from stationary-phase cultures of BW25113 WT and BW25113 ∆tnaA grown in four different media: LB, M9 (containing glucose as a carbon source), supplemented M9 (glucose, Vitamin B1, trace elements & casamino acids) and supplemented M9 + tryptophan. The average cytoplasmic pH (Fig. 1) is added for comparison. Supplementary Fig. S3: The cytoplasmic pH of E. coli BW25113 WT and ∆tnaA growing in LB, measured by flow cytometry . Cultures of BW25113 and BW25113 ∆tnaA in LB were sampled throughout exponential and stationary phase, and samples were analysed by flow cytometry. The cytoplasmic pH of each sample was determined by the average fluorescence intensity of pHluorin normalised to mCherry and compared to a standard curve. pH values are in agreement with those obtained by fluorescence spectroscopy (Fig. 1). Data are presented as means  SD, and the final point in each line is an overnight sample (~ 18 hrs incubation).
. Supplementary Fig. S4: Indole concentration in the LB supernatant of E. coli BW25113 WT during lag, exponential and stationary phase. An overnight culture of BW25113 WT in LB was subcultured into fresh LB (OD600 = 0.05). Samples were taken throughout the lag phase, which lasted ~ 20 min, the exponential phase and the stationary phase. For samples from lag-and exponential phase, Indole was measured by Kovacs assay after passing each sample through a C18 solid phase extraction column, which concentrated indole 10-fold. For samples from stationary phase (OD600 > 1), indole was measured by Kovacs assay without the pre-concentration step. 30 µM (± 10 µM) indole was made during the lag phase and this concentration remained constant throughout exponential phase. The first sample during the lag phase was taken immediately after sub-culturing (time after inoculation = 0 min) and had no detectable indole. Consistent with previous reports, (see, Figure 1 in Supplementary Reference 6) an increase in the supernatant indole concentration of 550 µM (± 50 µM) was seen during the transition from exponential to stationary phase, and this concentration remained constant throughout stationary phase.