Induction of the Stringent Response Underlies the Antimicrobial Action of Aliphatic Isothiocyanates

Abstract

Bacterial resistance to known antibiotics comprises a serious threat to public health. Propagation of multidrug-resistant pathogenic strains is a reason for undertaking a search for new therapeutic strategies, based on newly developed chemical compounds and the agents present in nature. Moreover, antibiotic treatment of infections caused by enterotoxin toxin-bearing strain-enterohemorrhagic Escherichia coli (EHEC) is considered hazardous and controversial due to the possibility of induction of bacteriophage-encoded toxin production by the antibiotic-mediated stress. The important source of potentially beneficial compounds are secondary plant metabolites, isothiocyanates (ITC), and phytoncides from the Brassicaceae family. We reported previously that sulforaphane and phenethyl isothiocyanate, already known for their chemopreventive and anticancer features, exhibit significant antibacterial effects against various pathogenic bacteria. The mechanism of their action is based on the induction of the stringent response and accumulation of its alarmones, the guanosine penta- and tetraphosphate. In this process, the amino acid starvation path is employed via the RelA protein, however, the precise mechanism of amino acid limitation in the presence of ITCs is yet unknown. In this work, we asked whether ITCs could act synergistically with each other to increase the antibacterial effect. A set of aliphatic ITCs, such as iberin, iberverin, alyssin, erucin, sulforaphen, erysolin, and cheirolin was tested in combination with sulforaphane against E. coli. Our experiments show that all tested ITCs exhibit strong antimicrobial effect individually, and this effect involves the stringent response caused by induction of the amino acid starvation. Interestingly, excess of specific amino acids reversed the antimicrobial effects of ITCs, where the common amino acid for all tested compounds was glycine. The synergistic action observed for iberin, iberverin, and alyssin also led to accumulation of (p)ppGpp, and the minimal inhibitory concentration necessary for the antibacterial effect was four- to eightfold lower than for individual ITCs. Moreover, the unique mode of ITC action is responsible for inhibition of prophage induction and toxin production, in addition to growth inhibition of EHEC strains. Thus, the antimicrobial effect of plant secondary metabolites by the stringent response induction could be employed in potential therapeutic strategies.

Keywords: (p)ppGpp; enterohemorrhagic Escherichia coli; isothiocyanate; stringent response; sulforaphane.

FIGURE 1

Chemical structure of isothiocyanates (ITCs) employed in this study.

FIGURE 2

Time-kill kinetics at a range of concentrations of ITCs. Escherichia coli MG1655 culture was challenged with compounds at 1 x, 2 x, and 4 x MIC levels and compared to untreated control (triangle, reversed triangle, diamond, and circle, respectively). Bactericidal activity was defined as a reduction of 99.9% (≥ 3 log₁₀) of the total number of CFU/ml in the original inoculum and marked as a dashed line on plots. Data are presented as mean and standard deviation of three independent replicates.

FIGURE 3

The effect of isothiocyanates on the (p)ppGpp alarmone accumulation in E. coli. Bacteria were grown overnight on LB agar plates at 30°C, then collected and washed with PBS buffer, concentrated, and resuspended in low phosphate MOPS labeling medium at OD₆₀₀ = 0.2 density. Cells were labeled with 5 μCi/ml ³²P for 20 min. (p)ppGpp synthesis was induced with 1 mg/ml of serine hydroxamate (SHX) for positive control; various isothiocyanates were used at 1 x MIC concentration for 20 min. Samples were spotted on PEI cellulose TLC plates, developed in 1.5 M potassium phosphate buffer and visualized with a Phosphoimager. The positions of guanosine nucleotides (GDP, GTP, ppGpp, and pppGpp) are indicated by arrows.

FIGURE 4

Analysis of synergistic interactions between IBR, IBN, ALN, and SFN. (A) Synergistic effects of sulforaphane in mixtures with iberin, iberverin, and alyssin are represented on isobolograms. Estimated FICI values are presented for combinations of SFN with other ITCs. A checkerboard technique was employed to delineate the Fractional Inhibitory Concentration Index (FICI). The treated cultures were screened for visual growth in a microplate reader. The FICI were then calculated as described in section “Materials and Methods” (FICI ≤ 0.5, synergy; 0.5 ≤ FICI ≤ 1.0, additivity; 1.1 ≤ FICI ≤ 2.0, indifference; FICI ≥ 2.0, antagonism) (B) ppGpp alarmone accumulation under treatment with synergistic combinations of SFN + ITCs. (C) The kinetics of relative ppGpp accumulation in treated cells. Relative ppGpp levels were assessed by densitometry using the QuantityOne Software. The ³³P incorporation method was used to evaluate stringent response induction like described previously. (p)ppGpp synthesis was induced with 1 mg/ml of serine hydroxamate (SHX) for positive control; various isothiocyanates were used in 1x MIC or FIC concentration in combination for 10, 15, 20, and 30 min. Samples were spotted on PEI cellulose TLC plates, developed in 1.5 M potassium phosphate buffer and visualized with a Phosphoimager. The results are from at least three independent experiments. The pooled (p)ppGpp and GTP amounts were taken as 100%. The statistical significance of differences in (p)ppGpp amount compared to its basal level of samples at the corresponding time of non-treated control was determined by t-test (*p ≤ 0.05, **p ≤ 0.01, and ***p ≤ 0.001) as indicated above or below plots.

FIGURE 5

The effect of specific amino acids on the stringent alarmone (p)ppGpp synthesis during SFN, IBR, IBN, and ALN treatment of E. coli MG1655. Relative (p)ppGpp accumulation after supplementation with specific amino acids in cultures treated with SFN (yellow), IBR (green), IBN (blue), and ALN (green). The assessment of intracellular level of (p)ppGpp alarmones was determined by [³²P]orthophosphoric acid incorporation and developed by TLC on PEI cellulose plates, followed by densitometry. The level of (p)ppGpp represents % of a sum of all G nucleotides visualized on TLC plate. The dotted line represents the mean level of alarmone induction in SFN, IBR, IBN, and ALN treated cells without amino acid supplementation (control). The results are from at least three independent experiments.

FIGURE 6

Inhibition of stx_p activity under ITC treatment of E. coli 86-24 O157:H7 strain. E. coli 86-24 O157:H7 Δstx:GFP was cultivated for 3 h in the presence of (A) not-treated control (B,G), SFN added at MIC value, (C,H) SFN and IBR at FIC values, (D,I) SFN and IBN added at FIC values, (E,J) SFN and ALN at FIC values. (F–J) Mitomycin C was added as the toxin production inducer. Bacteria were then stained with SynaptoRed to visualize membranes; GFP synthesis was analyzed by fluorescence microscopy. The activity of stx_p:GFP is marked by arrows. Pictures present merged green and red channels; the scale bar is valid for all panels.

FIGURE 7

N-source utilization upon sulforaphane treatment. (A) Kinetic curves of tetrazolium violet (TV) color development upon 48 h of sulforaphane (ITC) treatment. Each plate shows comparison of treated and untreated strains, as indicated. The yellow color reflects the situation where the tested treated strain utilizes nitrogen-rich compounds at the same level as untreated corresponding strain. The red color reflects reduced metabolic flow in the treated strains (most cases). Black frames indicate scored changes in kinetics that passed the reproducibility test. (B) The effect of ITC treatment on the wild-type and relA mutant strains. The more turquoise the color, the more ITC treatment suppresses utilization of a given nitrogen compound. Full arrows point to the wild-type strains and open arrows to the relA mutants. Descriptions in gray refer to the nitrogen compound plate position.