Studies on the Regulation of (p)ppGpp Metabolism and Its Perturbation Through the Over-Expression of Nudix Hydrolases in Escherichia coli

Abstract

Stringent response mediated by modified guanosine nucleotides is conserved across bacteria and is regulated through the Rel/Spo functions. In Escherichia coli, RelA and SpoT proteins synthesize the modified nucleotides ppGpp and pppGpp, together referred to as (p)ppGpp. SpoT is also the primary (p)ppGpp hydrolase. In this study, using hypomorphic relA alleles, we provide experimental evidence for SpoT-mediated negative regulation of the amplification of RelA-dependent stringent response. We investigated the kinetics of ppGpp degradation in cells recovering from stringent response in the complete absence of SpoT function. We found that, although greatly diminished, there was slow ppGpp degradation and growth resumption after a lag period, concomitant with decrease in ppGpp pool. We present evidence for reduction in the ppGpp degradation rate following an increase in pppGpp pool, during recovery from stringent response. From a genetic screen, the nudix hydrolases MutT and NudG were identified as over-expression suppressors of the growth defect of ΔspoT and ΔspoT ΔgppA strains. The effect of over-expression of these hydrolases on the stringent response to amino acid starvation and basal (p)ppGpp pool was studied. Over-expression of each hydrolase reduced the strength of the stringent response to amino acid starvation, and additionally, perturbed the ratio of ppGpp to pppGpp in strains with reduced SpoT hydrolase activity. In these strains that do not accumulate pppGpp during amino acid starvation, the expression of NudG or MutT supported pppGpp accumulation. This lends support to the idea that a reduction in the SpoT hydrolase activity is sufficient to cause the loss of pppGpp accumulation and therefore the phenomenon is independent of hydrolases that target pppGpp, such as GppA.

Keywords: (p)ppGpp; RelA; SpoT; nudix hydrolases; stringent response.

FIGURE 1

SMG resistance phenotype of hypomorphic RelA alleles was modulated by SpoT function. Cultures grown to saturation were washed, serially diluted, and spotted on minimal medium containing glucose with or without SMG (serine, methionine, and glycine) and photographed after 20 h incubation at 37°C. The relevant strain genotypes are indicated. *Growth in the presence of SMG was significantly retarded but not abolished. The strains from rows 1 to 13 are – RS1, RS8, RS9, RS17 white colony, RS53, RS92 white colony, RS54, RS420 white colony, RS11, RS18 white colony, RS303, RS316, RS361 white colony. Strains and their genotype are listed in Supplementary Table 1. The white colony of a strain refers to the derivative cured of plasmid P_lac-spoT⁺.

FIGURE 2

Hypomorphic relA alleles do not accumulate (p)ppGpp in response to isoleucine starvation. (p)ppGpp accumulation in response to isoleucine starvation was monitored in strains carrying the following relA alleles relA496:Tn10dTet (A), rlmD:Tn10dKan and rlmD:Tn10dKan ΔspoT (B), and ΔrlmD:FRT (C). Strains were cultured in MOPS glucose medium, labeled with P³², isoleucine starvation was induced with valine (arrow), and samples were collected immediately before valine addition and subsequently at the times indicated above the lanes and subjected to PEI-TLC (see section “Materials and Methods” for details). Data presented is a representative of experiments done at least 3 times. Strains used are RS9 (A), RS11 and RS18 white colony (B), and RS316 (C). The white colony of a strain refers to the derivative cured of plasmid P_lac-spoT⁺.

FIGURE 3

ppGpp turnover in the absence of SpoT function and its effect on growth. (A) The ΔrlmD:FRT ΔspoT strain (RS361, white colony) was cultured in MOPS minimal medium containing glucose, labeled with P³² and subjected to PEI-TLC. Isoleucine starvation was induced with valine (arrow), and samples were collected immediately before valine addition and subsequently at time points indicated above the lanes. Starvation was reversed by the addition of isoleucine (dotted arrow), and samples were collected at the time points indicated. The white colony of a strain refers to the derivative cured of plasmid P_lac-spoT⁺. Data presented is representative of experiments done 3 times. (B) The amount of (p)ppGpp over total [(p)ppGpp + GTP] at different time points after isoleucine starvation and after the reversal of starvation was plotted for the strains indicated using the data in Supplementary Tables 2, 3, 5. (C) ΔrlmD:FRT ΔspoT strain was grown in MOPS minimal medium containing glucose and subjected to isoleucine starvation by the addition of valine (solid arrow) and subsequently reversed by the addition of isoleucine (dotted arrow). Data from a representative experiment was plotted.

FIGURE 4

Effect of SpoT and/or GppA hydrolase activity on the synthesis and turnover of stringent nucleotides during amino acid starvation and recovery. Isoleucine starvation was induced by the addition of valine (solid arrow) and reversed by the addition of isoleucine (dotted arrow) in the wild type and ΔgppA (A), spoT1 (D), and spoT1 ΔgppA/P_lac-spoT⁺ strain after reducing the spoT expression by allowing growth in the absence of IPTG (E). A representative TLC is shown for each strain. (B) The concentration of ppGpp, pppGpp, or GTP over total [(p)ppGpp + GTP] at the time points indicated was plotted as bar graph using data from two independent experiments (Supplementary Table 5). Fraction of ppGpp or pppGpp or (p)ppGpp over total [(p)ppGpp + GTP] after the reversal of starvation was plotted for the wild type and gppA mutant with data from two independent experiments (C) the spoT1 and spoT1 ΔgppA/P_lac-spoT⁺ strains (F) with data from three independent experiments (Supplementary Tables 3, 4). The strains are wild type (RS1), ΔgppA (RS307 white colony), spoT1 (RS24), and spoT1 ΔgppA:FRT/P_lac-spoT⁺ (RS194). The white colony of a strain refers to the derivative cured of plasmid P_lac-spoT⁺.

FIGURE 5

Suppression of SpoT requirement by over-expression of mutT or nudG. Plasmid segregation assay was used to study the role of mutT and nudG genes in suppression of the growth defect of ΔspoT and ΔspoT ΔgppA strains in LB medium containing IPTG and Cm. A representative section from the plate has been included for each strain to show the color of the colonies after non-selective growth. The percentage of white colonies and the number of white colonies over the total number of colonies scored (white + blue) are provided for each panel. The ASKA plasmids carrying the mutT or nudG gene was used and the plasmid vector served as control. The ΔspoT/P_lac-spoT⁺ strain was grown in the presence of 0.1 mM IPTG (A) and the ΔspoT ΔgppA/P_lac-spoT⁺ strain was grown in the presence of 1 mM IPTG (B). The strains in panels (i–vi) are RS444, RS680, RS681, RS684, RS685, and RS686, respectively.

FIGURE 6

Increased expression of spoT or mutT or nudG alleviates the stringent response to isoleucine starvation. A representative TLC of MG1655 ΔlacZYAI:FRT strain carrying the ASKA plasmids indicated below each panel was cultured in MOPS minimal medium containing glucose Cm and 0.1 mM IPTG. The culture was labeled with P³² to follow the accumulation of stringent nucleotides after isoleucine starvation by the addition of valine (arrow). Samples were collected immediately before the addition of valine or at the time points indicated and subjected to PEI-TLC. The strains in panels (A–C) are RS688, RS760, RS689, and RS690.

FIGURE 7

Increased expression of mutT or nudG lowers the ppGpp pool and elevates pppGpp pool during stringent response in strains with reduced SpoT hydrolase activity. (A) Isoleucine starvation was induced by the addition of valine (arrow) to cultures of the spoT1 strain carrying the vector pCA24N (lanes 1–4), the vector with nudG (lanes 5–8), or mutT (lanes 9–12). (B) Isoleucine starvation was induced in the ΔspoT/P_lac-spoT⁺/pCA24N strain after reducing spoT + expression by growth in the absence of IPTG (lanes 1–4), in the ΔspoT/pCAnudG (lanes 5–8), and ΔspoT/pCAmutT (lanes 9–12) strains cultured in the presence of 0.1 mM IPTG. The accumulation of stringent nucleotides was followed with P³² labeled cultures as described in the methods. The strains used are, panel A, HR1348 (lanes 1–4), HR1350 (lanes 5–8), and HR1349 (lanes 9–12); Panel B, RS444 (lanes 1–4), RS460 (lanes 5–8), and RS459 (lanes 9–12).