Materials

Oligonucleotide primers for polymerase chain reaction (PCR) were purchased from FASMAC Co, Ltd (Atsugi, Japan). Other chemicals were all of analytical grade.

Bacterial strains and growth conditions

The strains used in this study were derivatives of E. coli K-12. BW25113 (rrnB3, ΔlacZ4787, hsdR514, Δ(araBAD)567, Δ(rhaBAD)568, rph-1) [1] and mutants of BW25113 in the Keio collection as a single-gene knockout library [3] were grown on plates or in liquid of modified Luria-Bertani (LB) medium (1% Bactotryptone, 0.5% yeast extract, and 0.5% NaCl) at various temperatures as described previously [1].

Effects of glucose and MgCl₂ and sensitivity to H₂O₂

To examine the effects of supplements, glucose (0.5% (w/v)) or MgCl₂ (20 mM) was added to the LB liquid culture as described in detail previously [1]. The experiments were performed three times, and the results were confirmed to be reproducible.

RT-PCR analysis

Cultures were grown in LB medium at 37°C until the exponential phase, and then the temperature was up-shifted to 46°C and incubation was continued for 8 min. Total RNA was immediately prepared from the heat-stressed cells by the hot phenol method [1, 4]. RT-PCR analysis was performed using a One-Step RNA PCR kit (AMV) (TAKARA BIO Inc, Otsu, Japan) to examine the expression of immediate downstream genes of disrupted genes as described in detail previously [1, 5]. The primer set used for each gene is shown in S1 Table. After the completion of PCR, the products were analyzed by 1.2% agarose gel electrophoresis and stained with ethidium bromide.

DNA manipulation

Recombinant plasmids for expression of thermotolerant genes, atpA, gntK, lpxL, fimG, yccM, yraN, cydD and nhaA, were constructed, generating pUC-atpA, pUC-gntK, pUC-lpxL, pUC-fimG, pUC-yccM, pUC-yraN, pUC-cydD and pUC-nhaA, respectively. The DNA fragment of each gene, including the sequence from 40-bp upstream from the translation initiation codon to the stop codon, was amplified and inserted into pUC19 by In-Fusion cloning (Clontech, USA). Amplification of the DNA fragment was performed by PCR using a specific primer set for each gene (S2 Table) and the genomic DNA of BW25113 as a template [6]. The DNA fragment of pUC19 [7] was also amplified by PCR using two primers, 5'-ACTCTTCCTTTTTCAATATTATTGAAGCA-3' and 5'-CTGTCAGACCAAGTTTACTCATATATAC-3', and pUC19 DNA as a template. The primer set for each thermotolerant gene and that for pUC19 shared 15 overlapping sequences that were required for In-Fusion cloning. Each gene inserted into the vector was designated to be transcribed from the lac promoter on the plasmid.

Complementation test

pUC-atpA, pUC-gntK, pUC-lpxL, pUC-fimG, pUC-yccM, pUC-yraN, pUC-cydD and pUC-nhaA DNAs were introduced into BW25113atpA::kan, BW25113gntK::kan, BW25113lpxL::kan, BW25113fimG::kan, BW25113yccM::kan, BW25113yraN::kan, BW25113cydD::kan and BW25113nhaA::kan cells, respectively. The empty plasmid DNA of pUC19 was also introduced into each of these mutant strains and BW25113, which were used as controls. These transformants were grown in LB medium at 37°C or 45°C for appropriate times. The experiments were performed three times, and the results were confirmed to be reproducible.

Identification of new thermotolerant genes

In a previous study, three successive screening steps were performed in a single-gene knockout library [3], and eventually 51 mutant strains that were sensitive to a CHT were found and reported [1]. In the screening process, 26 strains that exhibited relatively slow growth at 30°C on plates at the 1^st screening, which was examined at least twice for confirmation, were removed from further screenings by a spotting test and liquid culture. In this study, the 26 mutants were subjected to the 2^nd and 3^rd screenings of spotting and liquid culturing tests at a CHT as described previously [1] and careful growth experiments in liquid culture at 30°C, 37°C, 39°C, 44°C, 45°C and 46°C. All of the mutants were found to be significantly sensitive to a temperature of 46°C and some were even sensitive to 44°C (S1 Fig and Table 1). However, 5 of those mutants, atpB, atpC, atpE, atpF and atpH, that encode constituents in the complex of the F₁F_o-type ATP synthase showed little growth even at low temperatures, 30°C, 37°C and 39°C. Thus, out of the corresponding 26 genes, the 5 genes for the ATP synthase were excluded and 21 genes have become new thermotolerant genes in E. coli. In addition, the effects of supplements on cell growth of the sensitive mutants at a CHT were examined under the same conditions as those used previously [1] (S2 Fig and Table 1). Their sensitivity to H₂O₂ at 30°C was also tested because about 60% of disrupted mutants of thermotolerant genes are sensitive to H₂O₂ [1] (S2 Fig and Table 1). Moreover, according to the gene organization in the local region including the disrupted gene (S3 Fig), a possible polar effect of insertion of the aph cassette [3] was examined by RT-PCR with a specific primer set for each gene (S1 Table), but none of the mutants showed an obvious polar effect (S3 and S4 Figs), suggesting that the thermosensitive phenotype in the 21 mutants is due to disruption of the targeted gene. Furthermore, functional complementation was performed on 8 thermotolerant genes selected from five categories that are shown below (S5 Fig). As a result, all of the genes cloned on the plasmid were able to almost completely or partially complement the corresponding gene-knockout mutant. Notably, it was found that introduction of the empty vector, pUC19, and pUC-lpxL caused a negative and a positive effect, respectively, on growth at 37°C in BW25113 lpxL::kan. Similar but slightly weaker effects were found in the complementation of mutations of yraN, yccM and nhaA. It is likely that these mutations are sensitive to the presence of pUC19 and that each plasmid clone of these genes not only complements the corresponding mutation but also gives resistance to the vector. The mechanism of the growth influence by pUC19, however, is not obvious.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Thermotolerant genes identified in this study.

https://doi.org/10.1371/journal.pone.0189487.t001

Next, the 21 genes corresponding to the 21 thermosensitive mutants were classified on the basis of their functions by using public databases including the KEGG pathway database. Interestingly, most of the 21 thermotolerant genes were found to be classified into categories defined in a previous study: energy metabolism, outer membrane stabilization and tRNA modification (Table 1). It is noteworthy that this study allowed us to find lacking pieces in the following system or pathway. Six genes that are involved in the sulfur-relay system for tRNA modification are included in the previous list of thermotolerant genes [1], and yccM as the last one in the sulfur-relay system was newly identified. Similarly, lpxL, which encodes KDO2-lipid IV(A) lauroyltransferase, is a new thermotolerant gene in the lipopolysaccharide biosynthesis pathway.

The functions of the newly identified thermotolerant genes are summarized in Table 1. Most of them were categorized into group A, including genes for components or their synthesis in the respiratory chain and in the complex of the F₁F_o-type ATP synthase and for enzymes involved in the pentose phosphate pathway, the biosynthesis of lysine or phenylalanine/tyrosine/tryptophan and the metabolism of vitamin B6. The remaining genes were members of groups B, C and D. It is noteworthy that disruption of fimG for the outer membrane component in type 1 fimbriae, but not disruption of genes for its other components, causes thermosensitivity, suggesting the importance of outer membrane stability. As noticed previously, the growth of mutants that belong to group B or C was partially improved by the addition of Mg²⁺. Examination of the effect of H₂O₂ on growth showed that 14 mutants are sensitive to H₂O₂, being 67% of 21 newly identified thermotolerant genes (Table 1). This ratio is consistent with that of 51 previously identified ones, suggesting that the mechanism of thermotolerance partially overlaps with that of oxidative stress resistance in E. coli, as reported previously [1]. Additionally, DNA chip analysis [1] indicate that none of 21 newly identified thermotolerant genes are not up-regulated when cells are transiently exposed to a CHT, like most of previously identified thermotolerant genes.

Notably, functions of the members of group A suggest that ATP synthesis by substrate-level phosphorylation can support cell survival at temperatures up to 46°C but that ATP synthesis by oxidative phosphorylation is essential at a CHT. On the other hand, mutations of genes for subunits interacting with the membrane in the complex of the ATP synthase were found to be very sensitive even to low temperatures compared to other subunits of the complex. Presumably, mutations for the membrane-interacting subunits cause H⁺ leakage in the membrane, resulting in diminishment of membrane potential, which reduces cell survival of the organism. In addition, the dependence of oxidative phosphorylation at a CHT seems consistent with the essentiality of aceE, aceF, lpd, lipA and ackA at a CHT [1], which are mapped in the pyruvate metabolism pathway from pyruvate to acetyl CoA [8–11] that functions as energy metabolism for production of ATP. Therefore, these findings allow us to speculate that a metabolic flow at a CHT is toward the TCA cycle and increase in respiratory activity.

Possible mechanisms for survival at a CHT

In total, 72 thermotolerant genes were found in E. coli (Table 2). Intriguingly, thermotolerant genes in E. coli include only a small number of heat shock genes and are mostly genes responsible for functions to stabilize the membrane or to assist fundamental metabolism [1]. These functions are essential for survival in a 2–3°C range close to the CHT in E. coli. Genes categorized as genes responsible for energy metabolism seem to contribute to the synthesis of ATP, which may be required for repairing damage of macromolecules, protein, DNA or RNA. Many genes responsible for outer membrane stabilization and tRNA modification are involved in modification and stabilization of the structure of LPS and tRNA, respectively. Genes responsible for translation control or cell division may avoid or overcome accidents in translation or the cell division process at a CHT. On the other hand, the involvement of genes in DNA double-strand break repair and chaperone/proteinase suggests accumulation of reactive oxygen species (ROS) at a CHT, being consistent with evidence that a higher temperature results in accumulation of more oxidative stress [12] and the finding that mutants of all members in both groups exhibited sensitivity to H₂O₂ [1] and Table 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. List of thermotolerant genes in E. coli.

https://doi.org/10.1371/journal.pone.0189487.t002

Taken together, we propose a model of intracellular problems at a CHT and their protection mechanisms as shown in Fig 1. Under conditions at high temperatures close to a CHT, membrane fluidity may dramatically increase, causing leakage of electrons from the respiratory chain to generate ROS that give rise to oxidative damages to macromolecules. DNA double-strand breaks and protein denaturation by oxidative modifications seem to occur in the cytoplasm or periplasm because the corresponding genes become indispensable at a CHT. As protection mechanisms against these problems as thermotolerant mechanisms, LPS and membrane proteins may strengthen the membrane structure to prevent membrane fluidity. However, ROS scavenging genes were not found as thermotolerant genes. This may be due to the existence of homologues that can perform the same function in E. coli. Notably, overexpression of sodA for superoxide dismutase or katE for catalase significantly increased the number of viable and culturable cells at a CHT [12]. Similarly, overexpression of catalase and superoxide dismutase genes increased the degree of thermotolerance in Saccharomyces cerevisiae [13]. Notably, the generation of ROS might be enhanced by an increase in respiratory activity, as speculated above, for oxidative phosphorylation at a CHT. In addition to membrane stabilization and scavenging of ROS, there are protection mechanisms of stabilization of tRNA by modification and repair of DNA double-strand breaks or denatured proteins in E. coli. Moreover, there are other trouble-shooting mechanisms in translation or cell division at a CHT.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. A model of intracellular problems at a CHT and their protection mechanisms.

At a CHT, the level of ROS is increased as described in the text, resulting in damage of macromolecules. There are various possible protection mechanisms as thermotolerant mechanisms, such as stabilization of the membrane to protect electron leakage from the respiratory chain, scavenging ROS, and stabilization of tRNA by modification and repair of DNA double-strand breaks or denatured proteins. Abbreviations used are: OM, outer membrane; IM, inner membrane; O₂^•-, superoxide radical anion; H₂O₂, hydrogen peroxide; ^•OH, hydroxyl radical.

https://doi.org/10.1371/journal.pone.0189487.g001

Since the screening in this study was performed with a single-gene knockout library covering all non-essential genes in E. coli, knowledge of physiological functions of the 72 thermotolerant genes will be very beneficial for the genetic conversion of non-thermotolerant to thermotolerant bacteria.