Physiological Analysis of the Stringent Response Elicited in an Extreme Thermophilic Bacterium, Thermus thermophilus

Bacterial strains and culture conditions. Thermus thermophilus strain HB8 (= ATCC 27634) and its mutant strains were grown in MTM medium (see below) or in a chemically defined medium (medium 162) with shaking at 70°C. Strain KO-572, which requires phenylalanine for growth, was obtained by treating wild-type strain HB8 with the mutagen N-methyl-N′-nitro-N-nitrosoguanidine. The relA disruptants KO-571 and KO-652 were constructed from strains HB8 and KO-572, respectively, using a gene engineering technique (see below). To grow relA disruptants in the chemically defined medium, l-isoleucine, l-valine, and l-leucine (and l-phenylalanine for strain KO-652) (50 μg/ml each) were added to the medium to promote the growth. Bacillus stearothermophilus JCM 2501 and Bacillus subtilis 168 (Marlburg strain) were grown in NB medium (see below) at 50°C and 37°C, respectively. Escherichia coli strains W3110 and JM109 were used for in vitro ppGpp synthesis and for transformation of recombinant plasmids, respectively. Both E. coli strains were grown in Luria-Bertani medium (L medium). MTM medium is a modification of TM medium (28) and contained (per liter) 8 g of polypeptone, 4 g of yeast extract, 2 g of NaCl, 250 mg of MgSO₄′7H₂O, and 150 mg of CaCl₂ (adjusted to pH 7.5 with NaOH). Solid medium contained 2.8% agar. NB medium contained (per liter) 10 g of peptone, 10 g of meat extract, and 5 g of NaCl (adjusted to pH 7.0). The chemically defined medium, medium 162 (11), was described previously. Ampicillin (75 μg/ml) and kanamycin (25 μg/ml) were used for selection of transformants.

Preparation of thiostrepton-resistant mutants.Spontaneous thiostrepton-resistant mutants were obtained from colonies that grew within 3 days after cells were spread on MTM agar medium containing 0.3 μg/ml or 3 μg/ml thiostrepton. To detect the mutations, a 5′ part of the rplK gene was amplified by PCR using the oligonucleotide primers TTL11f (5′-ATGAAGAAAGTTGTTGCGG-3′) and TTL11r (5′-CGTGCTTAGGCATCCTTCAC-3′), which were based on the sequence described previously (18), since spontaneous rplK mutations conferring the resistance to thiostrepton (or thiopeptin) in bacteria have always been found in the 5′ part of the rplK gene (see Results). The reaction mixture (50 μl) contained GCI buffer (Takara), 20 ng of total DNA, 0.2 mM (each) deoxynucleoside triphosphate, 1 mM (each) primer, and 2.5 U of LA-Taq DNA polymerase (Takara). Amplification was carried out in a GeneAmp PCR system 9700 (Applied Biosystems) using a protocol that entailed initial denaturation at 98°C for 2 min followed by 30 cycles of denaturation at 98°C for 20 s, annealing at 58°C for 10 s, and elongation at 72°C for 30 s. The DNA was then sequenced using a DYEnamic ET terminator cycle sequencing premix kit (Amersham Biosciences) and an ABI PRISM 310 genetic analyzer (Applied Biosystems).

Genetic transformation of T. thermophilus. T. thermophilus wild-type strain HB8 was transformed to be thiostrepton resistant as described by Koyama et al. (28). Strain HB8 was cultured overnight, diluted 100-fold with fresh TM medium, and incubated with shaking at 70°C for 4 h. Then, the culture was mixed with chromosomal DNA (final concentration of 10 μg/ml) prepared from thiostrepton-resistant mutant cells. DNA was isolated by the method of Saito and Miura (50). The mixture was then incubated with shaking at 70°C for 1 h, followed by cooling on ice. Samples were appropriately diluted with TM medium and plated on a TM agar plate containing 1 μg/ml thiostrepton for the detection of thiostrepton-resistant transformants, which developed after 48 h of incubation at 70°C.

Chromosomal DNA isolation and manipulation.Standard methods for DNA isolation and manipulation were used as described by Sambrook et al. (51). Chromosomal DNA extraction for PCR amplification was carried out using an InstaGene Matrix resin (Bio-Rad) according to the instruction manual. DNA fragments were isolated from agarose gels using a QIAEX II gel extraction kit (QIAGEN).

Construction of RSH _Tt (relA) disruptants.The plasmids used to disrupt the T. thermophilus relA/spoT homolog (RSH _Tt ) through homologous recombination were constructed as follows. Based on the database nucleotide sequence, internal PCR primers for RSH _Tt and a thermostable kanamycin-resistant marker gene (HTK) were generated. A pBluescript SK(−) plasmid vector (Toyobo) was then used to clone PCR products in which an EcoRV site was introduced by PCR using the following primers (restriction sites are underlined, and modified sequences are in italics): TNP-020 (for the promoter region of RSH _Tt ) (5′-TAA TCTAGA CGGGTAGGGGTGCTACACTAAAAGTAGTGG-3′), TNP-021 (reverse primer for RSH _Tt ) (5′-CAAGATGGGAGTGGGGCATGGTTAGG-3′), TNP-024 (for the 5′ end of HTK) (5′-CCG CTCGAG CGTTGACGGCGGATATGGTA-3′), and TNP-025 (for the 3′ end of HTK) (5′-AAT CTCGAG CGTAACCAACATGATTAACA-3′). PCR with chromosomal DNA from T. thermophilus HB8 and the pUC18-HTK vector for HTK was carried out with TaKaRa PrimeSTAR HS DNA polymerase (Takara) as instructed by the manufacturer. The reactions were run in a GeneAmp PCR 9700 system (Applied Biosystems). The amplification protocol typically entailed 25 cycles of denaturation at 98°C for 10 s, annealing at 56°C for 5 s, and extension at 72°C for 2.5 min. The PCR fragment containing RSH _Tt was digested with HindIII-XbaI and cloned into the HindIII-XbaI sites of pBluescript SK(−), yielding pTN1018. The 2.3-kb HindIII-XbaI fragment containing RSH _Tt was subcloned into the HindIII-XbaI site of pUC119 (Toyobo), yielding pTN1022. To add the restriction site in the Kan^r gene cassette (HTK), PCR amplification was performed using primers TNP-024 and TNP-025 based on the pUC118-HTK, and then the 1.0-kb fragment was cloned into the pBluescript SK(−)/EcoRV site, yielding pTN1020. The 1.0-kb XhoI HTK(XhoI) fragment from pTN1020 was inserted into the XhoI sites (in RSH _Tt ), generating pTN1023. The respective plasmids for gene disruption were introduced into T. thermophilus strains HB8 (wild type) and KO-572 (Phe⁻), after which kanamycin (50 μg/ml)-resistant transformants were selected as described by Hashimoto et al. (17). An overnight culture was diluted 50-fold with one-half-diluted MTM medium (1/2 MTM) and shaken at 70°C until the optical density at 650 nm (OD₆₅₀) reached 0.45 to 0.55. This culture (400 μl) was mixed with 1.0 μg of gene disruption plasmid DNA, after which the mixture was incubated at 70°C for 2 h and then spread on 1/2 MTM agar medium containing 50 μg/ml kanamycin. The plates were then incubated at 70°C for 24 h. Disruption of RSH _Tt was confirmed with primers TNP-030 (5′-GGAGGCCCTCAAAGAGTTGG-3′) and TNP-031 (5′-GGAGTGGGGCATGGTTAGG-3′) by detecting the HTK cassette in the expected region (central region) of RSH _Tt .

Real-time quantitative PCR (RT-qPCR) analysis. T. thermophilus cultures (100 ml), grown to an OD₆₅₀ of 0.5 to 0.6, were treated with serine hydroxamate (final concentration, 10 mM) for the appropriate time, placed into ice-cold 50-ml tubes, and then centrifuged at 6,000 × g for 10 min at 4°C. The cell pellets were frozen and kept at −80°C until RNA extraction. At that time, the frozen pellet was incubated at room temperature for 30 min in lysozyme (0.5 mg/ml) solution. Total RNA was extracted using the ISOGEN-LS reagent (Nippon Gene) with chloroform, and the purification steps were carried out following the manufacturer's instructions. The extracted RNA was then purified further using an RNeasy kit with RNase-free DNase I (QIAGEN), after which the concentration of purified RNA was measured by spectroscopy at OD₂₆₀. cDNA was generated from a 1-μg sample of the purified total RNA by reverse transcription (ReverTra Ace, 100 U per reaction; Toyobo) at 45°C using the reverse transcription primers TNP-035 (for the 23Sa and 23Sb rRNA genes) (5′-CTGAGATGTTTCAGTTCCC-3′) and TNP-044 (for rpoD) (5′-GCCTTGTTCTCAATCTGCC-3′) and then subjected to RNase H treatment (2 U per reaction). Samples (50 ng per assay [for rpoD] or 0. 5 ng per assay [for 23Sa-23Sb]) were then analyzed using the 7300 real-time PCR system and Power SYBR Green PCR master mix (Applied Biosystems). The PCR primers used were TNP-032 (the forward primer for 23Sa and 23Sb [5′-TGCCCTGAGGGGGGTAGC-3′]), TNP-034 (the reverse primer for 23Sa and 23Sb [5′-CTGGCTTATCGCAGGTAGC-3′]), TNP-042 (rpoD forward primer [5′-CAACCTCCGGCTCGTGGTCTCC-3′]), and TNP-043 (rpoD reverse primer [5′-GGTGGCGTAGGTGGAGAAC-3′]). The standard curves for quantification were calculated by serial dilution of plasmid, generating pTN1024 and pTN1027, which contain an amplified 23Sa and 23Sb rRNA nucleotide sequence (ca. 260 bp) and a rpoD nucleotide sequence (ca. 700 bp), respectively. These plasmids were constructed based on pBluescript SK(−). The amplification protocol entailed 1 cycle of 96°C for 1 min, followed by 40 cycles of 96°C for 30 s (step 1), 58°C for 30 s (step 2), and 72°C for 35 s (step 3). Data collection and real-time analysis were done at step 3. After the final cycle, melt curve data were obtained using an additional stage of dissociation, beginning at 56°C for 10 s and then incrementally increasing the temperature until 96°C.

Measurement of RNA synthesis.The RNA synthesis rate in growing cells was determined as previously described (38) by measuring the incorporation of [2-¹⁴C]uracil into acid-precipitated material. Cells were grown to mid-exponential phase (OD₆₅₀ = 0.5) in a chemically defined medium (medium 162), after which [2-¹⁴C]uracil (0.1 μCi/ml, 100 μM), with or without serine hydroxamate (10 mM), was added to the culture and incubation continued with shaking. Samples (1 ml) were then removed at the indicated times and immediately added to 1 ml of ice-cold 10% trichloroacetic acid. The resultant precipitates were collected by filtration, washed with 5% trichloroacetic acid containing 10 mM cold uracil, dried, and counted in a liquid scintillation counter using clear-sol (Nakarai).

Assay of nucleotide pools.The intracellular concentrations of nucleotides, including ppGpp and pppGpp, were assayed by high-performance liquid chromatography (HPLC) after extraction with 1 M formic acid as previously described (38) and expressed as pmol per mg dry weight.

Ribosome-dependent in vitro ppGpp synthesis.Ribosomes were isolated from exponential-phase T. thermophilus (grown in MTM medium) or E. coli cells (grown in L medium) as described previously (30). The resultant pellets of ribosomes were resolved in buffer A (50 mM Tris acetate [pH 8.0], 15 mM magnesium acetate, 60 mM potassium acetate, 30 mM ammonium acetate, 1 mM dithiothreitol, 0.2 mM EDTA, 0.5 mM phenylmethylsulfonylfluoride), and the ribosome concentration was determined using a spectrophotometer (at 260 nm) and expressed in terms of A ₂₆₀ (U/ml). Ribosome-dependent synthesis of ppGpp or pppGpp was carried out according to the method of Martinez-Costa et al. (35). The reaction mixture (50 μl), containing 2 mM ATP, 1.3 mM GTP, 10 μCi/ml [α-³²P]GTP (3,000 Ci/mmol; Amersham Biosciences), 18% (vol/vol) methanol, and 60 A ₂₆₀ U/ml ribosome, was incubated for 30 min at the indicated temperature. The reaction was stopped by addition of 2 μl of 88% formic acid, after which the precipitated proteins were removed by centrifugation at 10,000 × g for 5 min. A 5-μl aliquot of the supernatant was subjected to thin-layer chromatography on polyethyleneimine-cellulose (Merck) with 1.5 M KH₂PO₄. Labeled mononucleotides were detected by autoradiography.

Construction of a plasmid for in vitro transcription assays.The region of the T. thermophilus 23S/5S rRNA operon (designated the 23Sb gene in the NCBI database [accession no. AP008226; previously accession no. M35674]) containing the promoter sequence was amplified by PCR using oligonucleotide primers TTrrn-1 (5′-CAGCA AAGCTT TGCCAGCGAAAAGGCCC-3′) and TTrrn-2 (5′-ACATCCAGGGATCCACGCCCGCTAC-3′), which were based on the sequence described previously (16). The PCR product was digested using the corresponding restriction enzymes at the HindIII site generated by TTrrn-1 (underlined sequence) and an internal BamHI site, and the resultant fragment was cloned into the HindIII and BamHI sites of the pSP64-polyA vector (Promega), yielding pTRP, which was then sequenced to ensure the correctness of the nucleotide sequence.

Purification of RNA polymerase.RNA polymerase was purified according to the method of Burgess and Jendrisak (5) or Xue et al. (66) with some modification. All steps were carried out at 4°C. TGED buffer (0.01 M Tris-HCl [pH 7.9], 5% glycerol, 0.1 mM EDTA, and 0.1 mM dithiothreitol) was used as the basal buffer throughout the purification procedures. Polyethylenimine P-70 solution (30% [vol/vol] solution) was purchased from Wako Pure Chemicals, and a 10% (vol/vol) solution titrated with HCl to pH 7.9 was prepared and subjected to filtration through Miracloth (Calbiochem) to remove insoluble materials. Cells (10 g) in 40 ml of grinding buffer (TGED buffer plus 23 μg/ml phenylmethylsulfonylfluoride and 2 mg/ml lysozyme) were then disrupted using a French press at 12,000 lb/in². After the resultant cell extract was centrifuged at 8,000 × g for 45 min and the supernatant was collected, 10% polyethylenimine P-70 solution (pH 7.9) was slowly added with stirring to a final concentration of 0.5%. The stirring was then continued for an additional 5 min before the mixture was centrifuged at 6,000 × g for 15 min. The resultant pellet was washed by suspending it in 40 ml of TGED buffer containing 0.2 M NaCl and centrifuged again at 6,000 × g for 15 min, after which the RNA polymerase was eluted with 40 ml of TGED buffer containing 1 M NaCl. The eluate was then centrifuged at 6,000 × g for 30 min, the supernatant collected, and solid ammonium sulfate added with stirring to 50% saturation. The mixture was then stirred for an additional 1 h and centrifuged at 8000 × g for 45 min, and the supernatant was discarded. To remove the nucleic acids, the pellet was washed with 40 ml of TGED buffer containing 2 M ammonium sulfate. After centrifugation, the pellet was resuspended in TGED buffer and dialyzed for 24 h against TGED buffer containing 0.15 M NaCl. Thereafter, a calf thymus DNA-cellulose (denatured; Amersham Biosciences) column (5 ml in volume) was equilibrated with 20 ml of TGED buffer containing 0.15 M NaCl. Each sample of dialyzate was pumped onto the column at a flow rate of 0.2 ml/min, washed with 20 ml of the same buffer, and then eluted with 40 ml of a 0.15 to 1 M NaCl linear salt gradient. Peak fractions were pooled and precipitated with 50% saturated ammonium sulfate at 0°C for 15 min with gentle stirring, centrifuged at 8,000 × g for 30 min, and dissolved in 1 ml of TGED buffer containing 0.5 M NaCl. The RNA polymerase sample thus obtained was then further fractionated using a superose 6 HR 10/30 gel filtration column with an AKTA explorer 10XT fast-protein liquid chromatography system (Amersham Biosciences). The sample was put onto the column, equilibrated with TGED buffer containing 0.5 M NaCl, and eluted at a flow rate of 0.1 ml/min. Peak fractions were collected and dialyzed for 24 h with storage buffer (the same as TGED buffer containing 0.1 M NaCl, except that the glycerol concentration was 50% instead of 5%), and the resultant enzyme solution was stored at −20°C until use.

Assay of RNA polymerase activity.RNA polymerase activity was assayed using the method of Barker et al. (4) with some modification. As template DNA, the 23S/5S rRNA operon region (23Sb gene) of plasmid pTRP was amplified by PCR using the oligonucleotide primers pSP64f (5′-ATTTAGGTGACACTATAGAATAC-3′) and pSPr (5′-ACAGCTATGACATGATTACGAATTC-3′) and then extracted twice with phenol-chloroform, precipitated with ethanol, and resuspended in 10 mM Tris-HCl (pH 8.0). Transcription was initiated by addition of RNA polymerase (5 μg) into 50 μl of reaction mixture containing 40 mM Tris-HCl (pH 8.0), 10 mM MgCl₂, 1 mM dithiothreitol, 0.1 μg/μl bovine serum albumin, 200 mM NaCl, 0.02 μg of template DNA, 0.5 mM (each) nucleotide triphosphate (ATP, GTP, CTP, and UTP), and 2 μCi/ml of [³H]UTP. After addition of the indicated amount of ppGpp, pppGpp, or ppApp to the mixture, the reaction was run at 60°C until 10-μl samples of the reaction mixture were collected and mixed with an equal volume of stop solution (1% sodium dodecyl sulfate [SDS] and 50 mM sodium pyrophosphate) to stop the reaction. The samples were then applied to DE81 DEAE filters (Whatman), which were then washed three times with 5% Na₂HPO₄ and twice with water, dried, and counted in a liquid scintillation counter using clear-sol (Nakarai).

To confirm the specific rRNA transcript, an in vitro transcription assay was also carried out as an alternative method using polyacrylamide gel electrophoresis. Transcription was performed in 20 μl of reaction mixture containing 40 mM Tris-HCl (pH 8.0), 10 mM MgCl₂, 1 mM dithiothreitol, 0.1 μg/μl bovine serum albumin, 100 mM NaCl, 1 μg of RNA polymerase, 8 ng of template DNA, 0.5 mM (each) ATP, GTP, CTP, and UTP, and 0.2 mCi/ml of [α-³²P]UTP. In some cases, ppGpp, ATP, or GTP was added. The reaction was run at 60°C for 20 min and mixed with 2 μl of 20% SDS to stop the reaction. The sample was treated with water-saturated phenol:chloroform:isoamyl alcohol (50:49:1, vol/vol/vol) and precipitated with ethanol, and the pellet was dissolved in 15 μl of water. After addition of the same volume of 2× formamide sample buffer (95% formamide, 18 mM EDTA, 0.025% SDS, 0.1% xylene cyanol, and 0.1% bromophenol blue), the transcripts were separated on 8 M urea-5% polyacrylamide gel. The radioactivity was detected with a BAS-2500 system (Fuji Film) and quantitated with Image Gauge version 3.41 software (Fuji Film).

Assay of IMP dehydrogenase activity.IMP dehydrogenase activity of T. thermophilus strain HB8 (grown to mid-exponential phase in MTM medium) was measured as described previously (37), except that the reaction was carried out at 50°C for 20 min in a reaction mixture with a pH adjusted to 8.5 instead of 8.0.

Reagents.ppGpp, pppGpp, and ppApp were prepared enzymatically in our laboratory using Streptomyces morookaense (44). HPLC analysis showed the purity of these samples to be >98% for ppGpp and pppGpp and 93% for ppApp.

Thermus thermophilus synthesizes ppGpp.To determine whether T. thermophilus, like other bacteria, has the ability to produce ppGpp, we first assessed intracellular ppGpp levels after treating the cells with serine hydroxamate, an amino acid analogue that provokes the stringent response by inhibiting tRNA^Ser aminoacylation (36). For instance, upon exposure to serine hydroxamate, B. subtilis and B. stearothermophilus accumulated a large amount (500 or 100 pmol/mg dry weight, respectively) of ppGpp, which was accompanied by marked inhibition of growth (Fig. 1). Likewise, T. thermophilus accumulated ppGpp when treated with serine hydroxamate while growing exponentially, though the level of ppGpp accumulated (30 pmol/mg dry weight) was much lower than was seen with B. subtilis (Fig. 1). Accumulation of ppGpp (34 pmol/mg dry weight) was also detected when cells growing exponentially in MTM medium were subjected to nutritional shift-down by transfer for 20 min to MTM medium lacking polypeptone and yeast extract (data not shown), as well as 2 h after cells entered stationary phase (5 pmol/mg dry weight). On the other hand, pppGpp was not detected (<3 pmol/mg dry weight) at any time during the study.

• Open in new tab
• Download powerpoint

FIG. 1.

Effect of serine hydroxamate on growth and ppGpp accumulation. T. thermophilus HB8 (grown in MTM medium at 70°C), B. stearothermophilus JCM 2501 (grown in NB medium at 50°C), and B. subtilis 168 (grown in NB medium at 37°C) were treated with serine hydroxamate (SH) (final concentration, 10 mM). Changes in intracellular ppGpp levels were measured using HPLC as described by Ochi (38). Arrows indicate the addition of serine hydroxamate. Open circles represent growth in the presence of serine hydroxamate.

Isolation of relC mutants.To further characterize ppGpp accumulation and the accompanying metabolic changes in T. thermophilus, we next attempted to isolate relaxed (rel) mutants from this thermophilic bacterium. To date, relC mutants have been detected in the genera Escherichia (13, 67), Bacillus (10, 15, 47, 61), and Streptomyces (15a, 27, 39) based on the ability of relC mutations to confer resistance to thiostrepton, an antibiotic that binds to the ribosome, or by means of the gene engineering technique. We found that thiostrepton-resistant T. thermophilus mutants did indeed develop on plates containing 0.3 to 3 μg/ml thiostrepton. Of these (we tested 15 isolates), two tsp strains (KO-564 and KO-565 [Table 1 ]) grew as well as the parental HB8 strain but were found to accumulate a markedly reduced amount of ppGpp when treated with serine hydroxamate. As expected, the mutants harbored a mutation within the rplK gene, which encodes the ribosomal protein L11 (Table 1). Notably, the point mutations found were both located within a well-conserved region near the N terminus, within which mutations conferring the RelC and/or Tsp^r phenotype had previously been reported for T. thermophilus (6) and for other microorganisms (27) (Fig. 2). Thereafter, we used the mutant KO-564 strain for further study.

• Open in new tab
• Download powerpoint

FIG. 2.

Comparison of the amino acid sequences of the ribosomal protein L11 (an rplK gene product) in bacteria. Arrowheads show the mutations conferring the RelC and/or Tsp^r phenotype in Thermus thermophilus (this study), Escherichia coli (67), Halobacterium halobium (48), and Streptomyces coelicolor A3(2) (64). Asterisks indicate identical amino acids.

To establish a causal relationship between the rplK mutation and the observed phenotype, we transformed wild-type T. thermophilus using DNA from the rplK mutant strain KO-564 (P21R) according to the method of Koyama et al. (28). All transformants exhibiting resistance to thiostrepton (we tested seven transformants), which developed at a high frequency of ∼1%, harbored a P21R mutation and displayed a diminished capacity to accumulate ppGpp upon exposure to serine hydroxamate, confirming that the rplK mutation was responsible for the observed phenotype.

Construction of relA disruptant. relA mutants, which harbor defects in their relA gene, encoding ppGpp synthetase I, together with relC mutants have been studied in great detail with respect to the stringent response represented by ppGpp accumulation (7). We therefore attempted to construct a T. thermophilus relA disruptant. T. thermophilus HB8 possesses a relA/spoT homologue (RSH _Tt ) encoding a 727-amino-acid protein (KEGG Genes Database [http://www.genome.jp/ ]). Based on a DBGET search, RSH _Tt was found to encode a HD His-Arp domain (amino acids 49 to 151), which is characteristic of metal-dependent phosphohydrolases (1), a conserved RelA/SpoT region (248 to 371), a TGS (for “ThrRS, GTPase, and SpoT”) domain (409 to 472), which is predicted to possess a predominant β-sheet structure (52, 63), and an aspartokinase domain (656 to 726), which binds specifically to a particular amino acid, leading to regulation of the linked enzyme (53). We successfully constructed T. thermophilus relA disruptants (KO-571 and KO-652) (see Materials and Methods) and found that they grew as well as the wild-type strain in nutritionally rich media (e.g., MTM medium) but lost completely the ability to accumulate ppGpp, as indicated by the absence of an effect of serine hydroxamate. For that reason, we will hereafter refer to RSH _Tt as relA.

Phenotypic characterization of relA and relC mutants.(i) Changes in nucleotide pool size.That the ability of the relA disruptant (KO-571) and relC mutant (KO-564) to accumulate ppGpp was severely impaired (100% and 80% reduction, respectively) was confirmed by monitoring the changes in nucleotide levels after serine hydroxamate treatment (Fig. 3A). Importantly, levels of GTP declined markedly (to less than half) in the wild-type strain (and also in the relC mutant), possibly due to potent inhibition of GMP synthesis by ppGpp (41), while its level even increased in the relA disruptant, possibly due to the complete absence of ppGpp synthesis under the growth-limiting (GTP consumption-limiting) condition caused by serine hydroxamate. In fact, an increase in the size of the GTP pool (about twofold) was also detected by treating the cells with a high concentration (25 μg/ml each) of streptomycin or chloramphenicol (data not shown). On the other hand, levels of ATP increased 3.5-fold in the wild-type strain, presumably due to direct and/or indirect inhibition of RNA synthesis (see below) by ppGpp (Fig. 3A). No such dramatic changes in ATP levels were observed in the relA disruptant. The intracellular concentrations of ATP and GTP (just before the addition of serine hydroxamate) and of ppGpp (5 min after the addition) were calculated as being 1.2 mM, 0.17 mM, and 10 μM, respectively, according to the method described previously (37).

• Open in new tab
• Download powerpoint

FIG. 3.

Changes in intracellular ppGpp, ATP, and GTP levels in the wild-type strain (HB8) and the relC and relA mutant strains after serine hydroxamate treatment. (A) Strains were grown to an OD₆₅₀ of 0.5 (mid-exponential phase) in MTM medium, after which serine hydroxamate was added to a final concentration of 10 mM. Concentrations of ppGpp, ATP, and GTP were determined by HPLC analysis. Circles, wild-type strain HB8; triangles, relC mutant KO-546; squares, relA disruptant KO-571. (B) Strains were grown in chemically defined medium (medium 162) containing phenylalanine (as required) and isoleucine, valine, and leucine (to promote the growth of the relA disruptant). When the OD₆₅₀ reached 0.5 (mid-exponential phase), cells were filtered with a membrane filter, transferred to the same chemically defined medium but lacking phenylalanine, and incubated for an additional 18 min. Nucleotide concentrations were determined as in panel A. Error bars indicate standard deviation. Circles, wild-type strain KO-572 (Phe⁻); squares, relA disruptant KO-652 (Phe⁻ relA).

The inability of the relA disruptant to accumulate ppGpp was further confirmed by depletion of a required amino acid in an auxotrophic strain (Fig. 3B). Strains KO-572 (Phe⁻ relA ⁺) and KO-652 (Phe⁻ relA) were grown in a chemically defined medium supplemented with phenylalanine, as required (see Materials and Methods). When cells reached mid-growth phase (OD₆₀₀ = 0.5), they were filtered and transferred to the same chemically defined medium lacking phenylalanine. In the absence of phenylalanine, wild-type cells immediately accumulated ppGpp to a level similar to that seen after serine hydroxamate treatment, accompanied by a marked decrease of the GTP level (Fig. 3B). By contrast, the relA disruptant cells showed no accumulation of ppGpp, in agreement with the results from serine hydroxamate addition.

(ii) RNA synthesis.One of the most characteristic components of the bacterial stringent response is suppression of RNA synthesis (7). Thus, RNA synthesis ([¹⁴C]uracil incorporation) was substantially reduced in wild-type T. thermophilus following treatment with serine hydroxamate (reduction rate = 60%) (Fig. 4). The same treatment reduced RNA synthesis in the relA disruptant by only 36% and in the relC mutant by an intermediate 52%.

• Open in new tab
• Download powerpoint

FIG. 4.

RNA synthesis by wild-type (strain HB8) or relC or relA mutant T. thermophilus. Cells were grown to an OD₆₅₀ of 0.5 (mid-exponential phase) in synthetic medium, after which [2-¹⁴C]uracil (0.1 μCi/ml, 100 μM) was added to the culture, with (open circles) or without (closed circles) serine hydroxamate (10 mM), and the incubation was continued for an additional 10 min with shaking.

RNA synthesis in rel ⁺ and relA mutant cells was further studied using RT-qPCR (see Materials and Methods), which enables one to measure the level of a specified RNA within cells. T. thermophilus HB8 harbors only two sets of rrn genes for rRNA synthesis (KEGG Genes Database [http://www.genome.jp/ ]). RT-qPCR was carried out focusing on two 23S-5S rRNA genes (23Sa and 23Sb), because these genes contain a putative discriminator sequence (GCGCTGC) in their promoter region and are thus considered to be under the control of the stringent response (16). A set of PCR primers (TNP-032) was designed to amplify the most 5′ region of the precursor transcript (see Materials and Methods), and the changes in the level of 23Sa-23Sb gene transcripts were monitored following serine hydroxamate treatment. As a reference, the rpoD gene product (encoding an essential sigma factor) was also monitored. As shown in Fig. 5, the level of 23Sa-23Sb transcripts in rel ⁺ cells declined markedly after serine hydroxamate treatment (by 78% at 5 to 20 min), whereas the relA mutant cells displayed a temporarily increased level of the transcripts. These features were always reproducible in separate experiments. Apparently, the reduction in the level of 23Sa-23Sb transcripts seen in rel ⁺ cells reflects a more pronounced reduction in the rate of RNA synthesis than is seen in relA mutant cells, as indicated by [¹⁴C]uracil incorporation (Fig. 4). It thus appears that like other bacterial relA mutants, the T. thermophilus relA mutant exhibits a typical Rel (relaxed) phenotype with respect to both ppGpp accumulation and RNA synthesis, while the relC mutant shows an intermediate phenotype (Fig. 4), apparently due to incomplete abolition of ppGpp synthesis (Fig. 3A). Although there was an unexpected increase of rpoD transcripts in relA mutant cells (Fig. 5), we have at present no explanation for this peculiar result.

• Open in new tab
• Download powerpoint

FIG. 5.

Determination of the levels of 23Sa-23Sb transcripts by RT-qPCR analysis following serine hydroxamate treatment. Culture conditions of strains are the same as in Fig. 3A, and 0.5 ng (for 23Sa-23Sb) or 50 ng (for rpoD) of total RNA was used for the reaction. Error bars indicate standard deviation from four experiments. Circles, wild-type strain HB8; squares, relA disruptant KO-571.

In order to resolve the question of whether the observed reduction in the rate of RNA synthesis in rel ⁺ cells is due to a direct effect of ppGpp or an indirect effect via the reduced level of GTP, we attempted to modulate the intracellular GTP level without an increase in the ppGpp level. However, unlike the case for B. subtilis and E. coli, neither mycophenolic acid nor decoyinine (known as inhibitors of IMP dehydrogenase and GMP synthetase, respectively) inhibited growth of T. thermophilus in chemically defined medium even at a high concentration (250 μg/ml and 2.8 mg/ml, respectively). In agreement with the whole-cell experiments, no inhibition of IMP dehydrogenase activity (see below) was detected in the presence of 200 μg/ml of mycophenolic acid, demonstrating insusceptibility of T. thermophilus IMP dehydrogenase to the drug.

In vitro ppGpp synthesis using the T. thermophilus ribosome.We next attempted to clarify the intrinsic mechanism of ppGpp synthesis in T. thermophilus by examining ribosome-dependent ppGpp synthesis in vitro (Fig. 6). E. coli W3110 served as a reference strain in these experiments. Ribosomes were isolated from exponentially growing cells, and the reaction mediating ppGpp synthesis was carried out according to the method of Martinez-Costa et al. (35). It is evident from Fig. 6A that ribosomes from wild-type T. thermophilus have the ability to produce ppGpp in addition to much greater amounts of pppGpp when the reaction mixture was incubated at 50°C; less than 10% of the product was ppGpp. Neither ppGpp nor pppGpp was detected in the absence of ribosome. Surprisingly, only trace amounts of ppGpp and pppGpp were produced at 70°C, although that temperature is optimal for growth of T. thermophilus. The observed low productivity was not due to the lability of ppGpp and pppGpp, as only 10% or less of a standard sample in water or buffer solution was degraded after 1 h at 70°C. Ribosomes from E. coli and T. thermophilus were also capable of producing ppGpp and pppGpp at 30°C but at a much lower level than was produced by the latter at 50°C (Fig. 6A). Thus, ribosomes from wild-type T. thermophilus show an enormous capacity to produce pppGpp at 50°C.

• Open in new tab
• Download powerpoint

FIG. 6.

Ribosome-dependent (p)ppGpp synthesis with in vitro assay system. (A) Synthesis of (p)ppGpp by purified ribosomes from T. thermophilus HB8 (and E. coli W3110 as a reference strain). The reaction was carried out for 45 min at various temperatures as described in Materials and Methods. (B) Comparison of in vitro (p)ppGpp synthesis between ribosomes from T. thermophilus wild-type (HB8), relC mutant (KO-564), and relA disruptant (KO-571) strains. Reactions were done at 50°C as for panel A. (C) Two-dimensional TLC analysis of nucleotides. Three microliters of reactions in panel B was subjected to analysis by the two-dimensional TLC method as described previously (25). (D) Effect of various antibiotics on in vitro pppGpp synthesis. Rifampin (Rif), tetracycline (Tet), or thiostrepton (Tsr) was added to the reaction mixture to a final concentration of 10 μg/ml, after which the reaction was run at 50°C for 30 min. As a negative control, dimethyl sulfoxide (DMSO), the solvent for the antibiotic solution, was also tested at a concentration of 4% (vol/vol). Intensities of the pppGpp spots were quantitated with Image Gauge version 3.41 software (Fuji Firm) and are indicated at the bottom.

In contrast to those from the wild-type strain, ribosomes from the KO-571 relA-null mutant had no ability to synthesize pppGpp, while those from the relC mutant retained reduced ability (Fig. 6B). These results were then confirmed using another analytical method: two-dimensional thin-layer chromatography (TLC) analysis (Fig. 6C). Antibiotics that block protein synthesis by binding to the ribosome are known to interfere with ribosome-dependent (p)ppGpp synthesis at a relatively low concentration (40). And as expected, in vitro pppGpp synthesis by ribosomes from the wild-type T. thermophilus HB8 was severely inhibited by addition of thiostrepton or tetracycline (10 μg/ml [each]) (Fig. 6D). By contrast, rifampin (tested as a reference drug), which blocks transcription by binding to RNA polymerase, had no effect on pppGpp synthesis. The results of Fig. 6D, together with the results from relA-relC mutant analyses (Fig. 6B and C), demonstrate unambiguously the existence of ribosome-dependent (p)ppGpp synthesis in T. thermophilus.

ppGpp inhibits IMP dehydrogenase activity.IMP dehydrogenase is the first enzyme of the pathway leading to GTP from IMP. Like the case for B. subtilis and Streptomyces (37, 41), IMP dehydrogenase from T. thermophilus was a target of (p)ppGpp. The activity of the T. thermophilus IMP dehydrogenase was measured with various concentrations of IMP (0.02 to 2 mM) in the presence or absence of (p)ppGpp (see Materials and Methods). The K_i values determined for ppGpp and pppGpp were 20 μM and 11 μM, respectively. These K_i values are somewhat lower than the previously reported K_i value (50 μM) for ppGpp for Streptomyces griseus (30), implying that ppGpp is capable of functioning more potently for T. thermophilus than for S. griseus. The marked and abrupt decline in the level of GTP in rel ⁺ cells accompanied by ppGpp accumulation (Fig. 3 A and B) can be accounted for in this way.

ppGpp inhibits RNA polymerase activity.To date, a number of biochemical and genetic events have been attributed to the activity of (p)ppGpp, mainly on the basis of results with E. coli (7). Among (p)ppGpp's functions, inhibition of RNA polymerase is the most prominent, leading to severe inhibition of certain genes, including the rrn genes for rRNA synthesis. We therefore undertook an in vitro analysis of the effects of ppGpp and pppGpp (and ppApp) on expression of the T. thermophilus 23Sb gene (driven by its own promoter) coding for 23S/5S rRNA, although the rRNA gene promoters from this organism have not yet been well characterized (16).

We purified RNA polymerase from T. thermophilus cells growing exponentially in MTM medium (see Materials and Methods); an SDS-polyacrylamide gel electrophoresis profile of the enzyme fraction at each purification step is shown in Fig. 7. RNA polymerase activity was determined by measuring [³H]UTP incorporation in the presence and absence of (p)ppGpp or ppApp. We found that RNA polymerase-catalyzed transcription of the 23S/5S rRNA gene was severely inhibited by increasing levels of ppGpp or pppGpp such that the activity was inhibited by 80% in the presence of 1 mM (p)ppGpp (Fig. 8A). It is worth mentioning that relatively low concentrations of (p)ppGpp (e.g., 0.05 mM) also negatively influenced RNA polymerase activity, though the observed inhibition was slight. It is thus concluded that in T. thermophilus both ppGpp and pppGpp could act by exerting an inhibitory effect on rRNA gene transcription. By contrast, ppApp displayed no effect, even at a high (1 mM) concentration. This conclusion was confirmed using an alternative method, the detection of the specific transcripts on the gel, in which the runoff transcripts from in vitro transcription reactions were separated (see Materials and Methods). The gel separation profile indicating the specific transcripts and a quantification of the effects of various concentrations of ppGpp on the level of the runoff transcripts are shown in Fig. 8B and C, respectively.

• Open in new tab
• Download powerpoint

FIG. 7.

SDS-polyacrylamide gel electrophoresis of the RNA polymerase fraction at each purification step. Lane 1, fraction eluted from the pellet after polyethylenimine P-70 precipitation; lane 2, peak fraction after calf thymus DNA-cellulose affinity chromatography; lane 3, peak fraction after superose 6 HR 10/30 gel filtration. α, β, and β′ represent subunits of RNA polymerase; σ represents σ-factor.

• Open in new tab
• Download powerpoint

FIG. 8.

Effect of ppGpp, pppGpp, and ppApp on in vitro transcription from the rRNA promoter. Effect of the indicated nucleotides on transcription of the 23S/5S rRNA operon (23Sb rRNA operon) was examined using an in vitro transcription system with purified T. thermophilus HB8 RNA polymerase (see Materials and Methods). (A) Effect of ppGpp, pppGpp, and ppApp on in vitro transcription, determined by measuring the radioactivity of the membrane filter as described in Materials and Methods. As a negative control, the reaction was also run without a template (dotted line). (B) Gel electrophoretic separation of runoff transcripts from in vitro transcription reactions with the 23S/5S rRNA operon promoter. ppGpp was added at the final concentration of 0, 0.05, 0.1, 0.2, 0.5 or 1 mM, and the products were separated on denaturing polyacrylamide gel as described in Materials and Methods. Runoff and paused products are indicated. (C) Quantitative evaluation of the band intensities of runoff transcripts shown in panel B.

The T. thermophilus 23S/5S rRNA promoter requires a high concentrations of GTP in vitro.Krasny and Gourse (29) reported recently that B. subtilis rrn P1 promoters require high concentrations of their initiating NTP for transcription in vitro. Therefore, we investigated potential mechanisms for regulation of T. thermophilus rRNA promoter activity. We measured the effects of GTP and ATP concentrations in vitro on the 23Sb gene promoter. As shown in Fig. 9, the 23Sb gene promoter required high levels of GTP, but not ATP, for maximal transcription. It is notable that 23Sb gene promoter activity increased in parallel with increasing GTP concentrations even up to 5 mM (Fig. 9B). Thus, like B. subtilis rRNA promoters, the T. thermophilus rRNA promoter displayed GTP dependence in vitro.

• Open in new tab
• Download powerpoint

FIG. 9.

Effects of GTP and ATP concentration on 23S/5S rRNA (23Sb gene) promoter activity in vitro. (A) In vitro transcription was carried out with 0.5 mM (each) ATP, GTP, CTP, and UTP, except when ATP or GTP was varied at the indicated concentrations. The transcripts were separated on denaturing polyacrylamide gel as for Fig. 8B. (B) Intensities of the runoff transcripts shown in panel A.