A RelA–SpoT homolog (Cr-RSH) identified in Chlamydomonas reinhardtii generates stringent factor in vivo and localizes to chloroplasts in vitro

Abstract

A gene encoding a putative guanosine 3′,5′-bispyrophosphate (ppGpp) synthase–degradase, designated Cr-RSH, was identified in the unicellular photosynthetic eukaryote Chlamydomonas reinhardtii. The encoded Cr-RSH protein possesses a putative chloroplast-targeting signal at its NH₂-terminus, and translocation of Cr-RSH into chloroplasts isolated from C.reinhardtii was demonstrated in vitro. The predicted mature region of Cr-RSH exhibits marked similarity to eubacterial members of the RelA–SpoT family of proteins. Expression of an NH₂-terminal portion of Cr-RSH containing the putative ppGpp synthase domain in a relA, spoT double mutant of Escherichia coli complemented the growth deficits of the mutant cells. Chromatographic analysis of ³²P-labeled cellular mononucleotides also revealed that expression of Cr-RSH in the mutant bacterial cells resulted in the synthesis of ppGpp. SpoT, which catalyzes (p)ppGpp degradation, is dispensable in E.coli only if cells also lack RelA, which possesses (p)ppGpp synthase activity. The complementation analysis thus indicated that Cr-RSH possesses both ppGpp synthase and degradase activities. These results represent the first demonstration of ppGpp synthase–degradase activities in a eukaryotic organism, and they suggest that eubacterial stringent control mediated by ppGpp has been conserved during evolution of the chloroplast from a photosynthetic bacterial symbiont.

INTRODUCTION

Guanosine 3′,5′-bispyrophosphate (ppGpp), also known as magic spot, alarmone and stringent factor, is a mononucleotide that is present in Escherichia coli and other eubacteria (1). Accumulation of ppGpp is often accompanied by pppGpp. Collectively, pppGpp and ppGpp are called (p)ppGpp. ppGpp mediates the stringent response in eubacteria, which results in a reduction in the abundance of stable RNAs and induction of stationary phase-specific gene expression (2–4). Amino acid deprivation and the consequent increase in the proportion of uncharged tRNA are thought to induce activation of ppGpp synthase, a ribosome-bound enzyme encoded by the gene relA (5). RelA mutants lacking ppGpp exhibit a ‘relaxed’ phenotype; that is, they do not undergo the stringent response (1). Genetic and biochemical studies have shown that ppGpp binds directly to RNA polymerase (6,7) and thereby stalls transcription at stringent promoters such as rrnB P1 (8). Thus, ppGpp rapidly signals a change in translational conditions to the transcriptional apparatus, allowing for adaptation to an environment with variable nutrient availability. The product of spoT degrades (p)ppGpp and thereby prevents prolongation of the stringent response (9). The (p)ppGpp generated in E.coli is thus rapidly degraded by SpoT, and SpoT is dispensable only if RelA activity is impaired (9). Under certain conditions, such as carbon or phosphate deficiency, SpoT also exhibits weak, but not insubstantial, ppGpp synthase activity, and this activity is essential for maintaining the basal level of amino acid biosynthesis (10,11). Although the ppGpp-mediated stringent response appears to be ubiquitous in the eubacterial kingdom (1), some eubacterial species, such as Bacillus subtilis, possess only a single relA–spoT homolog (12). The protein encoded by this B.subtilis gene exhibits both ppGpp synthase and degradase activities.

Transcription and translation in eubacteria are mediated by an α₂ββ′σ-type RNA polymerase and the 70S ribosome, respectively, both of which are conserved in the chloroplasts of photosynthetic organisms (13). The ribosomal proteins of spinach chloroplasts exhibit marked similarity to their bacterial counterparts (14,15). The function of the plastid α₂ββ′σ-type RNA polymerase has also been confirmed both genetically and biochemically (16–19). Therefore, we hypothesized that transcription and translation in chloroplasts might also be coupled by ppGpp signaling.

A database search revealed a candidate RelA–SpoT homolog in the unicellular photosynthetic eukaryote Chlamydomonas reinhardtii. We have now characterized this protein and here demonstrate its function as a ppGpp synthase–degradase, its chloroplast localization, and its ability to synthesize detectable levels of ppGpp in vivo.

MATERIALS AND METHODS

The nucleotide sequence of the Cr-RSH cDNA described in this work has been deposited in GenBank under the accession no. AB073639.

Strains and culture conditions

The C.reinhardtii wild-type strain cc-125 (mt⁺) was cultured in Tris–acetate–phosphate (TAP) medium (20) at 25°C under continuous white light with rotary shaking. The TW3 strain (thi10 cw15 mt⁺), which harbors the cell-wall deficiency mutation cw15 (21), was used for chloroplast isolation.

Isolation and analysis of DNA

Standard recombinant DNA techniques were performed basically as described (22). The polymerase chain reaction (PCR) was performed with total DNA from C.reinhardtii as template and the oligonucleotide primers Crrsh-1 (5′-CCG AACTGGCGTCAGTCTACTC-3′) and Crrsh-2 (5′-CGG CCGCACCTTCTTGTC-3′), which were based on the sequence of C.reinhardtii expressed sequence tags (ESTs) (DDBJ accession nos. AV629944 and AV631500) that show homology to bacterial relA and spoT genes. The 50-µl reaction mixture contained 1× GCI buffer (Takara), 20 ng of total DNA, 0.2 mM of each deoxynucleoside triphosphate, 1 mM of each primer and 2.5 U LA-Taq DNA polymerase (Takara). Amplification was performed in a GeneAmp PCR system 9700 (PE Applied Biosystems) with a protocol comprising an initial denaturation step at 96°C for 2 min followed by 30 cycles of denaturation at 95°C for 20 s, annealing at 58°C for 10 s and elongation at 72°C for 1 min. The PCR product was cloned into the pCR2.1 vector (Invitrogen) and sequenced. The resulting plasmid was digested with EcoRI, and the released Cr-RSH fragment was labeled with digoxigenin with the use of a DIG DNA Labeling Kit (Roche Diagnostics) to prepare a probe for gene screening and Southern hybridization.

Total RNA was prepared as described (23) from C.reinhardtii cc-125 that had been grown to a density of 1 × 10⁶ cells/ml at 25°C in the light. Polyadenylated [poly(A)⁺] RNA was purified from total RNA with a µMACS mRNA Isolation Kit (Miltenyi Biotec), and cDNA was synthesized from the poly(A)⁺ RNA with the use of a TimeSaver cDNA Synthesis Kit (Amersham Pharmacia Biotech). The synthesized cDNA was cloned into lambda-ZAP II (Stratagene) that had been digested with EcoRI and treated with calf intestinal alkaline phosphatase and was packaged with the use of a Gigapack III Gold Packaging Extract Kit (Stratagene). The resulting cDNA phage library was used to infect E.coli XLIblueMRF′ on NZY agar plates (235 × 235 mm; Nunc) supplemented with 10 mM MgSO₄. The ∼150 000 plaques formed on each plate were transferred to a Hybond-N⁺ membrane (220 × 220 mm; Amersham Pharmacia Biotech), and the phage DNA was fixed to the membrane with the use of an ultraviolet-activated cross-linker (UV Stratalinker; Stratagene). Plaque hybridization with the digoxigenin-labeled Cr-RSH probe was performed overnight at 50°C in Dig Easy Hyb solution (Roche Diagnostics). The membrane was washed sequentially twice (5 min each time) at room temperature with 1× standard saline citrate (SSC) containing 0.1% SDS, and once at 60°C for 15 min with 0.5× SSC containing 0.1% SDS. Hybridized probe was detected with a DIG Luminescent Detection Kit (Roche Diagnostics) and Hyperfilm (Amersham Pharmacia Biotech).

Genomic DNA prepared from C.reinhardtii was partially digested with Sau3AI, and the resulting fragments of ∼10– 15 kb were size-fractionated by Tris–acetate–EDTA (TAE) agarose gel (0.8%) electrophoresis and cloned into lambda-DASH II (Stratagene). The resulting genomic DNA library was used to infect E.coli XLIblueMRF′, and the plaques were screened as described above for cDNA library screening.

DNA sequencing was performed with a DYEnamic ET Terminator Cycle Sequencing Premix Kit (Amersham Pharmacia Biotech) and an ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems).

Genomic Southern hybridization

Genomic DNA was digested with restriction endonucleases, fractionated by electrophoresis through a TAE agarose gel, and transferred to a Hybond-N⁺ membrane, which was then subjected to hybridization with the digoxigenin-labeled Cr-RSH probe in DIG Easy Hyb for 14 h at 50°C. The membrane was washed for 15 min at 60°C with 0.1× SSC containing 0.1% SDS. Hybridized probe was detected as described above for plaque hybridization.

Northern hybridization

Northern blot analysis was performed with poly(A)⁺ RNA isolated from C.reinhardtii cc-125. The RNA was fractionated by electrophoresis through a 1.4% agarose gel in 10 mM sodium phosphate buffer (pH 7.0), denatured with 50 mM NaOH, and transferred to a Hybond-N⁺ membrane. A ³²P-labeled Cr-RSH probe was prepared with the use of a rediprime II DNA random-prime labeling system (Amersham Pharmacia Biotech). The membrane was subjected to hybridization for 14 h at 45°C with the probe in a solution containing 50% formamide, 5× SSC, 50 mM sodium phosphate (pH 7.0), 0.1% SDS, denatured salmon sperm DNA (50 µg/ml) (Wako) and 1× Denhardt’s solution. The blots were washed for 15 min at 50°C with 0.2× SSC containing 0.1% SDS, after which hybridization signals were detected by exposure for 24 h at –80°C to X-ray film (RX-U; Fuji Film) with an intensifying screen.

Chloroplast isolation

Chloroplasts were isolated as described (24), with slight modifications. The TW3 strain of C.reinhardtii was harvested after growth at 28°C in 2 l of TAP medium (with rotary shaking) under a cycle of 16 h of light and 8 h of darkness to a density of ∼2 × 10⁶ cells/ml. The cells were collected by centrifugation at 4000 g for 2 min at room temperature, and the pellets were rinsed and resuspended in 30 ml of P-medium [375 mM sorbitol, 35 mM HEPES–KOH (pH 7.7), 1 mM MnCl₂, 5 mM MgCl₂, 2 mM EDTA (K⁺ salt)]. The cells were disrupted in an ice-cold Parr Cell Disruption Bomb (model 4639) by incubation for 3 min at 5 bar with N₂ gas. After adjustment of the K⁺-EDTA concentration to 8 mM, the cell lysate was incubated for 10 min at 0°C with gentle shaking and then centrifuged at 5000 g for 10 s at 4°C. The pellet was suspended to a volume of 25 ml with P-medium, and an equal volume of 80% Percoll (Sigma) in P-medium was added to the suspension and mixed gently but thoroughly. The mixture was divided into two, and each half was layered on top of 15 ml of 60% Percoll in P-medium placed in centrifuge tubes. After centrifugation at 6000 g for 20 min at 4°C, the intact chloroplasts present at the interface of the two Percoll layers were collected, diluted with ∼4 vol of P-medium, and then separated by centrifugation at 5000 g for 10 s at 4°C. The pellet was washed with 5 ml of P-medium and resuspended in the same medium.

In vitro assay of protein import into chloroplasts

To produce the Cr-RSH precursor for in vitro import assays, the Cr-RSH cDNA, including the sequence for the putative transit peptide, was cloned into pEU3-N (Toyobo), a T7 transcription vector for in vitro translation (25). To isolate the Cr-RSH cDNA, we performed PCR with the primers p-eukks-a (5′-ACCCGGATATCTAGGATGGCCGCAACC-3′, which creates an EcoRV restriction site) and p-eukks-b (5′-TCCGTACTAGTCTAGCCCTGCGGCCGCAC-3′, which generates a SpeI recognition site and stop codon). The cDNA clone pCr-rsh6.1, which resulted from in vivo excision of the pBluescript SK(–) phagemid from the lambda-ZAP II clone CRrelA6.1, was used as the template. The PCR product was digested with EcoRV and SpeI, and the released fragment was ligated into the EcoRV–SpeI sites of pEU3-N. The resulting plasmid was designated pEUKKS3.2b.

The template RNA for the Cr-RSH precursor was synthesized by in vitro transcription with T7 RNA polymerase. The 100-µl reaction mixture, which contained 1× transcription buffer (Roche), 25 mM each of ATP, CTP, GTP and UTP (Roche), 2 µl of ribonuclease inhibitor (40 U/µl; Wako), 10 µg of pEUKKS3.2b and SP6 RNA polymerase (Roche), was incubated for 3 h at 37°C, after which the synthesized RNA was purified with a MicroSpin G-25 column (Amersham Pharmacia Biotech).

The ³⁵S-labeled Cr-RSH precursor was synthesized by in vitro translation in the presence of [³⁵S]methionine (0.4 mCi/ml) with a Proteios wheat germ cell-free protein synthesis kit (Toyobo). The reaction was performed for 3 h at 26°C.

Incorporation analysis was performed with the ³⁵S-labeled precursor as described (26), with some modifications. Chloroplasts were suspended in import buffer [300 mM sorbitol, 50 mM HEPES–KOH (pH 8.0)] at a density of 1 × 10⁸/ml. The translation products were diluted 4-fold in import buffer supplemented with 25 mM methionine. The import assay was carried out at 25°C for 20 min under white light in a 150-µl reaction mixture, which comprised 30 µl of diluted translation products, 100 µl of chloroplasts, 1 mM magnesium acetate, 1 mM dithiothreitol and 10 mM ATP. Assays were also performed in the dark in foil-wrapped tubes without ATP. After the addition of thermolysin (final cocncentration 100 µg/ml; Nacalai Tesque) and CaCl₂ (final concentration 1 mM), the reaction mixture was incubated further for 30 min on ice. Proteolysis was terminated by the addition of 3 µl of 0.5 mM EDTA. Intact chloroplasts were separated by centrifugation through 35% Percoll in import buffer, washed with import buffer containing 10 mM EDTA, isolated by centrifugation, and resuspended in 20 µl of 10 mM EDTA. After the addition of an equal volume of 2× SDS sample buffer, the chloroplasts were heated for 10 min at 67°C and portions (20 µl) were subjected to SDS–polyacrylamide gel electrophoresis (PAGE) through a 12% gel. The gel was dried and exposed to a Fuji imaging plate (BAS-III; Fuji Film), and radioactivity was detected with a Storm860 analyzer (Molecular Dynamics).

Complementation test of Cr-RSH in E.coli

The E.coli strains CF1652 (relA^–) and CF1678 (relA^–, spoT^–) were prepared as recipients of test plasmids by the lysogenic introduction of λ-DE3 phage (Novagen) into their chromosomal DNA in order to confer isopropyl-β-d-thiogalactopyranoside (IPTG)-inducible gene expression from the T7 promoter (27). The wild-type strain W3110 was used as a control.

For construction of an expression vector for Cr-RSH, the cDNA clone pCr-rsh6.1 was subjected to PCR with the primers Cr-rsh16 (5′-TCCAAGCTTCGCAGCCCTACCATGGC-3′) which generates HindIII and NcoI sites and Cr-rsh17 (5′-CATTCTTGCGCTGCTGGTTTTGATGC-3′). The PCR product was digested with HindIII and MluI, and the released fragment was substituted for the HindIII–MluI region of pCr-rsh6.1. This procedure introduced an NcoI site at the ATG start codon of the Cr-RSH cDNA; the resulting plasmid, pCr-rsh7.1, was digested with NotI, rendered blunt-ended with T4 DNA polymerase (Takara), digested with NcoI, and ligated into the NcoI site of the pET11d expression vector (Novagen) that had been cleaved with BamHI and rendered blunt-ended. The resulting plasmid, which contained a 1674-bp DNA fragment encoding the 558 NH₂-terminal residues of Cr-RSH, was designated pETKK1.3. Cells were transferred to MOPS agar medium [40 mM MOPS and 4 mM Tricine buffer (adjusted to pH 7.2 with KOH), 50 mM KCl, 10 mM NH₄Cl, 0.5 mM MgSO₄, 0.2 mM KH₂PO₄, 10 µM FeCl₃, 1.6% agar] (28) containing 0.4% glucose, and were grown at 37°C for 24 h.

Detection of ppGpp

Analysis of ppGpp accumulation in Cr-RSH-expressing E.coli mutant cells was performed as described (29), with minor modifications. Cells were grown in MOPS medium containing 0.4% glucose and 0.2% casamino acids at 37°C, with vigorous shaking, to an optical density at 650 nm of 0.05. The cells were then labeled with H₃³²PO₄ (100 µCi/ml; Amersham Pharmacia Biotech) for 3 h, after which the expression of Cr-RSH was induced by the addition of IPTG to the culture at a final concentration of 0.1 mM. After an appropriate incubation time, the ³²P-labeled cells were mixed with an equal volume of 8 M formic acid and subjected to extraction by three cycles of freezing and thawing. The extracts were centrifuged at 8000 g for 5 min at 4°C, and 5 µl of the resulting supernatant were analyzed by thin-layer chromatography on PEI-cellulose with 1.5 M KH₂PO₄. Labeled mononucleotides were detected by autoradiography.

RESULTS

Structure of the relA–spoT homolog of Chlamydomonas

A search of the Chlamydomonas EST database (http://www. kazusa.or.jp/en/plant/chlamy/EST/) with reference to the amino acid sequence of B.subtilis RelA yielded nine candidate clones out of a total of 37 990. These nine clones were classified into two groups (one comprising AV395927, AV627399, AV629944, AV630246, AV640334 and AV644170, and the other including AV631500, AV628611 and AV645291) on the basis of the observation that the members of each group appeared to be derived from the same cDNA fragment. The deduced amino acid sequences of these two groups of clones showed substantial homology to those of different regions of B.subtilis RelA. PCR analysis with primers based on these two sequence regions and with C.reinhardtii genomic DNA as a template revealed that the separated EST sequences were indeed derived from a single gene. The PCR product was then used as a probe to isolate a full-length cDNA from 150 000 independent clones in a phage library. Two clones, CRrelA5.1 and CRrelA6.1, that contained full-length open reading frames (ORFs) with identical nucleotide sequences were isolated. The insert of CRrelA5.1 comprised 3839 bp with a predicted ORF of 736 amino acids. We designated this relA–spoT homolog Cr-RSH (C.reinhardtii relA–spoT homolog).

We next characterized the genomic structure of Cr-RSH. The cDNA sequence was found to be derived from seven exons (Fig. 1A). Southern hybridization of fragmented genomic DNA under moderately stringent conditions with a cDNA probe indicated that Cr-RSH is a single-copy gene (Fig. 1B). The amino acid sequence encoded by Cr-RSH includes a typical chloroplast-targeting signal sequence, which is rich in serine and threonine and poor in aspartate and glutamate (30), in the NH₂-terminal region. Three programs that predict the cellular localization of proteins, iPSORT (31), TargetP ver. 1.0 (32) and ChloroP ver. 1.1 (33), indicated localization of Cr-RSH to chloroplasts (34). The putative mature protein shows marked homology to the RelA–SpoT family of proteins from eubacteria and land plants, being 33% identical to E.coli RelA (GenBank accession no. J04039), 37% identical to E.coli SpoT (M24503), 33% identical to B.subtilis RelA (U86377), 29% identical to Arabidopsis At-RSH1 (AF225702) and 36% identical to Arabidopsis At-RSH2 (AF225703) (Fig. 2). Cr-RSH also contains a glycine residue that is highly conserved among various bacterial RelA–SpoT homologs and whose mutation to glutamine in B.subtilis RelA gives rise to the relaxed relA1 mutant (12). The sequence motif thought to underlie the ppGpp degradation activity of SpoT homologs (12) is partially conserved in Cr-RSH (Fig. 2).

Figure 1.

Genomic structure of Cr-RSH. (A) Intronic–exonic organization of Cr-RSH. The gene contains seven exons (open boxes) and six introns (closed boxes). The ORF (open box) and 5′- and 3′-untranslated regions (gray boxes) of the cDNA are also indicated. (B) Genomic Southern blot analysis of Cr-RSH. Genomic DNA (3 µg) of C.reinhardtii was digested with NcoI (lane 1), SacI (lane 2), SmaI (lane 3), BamHI (lane 4) or SalI (lane 5), the recognition sites for which in Cr-RSH are shown in (A), and was then probed with digoxigenin-labeled Cr-RSH cDNA. The positions of molecular size standards are shown on the left.

Figure 2.

Comparison of the deduced ppGpp synthase region of Cr-RSH with those of B.subtilis RelA (Bs-RelA), E.coli RelA (Ec-RelA) and SpoT (Ec-SpoT), and Arabidopsis thaliana RSH1 and RSH2 (At-RSH1 and At-RSH2). The boxed region indicates a sequence motif associated with ppGpp degradation activity. The conserved glycine residue whose mutation abolishes ppGpp synthase activity in the B.subtilis relaxed strain IS56 is indicated by an asterisk.

Expression of Cr-RSH

The expression of Cr-RSH during the growth of C.reinhardtii was analyzed by northern hybridization (Fig. 3). Strain cc-125 of C.reinhardtii was grown in TAP synthetic medium under continuous light, and cells were harvested for preparation of RNA during the early and late growth phases. Although a single transcript of 3.9 kb, similar in size to the isolated cDNA, was detected at all times, its abundance was higher in the late phase of growth than in the early phase.

Figure 3.

Analysis of the expression of Cr-RSH during algal growth. (A) Growth curve of C.reinhardtii strain cc-125 cultured in TAP medium at 25°C with vigorous shaking and under white light. Arrows indicate sampling times for the isolation of total RNA. OD₇₅₀, optical density at 750 nm. (B) Northern blot analysis of Cr-RSH mRNA. Poly(A)⁺ RNA (1 µg) isolated from cells at 24-h intervals during culture was subjected to northern hybridization with a ³²P-labeled Cr-RSH cDNA probe. The positions of molecular size standards are indicated on the left. Lane numbers correspond to the sampling times indicated in (A). As a control, the filter was reprobed with CBLP cDNA (58).

Chloroplast localization of Cr-RSH

We next determined whether ³⁵S-labeled Cr-RSH produced by in vitro transcription and translation is incorporated into active chloroplasts isolated from C.reinhardtii. In Chlamydomonas, protein relocalization from the cytosol to chloroplasts depends on light and ATP (26). The in vitro protein import assay revealed that the precursor Cr-RSH was incorporated into chloroplasts and proteolytically processed in a light- and ATP-dependent manner (Fig. 4).

Figure 4.

Incorporation of Cr-RSH into Chlamydomonas chloroplasts in vitro. The [³⁵S]methionine-labeled Cr-RSH protein which lacks 177 amino acid residues of the C-terminal region was incubated with intact chloroplasts for 20 min at 25°C in the absence (lanes 2 and 3) or presence (lanes 4 and 5) of light and 10 mM ATP. After further incubation in the absence (lanes 2 and 4) or presence (lanes 3 and 5) of thermolysin, the chloroplasts were separated by centrifugation and subjected to SDS–PAGE and autoradiography. Lane 1 contains the ³⁵S-labeled precursor protein synthesized by in vitro transcription and translation. The positions of molecular size standards are shown on the left. The positions of the precursor and imported proteins are indicated on the right.

Functional complementation of an E.coli spoT mutant by Cr-RSH

In E.coli, both RelA and SpoT have been extensively characterized as key factors in the stringent response (35,36). Both relA mutants and relA, spoT double mutants of E.coli have been produced (37) and used for functional complementation analysis of the corresponding genes from other bacterial species (35,38). Disruption of spoT results in lethality in relA⁺ strains as a consequence of the loss of degradative activity required to quench the ppGpp signal for transcriptional repression. In contrast, spoT is dispensable in relA mutants that are not able to accumulate ppGpp (37). Escherichia coli relA, spoT double mutants are unable to grow on solid minimal medium, whereas relA mutants (spoT⁺) are able to grow on minimal medium plates but not on minimal medium supplemented with serine, methionine and glycine (SMG medium) (37,39). These phenotypes have provided insight into the catalytic functions of RelA–SpoT family proteins (35,36,38).

To examine whether Cr-RSH functions as a relA–spoT homolog, we cloned a region of the cDNA encoding the putative ppGpp synthase domain into the expression vector pET11d to yield pETKK1.3. Escherichia coli CF1652DE3 (λDE3, relA^–, spoT⁺), CF1678DE3 (λDE3, relA^–, spoT^–) and W3110DE3 (λDE3, relA⁺, spoT⁺), each harboring an IPTG-inducible T7 RNA polymerase gene in lysogenized λDE3, were transformed with the plasmids pETKK1.3 or pET11d. The exogenous Cr-RSH cDNA restored the growth of the relA, spoT double mutant of E.coli on MOPS minimal agar medium containing 0.4% glucose in the absence or presence of 0.01 mM IPTG (Fig. 5A); growth was not observed in the presence of 0.1 mM IPTG, however, suggesting that overexpression of Cr-RSH results in toxicity, as has previously been observed with E.coli relA (40). We also demonstrated that transformation with pETKK1.3 allowed the relA single mutant to grow on SMG medium (data not shown). These results indicate that a low level of Cr-RSH expression is sufficient to complement both the relA^–, spoT^– and relA^– phenotypes, and they therefore suggest that Cr-RSH functions as a ppGpp degradase similar to SpoT as well as a ppGpp synthase.

Figure 5.

Functional analysis of Cr-RSH in E.coli mutants. (A) Restoration of growth in an E.coli relA, spoT double mutant (CF1678) plated on minimal medium by expression of Cr-RSH. Escherichia coli strains CF1678 and W3110 (wild-type) were transformed with pETKK1.3 (encoding Cr-RSH) or the empty vector (pET11d) as described in Materials and Methods and then transferred to MOPS agar medium supplemented with 0.4% glucose, as indicated in (a). Plates were supplemented with IPTG at concentrations of 0 mM (b), 0.01 mM (c) or 0.1 mM (d) and incubated for 24 h at 37°C. (B) Accumulation of (p)ppGpp in CF1678 cells expressing Cr-RSH. ³²P-Labeled CF1678 cells transformed with pET11d (lanes 3–5) or pETKK1.3 (lanes 6–8) were incubated for 0, 15 or 30 min, respectively, with 0.1 mM IPTG, after which nucleotides were extracted and analyzed by thin-layer chromatography and autoradiography. ³²P-Labeled nucleotides extracted from E.coli W3110 before (lane 1) or after (lane 2) incubation for 10 min with serine hydroxamate (1 mg/ml) were used as controls.

ppGpp accumulation in an E.coli relA, spoT double mutant expressing Cr-RSH

To confirm the ppGpp synthase activity of Cr-RSH, we examined the effect of expression of Cr-RSH on the abundance of ppGpp in the E.coli relA^–, spoT^– double mutant. The expression of Cr-RSH in ³²P-labeled E.coli mutant cells was induced with 0.1 mM IPTG and nucleotides were subsequently analyzed by thin-layer chromatography. A substantial amount of ppGpp and a smaller amount of pppGpp were detected after incubation of cells with IPTG for 15 min but not after 30 min (Fig. 5B). No ppGpp or pppGpp signals were detected in double-mutant cells transformed with pET11d. These data thus indicate that Cr-RSH possesses ppGpp synthase activity and confirm that complementation of the relA^– phenotype in E.coli was attributable to this enzymatic activity of Cr-RSH.

DISCUSSION

Several studies have attempted to detect ppGpp in eukaryotic cells (41). Evidence suggests that lower eukaryotes synthesize ppGpp in response to various environmental stresses. Heat shock of Saccharomyces cerevisiae thus induced the synthesis of a compound identified as ppGpp on the basis of its comigration with the authentic nucleotide in a two-dimensional chromatographic system (42). A compound identified as ppGpp in the same manner was also detected in two auxotrophic mutants of C.reinhardtii that required arginine or acetate for growth in minimal medium (43). Attempts to detect ppGpp in cultured mammalian cells have failed (41). Attempts to demonstrate ppGpp synthesis with E.coli RelA and eukaryotic ribosomes isolated from yeast cytoplasm, rabbit reticulocytes, calf brain (44) or mouse embryos (45) have also failed. In contrast, ribosomes isolated from wheat germ (46) or yeast mitochondria (44) mediated a low level of ppGpp synthesis. Furthermore, ribosomes isolated from the chloroplasts, but not the cytoplasm, of C.reinhardtii mediated substantial ppGpp synthesis (47).

In Arabidopsis, the RelA–SpoT homolog At-RSH1 was identified by the yeast two-hybrid system as a protein that interacts with the pathogen resistance protein RPP5 (36). At-RSH1 complemented the growth defect of an E.coli relA mutant but not that of a relA, spoT double mutant. In addition, At-RSH1 restored antibiotic production in a Streptomyces coelicolor relA mutant. However, despite the proposal that At-RSH1 functions in a ppGpp-mediated defense response to pathogens or other stresses (36), neither the ability to synthesize or degrade ppGpp nor a chloroplast localization has been demonstrated for Arabidopsis RelA–SpoT homologs. The ability of a RelA–SpoT homolog to synthesize ppGpp has thus not previously been demonstrated in a photosynthetic organism.

We have now shown that Cr-RSH, a Chlamydomonas RelA–SpoT homolog, both possesses ppGpp synthase activity and is imported into chloroplasts in vitro. The ppGpp synthase activity of Cr-RSH was directly demonstrated by chromatographic analysis of nucleotides extracted from a relA, spoT double mutant of E.coli transformed with a Cr-RSH expression vector. We have also provided indirect evidence that Cr-RSH also possesses ppGpp degradase activity. In E.coli, spoT is indispensable if ppGpp synthase activity is not impaired (9). The growth conferred by expression of Cr-RSH in the relA, spoT double mutant of E.coli thus suggests that Cr-RSH possesses substantial ppGpp degradase activity as well as ppGpp synthase activity. This conclusion is also consistent with the presence of the conserved SpoT-like sequence in the NH₂-terminal region of Cr-RSH. The demonstration of import of the Cr-RSH precursor protein into chloroplasts in vitro is also consistent with the subcellular localization predicted for this protein by computer programs. The observed size of the processed protein, ∼44 kDa, indicated that the putative cleavage site is located around residue number 170–180 from the N-terminal end of Cr-RSH. This result is consistent with the peptide sequence alignment provided in Figure 2 that shows the conserved regions of the ppGpp synthase protein. In lane 5 of Figure 4, the protected protein after thermolysin treatment migrates slightly faster than the imported protein in lane 4 in the SDS–PAGE. This result was reproducible in four independent experiments. The reason why such a migration difference occurred is still unclear. As a possible explanation for this observation, we assume that a small portion of the sorted protein remained outside of the chloroplast membrane and was degraded by thermolysin.

Chloroplast counterparts of the E.coli ribosomal protein L11 have been identified in spinach (15), Arabidopsis (48) and Chlamydomonas (49). Genetic analyses in E.coli, B.subtilis and S.coelicolor have revealed that L11 is required for activation of RelA catalytic function in living cells. A mutation in E.coli rplK, which encodes L11, thus results in the relaxed phenotype, similar to that conferred by mutation of relA (50,51). Ribosomes containing the mutant L11 protein bind RelA but induce only 10% of the ppGpp synthetic activity apparent with wild-type ribosomes (50). The plastid peptide elongation factors EF-G and EF-Tu are also highly homologous to their bacterial counterparts (52,53). Mutation of EF-G perturbs the accumulation of (p)ppGpp in E.coli (54). Therefore, it is likely that chloroplasts also possess a mechanism for generating ppGpp that depends on the eubacterial-type 70S ribosome and peptide elongation factors. The transcriptional machinery of plastids, consisting of an RNA polymerase core subunit and associated sigma factors (13,19,55), also resembles that of eubacteria. These observations thus support the notion that stringent control exists in chloroplasts; that is, as in bacteria, ppGpp likely functions to regulate transcription in response to changes in translational conditions. Such coordinated regulation of transcription and translation in chloroplasts might be critical for maintaining the efficiency of important chloroplast functions such as photosynthesis.

Previously, it was suggested that chloroplast RNA synthesis in Chlamydomonas is controlled by stringent conditions (56). However, it is not clearly understood which type of RNA (e.g. rRNA or tRNA) is regulated. Also, there has been no particular evidence of the reduction of chloroplast rRNA synthesis under stringent response (56,57). To clarify whether Chlamydomonas shows stringent response to nutrient shift down, we further attempted to detect the induction of ppGpp synthesis and arrest of rRNA synthesis by depleting amino acids, phosphate or carbon sources from the culture medium (data not shown). However, we have not yet observed the expected results from these trials. The induction of ppGpp by stringent conditions could not be detected, even though the ppGpp spot itself was apparent as described previously (43). To elucidate the physiological roles of Cr-RSH as a ppGpp synthase in chloroplasts, we are now looking into what kind of conditions lead to the accumulation of ppGpp in Chlamydomonas chloroplasts.

To date, the existence of a RelA–SpoT homolog protein has been confirmed not only in Chlamydomonas, but also in higher plants such as Arabidopsis and rice (36; Y.Tozawa, H.Satsu, K.Kasai, Y.Ito, Y.Endo and K.Ochi, in preparation). There fore, we suggest that chloroplast requires the RelA–SpoT like protein as a ubiquitous component of its genetic system.

DDBJ/EMBL/GenBank accession no. AB073639