The effect of intracellular ppGpp levels on glutamate and lysine overproduction in Escherichia coli

Journal of Biotechnology 125 (2006) 328–337

The effect of intracellular ppGpp levels on glutamate and lysine overproduction in Escherichia coli Akira Imaizumi ∗ , Hiroyuki Kojima, Kazuhiko Matsui Laboratories of Fermentation and Biotechnology, Ajinomoto, CO., Inc., Kawasaki-shi, Kanagawa, 210-8681, Japan Received 14 December 2005; received in revised form 23 February 2006; accepted 13 March 2006

Abstract Although the enhancement of amino-acid synthesis by guanosine-3 ,5 -tetraphosphate (ppGpp) is well known, the effect of intracellular ppGpp levels on amino-acid overproduction in Escherichia coli has not been investigated. In this study, we demonstrate that overexpression of the relA gene, encoding ppGpp synthetase, increases the accumulation of amino acids, such as glutamate and lysine, in amino-acid-overproducing strains of E. coli. Elevation of intracellular ppGpp levels due to depletion of required amino acids also enhances glutamate overproduction. Moreover, the extent of overproduction is highly dependent on the intracellular ppGpp level. These results demonstrate that amino-acid overproduction in E. coli is closely connected to amino-acid auxotrophy via the accumulation of ppGpp. © 2006 Elsevier B.V. All rights reserved. Keywords: Escherichia coli; ppGpp; Stringent response; relA; l-glutamate; l-lysine

1. Introduction Amino acids are widely used as seasonings, nutritional supplements, animal feeds, and drug intermediates and so on. Several amino acids are produced by fermentation methods. Escherichia coli is commonly used for amino-acid production because of Abbreviations: OD, optical density; ppGpp, guanosine-3 ,5 tetraphosphate; rRNA, ribosomal RNA; tRNA, transfer RNA ∗ Corresponding author at: Institute of Life Sciences, Ajinomoto CO., Inc., 1-1,Suzuki-cho, Kawasaki-ku, Kawasaki-shi, Kanagawa, 210-8681, Japan. Tel.: +81 44 210 5928; fax: +81 44 244 4258. E-mail address: akira [email protected] (A. Imaizumi). 0168-1656/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.jbiotec.2006.03.015

the extensive knowledge of its metabolism (Herrmann and Somerville, 1983), well-established recombinant DNA technology (Swartz, 1996) and availability of the complete genome sequence (Blattner et al., 1997). Many studies of amino-acid production using E. coli have exploited these advantages (Aiba et al., 1980; Hashiguchi et al., 1999; Jetten and Sinskey, 1995; Kojima et al., 1995; Ogawa-Miyata et al., 2001; Tsujimoto et al., 1993; Usuda and Kurahashi, 2005), and standard breeding strategies have focused mainly on increasing the flux to target substances (Eggeling and Sahm, 1999; Hashiguchi et al., 1999; Jetten et al., 1995; Patnaik and Liao, 1994). Methods such as release from feedback inhibition and/or repression, amplifica-

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

tion of genes encoding enzymes of the relevant pathway and disruption of genes encoding degradative enzymes, have all been reported (Eggeling and Sahm, 1999; Hashiguchi et al., 1999; Jetten et al., 1995; Kikuchi et al., 1996; Ogawa-Miyata et al., 2001; Usuda et al., 2005). The resulting overproducing strains are often auxotrophic for some other amino acid because of inactivation or weakening of the corresponding biosynthetic pathway. Guanosine-3 ,5 -tetraphosphate (ppGpp) is a nucleotide that plays an important role in the signaltransduction pathway activated in response to several kinds of starvation (Cashel et al., 1996; Magnusson et al., 2005). A well-studied ppGpp-mediated response is that evoked by amino-acid starvation, the so-called ‘stringent response’ (Cashel et al., 1996); when amino acids are depleted, ribosome-bound RelA protein recognizes uncharged transfer RNA and synthesizes ppGpp (Hazeltine and Block, 1973), which binds to the RNA polymerase core enzyme (Chatterji et al., 1998; Toulokhonov et al., 2001) resulting in immediate various responses, such as growth arrest, repression of the synthesis of ribosomes, degradation of ribosomal proteins, retardation of translation speed, enhancement of amino-acid biosynthesis, and so on (reviewed in Cashel et al., 1996). Furthermore, ppGpp may act an important role on transcription of the rmf gene (Izutsu et al., 2001), encoding ribosome modulation factor, whose knockout results the increase of lysine production of lysine overproducing strain of E. coli (Imaizumi et al., 2005b).

329

It has been reported that knockout of relA enhanced exogenous protein production in E. coli because it eliminates repression by ppGpp (Dedhia et al., 1997). By contrast, overproduction of ppGpp enhanced production of secondary metabolites, such as antibiotics, in Streptomyces griseus and Bacillus subtilis (Ochi, 1987; Ochi and Ohsawa, 1984). Although there have been several demonstrations of upregulation of amino-acidsynthesizing genes by ppGpp (Barker et al., 2001a, 2001b; Cashel et al., 1996; Paul et al., 2005), there have been no reports of amino-acid overproduction in E. coli. Here, we investigated the effects of intracellular ppGpp levels on growth and amino-acid production in overproducing derivatives of E. coli. We used relA and spoT knockouts, and relA-overproducing mutants of amino-acid-overproducing strains of E. coli, to test whether accumulation of ppGpp is essential for aminoacid overproduction, and to determine how ppGpp accumulation influences growth and amino-acid overproduction when overproducing cells are starved of required amino acids.

2. Material and methods 2.1. Strains, plasmids and culture conditions The strains and plasmids used in this study are listed in Table 1. MG1655S is an l-glutamate-overproducing strain obtained by in-frame disruption of sucA gene

Table 1 List of bacterial strains and plasmids used in this study Strain or plasmid

Phenotype or gene

Source or reference

Strain MG1655 MG1655A MG1655S MG1655SA MG1655ST MG1655SAT WC196

Wild type MG1655
relA MG1655
sucA MG1655
sucA
relA MG1655
sucA
spoT MG1655
sucA
relA
spoT Lysine producing mutant derived from W3110

Our laboratory stock This study This study This study This study This study Kikuchi et al. (1996, 1997)

Plasmid pMAN997 pM15 pMrelA pMrelA pCAB1

pSC101 derivative, orits , Ampr Plasmid vector, Ampr pM15 carrying truncated relA gene pM15 carrying relA gene RSF1010 derivatives, carrying dapA*, dapB and lysC* Smr

Tanaka et al. (2001) Imaizumi et al. (2005a, 2005b) Imaizumi et al. (2005a, 2005b) Imaizumi et al. (2005a, 2005b) Kojima et al. (1995)

Ampr , ampicillin resistant; Smr , streptomycin resistant; orits , temperature-sensitive replication origin of plasmid.

330

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

from wild type strain, MG1655. WC196 an l-lysineoverproducing strain which is derived from W3110 as an S-[2-aminoethyl]-l-cysteine-resistant mutant (Kikuchi et al., 1997). Plasmid pM15 (Imaizumi et al., 2005a) carries the PR promoter from phage lambda (Walz and Pirrotta, 1975). Plasmids pMrelA and pMrelA carry a truncated relA open-reading frame (Schreiber et al., 1991) and a full-length relA openreading frame, respectively, each under the control of the Pr promoter (Imaizumi et al., 2005a). Plasmid pCAB1 carries a feedback insensitive lysC gene (lysC* ), a feedback insensitive dapA gene (dapA* ) and the dapB gene (Kojima et al., 1995). Cells were grown in Luria–Bertani medium (LB) containing 10 g l−1 bacto-tryptone (Difco), 5 g l−1 yeast extract (Difco) and 10 g l−1 NaCl, the LBG medium consisted of LB medium with 5 g l−1 glucose; the MS medium contained 40 g l−1 g lucose, 1 g l−1 MgSO4 , 1 g l−1 KH2 PO4 , 16 g l−1 (NH4 )2 SO4 and 2 g l−1 yeast extract (Difco) (Matsui et al., 2001); the modified MS medium consisted of 2 g l−1 glucose, 1 g l−1 MgSO4 , 1 g l−1 KH2 PO4 and 16 g l−1 (NH4 )2 SO4 in 50 mM MES-NaOH (pH 7.0). 18 amino acids (100 mg l−1 l-arginine, 100 mg l−1 l-alanine, 2 mg l−1 l-aspartate, 20 mg l−1 l-cystine, 100 mg l−1 l-glutamate, 100 mg l−1 l-glycine, 100 mg l−1 l-histidine, 100 mg l−1 lisoleucine, 100 mg l−1 l-leucine, 100 mg l−1 l-lysine, 100 mg l−1 l-methionine, 100 mg l−1 l-phenylalanine, 100 mg l−1 l-proline, 100 mg l−1 l-serine, 100 mg l−1 l-threonine, 50 mg l−1 l-tryptophan, 10 mg l−1 ltyrosine, and 100 mg l−1 l-valine) were added as required to the modified MS medium. For glutamate and lysine production, cells from 500 ␮l of overnight

culture in LB were collected and inoculated into 20 ml MS medium. For addition and omission test, cells from 50 ␮l of overnight culture in LB were subsequently collected, then inoculated into 5 ml of modified MS medium in a TN-5L test tube (Advantec) at 37 ◦ C aerobically. For analysis of the effects of amino acids supplementation on growth, glutamate production, and ppGpp accumulation, cells from 500 ␮l of overnight culture in LB were inoculated into 20 ml LBG in a 500-ml flask and grown at 37 ◦ C with shaking, then cells were subsequently collected, washed three times in 0.9% NaCl, resuspended in 20 ml modified MS medium in a 500-ml flask and grown at 37 ◦ C with shaking at 120 strokes per min. Ampicillin (100 ␮g ml−1 ) or streptomycin (25 ␮g ml−1 ) was added as required. 2.2. DNA manipulations The primers used in this study are listed in Table 2. Sequencing was carried out by the dideoxy-chaintermination method with a Taq DyeDeoxy terminator cycle sequencing kit and a 310 DNA sequencer (Applied Biosystems). Genomic DNA was extracted with a DNeasy Tissue Kit (Qiagen). Gene disruption was performed by a combination of crossover PCR and use of plasmid pMAN997 with a temperaturesensitive origin of replication, as described previously (Imaizumi et al., 2005b; Link et al., 1997; Tanaka et al., 2001). For gene disruption, two DNA fragments were amplified from the genomic DNA of MG1655 using the following primers: sucA–Co and sucA–Ci, and sucA–No and sucA–Ni for sucA; relA–Co and relA–Ci, and relA–No and relA–Ni for

Table 2 Primers used in this study sucA–Ni sucA–No sucA–Ci sucA–Co relA–Ni relA–No relA–Ci relA–Co spoT–Ni spoT–No spoT–Ci spoT–Co

5 -GCGAATTCCTGCCCCTGACACTAAGACA-3 5 -CGAGGTAACGTTCAAGACCT-3 5 -AGGTCTTGAACGTTACCTCGATCCATAACGGGCAGGGCGC-3 5 -GCGAATTCCCACTTTGTCAGTTTCGATT-3 5 -TGTTTAAGTTTAGTGGATGGGTGCGTCTGTTGCAGACAATAC-3 5 -GCGAATTC TTGAACTGGTACAGGCAACC-3 5 -CCCATCCACTAAACTTAAACATAGCGACACCAAACAGCAAC-3 5 -GCGAATTCAAGCACTTCACTACTGTTTTC-3 5 -CCCATCCACTAAACTTAAACAGCTTTCAAACAGATACAA-3 5 -GCGAATTCCGCGGAGTATCTTTATTTTAC-3 5 -TGTTTAAGTTTAGTGGATGGGACCCGAAACCGAAATTAA-3 5 -GCGAATTCTAAAGAATGAGGGCTGAGGC-3

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

relA; and spoT–Co and spoT–Ci, and spoT–No and spoT–Ni for spoT. After combining the first PCR products, second PCR reactions were performed using the following primers: sucA–No and sucA–Co for sucA; relA–No and relA–Co for relA; and spot–No and spot–Co for spoT. After blunting with a DNA blunting kit (Takara Bio), the resulting PCR products were cloned into the SmaI site of pMAN997. By repeating these procedures, double and triple disruptants were obtained from MG1655S. All other basic recombinant DNA procedures, such as isolation and purification of DNA, restriction enzyme digestion and E. coli transformation, were performed as described by literature (Sambrook et al., 1989). For confirmation of gene disruption, DNA was sequenced using the dideoxynucleotide method with a dye-terminator cycle-sequencing kit (Applied Biosystems) and a DNA sequencer (model 310; Applied Biosystems). 2.3. Extraction of ppGpp Cells grown in modified MS medium were rapidly resuspended in 10 ml pre-chilled 1 N formate and left for 30 min at 4 ◦ C to extract nucleic acids and derivatives, such as ppGpp. Cell debris was removed by centrifuge at 10,000 × g × 5 min and decantation. Then samples were freeze-dried and dissolved in 250 ␮l of water. 2.4. Measurements of cell growth and metabolites Growth was estimated from optical density (OD) at 562 nm measured with a Beckman DU-640 spectrometer. Dry cell weight was determined as the weight of freeze dried cells collected from 1 ml of culture. For

331

addition and omission test, OD at 660 nm was monitored automatically using TN-1506 incubator (Advantec). Glucose, glutamate and lysine were measured using Biotec Analyzer (Sakura Seiki Co). Yp/s was determined as the ratio of glutamate or lysine accumulation per consumed sugar (g g−1 ). Intracellular ppGpp levels were measured as described previously (Ochi, 1986), and calculated as concentration per mg of dry cell weight. All statistical analysis was performed using Microsoft Excel 2003.

3. Results 3.1. Knockout of the relA gene reduces l-glutamate overproduction and growth Cells of strain MG1655S, the in-frame sucA gene disruptant of MG1655, accumulated l-glutamate in the medium and consumed all the glucose within 24 h when cultivated in MS medium, whereas only slight accumulation of glutamate was detected in the parental strain MG1655 (data not shown). To examine the effect of the ppGpp-synthesizing enzymes, RelA and SpoT, we made knockouts of relA and spoT in MG1655S. Resulting strains (MG1655S, MG1655SA, and MG1655SAT) were cultured in MS medium for 24 h. The relA-knockout mutant, MG1655SA, showed decreased glutamate accumulation, growth, and glucose consumption (Table 3). And its glucose consumption retarded at that time (data not shown) and Yp/s was also substantially reduced (Table 3). Additionally, the relA spoT double knockout, MG1655SAT, behaved similarly to MG1655SA (Table 3). By contrast, in the spoT single knockout, MG1655ST, Yp/s was slightly increased (Table 3).

Table 3 Growth and glutamate production in MG1655S and its derivatives Strain

MG1655S MG1655SA MG1655ST MG1655SAT

OD600

16.6 (0.8) 10.8 (2.3) 17.9 (0.2) 8.6 (0.1)

l-glutamate

Residual glucose

g l−1

Yp/s

p

13.8 (1.4) 3.6 (0.1) 15.8 (0.2) 3.2 (0.0)

0.345 (0.035) 0.181 (0.017) 0.395 (0.004) 0.184 (0.003)

N.T.b 0.104 N.T.b

The numbers in parenthesis represent the standard deviations of three replicates. a p-value of Student’s t-test in which Y p/s was compared to control (MG1655S). b Not tested.

valuea

g l−1 0.0 (0.0) 20.2 (2.6) 0.0 (0.0) 22.6 (0.3)

332

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

3.2. Overexpression of relA gene increases glutamate overproduction

ppGpp might reduce the growth rate of MG1655S, no such effect was observed (data not shown).

To confirm that the phenotype of the relA knockout was complemented by introduction of the relA gene, we introduced plasmids pMrelA and pMrelA into MG1655SA and they were cultured in MS medium for 24 h (Imaizumi et al., 2005a). pMrelA carries a truncated relA gene (Schreiber et al., 1991); its product, RelA , the catalytic domain of RelA (1–455 amino acid residues), is metabolically unstable and lacks the ability to bind to ribosomes (Schreiber et al., 1991). And RelA protein produces ppGpp constitutively regardless of intracellular amino-acid levels because this protein is independent to ribosomes and does not require uncharged tRNA for activation (Svitil et al., 1993). We predicted that pMrelA would cause sustained elevation of intracellular ppGpp. By contrast, pMrelA carries the full-length relA gene under the Pr promoter of lambda phage (Walz and Pirrotta, 1975), and we expected it to cause sustained overexpression of RelA protein. Introduction of pMrelA did not complement the phenotype of MG1655SA—there was only a slight increase in glutamate accumulation and Yp/s (Table 4)—whereas the introduction of pMrelA not only completely reversed the poor growth of MG1655SA, but also significantly increased its glutamate accumulation and Yp/s (p < 0.01, Table 4). We also examined the effect of overexpression of the relA gene in MG1655S. There was, again, a significant increase of glutamate accumulation and yield when pMrelA was introduced (p < 0.01), whereas a little increase was detected with pMrelA (p < 0.05, Table 4). In addition, although we predicted that the overproduction of

3.3. Overexpression of relA gene also increases lysine overproduction To determine whether the effect of overexpression of relA could be reproduced in another amino-acidoverproducing strain, pMrelA was introduced into the lysine-overproducing strain WC196/pCAB1 (Kojima et al., 1995). In this strain, gene expression involving lysine biosynthetic pathway was enhanced by plasmid and enzymatic activities of aspartate kinase III (lysC product), dihydrodipicolinate synthase (dapA product), and dihydrodipicolinate reductase (dapB product) were relieved from feedback inhibition to lysine. Therefore, both the genetic background and mechanism of overproduction was different to the glutamate overproducing strain, MG1655S. WC196/pCAB1 harboring pM15 and WC196/pCAB1 harboring pMrelA were cultured in MS medium for 42 h. In this case, overexpression of relA gene led to the increase of lysine accumulation and Yp/s (significant at p < 0.01, Table 5). 3.4. Amino-acid auxotrophy of MG1655S and MG1655SA The reduced growth of MG1655SA and MG1655SAT in MS medium suggested that they were auxotrophic for some amino acids (Table 3). We therefore examined the amino-acid requirements of these strains. It was expected that MG1655S showed weak auxotrophy for lysine and methionine

Table 4 Growth and glutamate production in MG1655S, MG1655SA and their derivatives harboring a ppGpp-overproducing plasmid Strain

MG1655S MG1655SA MG1655S MG1655SA MG1655S MG1655SA

Plasmid

pM15 pM15 pMrelA pMrelA pMrelA pMrelA

OD600

17.81 (0.20) 7.49 (0.77) 17.92 (0.32) 7.01 (0.01) 17.49 (0.12) 16.91 (0.56)

Glutamate

Residual glucose

g l−1

Yp/s

p

15.80 (0.14) 1.92 (0.20) 16.67 (0.25) 2.70 (0.35) 19.23 (0.15) 18.47 (0.42)

0.395 (0.004) 0.178 (0.050) 0.417 (0.006) 0.370 (0.056) 0.481 (0.003) 0.462 (0.009)

N.T.b 0.023 N.T.b 0.000 0.000

The numbers in parenthesis represent the standard deviations of three replicates. a p-value of Student’s t-test in which Y p/s was compared to control (MG1655S harboring plasmid pM15). b Not tested.

valuea

g l−1 0.0 (0.0) 29.2 (3.8) 0.0 (0.0) 32.7 (0.1) 0.0 (0.0) 0.0 (0.0)

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

333

Table 5 Growth and lysine production of WC196/pCAB1 with ppGpp-overproducing plasmid Strain

WC196 WC196

Plasmid

pCAB1,pM15 pCAB1,pMrelA

OD600

12.33 (1.01) 13.35 (1.56)

Lysine g l−1

Yp/s

p valuea

14.25 (0.56) 15.71 (0.44)

0.356 (0.014) 0.393 (0.011)

0.012

The numbers in parenthesis represent the standard deviations of three replicates. a p-value of Student’s t-test in which Y p/s was compared to control (WC196 harboring plasmids pCAB1 and pM15).

because of shortage of succinyl-CoA supply due to the inactivation of 2-oxoglutarate dehydrogenase activity. The growth of MG1655S was relieved partially by addition of Lys, and almost completely by addition of Lys + Met (Fig. 1A). By contrast, the growth defect of MG1655SA was not relieved by addition of Lys + Met (Fig. 1B). To determine the required amino acids of MG1655SA, the omission test of each amino acid were performed (Fig. 1C). MG1655A showed the valine

auxotrophy in addition to the isoleucine auxotrophy under valine added condition, the characteristics of E. coli K-12 because of inactivation of ilvG gene (Desai and Polglase, 1966; De Felice et al., 1997) (Fig. 1C). This observation reproduced the previous study (Xiao et al., 1991). On the other hand, MG1655SA showed the requirement to lysine, methionine, isoleucine and valine (Fig. 1C). Therefore, this strain showed the characteristics both of MG1655S and MG1655A with amino acid requirement.

Fig. 1. Amino acids requirement of MG1655S, MG1655SA, and MG1655A. Cells were grown in modified MS medium with supplementation of amino acids. (A and B) Addition test of MG1655S (A) and MG1655SA (B). OD660 was monitored periodically using TN1506 incubator (Advantec). Cells were grown in modified MS medium with supplement no amino acid (closed square), 0.1 g l−1 of Lys (opened square), 0.1 g l−1 of Lys + Met (closed circle), and mixture of 18 amino acids (opened circle). (C) Omission test of MG1655A (closed) and MG1655SA (opened). Relative growth at 24 h to the control condition of each strain in which all 18 amino acids were supplemented (see Materials and methods) was indicated. OD660 was monitored using TN1506 incubator (Advantec).

334

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

3.5. Glutamate overproduction is dependent on the intracellular ppGpp level and requires amino acids According to the observation as described above, we next examined glutamate production under these conditions. MG1655S harboring pM15 or pMrelA, and MG1655SA harboring pM15 and pMrelA were cultured in modified MS medium. Both MG1655S/pM15 and MG1655S/pMrelA grew well in medium supplemented with Lys + Met and in medium supplemented with Lys + Met + Val + Ile (Fig. 2A). And in each strain, there was an inverse relationship between growth and glutamate production (Fig. 2A and B).

Fig. 2. Effects of the relA expression level and supplementation with required amino acids on growth (A), glutamate accumulation (B), and intracellar ppGpp accumulation (C) in glutamate-overproducing strains. Cells were grown in modified MS medium with supplement of no amino acid, 0.1 g l−1 of Lys + Met, 0.1 g l−1 of Lys + Met + Val, and 0.1 g l−1 of Lys + Met + Val + Ile. Error bars represent the standard deviations (S.D.s) of two replicates.

In addition, glutamate production might be also dependent on the copy number of relA expression: glutamate production was higher in the relAoverexpressing strain (MG1655S/pMrelA) than in the control strain (MG1655S/pM15) under any conditions (Fig. 2B). By contrast, in the relA disruptant (MG1655SA/pM15), glutamate production was almost same (Lys + Met) or lower than the control (other three conditions, Fig. 2B). The almost same result was observed in MG1655SA/pMrelA (Fig. 2B). Note that under Lys + Met condition, MG1655S and MG1655S/pMrelA could grow because required amino acids were supplemented while MG1655SA/pM15 and MG1655SA/pMrelA could not. Therefore, glutamate production increased in proportion to relA copy number and stringent response due to the depletion of required amino acids. Then we measured intracellular ppGpp levels 20 min after cells were resuspended in modified MS medium supplemented with various combinations of amino acids. In this measurement, a few amount of pppGpp (guanosine 3 -diphosphate 5 triphosphate) was also detected (data not shown). Since pppGpp is regarded that it is functionally equivalent to ppGpp (Cashel et al., 1996), the sum of pppGpp and ppGpp accumulation is hereafter designated intracellar ppGpp level. We confirmed that the ppGpp level rose rapidly, remained at a maximum level for ∼15–30 min after nutritional shift down, and then declined in each strain (data not shown). In MG1655S/pM15 and MG1655S/pMrelA, high levels of intracellular ppGpp were observed when growth was inhibited (no amino-acid supplementation and supplementation with Lys + Met + Val, respectively) and accumulation of ppGpp was low when the cells were able to grow (Lys + Met and Lys + Met + Val + Ile conditions, respectively; Fig. 2C). Higher ppGpp levels were present in MG1655S/pMrelA than in MG1655S under all conditions, consistent with the results described above (Fig. 2C). By contrast, only a small peak corresponding to ppGpp was detected in MG1655SA/pM15 under any conditions (Fig. 2C). Although higher intracellar ppGpp levels were observed in MG1655SA/pMrelA , no significant difference of ppGpp accumulation between amino acid sufficient condition (Lys + Met + Val + Ile) and insufficient conditions (other conditions) (Fig. 2C).

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

4. Discussion We have shown that overexpression of the relA gene, which encodes ppGpp synthetase, substantially enhances amino-acid overproduction. We have also demonstrated that relA is indispensable for amino-acid production, because knockout of relA greatly decreased both growth and amino-acid production. And we have also demonstrated that the overexpression of relA enhances glutamate and lysine production, but does not affect cell growth (Tables 4 and 5). The relA knockout showed requirement of multiple amino acids (Fig. 1C). It is suggested that the depletion of several amino acids has been occurred during the cultivation in MS medium because MG1655SA and MG1655SAT show the significant decrease of growth and amino acid production while MG1655S and MG1655S harboring pMrelA do not (Tables 3 and 4). Apparently the degree of glutamate overproduction is dependent on the intracellular level of ppGpp (Fig. 2). It is suggested that only few percent of the ribosomes have a bound RelA protein usually (Justesen et al., 1985). And it is also suggested that the more severe amino acid starvation, the more ppGpp is produced, the greater is the observed inhibition of rRNA transcription (Condon et al., 1995). Moreover, the overexpression of RelA using high copy number plasmid vector and strong promoter decreases growth rates significantly (Schreiber et al., 1991). Therefore, the inhibition level of rRNA transcription would be correlated to the intracellular ppGpp level. In this study, relA gene was expressed with the plasmid originated from the low copy number vector, pMW119 (Kobayashi et al., 2001). It is suggested that the expression level of RelA is so moderate that no growth inhibition is observed and also enough to complement the growth deficiency of the relA knockout strains caused by the multiple amino acids requirement. On the other hand, it has previously been reported that the genes controlling the biosynthesis of these amino acids are positively regulated by ppGpp (Cashel et al., 1996; Lloyd and Sharples, 1991; Patte et al., 1980; Saroja and Gowrishankar, 1996). For example, ppGpp might regulate the expression of gdhA for glutamate synthesis (Saroja and Gowrishankar, 1996) and lysA for lysine synthesis (Patte et al., 1980) positively. Recently, it was reported that DksA protein, crucial factor for negative control of rRNA transcription by

335

ppGpp, was also promotes the upregulation of some amino acid promoters suggesting the direct enhancement of amino acid transcription by ppGpp (Paul et al., 2005). On the contrary, enhancement of amino-acid biosynthesis has been considered a secondary effect of the inhibition of rRNA transcription and the resulting increase in availability of ‘free’ RNA polymerase (Zhou and Jin, 1998). Additionally, it was also shown that ppGpp decreased the lifetime of RNA polymerase – promoter open complex of rrn P1 open complex and this phenomenon also contributes to the increase of RNA polymerase – promoter complex formation of genes which are involved in amino acid biosynthesis (Barker et al., 2001a, 2001b). These ‘positive’ and ‘passive’ models might describe two sides of the same phenomenon or function cooperatively each other. Considering above, the increase of glutamate and lysine production observed in this study could be attributed to the increase of availability of RNA polymerase to transcribe the amino acids biosynthesizing genes in proportion to the intracellular ppGpp level. We also found that overexpression of a truncated relA gene (relA ) had no effect on growth, and only slight effect on glutamate production in MG1655S and MG1655SA, respectively (Table 4 and Fig. 2). Since RelA protein is independent to ribosomes and does not sense the uncharged tRNA (Svitil et al., 1993), it is suggested that overexpression of relA does not contribute to the increase of intracellar ppGpp level caused by the depletion of amino acids. Indeed, although the overexpression of relA increased the intracellar ppGpp level, the level was obviously low and almost stable regardless to required amino acids availability while the levels in the strains containing full-length relA gene varied in response to required amino acids availability (Fig. 2C). Our findings also imply that bradytrophy and/or auxotrophy for other amino acids play an essential role in amino-acid overproduction in conjunction with the accumulation of ppGpp. In ‘metabolic engineering’, creating auxotrophy has been considered an important aspect of process control in order to sustain adequate biomass and/or to block carbon flow into undesirable by-products and so increase target production (Eggeling and Sahm, 1999; Jetten and Sinskey, 1995). Apparently there was a reversible correlation between cell growth and glutamate production (Fig. 2A and B). However, accounting the carbon distribution to glutamate, biomass, and others from glucose, the dis-

336

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337

tribution to others was increased in relA knockouts under any conditions because that to glutamate and biomass was decreased (Fig. 2A and B). Therefore, it was suggested that full-length RelA-mediated ppGpp accumulation provoked by required amino acid depletion increased the carbon distribution to glutamate. Our findings show that the creating auxotrophy has effects not only as described previously (Eggeling and Sahm, 1999; Jetten and Sinskey, 1995) but also a greater effect induced by ppGpp accumulation on the overproduction of target amino acids. In conclusion, using a combination of gene knockout and amplification, we have demonstrated that intracellular accumulation of ppGpp is indispensable for the overproduction of amino acids, such as glutamate and lysine.

Acknowledgements We thank A. Briukhanov and R. Shaklov for fruitful discussions, and K. Ochi for helpful advice and instruction on ppGpp measurement. We are also grateful to N. Yoshizawa for technical assistance.

References Aiba, S., Imanaka, T., Tsunekawa, H., 1980. Enhancement of tryptophan production by Escherichia coli as an application of genetic engineering. Biotechnol. Lett. 2, 525–530. Barker, M.M., Gaal, T., Gourse, R.L., 2001a. Mechanism of regulation of transcription initiation by ppGpp. II. Models for positive control based on properties of RNAP mutants and competition for RNAP. J. Mol. Biol. 305, 689–702. Barker, M.M., Gaal, T., Josaitis, C.A., Gourse, R.L., 2001b. Mechanism of regulation of transcription initiation by ppGpp. I. Effects of ppGpp on transcription initiation in vivo and in vitro. J. Mol. Biol. 305, 673–688. Blattner, F.R., Plunkett, G., Bloch, C.A., Perna, N.T., Burland, V., Reley, M., Collado-Vides, J., Glasner, J.D., Rode, C.K., Mayhew, G.F., Gregor, J., Wayne Davis, N., Kirkpatrick, H.A., Goeden, M.A., Rose, D.J., Mau, B., Shao, Y., 1997. The complete genome sequence of Escherichia coli K-12. Science 277, 1453– 1462. Cashel, M., Gentry, D.R., Hernadez, V.J., Vinella, D., 1996. The stringent response. In: Neidhardt, F.C. (Ed.), Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd ed. ASM Press, Washington DC, pp. 1458–1496. Chatterji, D., Fujita, N., Ishihama, A., 1998. The mediator for stringent control, ppGpp, binds to the beta-subunit of Escherichia coli RNA polymerase. Genes Cells 3, 279–287.

Condon, C., Squres, C., Squires, C.L., 1995. Control of rRNA transcription in Escherichia coli. Microbiol. Rev. 59, 623–645. De Felice, M., Squires, C., Levinthal, M., Guardiola, J., Lamberti, A., Iaccarino, M., 1997. Growth inhibition of Escherichia coli K-12 by l-valine: a consequence of a regulatory pattern. Mol. Gen. Genet. 156, 1–7. Dedhia, N., Richins, R., Mesina, A., Chen, W., 1997. Improvement in recombinant protein production in ppGpp-deficient Escherichia coli. Biotechnol. Bioeng. 53, 379–386. Desai, I.D., Polglase, W.J., 1966. Valine sensitivity of Escherichia coli K-12. Can. J. Biochem. 44, 1403–1405. Eggeling, L., Sahm, H., 1999. Amino acid production: principles of metabolic engineering. In: Lee, S.Y., Papoutsakis, E.T. (Eds.), Metabolic Engineering. Marcel Dekker, Inc., New York, NY, pp. 153–176. Hashiguchi, K., Takesada, H., Suzuki, E., Matsui, H., 1999. Construction of an l-isoleucine overproducing strain of Escherichia coli K-12. Biosci. Biotechnol. Biochem. 63, 672–679. Herrmann, K.M., Somerville, R.L., 1983. Amino acids: biosynthesis and genetic regulation. Addison-Wesley Publishing Company, Reading, MA. Imaizumi, A., Shaklov, R. S., Briukhanov, A. L., 2005a. Method for producing l-amino acids, US patent 2005106688. Imaizumi, A., Takikawa, R., Koseki, C., Usuda, Y., Yasueda, H., Kojima, H., Matsui, K., Sugimoto, S., 2005b. Improved production of l-lysine by disruption of stationary phase-specific rmf gene in Escherichia coli. J. Biotechnol. 117, 111–118. Izutsu, K., Wada, A., Wada, C., 2001. Expression of ribosome modulation factor (RMF) in Escherichia coli requires ppGpp. Genes Cells 6, 665–676. Jetten, M.S.M., Sinskey, A.J., 1995. Recent advances in the physiology and genetics of amino acid-producing bacteria. Crit. Rev. Biotechnol. 15, 73–103. Justesen, J., Lund, T., Pedersen, F.S., Kjeldgaard, N.O., 1985. The Phisiology of stringent factor (ATP:GTP 3 diphosphotransferase) in Escherichia coli. Biochimie (Paris) 68, 715–722. Kikuchi, Y., Kojima, H., Tanaka, T., Takatsuka, Y., Kamio, Y., 1997. Characterization of second lysine decarboxylase isolated from Escherichia coli. J. Bacteriol. 179, 4486–4492. Kikuchi, Y., Suzuki, T. Kojima, H., 1996. Novel lysine decarboxylase gene and process for producing l-lysine. International Patent Cooperation Treaty patent WO9617930. Kobayashi, K., Tsukagoshi, N., Aono, R., 2001. Supression of hypersensitivity of Escherichia coli acrB mutant to organic solvents by integrational activation of the acrEF operon with the IS1 or IS2 element. J. Bacteriol. 183, 2646–2653. Kojima, H., Ogawa, Y., Kawamura, K., Sano, K., 1995. Process for producing l-lysine by fermentation. International Patent Cooperation Treaty patent WO9516042. Link, A.J., Philipps, D., Church, G.M., 1997. Methods for generating precise deletions and insertions in the genome of wild-type Escherichia coli: application to open reading frame characterization. J. Bacteriol. 179, 6228–6237. Lloyd, R.G., Sharples, G.J., 1991. Molecular organization and nucleotide sequences of the recG locus of Escherichia coli K12. J. Bacteriol. 173, 6837–6843.

A. Imaizumi et al. / Journal of Biotechnology 125 (2006) 328–337 Magnusson, L.U., Farewell, A., Nystrom, T., 2005. ppGpp: a global regulator in Escherichia coli. Trends Microbiol. 13, 236–242. Matsui, H., Kawasaki, H., Shimaoka, M., Kurahashi, O., 2001. Investigation of various genotype characterization for inosine accumulation in Escherichia coli W3110. Biosci. Biotechnol. Biochem. 65, 570–578. Ochi, K., 1986. Occurrence of the stringent response in Streptomyces sp. and its significance for the initiation of morphological and physiological differentiation. J. Gen. Microbiol. 132, 2621–2631. Ochi, K., Ohsawa, S., 1984. Initiation of antibiotic production by the stringent response of Bacillus subtilis Marburg. J. Gen. Microbiol. 130, 2473–2482. Ochi, K., 1987. Metabolic initiation of differentiation and secondary metabolism by Streptomyces griseus: significance of the stringent response (ppGpp) and GTP content in relation to A factor. J. Bacteriol. 169, 3608–3616. Ogawa-Miyata, Y., Kojima, H., Sano, K., 2001. Mutation analysis of the feedback inhibition site of aspartokinase III of Escherichia coli K-12 and its use in l-threonine production. Biosci. Biotechnol. Biochem. 65, 1149–1154. Patnaik, R., Liao, J.C., 1994. Engineering of Escherichia coli central metabolism for aromatic metabolite production with near theoretical yield. Appl. Environ. Microbiol. 60, 3903–3908. Patte, J., Morand, P., Boy, E., Richaud, C., Borne, F., 1980. Thr relA locus and the regulation of lysine biosynthesis in Escherichia coli. Mol. Gen. Genet. 179, 319–325. Paul, B.J., Berkmen, M.B., Gourse, R.L., 2005. DksA potentiates direct activation of amino acid promoters by ppGpp. Proc. Natl Acad. Sci. USA 102, 7823–7828. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Saroja, G.N., Gowrishankar, J., 1996. Roles of SpoT and FNR in NH4 + assimilation and osmoregulation in GOGAT (glutamate synthase)-deficient mutants of Escherichia coli. J. Bacteriol. 178, 4105–4114.

337

Schreiber, G., Metzger, S., Aizenman, E., Roza, S., Cashel, M., Glaser, G., 1991. Overexpression of the relA gene in Escherichia coli. J. Biol. Chem. 266, 3760–3767. Svitil, A.L., Cashel, M., Zyskind, J.W., 1993. Guanosine tetraphosphate inhibits protein synthesis in vivo. A possible protective mechanism for starvation stress in Escherichia coli. J. Biol. Chem. 268, 2307–2311. Swartz, J.R., 1996. Escherichia coli recombinant DNA technology. In: Neidhardt, F.C. (Ed.), Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd ed. ASM press, Washington DC, pp. 1458–1496. Tanaka, K., Matsuyama, S., Tokuda, H., 2001. Deletion of lolB, encoding an outer membrane lipoprotein, is lethal for Escherichia coli and causes accumulation of lipoprotein localization intermediates in the periplasm. J. Bacteriol. 183, 6538– 6542. Toulokhonov, I.I., Shulgina, I., Hernandez, V.J., 2001. Binding of the transcription effector ppGpp to Escherichia coli RNA polymerase is allosteric, modular, and occurs near the N terminus of the beta -subunit. J. Biol. Chem. 276, 1220–1225. Tsujimoto, N., Kikuchi, Y., Kurahashi, O., Kawahara, Y., 1993. Mutant Escherichia coli capable of enhanced l-glutamic acid production by fermentation. US patent 5378616. Usuda, Y., Kurahashi, O., 2005. Effects of deregulation of methionine biosynthesis on methionine excretion in Escherichia coli. Appl. Environ. Microbiol. 71, 3228–3234. Walz, A., Pirrotta, V., 1975. Sequence of the PR promoter of phage lambda. Nature 254, 118–121. Xiao, H., Kalman, M., Ikehara, K., Zemel, S., Glaser, G., Cashel, M., 1991. Residual guanosine 3 ,5 -bisphosphate synthetic activity of relA null mutants can be eliminated by spoT null mutations. J. Biol. Chem. 266, 5980–5990. Zhou, Y.N., Jin, D.J., 1998. The rpoB mutants destablizing initiation complexes at stringently controlled promoters behave like ‘stringent’ RNA polymerases in Esherichia coli. Proc. Natl. Acad. Sci. USA 95, 2908–2913.