Scholarly article on topic ' Production of 2-ketoisocaproate with C orynebacterium glutamicum strains devoid of plasmids and heterologous genes '

Production of 2-ketoisocaproate with C orynebacterium glutamicum strains devoid of plasmids and heterologous genes Academic research paper on "Industrial Biotechnology"

0
0
Share paper
Academic journal
Microbial Biotechnology
OECD Field of science
Keywords
{""}

Academic research paper on topic " Production of 2-ketoisocaproate with C orynebacterium glutamicum strains devoid of plasmids and heterologous genes "

UJ.UJJJ.U-U

Brief report

Production of 2-ketoisocaproate with Corynebacterium glutamicum strains devoid of plasmids and heterologous genes

Michael Vogt, Sabine Haas, Tino Polen, Jan van Ooyen* and Michael Bott**

Institute of Bio- and Geosciences, IBG-1: Biotechnology, Forschungszentrum Jülich, D-52425 Jülich, Germany.

Summary

2-Ketoisocaproate (KIC), the last intermediate in L-leucine biosynthesis, has various medical and industrial applications. After deletion of the ilvE gene for transaminase B in L-leucine production strains of Corynebacterium glutamicum, KIC became the major product, however, the strains were auxotrophic for L-isoleucine. To avoid auxotrophy, reduction of IlvE activity by exchanging the ATG start codon of ilvE by GTG was tested instead of an ilvE deletion. The resulting strains were indeed able to grow in glucose minimal medium without amino acid supplementation, but at the cost of lowered growth rates and KIC production parameters. The best production performance was obtained with strain MV-KICF1, which carried besides the ilvE start codon exchange three copies of a gene for a feedback-resistant 2-isopropylmalate synthase, one copy of a gene for a feedback-resistant acetohydroxyacid synthase and deletions of ltbR and iolR encoding transcriptional regulators. In the presence of 1 mM L-isoleucine, MV-KICF1 accumulated 47 mM KIC (6.1 g l-1) with a yield of 0.20 mol/mol glucose and a volumetric productivity of 1.41 mmol KIC l-1 h-1. Since MV-KICF1 is plasmid free and lacks heterologous genes, it is an interesting strain for industrial application and as platform for the production of KIC-derived compounds, such as 3-methyl-1-butanol.

Received 14 July, 2014; revised 7 October, 2014; accepted 7 October, 2014. For correspondence. *E-mail j.van.ooyen@fz-juelich .de; Tel. +49 2461 616205; Fax +49 2461 612710. **E-mail m.bott @fz-juelich.de; Tel. +49 2461 613294; Fax +49 2461 612710. Michael Vogt and Sabine Haas contributed equally to this work. Microbial Biotechnology (2015) 8(2), 351-360 doi:10.1111/1751-7915.12237

Funding Information No funding information provided.

Introduction

Corynebacterium glutamicum is the major host for biotechnological production of amino acids, the most important ones being the flavor enhancer L-glutamate and the feed additive L-lysine. In the past decades, C. glutamicum strains have been developed for the production of various other commercially interesting compounds (Becker and Wittmann, 2012), including organic acids (Okino etal., 2008; Litsanov etal., 2012a,b; Wieschalka etal., 2013), diamines (Mimitsuka etal., 2007; Kind and Wittmann, 2011; Schneider and Wendisch, 2011) or alcohols (Inui etal., 2004; Smith etal., 2010; Blombach etal., 2011). Besides small molecules, also heterologous proteins can be efficiently produced with this Gram-positive bacterium (Scheele etal., 2013, and references therein). Thus, C. glutamicum has become a production platform in white biotechnology. Three monographs (Eggeling and Bott, 2005; Burkovski, 2008; Yukawa and Inui, 2013) document the rapidly increasing knowledge on this species, which is based on the genome sequence (Ikeda and Nakagawa, 2003; Kalinowski etal., 2003) and efficient techniques for its genetic engineering (Kirchner and Tauch, 2003).

The spectrum of amino acids produced with C. glutamicum includes the essential branched-chain amino acids (BCAAs) L-valine, L-isoleucine and L-leucine, which are produced in quantities of up to 5000 tons per year in a steadily growing market (Becker and Wittmann, 2012). They have different applications in the food, feed and pharmaceutical industry (Park and Lee, 2010). The biosynthesis pathways of the BCAAs in C. glutamicum are overlapping and partly share the same precursors and enzymes (Fig. 1). The direct precursors of L-valine, L-isoleucine and L-leucine are 2-ketoisovalerate (KIV), 2-keto-3-methylvalerate (KMV), and 2-ketoisocaproate (KIC) respectively. These keto acids are predominantly transaminated to the respective amino acids by the transaminase IlvE (Radmacher etal., 2002; Marienhagen etal., 2005). Similar to their corresponding amino acids, KIV, KMV and KIC have a variety of applications in the medical, biological and food area, since they play an

Fig. 1. Biosynthesis pathways and their control by various regulatory mechanisms of the three branched-chain amino acids and the respective keto acids in C. glutamicum. Enzymes and their corresponding genes are shown in boxes. Lines with '+' indicate activation of gene expression; '-' indicates repression of gene expression (solid lines) or transcription attenuation or feedback inhibition (dashed lines). 'Leu', 'Val' and 'Ile' indicate the presence of l-leucine, l-valine and l-isoleucine respectively. Not shown is the avtA gene encoding the branched-chain amino acid transaminase AvtA, which predominantly transaminates 2-ketoisovalerate to l-valine. Abbreviations: AHAIR, acetohydroxyacid isomeroreductase; AHAS, acetohydroxyacid synthase; BCAA-E, branched-chain amino acid exporter (BrnFE); BCAA-T, branched-chain amino acid transaminase IlvE; DHAD, dihydroxyacid dehydratase; IPMD, 3-isopropylmalate dehydratase; IPMDH, 3-isopropylmalate dehydrogenase; IPMS, 2-isopropylmalate synthase; Lrp, leucine-responsive regulatory protein; LtbR, leucine and tryptophane biosynthesis regulator; TD, threonine dehydratase (threonine ammonia-lyase).

important role in living organisms as regulatory factors in metabolism and key intermediates in biosynthesis (Krause etal, 2010; Zhu etal, 2011; Buckle-Vallant etal., 2014). They are used, for example, in the therapy of chronic kidney disease patients (Aparicio etal., 2012). Similar to L-leucine, KIC has anti-catabolic properties through inhibition of muscle proteolysis and provokes

enhancement of protein synthesis, especially in the skeletal muscle (Escobar etal., 2010; Zanchi etal., 2011). Additionally, an insulin-releasing action of KIC (Heissig etal., 2005) and an inhibitory effect on glucagon release (Leclercq-Meyer etal., 1979) were discussed. It has been shown that KIC can also serve as a basis for the production of the biofuel isopentanol (Cann and Liao, 2010).

KIV, KMV and KIC are mainly produced by chemical synthesis using harsh reaction conditions and multiple purification steps resulting in plenty of waste (Cooper etal., 1983). The biotechnological production of these keto acids is thus an interesting alternative. Besides a biotransformation process with Rhodococcus opacus using L-leucine as substrate for KIC formation (Zhu etal., 2011), fermentative processes with glucose as substrate have recently been described for the production of KIV (Krause et al., 2010) and KIC (Bückle-Vallant etal, 2014), showing that deletion of ilvE in certain engineered C. glutamicum strains results in KIC formation. Whereas these strains contained plasmids and in part heterologous genes, the C. glutamicum KIC production strains developed in our work are plasmid free and lack heterologous genes.

Results and discussion

Initial studies on KIC production using plasmid-containing strains of C. glutamicum

Based on recently developed efficient production strains of C. glutamicum ATCC 13032 (Abe etal., 1967) for L-leucine (Vogt etal., 2014), we intended to modify these strains for the production of KIC. The conversion of KIC to L-leucine is catalysed by the transaminase IlvE, which also converts KIV to L-valine and KMV to L-isoleucine using L-glutamate as amino donor (Radmacher etal., 2002; Marienhagen etal., 2005). An ilvE deletion has been reported to cause auxotrophy for L-leucine and L-isoleucine, but not for L-valine, since the transaminase AvtA also effectively converts KIV to L-valine using L-alanine as amino donor (Marienhagen etal., 2005). According to this knowledge, deletion of ilvE in L-leucine production strains should lead to the accumulation of KIC and potentially also KMV. In a first series of experiments, we deleted ilvE in the wild-type C. glutamicum ATCC 13032 and transformed the AilvE mutant with plasmid pAN6-/euA_B018, carrying an IPTG-inducible leuA allele encoding a feedback-resistant 2-isopropylmalate synthase (IPMS) (Vogt etal., 2014). 2-Isopropylmalate synthase of C. glutamicum is strongly inhibited by L-leucine with a K of 0.4 mM (Patek etal., 1994) and the presence of a feedback-resistant variant is the key for L-leucine overproduction (Vogt etal., 2014). The AilvE mutant and the AilvE strain with plasmid pAN6-/euA_B018 were cultivated in 500 ml baffled Erlenmeyer flasks with 50 ml CGXII minimal medium (Keilhauer etal., 1993) with 4% (w/v) glucose, 1 mM L-leucine and 1 mM L-isoleucine at 30°C and 120 rpm on a rotary shaker. Keto acids and amino acids were quantified by highperformance liquid chromatography as described (Vogt etal., 2014). Chromosomal in-frame deletions and integrations of DNA fragments were performed by two-

step homologous recombination using the vector pK19mobsacB (Schäfer etal., 1994) and a method described previously (Niebisch and Bott, 2001).

Corynebacterium glutamicum AilvE exhibited a growth rate of 0.38 ± 0.01 h-1 and excreted up to 5 mM KIV but no detectable concentrations of KIC (detection limit < 0.1 mM), whereas C. glutamicum AilvE pAN6-leuA_B018 showed a growth rate of 0.30 ± 0.01 h-1 and accumulated 37 ± 0.7 mM KIC in the supernatant when induced with 0.1 mM IPTG, confirming that overexpression of the leuA allele encoding the feedback-resistant IPMS increased metabolic flux into the leucine pathway (Fig. 1). Surprisingly, C. glutamicum AilvE carrying pAN6-/euA_B018 also accumulated L-leucine (12.3 ± 0.4 mM) and in fact was only auxotrophic for L-isoleucine, but not for L-leucine. A possible limitation of L-valine due to a high metabolic flux from KIV towards KIC was excluded for this strain since additional supplementation of L-valine did not improve growth (data not shown). Accumulation of L-leucine was also reported for other KIC-producing AilvE strains and explained by the activity of unspecific transaminases (e.g. AlaT or AvtA) using KIC as substrate when it is present in high concentrations (Bückle-Vallant etal., 2014). Consequently, supplementation of the medium with L-leucine was omitted in the following cultivations. The formation of L-leucine as by-product additionally necessitates the presence of feedback-resistant IPMS for KIC overproduction. The results described above demonstrated that our previously described L-leucine producers (Vogt etal., 2014) can serve as basis for the construction of KIC production strains.

Deletion of ilvE in plasmid-free L-leucine production strains

Analogous to our strategy used for L-leucine strain development (Vogt etal., 2014), we intended to construct KIC production strains devoid of plasmids, heterologous genes and auxotrophies. Depending on the composition of the medium used in the fermentation process, auxotrophies can necessitate the addition of supplements, increasing the costs of the fermentation process. Plasmids usually necessitate the addition of antibiotics to the medium, which is undesirable for production strains applied in the food and feed industry and can be prohibited by regulatory authorities (Tauch etal., 2002). Moreover, the absence of plasmids, antibiotic resistance markers and heterologous genes often results in more stable producer strains (Patek, 2007). The use of heterologous genes is also an undesired trait for strains used in the food and feed industry. In a first attempt to construct a plasmid-free KIC producer, we deleted the ilvE gene in the previously constructed L-leucine producer MV-Leu20 (Table 1; Vogt etal., 2014), which contains a deletion of the ltbR gene, encoding a

Table 1. Strains and plasmids used in this studyab.

Strain or plasmid Relevant characteristics'3 Source or reference

C. glutamicum strains

Wild type ATCC 13032, biotin-auxotrophic Abe and colleagues (1967)

AilvE ATCC 13032 derivative with in-frame deletion of ilvE Marienhagen and

colleagues (2005)

MV-Leu20 Rationally designed C. glutamicum l-leucine producer (AltbR Vogt and colleagues (2014)

AleuA::Ptuf-leuA_B018)

MV-Leu20 AilvE MV-Leu20 derivative with in-frame deletion of ilvE This study

SH-KIC20 MV-Leu20 derivative with chromosomal replacement of ATG start codon of ilvE by This study

GTG start codon

MV-LeuF1 Rationally designed C. glutamicum l-leucine producer (AltbR::Ptuf-leuA_B018 Vogt and colleagues (2014)

AleuA::Ptuf-leuA_B018 IR(cg1121/1122)::Pfuf-leuA_B018 AiolR ilvN_fbr)

MV-KICF1 MV-LeuF1 derivative with chromosomal replacement of ATG start codon of ilvE by This study

GTG start codon

AilvE Acg0018 AilvE derivative with cg0018 in-frame deletion This study

AilvE Acg1121 AilvE derivative with cg1121 in-frame deletion This study

AilvE Acg1219 AilvE derivative with cg1219 in-frame deletion This study

AilvE Acg1419 AilvE derivative with cg1419 in-frame deletion This study

AilvE Acg1658 AilvE derivative with cg1658 in-frame deletion This study

AilvE Acg2557 AilvE derivative with cg2557 in-frame deletion This study

AilvE Acg2676 AilvE derivative with cg2676 in-frame deletion This study

AilvE Acg3334 AilvE derivative with cg3334 in-frame deletion This study

AilvE Acg1121::cg1121 AilvE Acg1121 derivative with re-integrated gene cg1121 into its wild-type locus This study

E. coli strains

DH5a F- Ф80lacZAM15 A(lacZYA-argF)U169 recA1 endA1 hsdR17 (rK-, mk+) phoA Invitrogen (Karlsruhe,

supE44 X- thi-1 gyrA96 relA1 Germany)

Plasmids

pAN6 Kanr; E. coli/C. glutamicum shuttle vector for inducible gene expression (Ptac, Frunzke and colleagues

lacIq, pBL1 pUC18 oriVE.coli, pBL1 oriVaglutamicum) (2008)

pAN6-leuA_B018 Kanr; pAN6 derivative containing leuA allele coding for feedback-resistant Vogt and colleagues (2014)

2-isopropylmalate synthase under control of the tac promoter

pAN6-leuA_B018-cg1121 Kanr; pAN6-leuA_B018 derivative carrying additional gene coding for cg1121 This study

along with its upstream (94 bp) and downstream (305 bp) regions

pK19mobsacB Kanr; vector for allelic exchange in C. glutamicum (pK18 oriVE.coii sacB lacZa) Schäfer and colleagues

(1994)

pK19mobsacB-AilvE Kanr, pK19mobsacB derivative for in-frame deletion of gene ilvE Marienhagen and

colleagues (2005)

pK19mobsacB-GTG-ilvE Kanr, pK19mobsacB derivative for replacement of ATG start codon of ilvE by GTG This study

pK19mobsacB-Acg0018 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg0018 This study

pK19mobsacB-Acg1121 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1121 This study

pK19mobsacB-Acg1219 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1219 This study

pK19mobsacB-Acg1419 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1419 This study

pK19mobsacB-Acg1658 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg1658 This study

pK19mobsacB-Acg2557 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg2557 This study

pK19mobsacB-Acg2676 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg2676 This study

pK19mobsacB-Acg3334 Kanr, pK19mobsacB derivative for in-frame deletion of gene coding for Cg3334 This study

pK19mobsacB-cg1121 Kanr, pK19mobsacB derivative for re-integration of gene coding for Cg1121 into This study

its wild-type locus

a. All constructed plasmids as well as chromosomal deletions and integrations in engineered strains were verified by DNA sequencing.

b. Plasmid constructions were performed in E. coli DH5a. Description of plasmid constructions and used DNA oligonucleotides (Table S1) can be found in the Supporting Information.

c. Kanr, kanamycin resistance

repressor of the L-leucine biosynthesis genes, and a replacement of the wild-type leuA gene by the feedback-resistant variant /euA_B018 under control of the strong tuf promoter (Vogt etal., 2014). In shake flask cultivations with CGXII medium containing 4% (w/v) glucose, MV-Leu20 accumulated about 20 mM L-leucine. When cultivated in the same medium supplemented with 1 mM L-isoleucine, the strain MV-Leu20 AilvEaccumulated 18.0 ± 1.6 mM KIC in the supernatant and formed as by-products 5.6 ± 0.3 mM L-leucine, 2.1 ± 0.5 mM KIV and 7.3 ± 1.7 mM KMV. Since KIV and KMV are substrates of the tran-

saminase IlvE (Marienhagen etal., 2005), the ilvE deletion leads to an accumulation of these keto acids. The presumably low concentrations of L-isoleucine and L-valine in strain MV-Leu20 AilvE may also contribute to overproduction of KIV and KMV by reducing the feedback-inhibition of threonine dehydratase (encoded by ilvA) by L-isoleucine (Möckel etal., 1992) and of acetohydroxyacid synthase (encoded by ilvBN) by L-valine and L-isoleucine (Eggeling etal., 1987).

As mentioned above, L-leucine formation in the absence of the transaminase IlvE is presumably due to

high cytoplasmic KIC concentrations, allowing its conversion by other transaminases such as AvtAthat have weak affinities for KIC (Marienhagen etal., 2005). To test this assumption, we measured the cytoplasmic KIC concentrations in the wild type and strain MV-Leu20 AilvE. Cells were grown in CGXII medium with 4% (w/v) glucose, harvested in the early exponential phase (optical density at 600 nm (OD600 = 5), and cytoplasmic concentrations were determined as described (Paczia etal., 2012). The internal KIC concentration of the wild type was below the detection limit of 5 |M, whereas MV-Leu20 AilvE accumulated approximately 2.8 mM KIC inside the cell, corresponding to a more than 500-fold increase.

Exchange of the ilvE start codon in plasmid-free L-leucine production strains

To avoid the L-isoleucine auxotrophy of strain MV-Leu20 A/'lvE, we intended to reduce IlvE activity to a value that was high enough to provide L-isoleucine for growth, but low enough to allow KIC overproduction. For this purpose, the ATG start codon of ilvE was exchanged against GTG, which should decrease the translation rate (Becker etal., 2010) of the ilvE transcript and thereby reduce the specific IlvE activity. The start codon exchange was performed in the L-leucine producers MV-Leu20 and MV-LeuF1 (Table 1). The latter strain contains (i) three copies of the feuA_B018 gene in the chromosome under control of the tuf promoter, two of them replacing ltbR and the native leuA gene, (ii) a deletion of iolR (Klaffl etal., 2013) for enhanced glucose uptake and (iii) a feedback-resistant acetohydroxyacid synthase encoded by /lvN_fbr (Vogt etal, 2014). The strains SH-KIC20 (from MV-Leu20) and MV-KICF1 (from MV-LeuF1) resulting from the ilvE start codon exchange (Table 1) were cultivated in CGXII medium with 4% (w/v) glucose to test for KIC accumulation in the supernatant (Fig. 2).

Without supplementation of L-isoleucine, SH-KIC20 showed a strongly reduced growth rate of 0.08 ± 0.01 h-1 compared with the ancestor strain MV-Leu20 (0.31 ± 0.01 h-1). This phenotype suggests that the ilvE start codon exchange reduced the availability of BCAAs and consequently the growth rate. Strain SH-KIC20 accumulated about 19 mM KIC in 49 h, which correlates with the KIC concentration produced by MV-Leu20 AilvE. The slower growth of SH-KIC20 led to a lowered volumetric productivity of 0.38 mmol l-1 h-1 in comparison to MV-Leu20 AilvE (approximately 0.8 mmol l-1 h-1). Strain MV-KICF1 showed a growth rate of only 0.03 ± 0.01 h-1, which is 85% lower than the one of the parent strain (| = 0.20 ± 0.01 h-1), and yielded a maximal KIC concentration of 33 mM after 95 h (Table 2). Supplementation of the medium with 1 mM L-isoleucine enabled MV-KICF1 to reach the same growth rate (| = 0.21 ±

0 10 20 30 40 50 Time (h)

0 10 20 30 40 50 60 70 80 90 100 Time (h)

Fig. 2. Growth and KIC formation of different C. glutamicum strains in shake flasks with CGXII minimal medium containing 4% (w/v) glucose.

A. MV-Leu20 without supplements (growth, ■; KIC, □), MV-Leu20 AilvE supplemented with 1 mM l-isoleucine (growth, •; KIC, O) and SH-KIC20 without supplements (growth, ▲; KIC, A).

B. MV-LeuF1 without supplements (growth, ▼; KIC, V), MV-KICF1 without supplements (growth, KIC, <) and MV-KICF1 with 1 mM l-isoleucine (growth, ♦; KIC, O). The data represent mean values and standard deviations obtained from three independent cultivations.

0.01 h-1) as its parent MV-LeuF1 and to form 47 ±q 4 mM KIC (6.1 gl-1) after 32 h with a yield of 0.20 ± 0.02 mol KIC per mol of glucose and a productivity of 1.41 ± 0.13 mmol KIC l-1 h-1 (Table 2). Both SH-KIC20 and MV-KICF1 formed KIV, KMV and L-leucine as by-products (Table 2).

The results described above demonstrate that the start codon exchange for reduction of IlvE activity was successful and allowed growth and KIC accumulation without supplementation of BCAAs; however, the production parameters were lower compared with supplementation with 1 mM L-isoleucine (Table 2). A successful industrial application depends on high product yields combined with sufficient cell growth, resulting in

Table 2. Growth and production parameters of strains MV-KICF1 and SH-KIC20 in shake flask cultivationsab.

Parameter MV-KICF1 + 1 mM l-isoleucine MV-KICF1 without l-isoleucine SH-KIC20d without l-isoleucine

Growth rate (h-1) 0.21 i 0.01 0.03 i 0.01 0.08 i 0.01

KIC (mM) 46.l i 4.1 31.8 i 2.1 18.8 i 0.6l

By-products'3:

KIV (mM) 13.3 i 2.2 19.0 i 4.1 2.6 i 0.3

KMV (mM) 8.8 i 1.3 4.9 i 0.6 8.l i 0.1

l-leucine (mM) 3.0 i 0.2 10.3 i 3.1 4.8 i 0.2

Molar product yield (mol KIC per mol glucose) 0.204 i 0.018 0.143 i 0.010 0.084 i 0.001

Volumetric productivity (mmol KIC l-1 h-1) 1.41 i 0.13 0.34 i 0.02 0.38 i 0.02

a. Cultivations were performed in 500 ml baffled shake flasks containing 50 ml CGXII minimal medium with 4% (w/v) glucose. Supplementation of l-isoleucine is indicated.

b. Mean values and standard deviations from three independent cultivations are shown.

c. Concentrations of l-valine and l-isoleucine were below 2 mM.

d. Cultivation of SH-KIC20 supplemented with 1 mM l-isoleucine was not tested.

competitive productivity. Therefore, the addition of L-isoleucine is still important to improve growth of the constructed strains to reach better productivity values. As an alternative approach to adjust the IlvE activity to an optimal value for prototrophic growth with simultaneous KIC production, ilvE gene expression could be fine-tuned by testing promoters with varying strength (Vasicová etal., 1999; Hammer etal., 2006).

Recently, Bückle-Vallant and colleagues (2014) described a plasmid-based C. glutamicum strain for the production of KIC. This strain is characterized by deletions of the genes ilvE, ItbR, prpCI and prpC2, an exchange of the two gltA promoters (van Ooyen etal., 2011) by the mutated dapA promoter L1 (Vasicová etal., 1999) to reduce citrate synthase activity (van Ooyen etal., 2012), and plasmid-based overexpression of ilvBNCD and a leuA allele of Escherichia coli encoding a feedback-resistant IPMS. This strain accumulated up to 71 mM KIC when cultivated with glucose plus acetate as carbon sources and supplemented with 2 mM each of L-isoleucine and L-valine. Under cultivation conditions comparable to ours, i.e. without acetate, this strain reached KIC titres

(54 i 4 mM) and yields (0.22 mol per mol of glucose) in a similar range as strain MV-KICF1 when supplemented with L-isoleucine. In comparison, the biotransformation with Rhodococcus opacus transcribed by Zhu and colleagues (2011) reached about 10 mM KIC using 39 mM L-leucine as substrate.

2-Ketoisocaproate transport

When MV-KICF1 was batch cultivated in a bioreactor as described (Vogt et ai., 2014) using a medium supplemented with 1 mM L-isoleucine, comparable KIC titres as in shake flasks of about 50 mM were achieved (data not shown). Interestingly, additional feeding of glucose in fed-batch experiments did not further increase the KIC concentrations and led to an arrest of cell growth and glucose consumption (data not shown). A possible explanation is a block of the L-leucine biosynthesis pathway by elevated KIC concentrations, as Bückle-Vallant and colleagues (2014) found competitive inhibition of IPMS by KIC and non-competitive inhibition of 3-isopropylmalate dehydratase by KIC. Additionally, a competitive inhibition

Table 3. Putative transporter genes showing increased expression in a KIC producer.

mRNA ratioa (MV-Leu20

Gene Annotation Ai/vE Wild type TMHb

cg0018 putative membrane protein, conserved 3.5 9

cg1121 putative permease of the major facilitator superfamily 2.2 l

cg1219 putative membrane protein 3.5 10

cg1419 putative Na+-dependent transporter, bile acid:Na+ symporter BASS family l.2 8

cg1658 putative permease of the major facilitator superfamily 35.6 12

cg255l putative secondary Na+/bile acid symporter, bile acid:Na+ symporter BASS family 2.5 8

cg26l6 putative ABC-type dipeptide/oligopeptide/nickel transport system, permease component 2.1 6

cg3334 putative arabinose efflux permease, MFS type 2.0 12

a. Transcriptome analyses of KIC producer MV-Leu20 AilvE in comparison to the wild type were performed using DNA microarrays as described (Vogt etal., 2014). Candidate transporter genes were chosen based on an mRNA ratio (MV-Leu20 AilvE/wild type) of > 2, an annotation as (putative) membrane or transporter proteins and the prediction of multiple transmembrane helices in the encoded proteins. Data represent mean values of at least two (maximum four) evaluable microarray experiments (P-value < 0.05).

b. TMH, number of transmembrane helices predicted with the sosui engine version 1.11.

of acetohydroxyacid synthase by KIV has been reported by Krause and colleagues (2010). To avoid a reduced flux into the leucine synthesis pathway by competitive inhibition of IPMS by KIC, the concentration of KIV within the cell needs to be increased and/or the concentration of KIC within the cell should be decreased. These concentrations are determined on one hand by the rates of synthesis and further metabolic conversion and on the other hand by the rates of export to and import from the supernatant.

Knowledge on KIC transport is very limited. Obviously, as shown by our studies and that of Bückle-Vallant and colleagues (2014), KIC can leave the cell, and previous studies by Groeger and Sahm (1987) demonstrated that KIC can enter the cell. As previously described for L-isoleucine (Zittrich and Krämer, 1994), passive diffusion, carrier-mediated uptake and carrier-mediated excretion must be considered as possibilities for an amphiphilic solute like KIC to cross the cytoplasmic membrane. Due to its similarity to L-isoleucine, it seems likely that also KIC is able to diffuse across the membrane. However, a necessity to possess carriers for KIC import or KIC export is not obvious, in contrast to the advantage of having importers for amino acids and exporters for non-catabolizable amino acids. In C. glutamicum, the export of BCAAs and methionine is catalysed by the exporter BrnFE (Kennerknecht etal., 2002; Trötschel etal., 2005; Xie etal., 2012), but evidence is available that BrnFE is not involved in KIC export (Radespiel, 2010). The identification of transporters can be beneficial for biotechnological processes to increase productivity since transport of desired products into the medium often represents a bottleneck. For example, export was identified as a limiting factor for L-isoleucine production with C. glutamicum (Morbach etal., 1996), and the production of this BCAAwas improved by overexpression of the respective transporter encoded by brnFE (Kennerknecht etal., 2002; Xie etal., 2012).

In order to test if a KIC exporter is present in C. glutamicum that might be useful to improve KIC overproduction, we searched for genes showing increased mRNA levels during KIC production by performing comparative transcriptome analyses using DNA microarrays as described previously (Vogt etal., 2014). KIC producer strain MV-Leu20 AilvE was compared with the wild type to determine differentially expressed genes, resulting in a list of eight candidate transporter genes (Table 3). Each of these genes was deleted in the C. glutamicum AilvE background, and the resulting double deletion mutants were transformed with pAN6-/eu^_B018. When cultivated in glucose minimal medium with 0.1 mM IPTG, one of the eight strains, which contained a deletion of cg1121 (annotated as putative permease of the major facilitator superfamily), showed a reduced growth rate (0.23 ± 0.01 h-1) and a reduced maximal KIC titre

Fig. 3. Complementation of the effects on growth and KIC accumulation caused by deletion of cg1121 in strain C. glutamicum AilvE carrying pAN6-leu^_B018.

A. Growth of strains AilvE with pAN6-leu^_B018 (■), AilvE Acg1121 with pAN6-leu^_B018 (•), AilvE Acg1121::cg1121 with pAN6-leu^_B018 (A) and AilvE Acg1121 with pAN6-leu^_B018-cg1121 (V) are shown.

B. Maximal KIC concentrations reached after 32 h cultivation in 500 ml baffled shake flasks with 50 ml CGXII minimal medium containing 4% (w/v) glucose and 0.1 mM IPTG at 30°C and 120 rpm on a rotary shaker. The deletion of cg1121 was complemented either by genomic reintegration of cg1121 (AilvE Acg1121::cg1121 pAN6-leu^_B018) or by plasmid-borne expression of cg1121 (AilvE Acg1121 pAN6-leu^_B018-cg1121). The data represent mean values and standard deviations obtained from three independent cultivations.

(22.4 mM) compared with the reference strain AilvE pAN6-leuA_B018 (0.30 ± 0.01 h-1, 37 mM KIC). The phenotype could be complemented by reintegration of cg1121 into the genome of strain AilvE Acg1121 or by plasmid-borne expression of cg1121 (Fig. 3). However, the specific KIC export rates (determined as described by Kennerknecht etal., 2002) of strain A/lvE Acg1121 (6.5 ± 1.0 nmol min-1 gCDW-1) were not significantly reduced compared with that of the reference strain A/lvE (8.0 ± 1.5 nmol min-1 gcDw-1) and the cytoplasmic KIC concentrations of the two strains at an OD600 of 5 were comparable. Therefore, the role of Cg1121 for growth and KIC production remains unclear and needs further investigations.

Acknowledgements

We thank Dr Nicole Paczia (IBG-1: Biotechnology, Forschungszentrum Jülich, Germany) for her help with the measurements of cytoplasmic 2-ketoisocaproate concentrations.

Conflict of interest

None declared.

References

Abe, S., Takayama, K.I., and Kinoshita, S. (1967) Taxonomi-cal studies on glutamic acid-producing bacteria. JGenAppl Microbiol 13: 279-301.

Aparicio, M., Bellizzi, V., Chauveau, P., Cupisti, A., Ecder, T., Fouque, D., etal. (2012) Keto acid therapy in predialysis chronic kidney disease patients: final consensus. J Ren Nutr 22: S22-S24.

Becker, J., and Wittmann, C. (2012) Systems and synthetic metabolic engineering for amino acid production - the heartbeat of industrial strain development. Curr Opin Biotechnol 23: 718-726.

Becker, J., Buschke, N., Bücker, R., and Wittmann, C. (2010) Systems level engineering of Corynebacterium glutamicum - reprogramming translational efficiency for superior production. Eng Life Sci 10: 430-438.

Blombach, B., Riester, T., Wieschalka, S., Ziert, C., Youn, J.W., Wendisch, V.F., and Eikmanns, B.J. (2011) Corynebacterium glutamicum tailored for efficient isobutanol production. Appl Environ Microbiol 77: 33003310.

Burkovski, A. (2008) Corynebacteria: Genomics and Molecular Biology. Norfolk, UK: Caister Academic Press.

Bückle-Vallant, V., Krause, F.S., Messerschmidt, S., and Eikmanns, B.J. (2014) Metabolic engineering of Corynebacterium glutamicum for 2-ketoisocaproate production. Appl Microbiol Biotechnol 98: 297-311.

Cann, A.F., and Liao, J.C. (2010) Pentanol isomer synthesis in engineered microorganisms. Appl Microbiol Biotechnol 85: 893-899.

Cooper, A.J., Ginos, J.Z., and Meister, A. (1983) Synthesis and properties of the a-keto acids. Chem Rev 83: 321358.

Eggeling, L., and Bott, M. (2005) Handbook of Corynebacterium glutamicum. Boca Raton, FL, USA: Taylor & Francis.

Eggeling, I., Cordes, C., Eggeling, L., and Sahm, H. (1987) Regulation of acetohydroxy acid synthase in Corynebacterium glutamicum during fermentation of a-ketobutyrate to L-isoleucine. Appl Microbiol Biotechnol 25: 346-351.

Escobar, J., Frank, J.W., Suryawan, A., Nguyen, H.V., Van Horn, C.G., Hutson, S.M., and Davis, T.A. (2010) Leucine and a-ketoisocaproic acid, but not norleucine, stimulate skeletal muscle protein synthesis in neonatal pigs. J Nutr 140: 1418-1424.

Frunzke, J., Engels, V., Hasenbein, S., Gätgens, C., and Bott, M. (2008) Co-ordinated regulation of gluconate catabolism and glucose uptake in Corynebacterium glutamicum by two functionally equivalent transcriptional regulators, GntR1 and GntR2. Mol Microbiol 67: 305-322.

Groeger, U., and Sahm, H. (1987) Microbial production of L-leucine from a-ketoisocaproate by Corynebacterium glutamicum. Appl Microbiol Biotechnol 25: 352-356.

Hammer, K., Mijakovic, I., and Jensen, P.R. (2006) Synthetic promoter libraries - tuning of gene expression. Trends Biotechnol 24: 53-55.

Heissig, H., Urban, K.A., Hastedt, K., Zunkler, B.J., and Panten, U. (2005) Mechanism of the insulin-releasing action of a-ketoisocaproate and related a-keto acid anions. Mol Pharmacol 68: 1097-1105.

Ikeda, M., and Nakagawa, S. (2003) The Corynebacterium glutamicum genome: features and impacts on biotechnological processes. Appl Microbiol Biotechnol 62: 99-109.

Inui, M., Kawaguchi, H., Murakami, S., Vertes, A.A., and Yukawa, H. (2004) Metabolic engineering of Corynebacterium glutamicum for fuel ethanol production under oxygen-deprivation conditions. J Mol Microbiol Biotechnol 8: 243-254.

Kalinowski, J., Bathe, B., Bartels, D., Bischoff, N., Bott, M., Burkovski, A., etal. (2003) The complete Corynebacterium glutamicum ATCC 13032 genome sequence and its impact on the production of L-aspartate-derived amino acids and vitamins. J Biotechnol 104: 5-25.

Keilhauer, C., Eggeling, L., and Sahm, H. (1993) Isoleucine synthesis in Corynebacterium glutamicum: molecular analysis of the ilvB-ilvN-ilvCoperon. J Bacteriol 175: 55955603.

Kennerknecht, N., Sahm, H., Yen, M.R., Patek, M., Saier Jr, M.H., Jr, and Eggeling, L. (2002) Export of L-isoleucine from Corynebacterium glutamicum: a two-gene-encoded member of a new translocator family. J Bacteriol 184: 3947-3956.

Kind, S., and Wittmann, C. (2011) Bio-based production of the platform chemical 1,5-diaminopentane. Appl Microbiol Biotechnol 91: 1287-1296.

Kirchner, O., and Tauch, A. (2003) Tools for genetic engineering in the amino acid-producing bacterium Corynebacterium glutamicum. J Biotechnol 104: 287299.

Klaffl, S., Brocker, M., Kalinowski, J., Eikmanns, B.J., and Bott, M. (2013) Complex regulation of the phosphoenolpyruvate carboxykinase gene pck and characterization of its GntR-type regulator lolR as a repressor of myo-inositol utilization genes in Corynebacterium glutamicum. J Bacteriol 195: 4283-4296.

Krause, F.S., Blombach, B., and Eikmanns, B.J. (2010) Metabolic engineering of Corynebacterium glutamicum for 2-ketoisovalerate production. Appl Environ Microbiol 76: 8053-8061.

Leclercq-Meyer, V., Marchand, J., Leclercq, R., and Malaisse, W.J. (1979) Interactions of a-ketoisocaproate, glucose and arginine in the secretion of glucagon and insulin from the perfused rat pancreas. Diabetologia 17: 121-126.

Litsanov, B., Brocker, M., and Bott, M. (2012a) Toward homosuccinate fermentation: metabolic engineering of Corynebacterium glutamicum for anaerobic production of succinate from glucose and formate. Appl Environ Microbiol 78: 3325-3337.

Litsanov, B., Kabus, A., Brocker, M., and Bott, M. (2012b) Efficient aerobic succinate production from glucose in minimal medium with Corynebacterium glutamicum. Microb Biotechnol 5: 116-128.

Marienhagen, J., Kennerknecht, N., Sahm, H., and Eggeling, L. (2005) Functional analysis of all aminotransferase proteins inferred from the genome sequence of Corynebacterium glutamicum. J Bacteriol 187: 76397646.

Mimitsuka, T., Sawai, H., Hatsu, M., and Yamada, K. (2007) Metabolic engineering of Corynebacterium glutamicum for cadaverine fermentation. Biosci Biotechnol Biochem 71: 2130-2135.

Morbach, S., Sahm, H., and Eggeling, L. (1996) L-Isoleucine production with Corynebacterium glutamicum: further flux increase and limitation of export. Appl Environ Microbiol 62: 4345-4351.

Möckel, B., Eggeling, L., and Sahm, H. (1992) Functional and structural analyses of threonine dehydratase from Corynebacterium glutamicum. J Bacteriol 174: 80658072.

Niebisch, A., and Bott, M. (2001) Molecular analysis of the cytochrome bc1-aa3 branch of the Corynebacterium glutamicum respiratory chain containing an unusual diheme cytochrome c1. Arch Microbiol 175: 282-294.

Okino, S., Noburyu, R., Suda, M., Jojima, T., Inui, M., and Yukawa, H. (2008) An efficient succinic acid production process in a metabolically engineered Corynebacterium glutamicum strain. Appl Microbiol Biotechnol 81: 459464.

van Ooyen, J., Emer, D., Bussmann, M., Bott, M., Eikmanns, B.J., and Eggeling, L. (2011) Citrate synthase in Corynebacterium glutamicum is encoded by two gltA transcripts which are controlled by RamA, RamB, and GlxR. J Biotechnol 154: 140-148.

van Ooyen, J., Noack, S., Bott, M., Reth, A., and Eggeling, L. (2012) Improved L-lysine production with Corynebacterium glutamicum and systemic insight into citrate synthase flux and activity. Biotechnol Bioeng 109: 2070-2081.

Paczia, N., Nilgen, A., Lehmann, T., Gätgens, J., Wiechert, W., and Noack, S. (2012) Extensive exometabolome analy-

sis reveals extended overflow metabolism in various microorganisms. Microb Cell Fact 11: 122.

Park, J.H., and Lee, S.Y. (2010) Fermentative production of branched chain amino acids: a focus on metabolic engineering. Appl Microbiol Biotechnol 85: 491-506.

Patek, M. (2007) Branched-chain amino acids. In Amino Acid Biosynthesis - Pathways, Regulation and Metabolic Engineering. Wendisch, V.F. (ed.). Berlin, Germany: Springer, pp. 129-162.

Patek, M., Krumbach, K., Eggeling, L., and Sahm, H. (1994) Leucine synthesis in Corynebacterium glutamicum: enzyme activities, structure of leuA, and effect of leuA inactivation on lysine synthesis. Appl Environ Microbiol60: 133-140.

Radespiel, T. (2010) Dissertation abstract. URL http:// kups.ub.uni-koeln.de/id/eprint/3156.

Radmacher, E., Vaitsikova, A., Burger, U., Krumbach, K., Sahm, H., and Eggeling, L. (2002) Linking central metabolism with increased pathway flux: L-valine accumulation by Corynebacterium glutamicum. Appl Environ Microbiol 68: 2246-2250.

Schäfer, A., Tauch, A., Jäger, W., Kalinowski, J., Thierbach,

G., and Pühler, A. (1994) Small mobilizable multi-purpose cloning vectors derived from the Escherichia coli plasmids pK18 and pK19: selection of defined deletions in the chromosome of Corynebacterium glutamicum. Gene 145: 69-73.

Scheele, S., Oertel, D., Bongaerts, J., Evers, S., Hellmuth,

H., Maurer, K.H., etal. (2013) Secretory production of an FAD cofactor-containing cytosolic enzyme (sorbitol-xylitol oxidase from Streptomyces coelicoloi) using the twin-arginine translocation (Tat) pathway of Corynebacterium glutamicum. Microb Biotechnol 6: 202-206.

Schneider, J., and Wendisch, V.F. (2011) Biotechnological production of polyamines by bacteria: recent achievements and future perspectives. Appl Microbiol Biotechnol 91: 17-30.

Smith, K.M., Cho, K.M., and Liao, J.C. (2010) Engineering Corynebacterium glutamicum for isobutanol production. Appl Microbiol Biotechnol 87: 1045-1055.

Tauch, A., Götker, S., Pühler, A., Kalinowski, J., and Thierbach, G. (2002) The alanine racemase gene alr is an alternative to antibiotic resistance genes in cloning systems for industrial Corynebacterium glutamicum strains. J Biotechnol 99: 79-91.

Trötschel, C., Deutenberg, D., Bathe, B., Burkovski, A., and Krämer, R. (2005) Characterization of methionine export in Corynebacterium glutamicum. J Bacteriol 187: 37863794.

Vasicova, P., Patek, M., Nesvera, J., Sahm, H., and Eikmanns, B. (1999) Analysis of the Corynebacterium glutamicum dapA promoter. J Bacteriol 181: 6188-6191.

Vogt, M., Haas, S., Klaffl, S., Polen, T., Eggeling, L., van Ooyen, J., and Bott, M. (2014) Pushing product formation to its limit: metabolic engineering of Corynebacterium glutamicum for L-leucine overproduction. Metab Eng 22: 40-52.

Wieschalka, S., Blombach, B., Bott, M., and Eikmanns, B.J. (2013) Bio-based production of organic acids with Corynebacterium glutamicum. Microb Biotechnol 6: 87-102.

Xie, X., Xu, L., Shi, J., Xu, Q., and Chen, N. (2012) Effect of transport proteins on L-isoleucine production with the L-isoleucine-producing strain Corynebacterium glutamicum YILW. J Ind Microbiol Biotechnol 39: 1549-1556.

Yukawa, H., and Inui, M. (2013) Corynebacterium glutamicum: Biology and Biotechnology. Heidelberg, Germany: Springer.

Zanchi, N.E., Gerlinger-Romero, F., Guimaraes-Ferreira, L., de Siqueira Filho, M.A., Felitti, V., Lira, F.S., et all. (2011) HMB supplementation: clinical and athletic performance-related effects and mechanisms of action. Amino Acids 40: 1015-1025.

Zhu, Y., Li, J., Liu, L., Du, G., and Chen, J. (2011) Production of a-ketoisocaproate via free-whole-cell biotransformation

by Rhodococcus opacus DSM 43250 with L-leucine as the substrate. Enzyme Microb Technol 49: 321-325. Zittrich, S., and Krämer, R. (1994) Quantitative discrimination of carrier-mediated excretion of isoleucine from uptake and diffusion in Corynebacterium glutamicum. J Bacteriol 176: 6892-6899.

Supporting information

Additional Supporting Information may be found in the online version of this article at the publisher's web-site:

Table S1. DNA oligonucleotides used in this study.

Copyright of Microbial Biotechnology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.