Increasing l-isoleucine production in Corynebacterium glutamicum by overexpressing global regulator Lrp and two-component export system BrnFE

Authors

  • L. Yin,

    1. State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi, China
    2. Key Laboratory of Industrial Biotechnology of Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China
    Search for more papers by this author
  • F. Shi,

    1. State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi, China
    Search for more papers by this author
  • X. Hu,

    1. State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi, China
    Search for more papers by this author
  • C. Chen,

    1. Key Laboratory of Industrial Biotechnology of Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China
    Search for more papers by this author
  • X. Wang

    Corresponding author
    1. Key Laboratory of Industrial Biotechnology of Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China
    • State Key Laboratory of Food Science and Technology, Jiangnan University, Wuxi, China
    Search for more papers by this author

Correspondence

Prof. Xiaoyuan Wang, Jiangnan University, 1800 Lihu Avenue, Wuxi 214000, China. E-mail: xiaoyuanwang@hotmail.com

Abstract

Aims

To increase the l-isoleucine production in Corynebacterium glutamicum by overexpressing the global regulator Lrp and the two-component export system BrnFE.

Methods and Results

The brnFE operon and the lrp gene were cloned into the shuttle vector pDXW-8 individually or in combination. The constructed plasmids were transformed into an l-isoleucine-producing strain C. glutamicum JHI3-156, and the l-isoleucine production in these different strains was analysed and compared. More l-isoleucine was produced when only Lrp was expressed than when only BrnFE was expressed. Significant increase in l-isoleucine production was observed when Lrp and BrnFE were expressed in combination. Compared to the control strain, l-isoleucine production in JHI3-156/pDXW-8-lrp-brnFE increased 63% in flask cultivation, and the specific yield of l-isoleucine increased 72% in fed-batch fermentation.

Conclusions

Both Lrp and BrnFE are important to enhance the l-isoleucine production in C. glutamicum.

Significance and Impact of the Study

The results provide useful information to enhance l-isoleucine or other branched-chain amino acid production in C. glutamicum.

Introduction

Corynebacterium glutamicum has been used to produce various amino acids including l-isoleucine. l-isoleucine biosynthetic pathway in C. glutamicum consists of 10 reaction steps (Park and Lee 2010). The key enzymes include aspartate kinase, homoserine dehydrogenase, homoserine kinase, threonine dehydrogenase and acetohydroxy acid synthase (AHAS). As the genome of C. glutamicum has been sequenced (Kalinowski et al. 2003), biosynthetic pathway of l-isoleucine became a major target for improving l-isoleucine production (Morbach et al. 1995, 1996a; Eggeling et al. 1996), but the export system, another major limiting step for l-isoleucine production (Morbach et al. 1996b), has been neglected. The two-component BrnFE encoded by the brnFE operon is the export system for branched-chain amino acids (BCAA) l-leucine, l-isoleucine and l-valine in C. glutamicum (Kennerknecht et al. 2002). The expression of the brnFE operon could be activated by high intracellular concentrations of BCAA (Hermann and Karmer 1996; Trötschel et al. 2005) and a leucine-responsive protein (Lrp) encoded by the gene lrp (Willins et al. 1991; Lange et al. 2012).

Lrp has been found in different bacteria as a global regulator controlling expression of various genes (Lintner et al. 2008). In Escherichia coli, Lrp regulates the expression of several operons involved in the metabolism of BCAA (Platko et al. 1990; Haney et al. 1992; Rhee et al. 1996; Kutukova et al. 2005). However, little is known about Lrp in C. glutamicum, the most important bacterium for BCAA production. Therefore, it is necessary to find out the effect of Lrp on the BCAA production in C. glutamicum. Recently, Lange et al. (2012) reported Lrp in C. glutamicum could activate the expression of the brnFE operon. Here, we demonstrate that brnFE and lrp overexpressed in an l-isoleucine-producing C. glutamicum subspecies lactofermentum JHI3-156 could lead to 63% increase in l-isoleucine production both in flask cultivation and in fed-batch fermentation. The results showed that both BrnFE and Lrp are important to enhance l-isoleucine production in C. glutamicum.

Materials and methods

Construction of strains and plasmids

Strains and plasmids used in this study are summarized in Table 1. Escherichia coli DH5α was grown at 37°C in LB media (5 g l−1 yeast extract, 10 g l−1 tryptone, and 10 g l−1 NaCl), and C. glutamicum subspecies lactofermentum ATCC13869 was grown at 30°C in LB medium supplemented with 5 g l−1 glucose. Corynebacterium glutamicum JHI3-156 is an l-isoleucine producer obtained by multiple random mutageneses (Peng et al. 2010). The shuttle vector pDXW-8 (Xu et al. 2010) between E. coli and C. glutamicum was used for gene expressions in C. glutamicum JHI3-156.

Table 1. Strains and plasmids used in the study
Strains/plasmidsDescriptionSources
Strains
DH5αWild-type Escherichia coliNovagen
ATCC13869Wild-type Corynebacterium glutamciumATCC
JHI3-156C. glitamicum l-isoleucine producerPeng et al. (2010)
JHI3-156/pDXW-8JHI3-156 harbouring pDXW-8This work
JHI3-156/pDXW-8-brnFEJHI3-156 harbouring pDXW-8-brnFEThis work
JHI3-156/pDXW-8-lrpJHI3-156 harbouring pDXW-8-lrpThis work
JHI3-156/ pDXW-8-lrp-brnFEJHI3-156 harbouring pDXW-8-lrp-brnFEThis work
Plasmids
pDXW-8Shutter expression vectorXu et al. (2010)
pDXW-8-brnFEpDXW-8 containing brnFEThis work
pDXW-8-lrppDXW-8 containing lrpThis work
pDXW-8-lrp-brnFEpDXW-8 containing both lrp and brnFEThis work

Primers used in this study are listed in Table 2. Most primer sequences were designed according to the genome sequence of C. glutamicum ATCC 13032 (Kalinowski et al. 2003). Genomic DNA of C. glutamicum subspecies lactofermentum ATCC13869 was isolated and used as templates for PCR-amplifying genes brnFE and lrp. The gene brnFE was amplified using primer pairs of brnFE-F and brnFE-R1, digested with EcoRI and NotI and ligated into pDXW-8 that was similarly digested, resulting the plasmid pDXW-8-brnFE. The gene lrp was amplified using primer pairs of lrp-F and lrp-R, digested with EcoRI and HindIII and ligated into pDXW-8 that was similarly digested, resulting the plasmid pDXW-8-lrp. The DNA fragment containing the tac promoter and brnFE in pDXW-8-brnFE was then amplified using primer pair brnFE-tac-F and brnFE-R2, digested with HindIII and ligated into pDXW-8-lrp that was similarly digested and treated with shrimp alkaline phosphatase, resulting the plasmid pDXW-8-lrp-brnFE. These plasmids were transformed into C. glutamicum JHI3-156, resulting strains JHI3-156/pDXW-8, JHI3-156/pDXW-8-lrp, JHI3-156/pDXW-8-brnFE and JHI3-156/pDXW-8-lrp-brnFE, respectively (Table 1).

Table 2. Primer sequences used in this study. Restriction enzyme sites are underlined
PrimersSequence (5′-3′)Restriction enzymePurpose
brnFE-FCCGGAATTCAGAAGGAGTAAATCGGTGCAAAAAACGCAAGAGATEcoRIPCR brnFE
brnFE-R1TTCCGCGGCCGCTTAGAAAAGATTCACCAGTCNotIPCR brnFE
lrp-FCCAGAATTCGAGAAGGAGATATACGATGAAGCTAGATTCCATTGAEcoRIPCR lrp
lrp-RCCGAAGCTTATCACACCTGGGGGCGAGCHindIIIPCR lrp
brnFE-tac-FCTCAAGCTTTCGGAAGCTGTGGTATGGHindIIIPCR brnFE with a tac promoter
brnFE-R2CCCAAGCTTTTAGAAAAGATTCACCAGTCHindIIIPCR brnFE with a tac promoter
16S rRNA-1ACCTGGAGAAGAAGCACCG RT-PCR 16S rRNA
16S rRNA-2TCAAGTTATGCCCGTATCG RT-PCR 16S rRNA
brnFE -1CTTATCGACGAAGCCTACG RT-PCR brnFE
brnFE-2AACTCTGCGATCGCCACTC RT-PCR brnFE
lrp-1TCATTTTGGGCTACAGCGC RT-PCR lrp
lrp-2GTACTTCATCATGCTGCGC RT-PCR lrp
lysC-1GATATGCTCCTGACTGCTGG RT-PCR lysC
lysC-2CAACAATGCGTGCGTTTCC RT-PCR lysC
hom-1GTGATGAACTTGCGCACCG RT-PCR hom
hom-2GAGTGCAAAAGCGTCCTCAG RT-PCR hom
thrB-1TATTCGTGCTGGCCTGAAG RT-PCR thrB
thrB-2GCCAAACCATTAGCTGCAGC RT-PCR thrB
ilvA-1GCAGGCACGAATTTCCTCC RT-PCR ilvA
ilvA-2CGGATCTTGTAGGAACGAAC RT-PCR ilvA
ilvBN-1CGGATGACAGGTGCAAAGG RT-PCR ilvBN
ilvBN-2TTGTGGAGGAATAGAGCGG RT-PCR ilvBN

Cultivation of different strains of Corynebacterium glutamicum JHI3-156

All cultivations of different strains of C. glutamicum JHI3-156 were carried out at 30°C. When appropriate, 30 μg l−1 of kanamycin was used for maintaining the plasmids, and 0·7 mmol l−1 IPTG was used for the induction.

Cells for inoculation were grown for 36 h on agar plates containing rich medium (5 g l−1 glucose, 10 g l−1 tryptone, 10 g l−1 beef extract, 5 g l−1 yeast extract and 5 g l−1 NaCl). One loop of colonies was served as the inoculum for the precultures. The precultures were grown for 18 h in 500-ml baffled shake flasks with 50 ml seed medium (30 g l−1 glucose, 5 g l−1 (NH4)2SO4, 1 g l−1 KH2PO4, 0·5 g l−1 MgSO4 and 30 g l−1 corn steep liquor) and used as the inoculum for the main cultivations with an initial optical density at 562 nm (OD562) adjusted to 0·2.

Flask cultivation was carried out at 200 rpm for 72 h in 500-ml baffled shake flasks with 50 ml fermentation medium (100 g l−1 glucose, 35 g l−1 (NH4)2SO4, 1 g l−1 KH2PO4, 0·5 g l−1 MgSO4 and 15 g l−1 corn steep liquor). Twenty grams per litre CaCO3 was used to adjust pH. Samples were taken every 12 h to analyse levels of glucose, biomass and amino acids.

For fermentor cultivation, 50-ml precultures were transferred into a 3-l fermentor (New Brunswick Scientific BioFlo 110, Enfield, CT, USA) containing 1·2 l fermentation medium. The pH was automatically controlled at 7·0 by adding 50% NH4OH solution. The dissolved oxygen level was controlled by adjusting agitation speeds (400 rpm in the first 4 h and 600 rpm thereafter) and aeration rate (1·5 vvm). The level of glucose was kept at 20 g l−1 by feeding concentrated glucose solution. Samples were taken every 4 h to analyse levels of glucose, biomass and amino acids.

The glucose concentration was determined by the dinitrosalicylic acid method (Miller 1959). Biomass was determined by measuring OD562 with UV-1800 spectrophotometer (Shimadzu, Tokyo, Japan). The dry cell weight (DCW) per litre was calculated according to an experimentally determined formula: DCW (g l−1) = 0·6495 × OD562-2·7925. The extracellular and intracellular fluids were separated using published method (Klingenberg and Pfaff 1977), and amino acids in the fluids were quantified using high-pressure liquid chromatography (Agilent Technologies 1200 series, Enfield, CT, USA) (Korös et al. 2008). To calculate the concentration of intracellular amino acids, 1 mg DCW was considered as 1·6 μl intracellular volume of cells (Kennerknecht et al. 2002).

Quantification of mRNA using real-time PCR

Real-time PCR (RT-PCR) combined with reverse transcription was used to quantify messenger RNA of genes lysC, hom, thrB, ilvA, ilvBN, brnFE and lrp in strains JHI3-156/pDXW-8, JHI3-156/pDXW-8-lrp, JHI3-156/pDXW-8-brnFE and JHI3-156/pDXW-8-lrp-brnFE. Total RNA was extracted from cultures harvested at the mid-exponential growth phase, using a total RNA extraction kit (BioFlux, Beijing, China). The DNA, existed in RNA extract sample, was disposed by DNase I. The quality and amount of RNA were judged and quantified by electrophoresis. Equal amounts of 500 ng RNA were transcribed into cDNA using a RevertAid™ First Strand cDNA synthesis kit (Fermentas, Shanghai, China) with the random hexamer primer. RT-PCR was carried out in an ABI StepOne real-time PCR system (Applied Biosystems, San Mateo, CA, USA) using Real Master Mix kit (TIANGEN, Beijing, China). Primers for detection of the previously mentioned genes are listed in Table 2. Programmes for RT-PCR were 94°C for 1 min followed by 40 cycles of 94°C for 10 s, 55°C for 30 s and 68°C for 15 s. All assays were performed in triplicate. The relative abundance of the targeted mRNAs was quantified based on the cycle threshold value, which is defined as the cycle number required to obtain a fluorescence signal above the background and was calculated by the Livak method (Livak and Schmittgen 2001; Nolden et al. 2001). To standardize the results, the relative abundance of 16S rRNA was used as the internal standard.

Assay of the export of l-isoleucine in different strains of Corynebacterium glutamicum

To determine the l-isoleucine export rates, cells of C. glutamicum JHI3-156/pDXW-8, JHI3-156/pDXW-8-lrp, JHI3-156/pDXW-8-brnFE and JHI3-156/pDXW-8-lrp-brnFE were grown in brain heart infusion medium, harvested and washed twice with ice-cold 0·9% NaCl. The washed cells were resuspended into MMI (Hermann and Karmer 1996) to give an initial OD562 of 10. After incubation for 20 min at 30°C, l-isoleucine-l-isoleucine was added, and the samples were collected every 10 min. Separation of external and internal fluid was performed by using silicone oil centrifugation (Klingenberg and Pfaff 1977).

Observed efflux rates of l-isoleucine are due to diffusion, active export and active import (Zittrich and Krämer 1994). The BrnFE-mediated export rate could be calculated as Vex = Veff − Kd × (Ilein − Ileout) + Vin. Veff, the observed efflux rate, could be calculated by using extracellular l-isoleucine concentrations. K× (Ilein − Ileout), the efflux due to diffusion, is dependent on the concentrations of internal and external l-isoleucine. The Kd value for l-isoleucine is 0·13 μl mg−1 min−1 (Krämer 1994; Milner et al. 1987). Vin, the uptake rate for l-isoleucine, is 1·1 nmol mg−1 min−1 (Ebbighausen et al. 1989; Kennerknecht et al. 2002).

Results

Overexpression of brnFE in Corynebacterium glutamicum JHI3-156 only slightly increased the l-isoleucine secretion

After 72-h cultivation of an l-isoleucine producer JHI3-156, the extracellular concentration of l-isoleucine reached 17 mmol l−1, but there was still 14 mmol l−1 l-isoleucine remained in the cells (Table 3; Fig. 1a). To increase the export of the intracellular l-isoleucine, the brnFE operon encoding the two-component export system BrnFE for BCAA in C. glutamicum was cloned into the shuttle vector pDXW-8 and overexpressed in JHI3-156. After 72-h cultivation, levels of intracellular and extracellular l-isoleucine and other related amino acids in JHI3-156/pDXW-8 and JHI3-156/pDXW-8-brnFE were analysed. Compared with the control JHI3-156/pDXW-8, the secretion of l-isoleucine in JHI3-156/pDXW-8-brnFE only slightly increased (Table 3; Fig. 1b,c). Similar patterns of growth and glucose consumption were observed for strains JHI3-156, JHI3-156/pDXW-8 and JHI3-156/pDXW-8-brnFE (Fig. 1d). Compared with the control JHI3-156/pDXW-8, the secretion of l-leucine and l-valine in JHI3-156/pDXW-8-brnFE also slightly increased (Fig. 1b,c). The results indicate that overexpressing the export system BrnFE does not significantly increase the secretion of BCAA in JHI3-156.

Table 3. Comparison of fermentation parameters of different strains in 72-h shake flask cultivation. Values for the calculation of standard deviations were derived from three independent measurements
ParametersJHI3-156JHI3-156/pDXW-8JHI3-156/pDXW-8-brnFEJHI3-156/pDXW-8-lrpJHI3-156/pDXW-8-lrp-brnFE
DCW (g l−1)18·7 ± 0·217·8 ± 0·318·9 ± 0·217·1 ± 0·412·8 ± 0·2
Glucose consumed (g l−1)49·8 ± 0·750·3 ± 0·951·4 ± 0·853·6 ± 0·760·2 ± 0·8
Extracellular l-isoleucine (mmol l−1)17·0 ± 0·416·3 ± 0·417·7 ± 0·721·5 ± 0·326·6 ± 0·8
Intracellular l-isoleucine (mmol l−1)14·1 ± 0·512·5 ± 0·313·1 ± 0·413·7 ± 0·59·7 ± 0·1
Specific yield (g g−1)0·119 ± 0·0050·120 ± 0·0020·123 ± 0·0070·164 ± 0·0020·272 ± 0·010
Yield on glucose (g g−1)0·045 ± 0·0020·042 ± 0·0010·045 ± 0·0020·052 ± 0·0000·058 ± 0·002
Productivity (g L−1 h−1)0·031 ± 0·0020·030 ± 0·0010·032 ± 0·0010·039 ± 0·0010·048 ± 0·002
Cell yield on glucose (g g−1)0·376 ± 0·0040·354 ± 0·0000·368 ± 0·0040·319 ± 0·0050·213 ± 0·000
Figure 1.

Comparison of intracellular and extracellular l-isoleucine and related amino acids in shake flask cultivation of strains JHI3-156 (a), JHI3-156/pDXW-8 (b) and JHI3-156/pDXW-8-brnFE (c). (d) Growth and glucose consumption of JHI3-156, JHI3-156/pDXW-8 and JHI3-156/pDXW-8-brnFE. Asp, l-aspartate; Thr, l-threonine; Val, l-valine; Met, l-methionine; Ile, l-isoleucine; Leu, l-leucine; Lys, l-lysine. Error bars in the figure indicate the standard deviations from three parallel samples. (image_n/jam12141-gra-0002.png) Extracellular; (image_n/jam12141-gra-0001.png) Intracellular; (image_n/jam12141-gra-0003.png) JHI3-156; (image_n/jam12141-gra-0004.png) JHI3-156/pDXW-8; (image_n/jam12141-gra-0005.png) JHI3-156/pDXW-8-brnFE.

Lrp is important for l-isoleucine production in Corynebacterium glutamicum JHI3-156

Considering that Lrp could activate the expression of brnFE in C. glutamicum (Lange et al. 2012), the gene lrp encoding Lrp was cloned into pDXW-8 and overexpressed in JHI3-156 to improve the l-isoleucine production. After 72-h cultivation, the extracellular levels of l-isoleucine in JHI3-156/pDXW-8-lrp reached 21·5 mmol l−1 (Table 3; Fig. 2a), which is 32% increase when compared with the control strain JHI3-156/pDXW-8 (Table 3). Interestingly, the ratio of extracellular and intracellular levels of l-threonine in JHI3-156/pDXW-8-lrp also significantly increased (Fig. 2a).

Figure 2.

Comparison of intracellular and extracellular l-isoleucine and related amino acids in shake flask cultivation of strains JHI3-156/pDXW-8-lrp (a) and JHI3-156/pDXW-8-lrp-brnFE (b). (c) Growth and glucose consumption of JHI3-156/pDXW-8-lrp and JHI3-156/pDXW-8-lrp-brnFE. Asp, l-aspartate; Thr, l-threonine; Val, l-valine; Met, l-methionine; Ile, l-isoleucine; Leu, l-leucine; Lys, l-lysine. Error bars in the figure indicate the standard deviations from three parallel samples. (image_n/jam12141-gra-0002.png) Extracellular; (image_n/jam12141-gra-0001.png) Intracellular; (image_n/jam12141-gra-0006.png) JHI3-156/pDXW-8-brn; (image_n/jam12141-gra-0007.png) JHI3-156/pDXW-8-brnFE.

When Lrp was overexpressed in combination with BrnFE in JHI3-156, the extracellular levels of l-isoleucine could reach 26·6 mmol l−1 in 72-h cultivation (Table 3; Fig. 2b), which is 63% increase compared with the control strain (Table 3). The ratio of extracellular and intracellular levels of l-isoleucine in JHI3-156/pDXW-8-lrp-brnFE increased to 2·8 from 1·3 in the control strain JHI3-156/pDXW-8 (Table 3). Meanwhile, the extracellular levels of l-valine, l-leucine and l-methionine also increased (Fig. 2b), which is consistent with the previous reports that BrnFE could excrete all three BCAA (Kennerknecht et al. 2002) and l-methionine (Trötschel et al. 2005) in C. glutamicum. These data suggest that Lrp is important for the export of l-isoleucine, and co-overexpression of BrnFE and Lrp could significantly increase l-isoleucine secretion in JHI3-156. Similar patterns of glucose consumptions were observed for both JHI3-156/pDXW-8-lrp and JHI3-156/pDXW-8-lrp-brnFE, but JHI3-156/pDXW-8-lrp-brnFE grew much slower than JHI3-156/pDXW-8-lrp (Fig. 2c), suggesting that more glucose was used to produce l-isoleucine in JHI3-156/pDXW-8-lrp-brnFE (Table 3).

To understand why l-isoleucine production was significantly increased when coexpressing the lrp and brnFE genes, the transcriptional levels of genes (lrp, brnFE, lysC, hom, thrB, ilvA and ilvBN) related to l-isoleucine biosynthesis (Morbach et al. 1996b) in strains JHI3-156/pDXW-8, JHI3-156/pDXW-8-brnFE, JHI3-156/pDXW-8-lrp and JHI3-156/pDXW-8-lrp-brnFE were investigated by RT-PCR analysis (Fig. 3a). In JHI3-156/pDXW-8-brnFE, only the transcriptional level of brnFE increased 3-fold. In JHI3-156/pDXW-8-lrp, however, not only the transcriptional level of lrp increased 14-fold, but the transcriptional levels of ilvA, hom and brnFE also increased about 3-folds. When lrp and brnFE were co-overexpressed in JHI3-156/pDXW-8-lrp-brnFE, the transcriptional levels lrp, brnFE, lysC, hom, thrB, ilvA and ilvBN increased 24·7, 13·6, 2·5, 9·6, 2·5, 15·5 and 3·7-folds, respectively. The increased transcriptional levels of brnFE, lysC, hom, thrB, ilvA and ilvBN in JHI3-156/pDXW-8-lrp-brnFE might play important roles in l-isoleucine production.

Figure 3.

(a) Transcriptional analysis of the key genes in the biosynthesis pathway of l-isoleucine in C. glutamicum JHI3-156/pDXW-8, JHI3-156/pDXW-8-brnFE, JHI3-156/pDXW-8-lrp and JHI3-156/pDXW-8-lrp-brnFE. Values are shown relative to the control strain JHI3-156/pDXW-8. (b) The changes of extracellular l-isoleucine in C. glutamicum JHI3-156/pDXW-8, JHI3-156/pDXW-8-brnFE, JHI3-156/pDXW-8-lrp and JHI3-156/pDXW-8-lrp-brnFE in the presence of l-isoleucine- l-isoleucine. The experiments were repeated three times, and standard deviations were determined. (image_n/jam12141-gra-0002.png) IWJ001/pDXW-8; (image_n/jam12141-gra-0008.png) IWJ001/pDXW-8-brnFE; (image_n/jam12141-gra-0009.png) IWJ001/pDXW-8-lrp; (image_n/jam12141-gra-0010.png) IWJ001/pDXW-8-lrp-brnFE; (image_n/jam12141-gra-0007.png) JHI3-156/pDXW-8; (image_n/jam12141-gra-0005.png) JHI3-156/pDXW-8-brnFE; (image_n/jam12141-gra-0011.png) JHI3-156/pDXW-8-lrp; (image_n/jam12141-gra-0012.png) JHI3-156/pDXW-8-lrp-brnFE.

To perform functional analyses of BrnFE, export rates of l-isoleucine were measured in the presence of l-isoleucine-l-isoleucine. Addition of this peptide leads to increased intracellular l-isoleucine steady-state levels. Because there were more BrnFE in JHI3-156/pDXW-8-lrp-brnFE than other strains (Fig. 3a), 10 mmol l−1 l-isoleucine- l-isoleucine was used for JHI3-156/pDXW-8-lrp-brnFE, and 3 mmol l−1 was used for strains JHI3-156/pDXW-8, JHI3-156/pDXW-8-brnFE and JHI3-156/pDXW-8-lrp to achieve comparable intracellular concentrations of l-isoleucine. The actual cytoplasmic concentrations of l-isoleucine are determined by the rates of import and hydrolysis of the peptide as well as the l-isoleucine efflux rate. The concentrations of extracellular l-isoleucine increased with time in all four strains, with the highest for JHI3-156/pDXW-8-lrp-brnFE cells and the lowest in JHI3-156/pDXW-8 cells (Fig. 3b). Both the l-isoleucine export rate and efflux rate in strains JHI3-156/pDXW-8, JHI3-156/pDXW-8-brnFE, JHI3-156/pDXW-8-lrp and JHI3-156/pDXW-8-lrp-brnFE were calculated (Table 4). The export rate of l-isoleucine in JHI3-156/pDXW-8-lrp-brnFE (22 nmol mg−1 min−1) was the highest, while that in JHI3-156/pDXW-8 (12 nmol mg−1 min−1) was the lowest. The export rate of l-isoleucine in JHI3-156/pDXW-8-brnFE (14 nmol mg−1 min−1) and JHI3-156/pDXW-8-lrp (14 nmol mg−1 min−1) was similar. These results are consistent with the data obtained from fermentation (Table 3) and RT-PCR analysis (Fig. 3a).

Table 4. Transport rates (nmol mg−1 min−1) of l-isoleucine in different strains of Corynebacterium glutamicum. Values for the calculation of standard deviations were derived from three independent measurements
StrainsDiffusion rate Kd (Ilein − Ileout)Export rate VexEfflux rate Veff
JHI3-156/pDXW-82·6 ± 0·212 ± 0·313·5 ± 0·4
JHI3-156/pDXW-8-brnFE2·2 ± 0·114·2 ± 0·515·5 ± 0·5
JHI3-156/pDXW-8-lrp2·1 ± 0·214·2 ± 0·415·2 ± 0·5
JHI3-156/pDXW-8-lrp-brnFE1·4 ± 0·121·8 ± 0·822·0 ± 0·7

Fed-batch fermentation of JHI3-156/pDXW-8-lrp-brnFE

As JHI3-156/pDXW-8-lrp-brnFE produced more l-isoleucine than the strains JHI3-156/pDXW-8-brnFE and JHI3-156/pDXW-8-lrp, it was further evaluated in fed-batch fermentation, using JHI3-156/pDXW-8 as a control (Fig. 4). After 72-h fermentation, l-isoleucine reached 205 mmol l−1 in JHI3-156/pDXW-8-lrp-brnFE, and the specific yield increased 72% when compared with the control strain (Table 5; Fig. 4a). Levels of l-valine and l-leucine were also increased in JHI3-156/pDXW-8-lrp-brnFE as expected (Fig. 4b).

Table 5. Comparison of parameters in 72-h fed-batch fermentation of JHI3-156/pDXW-8-lrp-brnFE and JHI3-156/pDXW-8. Data were collected from two parallel experiments
ParametersJHI3-156/pDXW-8JHI3-156/pDXW-8-lrp-brnFE
DCW (g l−1)69·7 ± 144·8 ± 0·7
Glucose consumed (g l−1)310 ± 3221 ± 2
Extracellular l-isoleucine (mmol l−1)185 ± 1205 ± 2
Specific yield (g g−1)0·349 ± 0·0070·600 ± 0·008
Yield on glucose (g g−1)0·078 ± 0·0000·122 ± 0·000
Productivity (g l−1 h−1)0·357 ± 0·0050·374 ± 0·007
Cell yield on glucose (g g−1)0·225 ± 0·0030·203 ± 0·002
Figure 4.

Changes in the levels of l-isoleucine (a), the related amino acids (b), the residual glucose (c) and the dry cell weight (d) at different time points in a fed-batch fermentation of JHI3-156/pDXW-8-lrp-brnFE. JHI3-156/pDXW-8 was used as a control. Asp, l-aspartate; Lys, l-lysine; Met, l-methionine; Thr, l-threonine; Ala, l-alanine; Leu, l-leucine; Val, l-valine. Error bars in the figure indicate the standard deviations from two parallel samples. (image_n/jam12141-gra-0005.png) JHI3-156/pDXW-8; (image_n/jam12141-gra-0013.png) JHI3-156/pDXW-8-lrp-brnFE; (image_n/jam12141-gra-0002.png) JHI3-156/pDXW-8; (image_n/jam12141-gra-0001.png) JHI3-156/pDXW-8-lrp-brnFE.

During the cell growth, JHI3-156/pDXW-8-lrp-brnFE consumed glucose slower (Fig. 4c) and grew slower (Fig. 4c) than the control strain JHI3-156/pDXW-8. The control strain reached the stationary phase after 28 h, while JHI3-156/pDXW-8-lrp-brnFE did not reach the stationary phase until 36 h. After 72-h fermentation, biomass of JHI3-156/pDXW-8-lrp-brnFE was 35% lower than that of the control, but the l-isoleucine production and its specific yield in JHI3-156/pDXW-8-lrp-brnFE increased 11% and 72%, respectively (Table 5). These results demonstrate that co-overexpression of brnFE and lrp in C. glutamicum could facilitate l-isoleucine production in fed-batch fermentation.

Discussion

Lrp could control the expression of hundreds of genes in various bacteria (Lintner et al. 2008). In E. coli, Lrp regulates several operons in the metabolism of BCAA, by binding to their upstream regions. Lrp represses transcription of the ilvIH operon encoding AHAS III (Platko et al. 1990; Wang and Calvo 1993) and the ilvGMEDA operon encoding four of the five enzymes in the common pathway for the biosynthesis of BCAA (Rhee et al. 1996). Lrp also represses the expression of the livJ and livK genes encoding the BCAA transporter (Haney et al. 1992) and the leuE gene encoding the l-leucine exporter (Kutukova et al. 2005). Interestingly, the expression of the ygaZH genes encoding the l-valine exporter could be activated by Lrp, and amplification of lrp and ygaZH led to the enhanced production of l-valine by 113% (Park et al. 2007). As different functions of Lrp were observed in different bacteria even though these proteins share >90% overall identity (Lintner et al. 2008), it is necessary to find out the effect of Lrp on the BCAA production in C. glutamicum, the most important bacterium for BCAA production. In this study, brnFE and lrp were overexpressed in an l-isoleucine-producing C. glutamicum JHI3-156. Compared with the control strain, the specific yield of l-isoleucine production in JHI3-156/pDXW-8-lrp-brnFE increased 72%, but the cell growth of JHI3-156/pDXW-8-lrp-brnFE was slower.

Compared with the control JHI3-156/pDXW-8, the secretion of l-isoleucine in JHI3-156/pDXW-8-brnFE only slightly increased (Table 3; Fig. 1b,c), even though the transcriptional levels of brnFE increased 3-fold (Fig. 3a). This might be due to BrnQ which could uptake l-isoleucine (Tauch et al. 1998). When the gene brnQ encoding BrnQ was deleted in the chromosome of C. glutamicum, overexpressing BrnFE could significantly increase the l-isoleucine production (Xie et al. 2012). One important finding in this study is that overexpressing Lrp could produce more l-isoleucine than overexpressing BrnFE in Cglutamicum JHI3-156, because the overexpressed Lrp increased not only the expression of brnFE but also that of ilvA and hom (Fig. 3a), the two key genes in the biosynthesis pathway of l-isoleucine. When lrp and brnFE were overexpressed together, transcriptional levels of brnFE, lysC, hom, thrB, ilvA and ilvBN all significantly increased due to the regulation of Lrp (Fig. 3a). Therefore, the significantly increased l-isoleucine production in JHI3-156/pDXW-8-lrp-brnFE was caused not only by the overexpressed BrnFE but also by the enhanced biosynthesis pathway of l-isoleucine.

The slow growth of JHI3-156/pDXW-8-lrp-brnFE might be due to the limitation of intracellular l-methionine, considering that the overexpressed BrnFE in JHI3-156/pDXW-8-lrp-brnFE exported most of the l-methionine required for cell growth (Lange et al. 2012). More l-isoleucine production should be expected if the growth of the JHI3-156/pDXW-8-lrp-brnFE could be improved. The l-isoleucine production could be further increased by overexpressing the key genes in its biosynthesis pathway (Yin et al. 2012). This work showed the potential of BrnFE transport engineering and Lrp regulatory engineering in the production of BCAA in C. glutamicum.

Acknowledgements

This project was financially supported by grants from National key Basic Research Program of China (2012CB725202), the Basic Research Programs of Jiangsu Province (BK2009003) and the 111 Project (No. 111-2-06).

Ancillary