ARTICLE Metabolic Engineering of Corynebacterium glutamicum for Efficient Production of 5-Aminolevulinic Acid Lili Feng,1,2,3 Ya Zhang,1,2,3 Jing Fu,1,2,3 Yufeng Mao,1,2,3 Tao Chen,1,2,3 Xueming Zhao,1,2,3 Zhiwen Wang1,2,3 1
Department of Biochemical Engineering, School of Chemical Engineering and Technology, Tianjin University, Tianjin, China 2 Key Laboratory of Systems Bioengineering (Ministry of Education), Tianjin University, Tianjin, China 3 SynBio Research Platform, Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China; telephone: 86-22-2740-6582; fax: 86-22-2740-6582; e-mail: [email protected]
ABSTRACT: 5-Aminolevulinic acid (5-ALA) has recently attracted attention for its potential applications in the fields of medicine and agriculture. In this study, Corynebacterium glutamicum was firstly engineered for 5-ALA production via the C4 pathway. HemA encoding 5-aminolevulinic acid synthase from Rhodobacter sphaeroides was codon optimized and expressed in C. glutamicum ATCC13032, resulting in accumulation of 5-ALA. Deletion of all known genes responsible for the formation of acetate and lactate further enhanced production of 5-ALA. Overexpression of ppc gene encoding phoenolpyruvate carboxylase resulted in an accumulation of 5-ALA up to 2.06 0.05 g/L. Furthermore, deletion of highmolecular-weight penicillin-binding proteins (HMW-PBPs) genes pbp1a, pbp1b, and pbp2b led to an increase in 5-ALA production of 13.53%, 29.47%, and 22.22%, respectively. Finally, 5-ALA production was enhanced to 3.14 0.02 g/L in shake flask by heterologously expressing rhtA encoding threonine/homoserine exporter, and 86.77% of supplemented glycine was channeled toward 5-ALA production in shake flask. The engineered C. glutamicum ALA7 strain produced 7.53 g/L 5-ALA in a 5 L bioreactor. This study demonstrated the potential utility of C. glutamicum as a platform for metabolic production of 5-ALA. Change of cell permeability by metabolic engineering HMW-PBPs
Correspondence to: Z. Wang Contract grant sponsor: National Program on Key Basic Research Project Contract grant number: 2012CB725203 Contract grant sponsor: National Natural Science Foundation of China Contract grant numbers: NSFC-21206112; NSFC-21576200; NSFC-21176182; NSFC21390201 Contract grant sponsor: National High-tech R&D Program of China Contract grant numbers: 2012AA022103; 2012AA02A702 Contract grant sponsor: Natural Science Foundation of Tianjin Contract grant number: 15JCQNJC06000 Received 3 August 2015; Revision received 10 November 2015; Accepted 19 November 2015 Accepted manuscript online 30 November 2015; Article first published online 9 December 2015 in Wiley Online Library (http://onlinelibrary.wiley.com/doi/10.1002/bit.25886/abstract). DOI 10.1002/bit.25886
1284
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
may provide a new strategy for biochemicals production in Corynebacterium glutamicum. Biotechnol. Bioeng. 2016;113: 1284–1293. ß 2015 Wiley Periodicals, Inc. KEYWORDS: 5-aminolevulinic acid; Corynebacterium glutamicum; C4 pathway; penicillin-binding proteins; cell permeability
Introduction 5-Aminolevulinic acid (5-ALA), a non-proteinogenic amino acid, plays a key role in the biosynthesis of tetrapyrrole, such as heme, chlorophyll, and vitamin B12. 5-ALA has been used in photodynamic diagnosis and therapy for various cancers (Cornelius et al., 2014), and applied as a selective biodegradable herbicide, insecticide, or a growth regulator with nontoxicity to crops, animals, and humans (Akram and Ashraf, 2013; Sasaki et al., 2002). Since 5-ALAproduction via chemical synthesis is low-yielding and high-cost (Kang et al., 2012), alternative microbial production of 5-ALA from renewable carbon sources has received much attention. Two natural pathways for the biosynthesis of 5-ALA exist in living organisms. One is the C4 pathway, through which glycine and succinyl-CoA are condensed to form 5-ALA by 5-aminolevulinate synthase (ALAS) in the presence of the essential cofactor pyridoxal 50 -phosphate (PLP) (Liu et al., 2014). This pathway is present in mammals, birds, yeasts, and purple non-sulfur photosynthetic bacteria, such as in Rhodobacter sphaeroides and Rhodopseudomonas. The alternative C5 pathway is tightly regulated by feedback inhibition of heme (Levican et al., 2007). In this pathway, 5-ALA is synthesized from glutamate via three steps, and the corresponding enzymes involved in this process are glutamyl-tRNA synthetase ß 2015 Wiley Periodicals, Inc.
(encoded by gltx), NADPH-dependent glutamyl-tRNA reductase (encoded by hemA), and glutamate-1-semialdehyde aminotransferase (encoded by hemL) (Liu et al., 2014). This pathway exists in higher plants, algae, and various bacteria, such as Escherichia coli and Corynebacterium glutamicum. Metabolic engineering of microbial organisms for 5-ALA production has mainly focused on E. coli. Recombinant E. coli heterologously expressing ALAS from R. sphaeroides could accumulate 5-ALA via C4 pathway, and the expression of ALAS in E. coli was further optimized to improve 5-ALA production using factorial design (Liu et al., 2013). In addition, a rare codon optimizer strain was used to enhance ALAS activity, resulting in production of 6.56 g/L 5-ALA in a 15 L fermenter (Fu et al., 2010). As succinyl-CoA is a precursor for 5-ALA, production was enhanced fourfold by expressing ALAS from R. sphaeroides ALAS in an E. coli strain engineered for succinate production (Kang et al., 2011b). Moreover, 5-ALA degradation can be prevented by adding D-xylose and D-glucose, effective inhibitors of 5-ALA dehydratase (Lee et al., 2003; Lin et al., 2009). Fermentation process engineering using short-term dissolved oxygen shock has also been applied to enhance 5-ALA production up to 9.44 g/L during fed-batch fermentation (Yang et al., 2013). Glutamyl-tRNA reductase and glutamate-1-semialdehyde aminotransferase were found to have a synergistic effect on 5-ALA production via the C5 pathway (Layer et al., 2010). 5-ALA production reached 2.05 g/L by co-expressing a mutant of hemA (Salmonella arizona) together with hemL (E. coli), and production was further increased to 2.96 g/L after overexpressing RhtA transporter involved in the threonine and homoserine efflux system (Kang et al., 2011a). In the C5 pathway, 5-ALA biosynthesis is regulated through a feedback inhibition mechanism by heme (Zhang et al., 2015), whose biosynthesis is involved in iron consumption. Recombinant E. coli produced 16% higher 5-ALA by overexpressing ryhB, a small RNA regulator of iron metabolism in E. coli (Liu et al., 2013). Besides E. coli, C. glutamicum, one of the most important commercial amino acid-producing microorganism with generally regarded as safe (GRAS) status (Yang et al., 2015), seems to be a promising alternative for 5-ALA production. Although C. glutamicum utilizes the C5 pathway for synthesis of 5-ALA, the complex regulation of this biosynthetic route suggests that metabolic engineering for efficient 5-ALA production via the C5 pathway might be problematic. Therefore, we focused on exploring the potential of 5-ALA production via the C4 pathway. Moreover, there appears to be a higher potential for 5-ALA production via the C4 pathway using C. glutamicum compared to E. coli, due to the lack of glycine cleavage systems found in E. coli (Jørgensen et al., 2012). In the present work, we described the production of 5-ALA via the C4 pathway using C. glutamicum (Fig. 1). Metabolic engineering strategies included expressing of heterologous ALAS, reducing byproduct formation, and increasing flux to 5-ALA. Additionally, engineering of high-molecular-weight penicillin-binding proteins (HMW-PBPs) and 5-ALA export was found to have a significant effect on the 5-ALA production. This study demonstrates the potential application of C. glutamicum as a platform for 5-ALA production and provides evidence that modification of different
HMW-PBPs may be a universal strategy for some biochemical production such as glutamate and 2-(3-hydroxy-1-adamantyl)(2S)-amino ethanoic acid.
Materials and Methods Bacteria Strains, Media, and Growth Conditions All C. glutamicum strains used were derived from the wild type C. glutamicum WT 13032. The bacterial strains and plasmids used in this work were listed in Table I. The construction of plasmids and strains was described in Supporting Information 1. 5-ALA production was tested in CGIIIM medium containing 10 g/L yeast extract, 10 g/L tryptone, 2.5 g/L NaCl, 10 g/L glucose, and 21 g/L 3-morpholinopropanesulfonic acid. Physiological characterizations were tested in defined CGXII medium (Unthan et al., 2014) supplemented with 5.5 g/L glucose and 7.5 g/L glycine. For precultivation of C. glutamicum ATCC 13032 and its recombinant derivatives, a single clone was grown in 5 mL of BHI medium (Vogt et al., 2014). After incubation for approximately 15 h, cells were inoculated into 500-mL shake flask containing 50 mL shake cultivation medium with an initial OD600 of 0.1. The cells were cultivated at 30 C in a rotatory shaker at 220 rpm. The medium for strains with the episomal replicating plasmid was additionally supplemented with 25.0 mg/mL kanamycin and/or 10 mg/mL chloroamphenicol when needed. When OD600 reached to 1, 0.5 mM isopropyl b-D-1-thiogalactopyranoside (IPTG) was added into the medium to induce the expression of genes under control of IPTG-inducible promoter. For fed-batch culture, a 50 mL seed cultured in CGIIIM medium for 8 h was transferred into a 5 L bioreactor containing 2.5 L initial fermentation medium composed of 10 g/L tryptone, 15 g/L yeast extract, 2.5 g/L NaCl, 7.5 g/L glycine, and 20 g/L glucose. Initially, ammonia and 3 M H2SO4 were added automatically to maintain a pH of 7.0, and after 15 h fermentation the pH was maintained at 6.5. The aeration rate was 1 vvm and dissolved oxygen was maintained between 25% and 35% saturation by controlling the stirrer speed. When glucose was lower than 1 g/L, the feeding process was initiated to maintain a basal glucose concentration. The feed solution was composed of 10 g/L tryptone, 15 g/L yeast extract, 2.5 g/L NaCl, 5 g/L K2HPO4, 5 g/L KH2PO4, 2.5 g/L MgSO4 7H2O, and 600 g/L glucose. Scanning Electron Microscopy (SEM) Cells from exponential growth phase were harvested by centrifugation, washed twice using pH 7.4 phosphate buffer solution (PBS), and fixed in 2.5% glutaraldehyde solution for 12 h at 4 C. The cells were washed twice in PBS (pH 7.4) and successively dehydrated following a stepwise treatment in 20, 50, 80, and 100% ethanol, respectively. Finally, the cells were treated with tert-butyl alcohol, freeze-dried, and analyzed using SEM. Lysozyme Sensitivity Test of C. glutamicum Lysozyme sensitivities of C. glutamicum were determined by growth assay in LB medium supplemented with 1 g/L glucose. 25 mg/mL
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1285
Figure 1.
Metabolic pathways associated with 5-ALA biosynthesis in C. glutamicum and the metabolic engineering strategies for aerobic 5-ALA overproduction. The red indicated that the pathways were disrupted. The blue font indicated that pathways were overexpressed. Relevant reactions were represented by the genes encoding for corresponding enzymes: ldhA, encoding L-lactate dehydrogenase; pqo, encoding pyruvate: menaquinone oxidoreductase; pta, encoding phosphotransacetylase; ackA, encoding acetate kinase; cat, encoding acetyl-CoA:CoA transferase; ppc, encoding phosphoenolpyruvate carboxylase; pyc, encoding pyruvate carboxylase; pck, encoding phosphoenolpyruvate carboxykinase; gltA, encoding citrate synthase; hemA, encoding 5-aminolevulinic acid synthease; hemB, encoding aminolevulinic acid dehydratase; rhtA, encoding gene encoding inner membrane transporter for L-threonine; and L-homoserine; pbp1a, pbp1b, pbp2a, and pbp2b, genes encoding HMW-PBPs.
lysozyme was added into medium when the cells were in the exponential phase. Analytical Methods Cell concentration was determined by the optical density at 600 nm (OD600). The biomass concentration was calculated from OD600 values using an experimentally determined correlation 1 OD600 unit is equal to 0.25 g/L cell dry weight (CDW) (Meiswinkel et al., 2013). Glucose concentration was measured using a SBA sensor machine
1286
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
(SBA-40E, Institute of Microbiology, Shandong, China). Intracellular 5-ALA was obtained with ultrasonic treatment as described previously (Kang et al., 2011a). Concentration of extracellular and intracellular 5-ALA was measured using modified Ehrlich’s reagent (Kang et al., 2011a). Concentrations of extracellular organic acids were determined by high performance liquid chromatography (HPLC) equipped with a cation exchange column (Aminex HPX87-H, Bio-Rad, Hercules, CA), a UV absorbance detector (Agilent Technologies, G1315D), and a refractive index (RI) detector (Agilent Technologies, HP1047A). A mobile phase of 5 mM H2SO4 solution at
Table I. Bacterial strains and plasmids used in this study. Relevant characteristicsa
Strains and plasmids Strains C. glutamicum WT 13032 C. glutamicum SAZ1 C. glutamicum SAZ2 C. glutamicum SAZ3 C. glutamicum ALA1 C. glutamicum ALA2 C. glutamicum ALA3 C. glutamicum ALA4 C. glutamicum ALA5 C. glutamicum ALA6 C. glutamicum AP1 C. glutamicum AP2 C. glutamicum AP3 C. glutamicum AP4 C. glutamicum ALA7 C. glutamicum ALA8 C. glutamicum C1 Plasmids pXMJ19 pXA pEC-XK99E pECacsAgltA pDsacB pD-pck pD-pbp1a pD-pbp1b pD-pbp2a pD-pbp2b pEP2 pEP2-rhtA a
Sources or references
Wild-type strain, biotin auxotrophic WT13032DldhADpqoDcatDptaDackA WT13032DldhADpqoDcatDptaDackA Pppc::Psod WT13032DldhADpqoDcatDptaDackA Pppc::Psod Ppyc::Psod Wild-type ATCC 13032 harboring pXA WT13032DldhADpqoDcatDptaDackA harboring pXA WT13032DldhADpqoDcatDptaDackA harboring pXA and pECacsAgltA WT13032DldhADpqoDcatDptaDackA Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackA Pppc::Psod Ppyc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpck Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp1a Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp1b Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp2a Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp2b Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp1b Pppc::Psod harboring pXA and pEP2-rhtA WT13032DldhADpqoDcatDptaDackADpck Pppc::Psod harboring pXA and pEP2-rhtA WT13032DldhADpqoDcatDpta harboring pXA and pEC-XK99E Cmr, C. glutamicum/E. coli shuttle vector Derived from pXMJ19, for overexpress hemA under the control of tac promoter Kanr; C. glutamicum/E. coli shuttle vector derived from pEC-XK99E, for overexpression of acsA and gltA under the control of trc promoter derived from pK18mobsacB, for increasing expression of sacB under trc promoter Integrative transformation vector for deletion of the of pck gene Integrative transformation vector for deletion of the pbp1a gene. Integrative transformation vector for deletion of the pbp1b gene Integrative transformation vector for deletion of the pbp2a gene Integrative transformation vector for deletion of the pbp2b gene Kanr, C. glutamicum/E. coli shuttle vector Derived from pEP2, for overexpression of rhtA from E.coli under control of tuf promoter
ATCCb Lab stock (Zhu et al. 2014) (Zhu et al., 2013) This study This study This study This study This study This study This study This study This study This study This study This study This study (Jakoby et al. 1999) This study (Zhu et al. 2013) (Zhu et al. 2013) (Ma et al. 2015) This study This study This study This study This study (Zhang et al. 1994) This study
Abbreviations: Cm, chloramphenicol; Kanr, kanamycin; r, resistance. American type culture collection.
b
a 0.4 mL/min flow rate was used. The column was operated at 65 C. The concentration of amino acids was measured using a C18 column (Luna 150 4.6 mm) with 20 mM sodium acetate/acetonitrile (97:3, v/v) (buffer A), 20 mM sodium acetate solution/acetonitrile (50:50, v/v) (buffer B) as mobile phase at 0.6 mL/min, and derivative reagent used was consisted of 6.0 mg o-phthalaldehyde, 0.1 mL anhydrous ethanol, 0.9 mL 0.1 mM pH 10.4 borate buffer solution, and 5.0 mL 3-mercaptopropionic acid (MPA).
Enzyme Assays Cells from exponential phase were harvested by centrifugation (20 min, 6500 rpm, 4 C). Following centrifugation, cells were washed twice with 15 mL 50 mM Tris-HCl (pH 7.5) containing 0.1 mM EDTA and 2 mM dithiothreitol (DTT), centrifuged again, and resuspended in the same buffer. The cells were disrupted using sonication for 30 min in an ice bath (130 W, 20 kHz, pulse: 5 s on; 5 s off), and the cell lysate was clarified by centrifugation (15 min, 6500 rpm, 4 C). Total protein concentrations were quantified by the Bradford method using bovine serum albumin as standard. ALAS activity was determined as described previously (Fu et al., 2007). One unit of ALAS activity represents the amount of enzyme required to produce 1 nmol of 5-ALA per minute.
Results Establishing the C4 Pathway for 5-ALA Production by Introducing the HemA Gene (Encoding ALAS) From R. sphaeroides in C. glutamicum ALAS (encoded by hemA) has been isolated and characterized in several bacteria, such as R. sphaeroides and Rhodobacter capsulatus (Kang et al., 2012). We favored utilization of ALAS from R. sphaeroides due to its higher specific activity and lower Km for glycine and succinyl-CoA (van der Werf and Zeikus, 1996). Therefore, the codon-optimized version of hemA from R. sphaeroides was synthesized and expressed under the control of IPTG-inducible tac promoter from a medium-copy plasmid pXMJ19 in C. glutamicum. It was found that specific ALAS activity of C. glutamicum ALA1 in the middle of exponential growth phase was 19.49 0.19 U/mg protein, whereas no activity was detected in wild-type C. glutamicum 13032 harboring empty plasmid pXMJ19, indicating successful expression of ALAS. However, only a small amount 5-ALA was accumulated by C. glutamicum ALA1. Although glycine can be synthesized from glucose by C. glutamicum itself, the amount of intracellular glycine may not be sufficient for achieving high titers of 5-ALA. The glycine was supplemented with different
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1287
concentrations. Although addition of glycine resulted in reduction of biomass accumulation, 5-ALA production increased tremendously, reaching 1.44 0.03 g/L in the presence of 5 and 7.5 g/L glycine (Fig. 2), and CGIIIM medium supplemented with 7.5 g/L glycine was used for subsequent experiments in this study. Improving 5-ALA Production by Reducing Acetate Formation Under aerobic conditions, 1.70 0.03 g/L acetate was produced by C. glutamicum ALA1 (Table II). To reduce acetate formation, genes responsible for acetate accumulation including pqo, pta, ackA, and cat encoding for pyruvate:menaquinone oxidoreductase, phosphotransacetylase, acetate kinase, and acetyl-CoA: CoA transferase, respectively, were all deleted. Although lactate was not detected in the shake flask, deletion of ldhA encoding L-lactate dehydrogenase may exert positive effect on 5-ALA production during high-density fermentation in fed-batch cultivation. Therefore, ldhA was also deleted. The deletion of the above-mentioned five genes in C. glutamicum ALA1 led to an 31.18% reduction in acetate formation (Table II). 5-ALA production in the supernatant increased to 1.92 0.02 g/L at the end of fermentation. Although all identified genes responsible for acetic acid formation in C. glutamicum were deleted, 1.17 0.05 g/L acetate was still detected in the culture. To further reduce acetate formation and redirect carbon flux toward 5-ALA biosynthesis, acsA encoding acetyl-CoA synthetase from Bacillus subtilis, which catalyzes a reaction for converting acetate to acetyl-CoA (Lin et al., 2006), was expressed in combination with endogenous gltA gene encoding citrate synthase from a medium-copy plasmid pEC-XK99E under the control of the IPTG-inducible trc promoter. Acetic acid production of resulting strain C. glutamicum ALA3 was reduced by 58.40%, demonstrating that acetate was effectively assimilated. Nevertheless,5-ALA production dramaticly dropped to 0.65 0.04 g/L (Table II). In light of this result, overexpression of
acetyl-CoA synthetase and citrate synthase seemed to work against the production of 5-ALA in C. glutamicum.
Effect of Modification of Anaplerotic Reactions on 5-ALA Production Oxaloacetate supply was reported to be the bottleneck for the production of compounds derived from the TCA cycle (Balzer et al., 2013; Litsanov et al., 2012). To explore whether oxaloacetate supply contributes to 5-ALA production, the flux to oxaloacetate was engineered by overexpressing ppc (encoding phosphoenolpyruvate carboxylase) and pyc (encoding pyruvate carboxylase) as well as deleting pck (encoding phosphoenolpyruvate carboxykinase). Phosphoenolpyruvate carboxylase, catalyzing the carboxylation of phosphoenolpyruvate to oxaloacetate, was reported for oxaloacetate replenishment in C. glutamicum (Litsanov et al., 2012). Therefore, we replaced the native promoter of ppc gene in C. glutamicum ALA2 with the strong sod promoter from C. glutamicum, resulting in C. glutamicum ALA4. In response to this modification, biomass of C. glutamicum ALA4 increased by 8.98% and 5-ALA production increased by 7.29%, reaching to 2.06 0.05 g/L (Table II). To further enhance oxaloacetate supply, the native promoter of pyc in C. glutamicum ALA4 was also replaced by the sod promoter, resulting in C. glutamicum ALA5. Nevertheless, this modification did not further enhance 5-ALA production. We speculated that the increased supply of oxaloacetate may not be efficiently channeled into the TCA pathway, and may be redirected back into glycolysis via phosphoenolpyruvate carboxykinase (Becker et al., 2011). Therefore, the phosphoenolpyruvate carboxykinase was inactivated in C. glutamicum ALA4. However, the resulting C. glutamicum ALA6 produced 2.07 0.01 g/L 5-ALA, almost identical to that produced by C. glutamicum ALA4 (Table II).
Enhancing 5-ALA Production by Deletion of Different HMW-PBPs
Figure 2.
Effect of glycine addition on cell growth and 5-ALA production. C. glutamicum ALA1 was cultured in CGIIIM medium supplemented with different concentration of glycine at 30 C. 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
1288
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
In the previous study, inhibition of cell wall synthesis through the addition of penicillin has proven to be an efficient strategy for enhancing production of L-glutamate by C. glutamicum (Laneelle et al., 2013). Herein, we investigated the effect of penicillin addition on 5-ALA production. Extracellular 5-ALA was tremendously increased with the addition of 2.5 mg/mL penicillin and intracellular 5-ALA decreased by 80.85% (Fig. 3). Unfortunately, addition of penicillin to the medium results in a more complex and expensive production process as well as presenting a serious potential for environmental polution. An alternative approach is to deregulate the activity of PBPs targeted by penicillin rather than to directly add penicillin. Penicillin hinders cell wall biosynthesis by inhibiting the activity of PBPs. C. glutamicum has at least nine PBPs, and five of them are HMW-PBPs, which play a major role in PG synthesis. Among the HMW-PBPs, only FtsI was reported to be essential for cell growth (Valbuena et al., 2006). Thus, the other four HMW-PBP genes including class A HMW-PBP genes (pbp1a and pbp1b) and class B HMW-PBP genes (pbp2a and pbp2b) were deleted.
Table II. Growth and production performance of various strains constructed in the study. Strains
Biomass (g/L)
Consumed glucose (g/L)
5-ALA (g/L)
Consumed glycine (g/L)
Acetate (g/L)
ALA1 ALA2 C1 ALA3 ALA4 ALA5 ALA6 AP1 AP2 AP3 AP4 ALA7 ALA8
5.57 0.08 5.90 0.11 5.95 0.11 5.89 0.12 6.43 0.13 6.09 0.03 6.12 0.04 5.22 0.17 6.78 0.10 5.45 0.10 6.25 0.07 6.68 0.14 6.46 0.08
10.15 0.05 10.10 0.10 9.85 0.05 10.05 0.05 10.15 0.15 9.90 0.00 10.00 0.10 9.90 0.00 10.15 0.15 10.15 0.05 9.90 0.10 10.05 0.15 9.90 0.00
1.44 0.03 1.92 0.02 1.90 0.01 0.65 0.04 2.06 0.05 1.97 0.02 2.07 0.01 2.35 0.00 2.68 0.05 2.04 0.02 2.53 0.03 3.14 0.02 2.78 0.04
1.00 0.05 1.45 0.11 1.51 0.07 0.45 0.00 1.68 0.10 1.64 0.08 1.66 0.11 1.67 0.10 1.76 0.05 1.54 0.04 1.58 0.07 1.86 0.06 1.54 0.07
1.70 0.03 1.17 0.05 1.25 0.04 0.52 0.02 0.85 0.05 0.77 0.03 0.73 0.04 0.56 0.03 0.36 0.02 0.21 0.04 0.42 0.05 0.32 0.02 0.45 0.05
Batch cultures of recombinant strains were carried out in CGIIIM medium supplemented with 7.5 g/L glycine at 30 C. 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
The pbp1a, pbp1b, and pbp2b deletions all showed a positive effect on 5-ALA production. The corresponding mutants C. glutamicum AP1, C. glutamicum AP2, and C. glutamicum AP4 produced 2.35 0.00, 2.68 0.05, and 2.53 0.03 g/L 5-ALA, respectively, which were 13.53%, 29.47%, and 22.22% higher compared to the parent strain C. glutamicum ALA6 (Fig. 4). However, the pbp2a deletion mutant strain C. glutamicum AP3 produced almost the same amount of 5-ALA compared to C. glutamicum ALA6. Moreover, intracellular 5-ALA concentrations in HMW-PBPs mutants were determined. The intracellular level of 5-ALA in C. glutamicum AP1, C. glutamicum AP2, and C. glutamicum AP4 decreased to some extent compared to C. glutamicum ALA6 (Fig. 4). HMW-PBPs play a role in the biosynthesis of bacterial cell wall, and mutations in PBPs may affect cell growth and structural integrity of the cell wall. We assessed physiological characterizations and lysozyme sensitivity of different PBPs mutants. As shown in Figure 5, most of the mutants cultured in minimal medium
exhibited a similar growth rate as the parent strain. However, the growth rate of pbp2b mutant C. glutamicum AP4 was a little lower than wild-type strain, resulting in a lower biomass yield. For lysozyme sensitivity, C. glutamicum ALA6 was not affected by lysozyme treatment, indicating the high resistance to lysozyme. Nevertheless, the growth of the four HMW-PBP mutants decreased after the addition of lysozyme, resulting in lower biomass yield (Fig. 6). These results implied that deleting HMW-PBPs indeed affect cell surface integrity.
Figure 3.
Figure 4.
Comparison of cell growth and 5-ALA production with or without penicillin. C. glutamicum ALA6 was cultured in CGIIIM medium supplemented with 7.5 g/L glycine at 30 C. 0.5 mM IPTG was added into the medium when OD600 reached to 1. (A) C. glutamicum ALA6 was cultured without penicillin; (B) C. glutamicum ALA6 was cultured with the addition of 2.5 mg/mL penicillin when OD600 ¼ 5.79 0.11.
Improving 5-ALA Production by Heterologous Expression of rhtA From E. coli As decreasing intracellular 5-ALA is an effective strategy for increasing 5-ALA production, we further tried to decrease intracelluar 5-ALA by improving its transportation. RhtA transporter involved in threonine and homoserine efflux system
Distribution of intracellular and extracellular 5-ALA by C. glutamicum ALA6 and HWM-PBPs defect mutants. Batch cultures of recombinant strains were carried out in CGIIIM medium supplemented with 7.5 g/L glycine at 30 C, 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1289
Figure 5. Influence of deleting different HMW-PBPs on cell growth. (A) Biomass formation. (B) Glucose consumption. The cells were cultivated in CGXII medium supplemented with 5.5 g/L glucose and 7.5 g/L glycine and 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
Figure 6. Effect of lysozyme on growth of C. glutamicum strains. The strains were cultured in LB medium supplemented with 1.0 g/L glucose. 0.5 mM IPTG was added into the medium when OD600 reached to 1. (A) The cells were cultured without lysozyme. (B) Lysozyme with concentration of 25 mg/ml was added to the medium at the time as indicated by an arrow.
could enhance 5-ALA export in E. coli due to its broad substrate specificity (Kang et al., 2011a). Herein, rhtA gene was expressed in C. glutamicum ALA7 from a low copy plasmid pEP2. Plasmid pEP2 carrying a different derivative replication system of pNG2 derived from Corynebacterium diphtheriae (Zhang et al., 1994) can be compatible with pXMJ19 carrying a mutant pBL1 ori (Jakoby et al., 1999). This strain produced 0.32 0.00 g/L 5-ALA without addition of glycine and 3.14 0.02 g/L 5-ALA with consumption of 1.86 0.06 g/L glycine. About 86.77% of supplemented glycine was channeled toward 5-ALA production in shake flask. Intracellular 5-ALA concentration of C. glutamicum ALA7 decreased by 16.43% compared to C. glutamicum AP2 (Fig. 4). These results demonstrated that heterologous expression of rhtA from E. coli effectively accelerated 5-ALA transport out of C. glutamicum. Plasmids stability of these two plasmids in C. glutamicum ALA7 was tested, about 42% plasmids could stay in C. glutamicum ALA7 at the end of shake flask cultivation.
Fed-Batch Culture for Aerobic 5-ALA Production
1290
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
The performance of C. glutamicum ALA7 was investigated in a fed-batch process. 5-ALA started to accumulate when IPTG was added into the medium, reaching to 7.53 g/L in 33 h with consumption of 4.70 g/L glycine and the overall 5-ALA productivity was 0.23 g/L/h (Fig. 7). Although C. glutamicum ALA7 produced 2.00 g/L 5-ALA without addition of glycine in fed-batch cultivation. About 67.27% of supplemented glycine was used for 5-ALA production in fed-batch cultivation.
Discussion C. glutamicum, a dominating amino acid producer, was firstly engineered for 5-ALA production via C4 pathway. Heterologous expression of ALAS from R. sphaeroides in C. glutamicum led to 5-ALA accumulation. To increase 5-ALA production, the known pathways for acetate and lactate formation were disrupted. A
Figure 7. Production performance of C. glutamicum ALA7 in a 5 L fed-batch cultivation. 25 mg/mL kanamycin and 10 mg/mL chloroamphenicol were added initially to provide selective pressure. 0.5 mM IPTG was added into the medium when OD600 reached to 1.
second step to improve 5-ALA production was to enhance the flux toward oxaloacetate, which was reported to effectively improve production of compounds derived from TCA cycle (Song et al., 2013). However, overexpression of pyc gene did not exert a positive effect on 5-ALA production, whereas overexpression of ppc was found to increase 5-ALA production. This was not consistent with previous research, which showed that pyruvate carboxylase
contributed to about 90% oxaloacetate replenishment of C. glutamicum when grown on glucose under aerobic condition, whereas phosphoenolpyruvate carboxylase was responsible for 10% oxaloacetate replenishment (Litsanov et al., 2012). One possible reason for this inconsistent result is that the disruption of flux through the acetate-forming pathway may result in the accumulation of acetyl-CoA, which regulates the two anaplerotic enzymes in a contrary manner. Pyruvate carboxylase is inhibited by 110 mM acetyl-CoA, whereas 100 mM acetyl-CoA activates phosphoenolpyruvate carboxylase (Petersen et al., 2000). Similar results were reported by Litsanov et al. (2012) for succinate production following deletion of an acetate-forming pathway in C. glutamicum. In this study, we made several attempts to avoid accumulation of high level of intracellular 5-ALA. Intracellular 5-ALA decreased by 80.85% with the addition of penicillin, which was reported to inhibit the activity of PBPs for PG formation. However, penicillin addition results in an increased cost and a more complex production process. What is worse, besides the PG formation related genes, penicillin also inhibited the activity of 2-oxoglutarate dehydrogenase (Kim et al., 2010), which is unfavorable for 5-ALA production using C4 pathway. Based on this, we investigated the effect of deleting specific genes for PG synthesis on 5-ALA production. HMW-PBPs play a major role in PG synthesis. We screened the mutations of different nonessential HMW-PBPs. Deletion of pbp1a and pbp1b both promoted 5-ALA production significantly, whereas pbp1b mutant produced higher 5-ALA than pbp1a mutant. The different function in cell wall synthesis may explain the higher 5-ALA production in pbp1b mutant than that in pbp1a mutant.
Figure 8.
Scanning electron micrographs of C. glutamicum wild type and the HWM-PBPs defect mutants. The cells were cultured in liquid CGIIIM medium to exponential growth phase and treated as described in Materials and Methods section.
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1291
PBP1a interacts with the polar scaffold protein DivIVA (Letek et al., 2008). PBP1b not only interacts with RodA involving in the translocation of PG lipid precursors (Sieger et al., 2013) but also interacts with PBP2a and PBP2b (Valbuena et al., 2007). Moreover, pbp1b deletion mutant exhibited a reduced biofilm formation in E. coli (Kumar et al., 2012). Similar phenomena may also occur in C. glutamicum, thus enhancing cell permeability. Deletion of pbp2b also tremendously promoted 5-ALA production. Two reasons may account for this high production performance. Compared to the parent strain and other nonessential HMW-PBPs deletion mutants, the shape of pbp2b mutant cells became longer (Fig. 8). The apparent reduction of intracellular 5-ALA concentration may be due to the elongation of cells, which may result in reduced ALAS inhibition. Furthermore, the increased sensitivity to lysozyme in pbp2b deletion mutant demonstrated that integrity of the cell wall has been affected, which in turn results in enhanced cell wall permeability. The deletion of pbp1b in C. glutamicum ALA7(overexpression of RhtA transporter in C. glutamicum AP2) contributed to a 12.95% increase in 5-ALA production compared to C. glutamicum ALA8 (overexpression of RhtA transporter in C. glutamicum ALA6). While deletion of pbp1b in C. glutamicum AP2 caused a 29.47% increase in 5-ALA production compared to C. glutamicum ALA6. These data disclosed the significant effect of deletion of pbp1b on 5-ALA production especially when 5-ALA export was limited. The increased cell permeability in HMW-PBPs mutants may be beneficial for the production of biochemicals, especially for those involving in rigorous feedback or allosteric regulation. In the fed-batch process, C. glutamicum ALA7 produced 7.53 g/L 5-ALA with addition of glycine, however, only 2.00 g/L 5-ALA was produced when the medium was not supplemented with glycine. Therefore, metabolic engineering of glycine pathways will become an interesting target for 5-ALA production from renewable resources. This work was supported by National Program on Key Basic Research Project (2012CB725203), National Natural Science Foundation of China (NSFC21206112, NSFC-21576200, NSFC-21176182, and NSFC-21390201), National High-tech R&D Program of China (2012AA022103, 2012AA02A702), Natural Science Foundation of Tianjin (No. 15JCQNJC06000).
References Akram NA, Ashraf M. 2013. Regulation in plant stress tolerance by a potential plant growth regulator, 5-aminolevulinic acid. J Plant Growth Regul 32:663–679. Balzer GJ, Thakker C, Bennett GN, San KY. 2013. Metabolic engineering of Escherichia coli to minimize byproduct formate and improving succinate productivity through increasing NADH availability by heterologous expression of NAD(þ)-dependent formate dehydrogenase. Metab Eng 20:1–8. Becker J, Zelder O, Hafner S, Schroder H, Wittmann C. 2011. From zero to herodesign-based systems metabolic engineering of Corynebacterium glutamicum for L-lysine production. Metab Eng 13:159–168. Cornelius JF, Slotty PJ, El Khatib M, Giannakis A, Senger B, Steiger HJ. 2014. Enhancing the effect of 5-aminolevulinic acid based photodynamic therapy in human meningioma cells. Photodiagnosis Photodyn Ther 11:1–6. Fu W, Lin J, Cen P. 2007. 5-Aminolevulinate production with recombinant Escherichia coli using a rare codon optimizer host strain. Appl Microbiol Biotechnol 75:777–782. Fu W, Lin J, Cen P. 2010. Expression of a hemA gene from Agrobacterium radiobacter in a rare codon optimizing Escherichia coli for improving 5-aminolevulinate production. Appl Biochem Biotech 160:456–466. Jakoby M, Ngouoto-Nkili C-E, Burkovski A. 1999. Construction and application of new Corynebacterium glutamicum vectors. Biotechnol Tech 13:437–441.
1292
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
Jørgensen MG, Nielsen JS, Boysen A, Franch T, Møller-Jensen J, Valentin-Hansen P. 2012. Small regulatory RNAs control the multi-cellular adhesive lifestyle of Escherichia coli. Mol Microbiol 84:36–50. Kang Z, Wang Y, Gu P, Wang Q, Qi Q. 2011a. Engineering Escherichia coli for efficient production of 5-aminolevulinic acid from glucose. Metab Eng 13:492–498. Kang Z, Wang Y, Wang Q, Qi Q. 2011b. Metabolic engineering to improve 5-aminolevulinic acid production. Bioeng Bugs 2:342–345. Kang Z, Zhang J, Zhou J, Qi Q, Du G, Chen J. 2012. Recent advances in microbial production of delta-aminolevulinic acid and vitamin B12. Biotechnol Adv 30:1533–1542. Kim J, Fukuda H, Hirasawa T, Nagahisa K, Nagai K, Wachi M, Shimizu H. 2010. Requirement of de novo synthesis of the OdhI protein in penicillin-induced glutamate production by Corynebacterium glutamicum. Appl Microbiol Biotechnol 86:911–920. Kumar A, Sarkar SK, Ghosh D, Ghosh AS. 2012. Deletion of penicillin-binding protein 1b impairs biofilm formation and motility in Escherichia coli. Res Microbiol 163:254–257. Laneelle M-A, Tropis M, Daffe M. 2013. Current knowledge on mycolic acids in Corynebacterium glutamicum and their relevance for biotechnological processes. Appl Microbiol Biotechnol 97:9923–9930. Layer G, Reichelt J, Jahn D, Heinz DW. 2010. Structure and function of enzymes in heme biosynthesis. Protein Sci 19:1137–1161. Lee D-H, Jun W-J, Kim K-M, Shin D-H, Cho H-Y, Hong B-S. 2003. Inhibition of 5-aminolevulinic acid dehydratase in recombinant Escherichia coli using D-glucose. Enzyme Microb Tech 32:27–34. Letek M, Ordonez E, Vaquera J, Margolin W, Flardh K, Mateos LM, Gil JA. 2008. DivIVA is required for polar growth in the MreB-lacking rod-shaped actinomycete Corynebacterium glutamicum. J Bacteriol 190:3283–3292. Levican G, Katz A, de Armas M, Nunez H, Orellana O. 2007. Regulation of a glutamyl-tRNA synthetase by the heme status. Proc Natl Acad Sci USA 104: 3135–3140. Li F, Wang Y, Gong K, Wang Q, Liang Q, Qi Q. 2013. Constitutive expression of RyhB regulates the heme biosynthesis pathway and increases the 5-aminolevulinic acid accumulation in Escherichia coli. FEMS Microbiol Lett 350:209–215. Lin H, Castro NM, Bennett GN, San KY. 2006. Acetyl-CoA synthetase overexpression in Escherichia coli demonstrates more efficient acetate assimilation and lower acetate accumulation: A potential tool in metabolic engineering. Appl Microbiol Biotechnol 71:870–874. Lin J, Fu W, Cen P. 2009. Characterization of 5-aminolevulinate synthase from Agrobacterium radiobacter, screening new inhibitors for 5-aminolevulinate dehydratase from Escherichia coli and their potential use for high 5-aminolevulinate production. Bioresour Technol 100:2293–2297. Litsanov B, Kabus A, Brocker M, Bott M. 2012. Efficient aerobic succinate production from glucose in minimal medium with Corynebacterium glutamicum. Microb Biotechnol 5:116–128. Liu S, Zhang G, Li X, Zhang J. 2014. Microbial production and applications of 5-aminolevulinic acid. Appl Microbiol Biotechnol 98:7349–7357. Ma W, Wang X, Mao Y, Wang Z, Chen T, Zhao X. 2015. Development of a markerless gene replacement system in Corynebacterium glutamicum using upp as a counter-selection marker. Biotechnol Lett 37:609–617. Meiswinkel TM, Rittmann D, Lindner SN, Wendisch VF. 2013. Crude glycerol-based production of amino acids and putrescine by Corynebacterium glutamicum. Bioresource Technol 145:254–258. Petersen S, de Graaf AA, Eggeling L, Mollney M, Wiechert W, Sahm H. 2000. In vivo quantification of parallel and bidirectional fluxes in the anaplerosis of Corynebacterium glutamicum. J Biol Chem 275:35932–35941. Sasaki K, Watanabe M, Tanaka T, Tanaka T. 2002. Biosynthesis, biotechnological production and applications of 5-aminolevulinic acid. Appl Microbiol Biotechnol 58:23–29. Sieger B, Schubert K, Donovan C, Bramkamp M. 2013. The lipid II flippase RodA determines morphology and growth in Corynebacterium glutamicum. Mol Microbiol 90:966–982. Song CW, Kim DI, Choi S, Jang JW, Lee SY. 2013. Metabolic engineering of Escherichia coli for the production of fumaric acid. Biotechnol Bioeng 110: 2025–2034. Unthan S, Gr€unberger A, Van Ooyen J, G€atgens J, Heinrich J, Paczia N, Wiechert W, Kohlheyer D, Noack S. 2014. Beyond growth rate 0.6: What drives Corynebacterium glutamicum to higher growth rates in defined medium. Biotechnol Bioeng 111:359–371.
Valbuena N, Letek M, Ordonez E, Ayala J, Daniel RA, Gil JA, Mateos LM. 2007. Characterization of HMW-PBPs from the rod-shaped actinomycete Corynebacterium glutamicum: Peptidoglycan synthesis in cells lacking actin-like cytoskeletal structures. Mol Microbiol 66:643–657. Valbuena N, Letek M, Ramos A, Ayala J, Nakunst D, Kalinowski J, Mateos LM, Gil JA. 2006. Morphological changes and proteome response of Corynebacterium glutamicum to a partial depletion of FtsI. Microbiology 152:2491–2503. van der Werf MJ, Zeikus JG. 1996. 5-Aminolevulinate production by Escherichia coli containing the Rhodobacter sphaeroides hemA gene. Appl Environ Microbiol 62:3560–3566. Vogt M, Haas S, Klaffl S, Polen T, Eggeling L, van Ooyen J, Bott M. 2014. Pushing product formation to its limit: Metabolic engineering of Corynebacterium glutamicum for L-leucine overproduction. Metab Eng 22:40–52. Yang J, Zhu L, Fu W, Lin Y, Lin J, Cen P. 2013. Improved 5-aminolevulinic acid production with recombinant Escherichia coli by a short-term dissolved oxygen shock in fed-batch fermentation. Chin J Chem Eng 21:1291–1295. Yang J, Zhu Y, Li J, Men Y, Sun Y, Ma Y. 2015. Biosynthesis of rare ketoses through constructing a recombination pathway in an engineered Corynebacterium glutamicum. Biotechnol Bioeng 112:168–180.
Zhang J, Kang Z, Chen J, Du G. 2015. Optimization of the heme biosynthesis pathway for the production of 5-aminolevulinic acid in Escherichia coli. Sci Rep 5:8584. Zhang Y, Praszkier J, Hodgson A, Pittard A. 1994. Molecular analysis and characterization of a broad-host-range plasmid, p EP2. J Bacteriol 176: 5718–5728. Zhu N, Xia H, Wang Z, Zhao X, Chen T. 2013. Engineering of acetate recycling and citrate synthase to improve aerobic succinate production in Corynebacterium glutamicum. PLoS ONE 8:e60659. Zhu N, Xia H, Yang J, Zhao X, Chen T. 2014. Improved succinate production in Corynebacterium glutamicum by engineering glyoxylate pathway and succinate export system. Biotechnol Lett 36:553–560.
Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web-site.
Feng et al.: Production of 5-aminolevulinic acid
1293
Department of Biochemical Engineering, School of Chemical Engineering and Technology, Tianjin University, Tianjin, China 2 Key Laboratory of Systems Bioengineering (Ministry of Education), Tianjin University, Tianjin, China 3 SynBio Research Platform, Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), School of Chemical Engineering and Technology, Tianjin University, Tianjin 300072, China; telephone: 86-22-2740-6582; fax: 86-22-2740-6582; e-mail: [email protected]
ABSTRACT: 5-Aminolevulinic acid (5-ALA) has recently attracted attention for its potential applications in the fields of medicine and agriculture. In this study, Corynebacterium glutamicum was firstly engineered for 5-ALA production via the C4 pathway. HemA encoding 5-aminolevulinic acid synthase from Rhodobacter sphaeroides was codon optimized and expressed in C. glutamicum ATCC13032, resulting in accumulation of 5-ALA. Deletion of all known genes responsible for the formation of acetate and lactate further enhanced production of 5-ALA. Overexpression of ppc gene encoding phoenolpyruvate carboxylase resulted in an accumulation of 5-ALA up to 2.06 0.05 g/L. Furthermore, deletion of highmolecular-weight penicillin-binding proteins (HMW-PBPs) genes pbp1a, pbp1b, and pbp2b led to an increase in 5-ALA production of 13.53%, 29.47%, and 22.22%, respectively. Finally, 5-ALA production was enhanced to 3.14 0.02 g/L in shake flask by heterologously expressing rhtA encoding threonine/homoserine exporter, and 86.77% of supplemented glycine was channeled toward 5-ALA production in shake flask. The engineered C. glutamicum ALA7 strain produced 7.53 g/L 5-ALA in a 5 L bioreactor. This study demonstrated the potential utility of C. glutamicum as a platform for metabolic production of 5-ALA. Change of cell permeability by metabolic engineering HMW-PBPs
Correspondence to: Z. Wang Contract grant sponsor: National Program on Key Basic Research Project Contract grant number: 2012CB725203 Contract grant sponsor: National Natural Science Foundation of China Contract grant numbers: NSFC-21206112; NSFC-21576200; NSFC-21176182; NSFC21390201 Contract grant sponsor: National High-tech R&D Program of China Contract grant numbers: 2012AA022103; 2012AA02A702 Contract grant sponsor: Natural Science Foundation of Tianjin Contract grant number: 15JCQNJC06000 Received 3 August 2015; Revision received 10 November 2015; Accepted 19 November 2015 Accepted manuscript online 30 November 2015; Article first published online 9 December 2015 in Wiley Online Library (http://onlinelibrary.wiley.com/doi/10.1002/bit.25886/abstract). DOI 10.1002/bit.25886
1284
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
may provide a new strategy for biochemicals production in Corynebacterium glutamicum. Biotechnol. Bioeng. 2016;113: 1284–1293. ß 2015 Wiley Periodicals, Inc. KEYWORDS: 5-aminolevulinic acid; Corynebacterium glutamicum; C4 pathway; penicillin-binding proteins; cell permeability
Introduction 5-Aminolevulinic acid (5-ALA), a non-proteinogenic amino acid, plays a key role in the biosynthesis of tetrapyrrole, such as heme, chlorophyll, and vitamin B12. 5-ALA has been used in photodynamic diagnosis and therapy for various cancers (Cornelius et al., 2014), and applied as a selective biodegradable herbicide, insecticide, or a growth regulator with nontoxicity to crops, animals, and humans (Akram and Ashraf, 2013; Sasaki et al., 2002). Since 5-ALAproduction via chemical synthesis is low-yielding and high-cost (Kang et al., 2012), alternative microbial production of 5-ALA from renewable carbon sources has received much attention. Two natural pathways for the biosynthesis of 5-ALA exist in living organisms. One is the C4 pathway, through which glycine and succinyl-CoA are condensed to form 5-ALA by 5-aminolevulinate synthase (ALAS) in the presence of the essential cofactor pyridoxal 50 -phosphate (PLP) (Liu et al., 2014). This pathway is present in mammals, birds, yeasts, and purple non-sulfur photosynthetic bacteria, such as in Rhodobacter sphaeroides and Rhodopseudomonas. The alternative C5 pathway is tightly regulated by feedback inhibition of heme (Levican et al., 2007). In this pathway, 5-ALA is synthesized from glutamate via three steps, and the corresponding enzymes involved in this process are glutamyl-tRNA synthetase ß 2015 Wiley Periodicals, Inc.
(encoded by gltx), NADPH-dependent glutamyl-tRNA reductase (encoded by hemA), and glutamate-1-semialdehyde aminotransferase (encoded by hemL) (Liu et al., 2014). This pathway exists in higher plants, algae, and various bacteria, such as Escherichia coli and Corynebacterium glutamicum. Metabolic engineering of microbial organisms for 5-ALA production has mainly focused on E. coli. Recombinant E. coli heterologously expressing ALAS from R. sphaeroides could accumulate 5-ALA via C4 pathway, and the expression of ALAS in E. coli was further optimized to improve 5-ALA production using factorial design (Liu et al., 2013). In addition, a rare codon optimizer strain was used to enhance ALAS activity, resulting in production of 6.56 g/L 5-ALA in a 15 L fermenter (Fu et al., 2010). As succinyl-CoA is a precursor for 5-ALA, production was enhanced fourfold by expressing ALAS from R. sphaeroides ALAS in an E. coli strain engineered for succinate production (Kang et al., 2011b). Moreover, 5-ALA degradation can be prevented by adding D-xylose and D-glucose, effective inhibitors of 5-ALA dehydratase (Lee et al., 2003; Lin et al., 2009). Fermentation process engineering using short-term dissolved oxygen shock has also been applied to enhance 5-ALA production up to 9.44 g/L during fed-batch fermentation (Yang et al., 2013). Glutamyl-tRNA reductase and glutamate-1-semialdehyde aminotransferase were found to have a synergistic effect on 5-ALA production via the C5 pathway (Layer et al., 2010). 5-ALA production reached 2.05 g/L by co-expressing a mutant of hemA (Salmonella arizona) together with hemL (E. coli), and production was further increased to 2.96 g/L after overexpressing RhtA transporter involved in the threonine and homoserine efflux system (Kang et al., 2011a). In the C5 pathway, 5-ALA biosynthesis is regulated through a feedback inhibition mechanism by heme (Zhang et al., 2015), whose biosynthesis is involved in iron consumption. Recombinant E. coli produced 16% higher 5-ALA by overexpressing ryhB, a small RNA regulator of iron metabolism in E. coli (Liu et al., 2013). Besides E. coli, C. glutamicum, one of the most important commercial amino acid-producing microorganism with generally regarded as safe (GRAS) status (Yang et al., 2015), seems to be a promising alternative for 5-ALA production. Although C. glutamicum utilizes the C5 pathway for synthesis of 5-ALA, the complex regulation of this biosynthetic route suggests that metabolic engineering for efficient 5-ALA production via the C5 pathway might be problematic. Therefore, we focused on exploring the potential of 5-ALA production via the C4 pathway. Moreover, there appears to be a higher potential for 5-ALA production via the C4 pathway using C. glutamicum compared to E. coli, due to the lack of glycine cleavage systems found in E. coli (Jørgensen et al., 2012). In the present work, we described the production of 5-ALA via the C4 pathway using C. glutamicum (Fig. 1). Metabolic engineering strategies included expressing of heterologous ALAS, reducing byproduct formation, and increasing flux to 5-ALA. Additionally, engineering of high-molecular-weight penicillin-binding proteins (HMW-PBPs) and 5-ALA export was found to have a significant effect on the 5-ALA production. This study demonstrates the potential application of C. glutamicum as a platform for 5-ALA production and provides evidence that modification of different
HMW-PBPs may be a universal strategy for some biochemical production such as glutamate and 2-(3-hydroxy-1-adamantyl)(2S)-amino ethanoic acid.
Materials and Methods Bacteria Strains, Media, and Growth Conditions All C. glutamicum strains used were derived from the wild type C. glutamicum WT 13032. The bacterial strains and plasmids used in this work were listed in Table I. The construction of plasmids and strains was described in Supporting Information 1. 5-ALA production was tested in CGIIIM medium containing 10 g/L yeast extract, 10 g/L tryptone, 2.5 g/L NaCl, 10 g/L glucose, and 21 g/L 3-morpholinopropanesulfonic acid. Physiological characterizations were tested in defined CGXII medium (Unthan et al., 2014) supplemented with 5.5 g/L glucose and 7.5 g/L glycine. For precultivation of C. glutamicum ATCC 13032 and its recombinant derivatives, a single clone was grown in 5 mL of BHI medium (Vogt et al., 2014). After incubation for approximately 15 h, cells were inoculated into 500-mL shake flask containing 50 mL shake cultivation medium with an initial OD600 of 0.1. The cells were cultivated at 30 C in a rotatory shaker at 220 rpm. The medium for strains with the episomal replicating plasmid was additionally supplemented with 25.0 mg/mL kanamycin and/or 10 mg/mL chloroamphenicol when needed. When OD600 reached to 1, 0.5 mM isopropyl b-D-1-thiogalactopyranoside (IPTG) was added into the medium to induce the expression of genes under control of IPTG-inducible promoter. For fed-batch culture, a 50 mL seed cultured in CGIIIM medium for 8 h was transferred into a 5 L bioreactor containing 2.5 L initial fermentation medium composed of 10 g/L tryptone, 15 g/L yeast extract, 2.5 g/L NaCl, 7.5 g/L glycine, and 20 g/L glucose. Initially, ammonia and 3 M H2SO4 were added automatically to maintain a pH of 7.0, and after 15 h fermentation the pH was maintained at 6.5. The aeration rate was 1 vvm and dissolved oxygen was maintained between 25% and 35% saturation by controlling the stirrer speed. When glucose was lower than 1 g/L, the feeding process was initiated to maintain a basal glucose concentration. The feed solution was composed of 10 g/L tryptone, 15 g/L yeast extract, 2.5 g/L NaCl, 5 g/L K2HPO4, 5 g/L KH2PO4, 2.5 g/L MgSO4 7H2O, and 600 g/L glucose. Scanning Electron Microscopy (SEM) Cells from exponential growth phase were harvested by centrifugation, washed twice using pH 7.4 phosphate buffer solution (PBS), and fixed in 2.5% glutaraldehyde solution for 12 h at 4 C. The cells were washed twice in PBS (pH 7.4) and successively dehydrated following a stepwise treatment in 20, 50, 80, and 100% ethanol, respectively. Finally, the cells were treated with tert-butyl alcohol, freeze-dried, and analyzed using SEM. Lysozyme Sensitivity Test of C. glutamicum Lysozyme sensitivities of C. glutamicum were determined by growth assay in LB medium supplemented with 1 g/L glucose. 25 mg/mL
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1285
Figure 1.
Metabolic pathways associated with 5-ALA biosynthesis in C. glutamicum and the metabolic engineering strategies for aerobic 5-ALA overproduction. The red indicated that the pathways were disrupted. The blue font indicated that pathways were overexpressed. Relevant reactions were represented by the genes encoding for corresponding enzymes: ldhA, encoding L-lactate dehydrogenase; pqo, encoding pyruvate: menaquinone oxidoreductase; pta, encoding phosphotransacetylase; ackA, encoding acetate kinase; cat, encoding acetyl-CoA:CoA transferase; ppc, encoding phosphoenolpyruvate carboxylase; pyc, encoding pyruvate carboxylase; pck, encoding phosphoenolpyruvate carboxykinase; gltA, encoding citrate synthase; hemA, encoding 5-aminolevulinic acid synthease; hemB, encoding aminolevulinic acid dehydratase; rhtA, encoding gene encoding inner membrane transporter for L-threonine; and L-homoserine; pbp1a, pbp1b, pbp2a, and pbp2b, genes encoding HMW-PBPs.
lysozyme was added into medium when the cells were in the exponential phase. Analytical Methods Cell concentration was determined by the optical density at 600 nm (OD600). The biomass concentration was calculated from OD600 values using an experimentally determined correlation 1 OD600 unit is equal to 0.25 g/L cell dry weight (CDW) (Meiswinkel et al., 2013). Glucose concentration was measured using a SBA sensor machine
1286
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
(SBA-40E, Institute of Microbiology, Shandong, China). Intracellular 5-ALA was obtained with ultrasonic treatment as described previously (Kang et al., 2011a). Concentration of extracellular and intracellular 5-ALA was measured using modified Ehrlich’s reagent (Kang et al., 2011a). Concentrations of extracellular organic acids were determined by high performance liquid chromatography (HPLC) equipped with a cation exchange column (Aminex HPX87-H, Bio-Rad, Hercules, CA), a UV absorbance detector (Agilent Technologies, G1315D), and a refractive index (RI) detector (Agilent Technologies, HP1047A). A mobile phase of 5 mM H2SO4 solution at
Table I. Bacterial strains and plasmids used in this study. Relevant characteristicsa
Strains and plasmids Strains C. glutamicum WT 13032 C. glutamicum SAZ1 C. glutamicum SAZ2 C. glutamicum SAZ3 C. glutamicum ALA1 C. glutamicum ALA2 C. glutamicum ALA3 C. glutamicum ALA4 C. glutamicum ALA5 C. glutamicum ALA6 C. glutamicum AP1 C. glutamicum AP2 C. glutamicum AP3 C. glutamicum AP4 C. glutamicum ALA7 C. glutamicum ALA8 C. glutamicum C1 Plasmids pXMJ19 pXA pEC-XK99E pECacsAgltA pDsacB pD-pck pD-pbp1a pD-pbp1b pD-pbp2a pD-pbp2b pEP2 pEP2-rhtA a
Sources or references
Wild-type strain, biotin auxotrophic WT13032DldhADpqoDcatDptaDackA WT13032DldhADpqoDcatDptaDackA Pppc::Psod WT13032DldhADpqoDcatDptaDackA Pppc::Psod Ppyc::Psod Wild-type ATCC 13032 harboring pXA WT13032DldhADpqoDcatDptaDackA harboring pXA WT13032DldhADpqoDcatDptaDackA harboring pXA and pECacsAgltA WT13032DldhADpqoDcatDptaDackA Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackA Pppc::Psod Ppyc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpck Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp1a Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp1b Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp2a Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp2b Pppc::Psod harboring pXA WT13032DldhADpqoDcatDptaDackADpckDpbp1b Pppc::Psod harboring pXA and pEP2-rhtA WT13032DldhADpqoDcatDptaDackADpck Pppc::Psod harboring pXA and pEP2-rhtA WT13032DldhADpqoDcatDpta harboring pXA and pEC-XK99E Cmr, C. glutamicum/E. coli shuttle vector Derived from pXMJ19, for overexpress hemA under the control of tac promoter Kanr; C. glutamicum/E. coli shuttle vector derived from pEC-XK99E, for overexpression of acsA and gltA under the control of trc promoter derived from pK18mobsacB, for increasing expression of sacB under trc promoter Integrative transformation vector for deletion of the of pck gene Integrative transformation vector for deletion of the pbp1a gene. Integrative transformation vector for deletion of the pbp1b gene Integrative transformation vector for deletion of the pbp2a gene Integrative transformation vector for deletion of the pbp2b gene Kanr, C. glutamicum/E. coli shuttle vector Derived from pEP2, for overexpression of rhtA from E.coli under control of tuf promoter
ATCCb Lab stock (Zhu et al. 2014) (Zhu et al., 2013) This study This study This study This study This study This study This study This study This study This study This study This study This study (Jakoby et al. 1999) This study (Zhu et al. 2013) (Zhu et al. 2013) (Ma et al. 2015) This study This study This study This study This study (Zhang et al. 1994) This study
Abbreviations: Cm, chloramphenicol; Kanr, kanamycin; r, resistance. American type culture collection.
b
a 0.4 mL/min flow rate was used. The column was operated at 65 C. The concentration of amino acids was measured using a C18 column (Luna 150 4.6 mm) with 20 mM sodium acetate/acetonitrile (97:3, v/v) (buffer A), 20 mM sodium acetate solution/acetonitrile (50:50, v/v) (buffer B) as mobile phase at 0.6 mL/min, and derivative reagent used was consisted of 6.0 mg o-phthalaldehyde, 0.1 mL anhydrous ethanol, 0.9 mL 0.1 mM pH 10.4 borate buffer solution, and 5.0 mL 3-mercaptopropionic acid (MPA).
Enzyme Assays Cells from exponential phase were harvested by centrifugation (20 min, 6500 rpm, 4 C). Following centrifugation, cells were washed twice with 15 mL 50 mM Tris-HCl (pH 7.5) containing 0.1 mM EDTA and 2 mM dithiothreitol (DTT), centrifuged again, and resuspended in the same buffer. The cells were disrupted using sonication for 30 min in an ice bath (130 W, 20 kHz, pulse: 5 s on; 5 s off), and the cell lysate was clarified by centrifugation (15 min, 6500 rpm, 4 C). Total protein concentrations were quantified by the Bradford method using bovine serum albumin as standard. ALAS activity was determined as described previously (Fu et al., 2007). One unit of ALAS activity represents the amount of enzyme required to produce 1 nmol of 5-ALA per minute.
Results Establishing the C4 Pathway for 5-ALA Production by Introducing the HemA Gene (Encoding ALAS) From R. sphaeroides in C. glutamicum ALAS (encoded by hemA) has been isolated and characterized in several bacteria, such as R. sphaeroides and Rhodobacter capsulatus (Kang et al., 2012). We favored utilization of ALAS from R. sphaeroides due to its higher specific activity and lower Km for glycine and succinyl-CoA (van der Werf and Zeikus, 1996). Therefore, the codon-optimized version of hemA from R. sphaeroides was synthesized and expressed under the control of IPTG-inducible tac promoter from a medium-copy plasmid pXMJ19 in C. glutamicum. It was found that specific ALAS activity of C. glutamicum ALA1 in the middle of exponential growth phase was 19.49 0.19 U/mg protein, whereas no activity was detected in wild-type C. glutamicum 13032 harboring empty plasmid pXMJ19, indicating successful expression of ALAS. However, only a small amount 5-ALA was accumulated by C. glutamicum ALA1. Although glycine can be synthesized from glucose by C. glutamicum itself, the amount of intracellular glycine may not be sufficient for achieving high titers of 5-ALA. The glycine was supplemented with different
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1287
concentrations. Although addition of glycine resulted in reduction of biomass accumulation, 5-ALA production increased tremendously, reaching 1.44 0.03 g/L in the presence of 5 and 7.5 g/L glycine (Fig. 2), and CGIIIM medium supplemented with 7.5 g/L glycine was used for subsequent experiments in this study. Improving 5-ALA Production by Reducing Acetate Formation Under aerobic conditions, 1.70 0.03 g/L acetate was produced by C. glutamicum ALA1 (Table II). To reduce acetate formation, genes responsible for acetate accumulation including pqo, pta, ackA, and cat encoding for pyruvate:menaquinone oxidoreductase, phosphotransacetylase, acetate kinase, and acetyl-CoA: CoA transferase, respectively, were all deleted. Although lactate was not detected in the shake flask, deletion of ldhA encoding L-lactate dehydrogenase may exert positive effect on 5-ALA production during high-density fermentation in fed-batch cultivation. Therefore, ldhA was also deleted. The deletion of the above-mentioned five genes in C. glutamicum ALA1 led to an 31.18% reduction in acetate formation (Table II). 5-ALA production in the supernatant increased to 1.92 0.02 g/L at the end of fermentation. Although all identified genes responsible for acetic acid formation in C. glutamicum were deleted, 1.17 0.05 g/L acetate was still detected in the culture. To further reduce acetate formation and redirect carbon flux toward 5-ALA biosynthesis, acsA encoding acetyl-CoA synthetase from Bacillus subtilis, which catalyzes a reaction for converting acetate to acetyl-CoA (Lin et al., 2006), was expressed in combination with endogenous gltA gene encoding citrate synthase from a medium-copy plasmid pEC-XK99E under the control of the IPTG-inducible trc promoter. Acetic acid production of resulting strain C. glutamicum ALA3 was reduced by 58.40%, demonstrating that acetate was effectively assimilated. Nevertheless,5-ALA production dramaticly dropped to 0.65 0.04 g/L (Table II). In light of this result, overexpression of
acetyl-CoA synthetase and citrate synthase seemed to work against the production of 5-ALA in C. glutamicum.
Effect of Modification of Anaplerotic Reactions on 5-ALA Production Oxaloacetate supply was reported to be the bottleneck for the production of compounds derived from the TCA cycle (Balzer et al., 2013; Litsanov et al., 2012). To explore whether oxaloacetate supply contributes to 5-ALA production, the flux to oxaloacetate was engineered by overexpressing ppc (encoding phosphoenolpyruvate carboxylase) and pyc (encoding pyruvate carboxylase) as well as deleting pck (encoding phosphoenolpyruvate carboxykinase). Phosphoenolpyruvate carboxylase, catalyzing the carboxylation of phosphoenolpyruvate to oxaloacetate, was reported for oxaloacetate replenishment in C. glutamicum (Litsanov et al., 2012). Therefore, we replaced the native promoter of ppc gene in C. glutamicum ALA2 with the strong sod promoter from C. glutamicum, resulting in C. glutamicum ALA4. In response to this modification, biomass of C. glutamicum ALA4 increased by 8.98% and 5-ALA production increased by 7.29%, reaching to 2.06 0.05 g/L (Table II). To further enhance oxaloacetate supply, the native promoter of pyc in C. glutamicum ALA4 was also replaced by the sod promoter, resulting in C. glutamicum ALA5. Nevertheless, this modification did not further enhance 5-ALA production. We speculated that the increased supply of oxaloacetate may not be efficiently channeled into the TCA pathway, and may be redirected back into glycolysis via phosphoenolpyruvate carboxykinase (Becker et al., 2011). Therefore, the phosphoenolpyruvate carboxykinase was inactivated in C. glutamicum ALA4. However, the resulting C. glutamicum ALA6 produced 2.07 0.01 g/L 5-ALA, almost identical to that produced by C. glutamicum ALA4 (Table II).
Enhancing 5-ALA Production by Deletion of Different HMW-PBPs
Figure 2.
Effect of glycine addition on cell growth and 5-ALA production. C. glutamicum ALA1 was cultured in CGIIIM medium supplemented with different concentration of glycine at 30 C. 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
1288
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
In the previous study, inhibition of cell wall synthesis through the addition of penicillin has proven to be an efficient strategy for enhancing production of L-glutamate by C. glutamicum (Laneelle et al., 2013). Herein, we investigated the effect of penicillin addition on 5-ALA production. Extracellular 5-ALA was tremendously increased with the addition of 2.5 mg/mL penicillin and intracellular 5-ALA decreased by 80.85% (Fig. 3). Unfortunately, addition of penicillin to the medium results in a more complex and expensive production process as well as presenting a serious potential for environmental polution. An alternative approach is to deregulate the activity of PBPs targeted by penicillin rather than to directly add penicillin. Penicillin hinders cell wall biosynthesis by inhibiting the activity of PBPs. C. glutamicum has at least nine PBPs, and five of them are HMW-PBPs, which play a major role in PG synthesis. Among the HMW-PBPs, only FtsI was reported to be essential for cell growth (Valbuena et al., 2006). Thus, the other four HMW-PBP genes including class A HMW-PBP genes (pbp1a and pbp1b) and class B HMW-PBP genes (pbp2a and pbp2b) were deleted.
Table II. Growth and production performance of various strains constructed in the study. Strains
Biomass (g/L)
Consumed glucose (g/L)
5-ALA (g/L)
Consumed glycine (g/L)
Acetate (g/L)
ALA1 ALA2 C1 ALA3 ALA4 ALA5 ALA6 AP1 AP2 AP3 AP4 ALA7 ALA8
5.57 0.08 5.90 0.11 5.95 0.11 5.89 0.12 6.43 0.13 6.09 0.03 6.12 0.04 5.22 0.17 6.78 0.10 5.45 0.10 6.25 0.07 6.68 0.14 6.46 0.08
10.15 0.05 10.10 0.10 9.85 0.05 10.05 0.05 10.15 0.15 9.90 0.00 10.00 0.10 9.90 0.00 10.15 0.15 10.15 0.05 9.90 0.10 10.05 0.15 9.90 0.00
1.44 0.03 1.92 0.02 1.90 0.01 0.65 0.04 2.06 0.05 1.97 0.02 2.07 0.01 2.35 0.00 2.68 0.05 2.04 0.02 2.53 0.03 3.14 0.02 2.78 0.04
1.00 0.05 1.45 0.11 1.51 0.07 0.45 0.00 1.68 0.10 1.64 0.08 1.66 0.11 1.67 0.10 1.76 0.05 1.54 0.04 1.58 0.07 1.86 0.06 1.54 0.07
1.70 0.03 1.17 0.05 1.25 0.04 0.52 0.02 0.85 0.05 0.77 0.03 0.73 0.04 0.56 0.03 0.36 0.02 0.21 0.04 0.42 0.05 0.32 0.02 0.45 0.05
Batch cultures of recombinant strains were carried out in CGIIIM medium supplemented with 7.5 g/L glycine at 30 C. 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
The pbp1a, pbp1b, and pbp2b deletions all showed a positive effect on 5-ALA production. The corresponding mutants C. glutamicum AP1, C. glutamicum AP2, and C. glutamicum AP4 produced 2.35 0.00, 2.68 0.05, and 2.53 0.03 g/L 5-ALA, respectively, which were 13.53%, 29.47%, and 22.22% higher compared to the parent strain C. glutamicum ALA6 (Fig. 4). However, the pbp2a deletion mutant strain C. glutamicum AP3 produced almost the same amount of 5-ALA compared to C. glutamicum ALA6. Moreover, intracellular 5-ALA concentrations in HMW-PBPs mutants were determined. The intracellular level of 5-ALA in C. glutamicum AP1, C. glutamicum AP2, and C. glutamicum AP4 decreased to some extent compared to C. glutamicum ALA6 (Fig. 4). HMW-PBPs play a role in the biosynthesis of bacterial cell wall, and mutations in PBPs may affect cell growth and structural integrity of the cell wall. We assessed physiological characterizations and lysozyme sensitivity of different PBPs mutants. As shown in Figure 5, most of the mutants cultured in minimal medium
exhibited a similar growth rate as the parent strain. However, the growth rate of pbp2b mutant C. glutamicum AP4 was a little lower than wild-type strain, resulting in a lower biomass yield. For lysozyme sensitivity, C. glutamicum ALA6 was not affected by lysozyme treatment, indicating the high resistance to lysozyme. Nevertheless, the growth of the four HMW-PBP mutants decreased after the addition of lysozyme, resulting in lower biomass yield (Fig. 6). These results implied that deleting HMW-PBPs indeed affect cell surface integrity.
Figure 3.
Figure 4.
Comparison of cell growth and 5-ALA production with or without penicillin. C. glutamicum ALA6 was cultured in CGIIIM medium supplemented with 7.5 g/L glycine at 30 C. 0.5 mM IPTG was added into the medium when OD600 reached to 1. (A) C. glutamicum ALA6 was cultured without penicillin; (B) C. glutamicum ALA6 was cultured with the addition of 2.5 mg/mL penicillin when OD600 ¼ 5.79 0.11.
Improving 5-ALA Production by Heterologous Expression of rhtA From E. coli As decreasing intracellular 5-ALA is an effective strategy for increasing 5-ALA production, we further tried to decrease intracelluar 5-ALA by improving its transportation. RhtA transporter involved in threonine and homoserine efflux system
Distribution of intracellular and extracellular 5-ALA by C. glutamicum ALA6 and HWM-PBPs defect mutants. Batch cultures of recombinant strains were carried out in CGIIIM medium supplemented with 7.5 g/L glycine at 30 C, 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1289
Figure 5. Influence of deleting different HMW-PBPs on cell growth. (A) Biomass formation. (B) Glucose consumption. The cells were cultivated in CGXII medium supplemented with 5.5 g/L glucose and 7.5 g/L glycine and 0.5 mM IPTG was added into the medium when OD600 reached to 1. Standard deviations were calculated from the results of three independent experiments.
Figure 6. Effect of lysozyme on growth of C. glutamicum strains. The strains were cultured in LB medium supplemented with 1.0 g/L glucose. 0.5 mM IPTG was added into the medium when OD600 reached to 1. (A) The cells were cultured without lysozyme. (B) Lysozyme with concentration of 25 mg/ml was added to the medium at the time as indicated by an arrow.
could enhance 5-ALA export in E. coli due to its broad substrate specificity (Kang et al., 2011a). Herein, rhtA gene was expressed in C. glutamicum ALA7 from a low copy plasmid pEP2. Plasmid pEP2 carrying a different derivative replication system of pNG2 derived from Corynebacterium diphtheriae (Zhang et al., 1994) can be compatible with pXMJ19 carrying a mutant pBL1 ori (Jakoby et al., 1999). This strain produced 0.32 0.00 g/L 5-ALA without addition of glycine and 3.14 0.02 g/L 5-ALA with consumption of 1.86 0.06 g/L glycine. About 86.77% of supplemented glycine was channeled toward 5-ALA production in shake flask. Intracellular 5-ALA concentration of C. glutamicum ALA7 decreased by 16.43% compared to C. glutamicum AP2 (Fig. 4). These results demonstrated that heterologous expression of rhtA from E. coli effectively accelerated 5-ALA transport out of C. glutamicum. Plasmids stability of these two plasmids in C. glutamicum ALA7 was tested, about 42% plasmids could stay in C. glutamicum ALA7 at the end of shake flask cultivation.
Fed-Batch Culture for Aerobic 5-ALA Production
1290
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
The performance of C. glutamicum ALA7 was investigated in a fed-batch process. 5-ALA started to accumulate when IPTG was added into the medium, reaching to 7.53 g/L in 33 h with consumption of 4.70 g/L glycine and the overall 5-ALA productivity was 0.23 g/L/h (Fig. 7). Although C. glutamicum ALA7 produced 2.00 g/L 5-ALA without addition of glycine in fed-batch cultivation. About 67.27% of supplemented glycine was used for 5-ALA production in fed-batch cultivation.
Discussion C. glutamicum, a dominating amino acid producer, was firstly engineered for 5-ALA production via C4 pathway. Heterologous expression of ALAS from R. sphaeroides in C. glutamicum led to 5-ALA accumulation. To increase 5-ALA production, the known pathways for acetate and lactate formation were disrupted. A
Figure 7. Production performance of C. glutamicum ALA7 in a 5 L fed-batch cultivation. 25 mg/mL kanamycin and 10 mg/mL chloroamphenicol were added initially to provide selective pressure. 0.5 mM IPTG was added into the medium when OD600 reached to 1.
second step to improve 5-ALA production was to enhance the flux toward oxaloacetate, which was reported to effectively improve production of compounds derived from TCA cycle (Song et al., 2013). However, overexpression of pyc gene did not exert a positive effect on 5-ALA production, whereas overexpression of ppc was found to increase 5-ALA production. This was not consistent with previous research, which showed that pyruvate carboxylase
contributed to about 90% oxaloacetate replenishment of C. glutamicum when grown on glucose under aerobic condition, whereas phosphoenolpyruvate carboxylase was responsible for 10% oxaloacetate replenishment (Litsanov et al., 2012). One possible reason for this inconsistent result is that the disruption of flux through the acetate-forming pathway may result in the accumulation of acetyl-CoA, which regulates the two anaplerotic enzymes in a contrary manner. Pyruvate carboxylase is inhibited by 110 mM acetyl-CoA, whereas 100 mM acetyl-CoA activates phosphoenolpyruvate carboxylase (Petersen et al., 2000). Similar results were reported by Litsanov et al. (2012) for succinate production following deletion of an acetate-forming pathway in C. glutamicum. In this study, we made several attempts to avoid accumulation of high level of intracellular 5-ALA. Intracellular 5-ALA decreased by 80.85% with the addition of penicillin, which was reported to inhibit the activity of PBPs for PG formation. However, penicillin addition results in an increased cost and a more complex production process. What is worse, besides the PG formation related genes, penicillin also inhibited the activity of 2-oxoglutarate dehydrogenase (Kim et al., 2010), which is unfavorable for 5-ALA production using C4 pathway. Based on this, we investigated the effect of deleting specific genes for PG synthesis on 5-ALA production. HMW-PBPs play a major role in PG synthesis. We screened the mutations of different nonessential HMW-PBPs. Deletion of pbp1a and pbp1b both promoted 5-ALA production significantly, whereas pbp1b mutant produced higher 5-ALA than pbp1a mutant. The different function in cell wall synthesis may explain the higher 5-ALA production in pbp1b mutant than that in pbp1a mutant.
Figure 8.
Scanning electron micrographs of C. glutamicum wild type and the HWM-PBPs defect mutants. The cells were cultured in liquid CGIIIM medium to exponential growth phase and treated as described in Materials and Methods section.
Feng et al.: Production of 5-aminolevulinic acid Biotechnology and Bioengineering
1291
PBP1a interacts with the polar scaffold protein DivIVA (Letek et al., 2008). PBP1b not only interacts with RodA involving in the translocation of PG lipid precursors (Sieger et al., 2013) but also interacts with PBP2a and PBP2b (Valbuena et al., 2007). Moreover, pbp1b deletion mutant exhibited a reduced biofilm formation in E. coli (Kumar et al., 2012). Similar phenomena may also occur in C. glutamicum, thus enhancing cell permeability. Deletion of pbp2b also tremendously promoted 5-ALA production. Two reasons may account for this high production performance. Compared to the parent strain and other nonessential HMW-PBPs deletion mutants, the shape of pbp2b mutant cells became longer (Fig. 8). The apparent reduction of intracellular 5-ALA concentration may be due to the elongation of cells, which may result in reduced ALAS inhibition. Furthermore, the increased sensitivity to lysozyme in pbp2b deletion mutant demonstrated that integrity of the cell wall has been affected, which in turn results in enhanced cell wall permeability. The deletion of pbp1b in C. glutamicum ALA7(overexpression of RhtA transporter in C. glutamicum AP2) contributed to a 12.95% increase in 5-ALA production compared to C. glutamicum ALA8 (overexpression of RhtA transporter in C. glutamicum ALA6). While deletion of pbp1b in C. glutamicum AP2 caused a 29.47% increase in 5-ALA production compared to C. glutamicum ALA6. These data disclosed the significant effect of deletion of pbp1b on 5-ALA production especially when 5-ALA export was limited. The increased cell permeability in HMW-PBPs mutants may be beneficial for the production of biochemicals, especially for those involving in rigorous feedback or allosteric regulation. In the fed-batch process, C. glutamicum ALA7 produced 7.53 g/L 5-ALA with addition of glycine, however, only 2.00 g/L 5-ALA was produced when the medium was not supplemented with glycine. Therefore, metabolic engineering of glycine pathways will become an interesting target for 5-ALA production from renewable resources. This work was supported by National Program on Key Basic Research Project (2012CB725203), National Natural Science Foundation of China (NSFC21206112, NSFC-21576200, NSFC-21176182, and NSFC-21390201), National High-tech R&D Program of China (2012AA022103, 2012AA02A702), Natural Science Foundation of Tianjin (No. 15JCQNJC06000).
References Akram NA, Ashraf M. 2013. Regulation in plant stress tolerance by a potential plant growth regulator, 5-aminolevulinic acid. J Plant Growth Regul 32:663–679. Balzer GJ, Thakker C, Bennett GN, San KY. 2013. Metabolic engineering of Escherichia coli to minimize byproduct formate and improving succinate productivity through increasing NADH availability by heterologous expression of NAD(þ)-dependent formate dehydrogenase. Metab Eng 20:1–8. Becker J, Zelder O, Hafner S, Schroder H, Wittmann C. 2011. From zero to herodesign-based systems metabolic engineering of Corynebacterium glutamicum for L-lysine production. Metab Eng 13:159–168. Cornelius JF, Slotty PJ, El Khatib M, Giannakis A, Senger B, Steiger HJ. 2014. Enhancing the effect of 5-aminolevulinic acid based photodynamic therapy in human meningioma cells. Photodiagnosis Photodyn Ther 11:1–6. Fu W, Lin J, Cen P. 2007. 5-Aminolevulinate production with recombinant Escherichia coli using a rare codon optimizer host strain. Appl Microbiol Biotechnol 75:777–782. Fu W, Lin J, Cen P. 2010. Expression of a hemA gene from Agrobacterium radiobacter in a rare codon optimizing Escherichia coli for improving 5-aminolevulinate production. Appl Biochem Biotech 160:456–466. Jakoby M, Ngouoto-Nkili C-E, Burkovski A. 1999. Construction and application of new Corynebacterium glutamicum vectors. Biotechnol Tech 13:437–441.
1292
Biotechnology and Bioengineering, Vol. 113, No. 6, June, 2016
Jørgensen MG, Nielsen JS, Boysen A, Franch T, Møller-Jensen J, Valentin-Hansen P. 2012. Small regulatory RNAs control the multi-cellular adhesive lifestyle of Escherichia coli. Mol Microbiol 84:36–50. Kang Z, Wang Y, Gu P, Wang Q, Qi Q. 2011a. Engineering Escherichia coli for efficient production of 5-aminolevulinic acid from glucose. Metab Eng 13:492–498. Kang Z, Wang Y, Wang Q, Qi Q. 2011b. Metabolic engineering to improve 5-aminolevulinic acid production. Bioeng Bugs 2:342–345. Kang Z, Zhang J, Zhou J, Qi Q, Du G, Chen J. 2012. Recent advances in microbial production of delta-aminolevulinic acid and vitamin B12. Biotechnol Adv 30:1533–1542. Kim J, Fukuda H, Hirasawa T, Nagahisa K, Nagai K, Wachi M, Shimizu H. 2010. Requirement of de novo synthesis of the OdhI protein in penicillin-induced glutamate production by Corynebacterium glutamicum. Appl Microbiol Biotechnol 86:911–920. Kumar A, Sarkar SK, Ghosh D, Ghosh AS. 2012. Deletion of penicillin-binding protein 1b impairs biofilm formation and motility in Escherichia coli. Res Microbiol 163:254–257. Laneelle M-A, Tropis M, Daffe M. 2013. Current knowledge on mycolic acids in Corynebacterium glutamicum and their relevance for biotechnological processes. Appl Microbiol Biotechnol 97:9923–9930. Layer G, Reichelt J, Jahn D, Heinz DW. 2010. Structure and function of enzymes in heme biosynthesis. Protein Sci 19:1137–1161. Lee D-H, Jun W-J, Kim K-M, Shin D-H, Cho H-Y, Hong B-S. 2003. Inhibition of 5-aminolevulinic acid dehydratase in recombinant Escherichia coli using D-glucose. Enzyme Microb Tech 32:27–34. Letek M, Ordonez E, Vaquera J, Margolin W, Flardh K, Mateos LM, Gil JA. 2008. DivIVA is required for polar growth in the MreB-lacking rod-shaped actinomycete Corynebacterium glutamicum. J Bacteriol 190:3283–3292. Levican G, Katz A, de Armas M, Nunez H, Orellana O. 2007. Regulation of a glutamyl-tRNA synthetase by the heme status. Proc Natl Acad Sci USA 104: 3135–3140. Li F, Wang Y, Gong K, Wang Q, Liang Q, Qi Q. 2013. Constitutive expression of RyhB regulates the heme biosynthesis pathway and increases the 5-aminolevulinic acid accumulation in Escherichia coli. FEMS Microbiol Lett 350:209–215. Lin H, Castro NM, Bennett GN, San KY. 2006. Acetyl-CoA synthetase overexpression in Escherichia coli demonstrates more efficient acetate assimilation and lower acetate accumulation: A potential tool in metabolic engineering. Appl Microbiol Biotechnol 71:870–874. Lin J, Fu W, Cen P. 2009. Characterization of 5-aminolevulinate synthase from Agrobacterium radiobacter, screening new inhibitors for 5-aminolevulinate dehydratase from Escherichia coli and their potential use for high 5-aminolevulinate production. Bioresour Technol 100:2293–2297. Litsanov B, Kabus A, Brocker M, Bott M. 2012. Efficient aerobic succinate production from glucose in minimal medium with Corynebacterium glutamicum. Microb Biotechnol 5:116–128. Liu S, Zhang G, Li X, Zhang J. 2014. Microbial production and applications of 5-aminolevulinic acid. Appl Microbiol Biotechnol 98:7349–7357. Ma W, Wang X, Mao Y, Wang Z, Chen T, Zhao X. 2015. Development of a markerless gene replacement system in Corynebacterium glutamicum using upp as a counter-selection marker. Biotechnol Lett 37:609–617. Meiswinkel TM, Rittmann D, Lindner SN, Wendisch VF. 2013. Crude glycerol-based production of amino acids and putrescine by Corynebacterium glutamicum. Bioresource Technol 145:254–258. Petersen S, de Graaf AA, Eggeling L, Mollney M, Wiechert W, Sahm H. 2000. In vivo quantification of parallel and bidirectional fluxes in the anaplerosis of Corynebacterium glutamicum. J Biol Chem 275:35932–35941. Sasaki K, Watanabe M, Tanaka T, Tanaka T. 2002. Biosynthesis, biotechnological production and applications of 5-aminolevulinic acid. Appl Microbiol Biotechnol 58:23–29. Sieger B, Schubert K, Donovan C, Bramkamp M. 2013. The lipid II flippase RodA determines morphology and growth in Corynebacterium glutamicum. Mol Microbiol 90:966–982. Song CW, Kim DI, Choi S, Jang JW, Lee SY. 2013. Metabolic engineering of Escherichia coli for the production of fumaric acid. Biotechnol Bioeng 110: 2025–2034. Unthan S, Gr€unberger A, Van Ooyen J, G€atgens J, Heinrich J, Paczia N, Wiechert W, Kohlheyer D, Noack S. 2014. Beyond growth rate 0.6: What drives Corynebacterium glutamicum to higher growth rates in defined medium. Biotechnol Bioeng 111:359–371.
Valbuena N, Letek M, Ordonez E, Ayala J, Daniel RA, Gil JA, Mateos LM. 2007. Characterization of HMW-PBPs from the rod-shaped actinomycete Corynebacterium glutamicum: Peptidoglycan synthesis in cells lacking actin-like cytoskeletal structures. Mol Microbiol 66:643–657. Valbuena N, Letek M, Ramos A, Ayala J, Nakunst D, Kalinowski J, Mateos LM, Gil JA. 2006. Morphological changes and proteome response of Corynebacterium glutamicum to a partial depletion of FtsI. Microbiology 152:2491–2503. van der Werf MJ, Zeikus JG. 1996. 5-Aminolevulinate production by Escherichia coli containing the Rhodobacter sphaeroides hemA gene. Appl Environ Microbiol 62:3560–3566. Vogt M, Haas S, Klaffl S, Polen T, Eggeling L, van Ooyen J, Bott M. 2014. Pushing product formation to its limit: Metabolic engineering of Corynebacterium glutamicum for L-leucine overproduction. Metab Eng 22:40–52. Yang J, Zhu L, Fu W, Lin Y, Lin J, Cen P. 2013. Improved 5-aminolevulinic acid production with recombinant Escherichia coli by a short-term dissolved oxygen shock in fed-batch fermentation. Chin J Chem Eng 21:1291–1295. Yang J, Zhu Y, Li J, Men Y, Sun Y, Ma Y. 2015. Biosynthesis of rare ketoses through constructing a recombination pathway in an engineered Corynebacterium glutamicum. Biotechnol Bioeng 112:168–180.
Zhang J, Kang Z, Chen J, Du G. 2015. Optimization of the heme biosynthesis pathway for the production of 5-aminolevulinic acid in Escherichia coli. Sci Rep 5:8584. Zhang Y, Praszkier J, Hodgson A, Pittard A. 1994. Molecular analysis and characterization of a broad-host-range plasmid, p EP2. J Bacteriol 176: 5718–5728. Zhu N, Xia H, Wang Z, Zhao X, Chen T. 2013. Engineering of acetate recycling and citrate synthase to improve aerobic succinate production in Corynebacterium glutamicum. PLoS ONE 8:e60659. Zhu N, Xia H, Yang J, Zhao X, Chen T. 2014. Improved succinate production in Corynebacterium glutamicum by engineering glyoxylate pathway and succinate export system. Biotechnol Lett 36:553–560.
Supporting Information Additional supporting information may be found in the online version of this article at the publisher’s web-site.
Feng et al.: Production of 5-aminolevulinic acid
1293
