From zero to hero - production of bio-based nylon from renewable resources using engineered Corynebacterium glutamicum.

Metabolic Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Q1 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 Q3 58 59 60 61 62Q10 63 64 65 66

Contents lists available at ScienceDirect

Metabolic Engineering journal homepage: www.elsevier.com/locate/ymben

From zero to hero – Production of bio-based nylon from renewable resources using engineered Corynebacterium glutamicum Stefanie Kind a, Steffi Neubauer a, Judith Becker a,b, Motonori Yamamoto c, Martin Vö lkert c, c c a,b,n Gregory von Abendroth , Oskar Zelder , Christoph Wittmann a

Institute of Biochemical Engineering, Technische Universität Braunschweig, Braunschweig, Germany Institute of Systems Biotechnology, Saarland University, Saarbrücken, Germany c BASF SE, Fine Chemicals and Biotechnology, Ludwigshafen, Germany b

art ic l e i nf o

a b s t r a c t

Article history: Received 28 January 2014 Received in revised form 3 April 2014 Accepted 5 May 2014

Polyamides are important industrial polymers. Currently, they are produced exclusively from petrochemical monomers. Herein, we report the production of a novel bio-nylon, PA5.10 through an integration of biological and chemical approaches. First, systems metabolic engineering of Corynebacterium glutamicum was used to create an effective microbial cell factory for the production of diaminopentane as the polymer building block. In this way, a hyper-producer, with a high diaminopentane yield of 41% in shake flask culture, was generated. Subsequent fed-batch production of C. glutamicum DAP-16 allowed a molar yield of 50%, a productivity of 2.2 g L 1 h 1, and a final titer of 88 g L 1. The streamlined producer accumulated diaminopentane without generating any by-products. Solvent extraction from alkalized broth and two-step distillation provided highly pure diaminopentane (99.8%), which was then directly accessible for poly-condensation. Chemical polymerization with sebacic acid, a ten-carbon dicarboxylic acid derived from castor plant oil, yielded the bio-nylon, PA5.10. In pure form and reinforced with glass fibers, the novel 100% bio-polyamide achieved an excellent melting temperature and the mechanical strength of the well-established petrochemical polymers, PA6 and PA6.6. It even outperformed the oil-based products in terms of having a 6% lower density. It thus holds high promise for applications in energy-friendly transportation. The demonstration of a novel route for generation of bio-based nylon from renewable sources opens the way to production of sustainable biopolymers with enhanced material properties and represents a milestone in industrial production. & 2014 International Metabolic Engineering Society. Published by Elsevier Inc.

Keywords: Bio-based polyamide Cadaverine Corynebacterium glutamicum Diaminopentane PA5.10 Systems metabolic engineering

1. Introduction Polyamide (PA), commonly known as nylon, is a polymer with a myriad of pharmaceutical and industrial applications. Chemically, the polymer backbone is composed of repetitive units of diamines and dicarboxylic acids that contain different numbers of carbon atoms, imparting a variety of material properties. Commercialized in the 1940s, polyamides have entered the market place on a large scale for the manufacturing of fibers for clothing or thermoplastics for carpets, cogs, car parts, tire reinforcements, and other products. Currently, the global market requires 6.6 million tons per annum, making polyamides one of the most important industrial polymers. Its good biocompatibility has further led to its implementation in medical applications, such as providing a scaffold for tissue cultures (Yoo et al., 2011), foil for orbital implants (Park et al., 2008), and bone support in arthroplasty (Edwards et al., 2011).

n Corresponding author at: Institute of Systems Biotechnology, Saarland University, Campus A1.5, 66123 Saarbrücken, Germany. Fax: þ 49 681 302 71972. E-mail address: [email protected] (C. Wittmann).

Novel developments extend the application range of polyamides to diagnostics (Cox, 2001) and cellular control devices in molecular medicine (Schmitz and Schepers, 2004). Interest in a “green” route for polyamides has arisen due to the inevitable stoichiometric wastes of classical petrochemical production routes, commonly thought to cause global warming and ozone depletion, as well as acid rain and smog (Sato et al., 1998). Compared to other industrial polymers, the production of petrochemical nylon has a markedly severe impact on global climate change, expressed as carbon dioxide equivalents released (Vink et al., 2003). Moreover, it exhibits an exceptionally high requirement for fossil energy, which is particularly unfavorable in the light of the increasing shortage and rising price of fossil resources. Alternative chemical routes, such as the direct oxidation of cyclohexene to the nylon precursor adipic acid (Sato et al., 1998), promise attractive solutions for the reduction of the carbon footprint of polyamides, but these methods still rely on fossil fuels. Without doubt, a highly promising bio-block for polyamides is 1,5-diaminopentane (cadaverine, diaminopentane), a five-carbon diamine (Qian et al., 2010). Polyamides based on diaminopentane exhibit excellent and well-known material properties (Carothers,

http://dx.doi.org/10.1016/j.ymben.2014.05.007 1096-7176/& 2014 International Metabolic Engineering Society. Published by Elsevier Inc.

Please cite this article as: Kind, S., et al., From zero to hero – Production of bio-based nylon from renewable resources using engineered Corynebacterium glutamicum. Metab. Eng. (2014), http://dx.doi.org/10.1016/j.ymben.2014.05.007i

S. Kind et al. / Metabolic Engineering ∎ (∎∎∎∎) ∎∎∎–∎∎∎

2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Q4 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Table 1 Lysine and diaminopentane-producing Corynebacterium glutamicum strains used and constructed in the present study. Strain

Modification

Reference

LYS-12

Wild-type ATCC 13032 þ nucleotide exchange resulting in the amino acid exchange T311I in lysC, encoding aspartokinase (lysCT311I); over expression of lysCT311I by replacement of the native promoter with the promoter of sod, encoding superoxide dismutase (PsodlysCT311I); overexpression of dapB, encoding dihydropicolinate reductase, under the sod promoter (PsoddapB); duplication of ddh, encoding diaminopimelate dehydrogenase (2xddh); duplication of lysA, encoding diaminopimelate decarboxylase (2xlysA); deletion of pepck, encoding PEP-carboxykinase (Δpepck); nucleotide exchange resulting in a V59A change in hom, encoding homoserine dehydrogenase (homV59A); overexpression of pycA, encoding pyruvate carboxylase, by replacement of the native promoter with the sod promoter (PsodpycA); nucleotide exchange resulting in the P458S change in pycA (PsodpycAP458S); replacement of the ATG start codon with the rare GTG in icd, encoding isocitrate dehydrogenase (icdgtg); replacement of the natural promoter of fbp by the tuf promoter, encoding elongation factor tu (Ptuffbp); replacement of the natural promoter of the tkt-operon, comprising the genes zwf and tal, encoding transaldolase, tkt, encoding transketolase, opcA, encoding a putative subunit of glucose 6-phosphate dehydrogenase, and pgl, encoding 6-phosphogluconolactonase, by replacement of the native promoter with the sod promoter (Psodtkt) LYS-12 þheterologous, genome-based expression of a codon-optimized variant of E. coli ldcC, encoding lysine decarboxylase, with replacement of the native promoter by the tuf promoter (PtufldcCopt), insertion at the bioD locus, encoding dithiobiotin synthetase DAP-13þ deletion of NCgl1469, encoding a N-acetyltransferase (ΔNCgl1469) DAP-14þ deletion of lysE, encoding the lysine exporter (ΔlysE) DAP-15þ overexpression of cg2893, encoding a major facilitator permease, by replacement of the native promoter by the sod promoter (Psod cg2893)

Becker et al. (2011)

DAP-13 DAP-14 DAP-15 DAP-16

1938). However, due to the lack of efficient petrochemical routes for production of this monomer, the corresponding polymers are not considered industrially valuable. In particular, bio-based diaminopentane would open up green routes to production of novel bio-nylons, such as PA5.10 and PA5.4, copolymerized with sebacic acid from natural castor oil (Ogunniyi, 2006) and succinic acid from microbial fermentation (Hong et al., 2004), respectively (Kind and Wittmann, 2011). Moreover, systems metabolic engineering is currently available to facilitate design and improvement of the performance of microorganisms for fermentative production of an increasing number of chemicals, materials, and fuels from low-cost renewable resources (Becker and Wittmann, 2012a, 2012b; Jang et al., 2012). This opens new possibilities for the creation of biobased polymers at the required titers, yields, and productivities. Recently, pioneering studies have enabled diaminopentane production from sugar in engineered cells of the soil bacterium Corynebacterium glutamicum (Mimitsuka et al., 2007). These synthetize the desired chemical from the natural amino acid lysine through heterologous expression of the E. coli lysine decarboxylase CadA (Mimitsuka et al., 2007) or LdcC (Kind et al., 2010a). Further rounds of systems metabolic engineering have allowed improvement of the biosynthetic capacity and supporting reactions (Kind et al., 2010a) and elimination of competing pathways generating undesired by-products (Kind et al., 2010b), and have increased product secretion (Kind et al., 2011) and extended the substrate spectrum to starch (Tateno et al., 2009) and xylose (Buschke et al., 2011). Currently, the reported product yields of the best strains are still far from economic applicability, but at least demonstrate promising proof-of-concepts (Becker and Wittmann, 2012a, 2012b). In this work, we describe a sustainable value chain, extending from renewable resources to a novel, bio-based polyamide PA5.10. The development integrated engineering of C. glutamicum at the cellular and the process level to ensure fermentative supply of the polyamide building block, diaminopentane. Thereafter, we further developed processes for the fermentation and down-stream purification of the monomer, poly-condensation into a novel bio-nylon, and further conditioning by generating enforced industrial polymers.

2. Materials and methods 2.1. Strains and plasmids The lysine producer C. glutamicum LYS-12 (Table 1) was used as parent strain (Becker et al., 2011). For genetic engineering work,

This study This study This study This study

the Escherichia coli strains DH5α and NM522 (Invitrogen, Karlsruhe, Germany) and the plasmids pClik int sacB and pTc were applied as described previously (Kind et al., 2010a). 2.2. Genetic engineering All modifications were introduced into the genome of C. glutamicum using homologous recombination and a two-step selection, using kanamycin resistance and the sacB system (Becker et al., 2005; Jäger et al., 1992). Construction, purification, and analysis of plasmid DNA, as well as transformation of E. coli and C. glutamicum were performed as described previously (Becker et al., 2011; Kind et al., 2010a). Targeted gene deletion was carried out by replacement of the coding region of the gene of interest by a shortened gene fragment. For overexpression, the strong sod and tuf promoter was inserted in front of the candidate gene. The primers used for construction and verification of the introduced genetic changes have been described previously (Becker et al., 2011; Kind et al., 2010a; Kind et al., 2010b). 2.3. Chemicals Tryptone, beef extract, yeast extract, brain heart infusion (BHI), and agar were obtained from Difco Laboratories (Detroit, MI, USA). All other chemicals were of analytical grade and were obtained from Sigma-Aldrich (Steinheim, Germany), Merck (Darmstadt, Germany), or Fluka (Buchs, Switzerland). Ultramid B27 and Ultramid A27 were supplied by BASF SE (Ludwigshafen, Germany). 2.4. Media For shake flask studies, a complex medium was applied for the first pre-culture, whereas the second pre-culture and the main culture were carried out in glucose-based minimal medium (Kind et al., 2010a). For fed-batch production in the bioreactor, the preculture was grown in shake flasks in a complex medium containing 37 g L 1 BHI and 20 g L 1 glucose. The production process started with a batch medium that contained the following amounts of substances per liter: 90 g glucose, 15 g yeast extract, 2 g citric acid, 25 g (NH4)2SO4, 1.25 g KH2PO4, 1.25 g Na2HPO4, 1.25 g MgSO4 7H2O, 70 mg FeSO4 7H2O, 30 mg ZnSO4 7H2O, 14 mg MnSO4 6H2O, 168 mg CaSO4 2H2O, 0.43 mg boric acid, 0.34 mg CoSO4, 0.42 mg CuSO4, 0.07 mg Na2MoO4, 4.5 mg biotin, 7.5 mg thiamin, 9 mg nicotinamide, 30 mg pantothenic acid, 10 mg pyridoxal hydrochloride, and 1 mL antifoam (polyoxyethylene



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Q5 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

polyoxypropylene copolymer). The feed solution contained the following amounts of substances per liter: 600 g glucose, 15 g yeast extract, 275 g (NH4)2SO4, 14 g urea, 2 g citric acid, 1.25 g KH2PO4, 1.25 g Na2HPO4, 1.25 g MgSO4 7H2O, 70 mg FeSO4 7H2O, 30 mg ZnSO4 7H2O, 14 mg MnSO4 6H2O, 168 mg CaSO4 2H2O, 0.43 mg boric acid, 0.34 mg CoSO4, 0.42 mg CuSO4, 0.07 mg Na2MoO4, 4.5 mg biotin, 7.5 mg thiamin, 9 mg nicotinamide, 30 mg pantothenic acid, 10 mg pyridoxal hydrochloride, and 5 mL antifoam agent. 2.5. Batch cultivation in shake flasks Cultivation experiments in shake flasks were performed in triplicate with two successive pre-cultures, as described previously (Kind et al., 2010a). Dissolved oxygen was measured using immobilized sensor spots that contain a fluorophore with an O2dependent luminescent decay time (Wittmann et al., 2003) and the SFR shake flask reader system (PreSens Precision Sensing GmbH, Regensburg, Germany) (Schneider et al., 2010). 2.6. Fed-batch production of diaminopentane Fermentation was performed in a 1-L DASGIP fermenter, equipped with a six-blade Rushton impeller and control software for online monitoring and control of process data (DASGIP AG, Jülich, Germany). The start volume of the fermentation was 300 mL. The medium was inoculated with cells from the preculture grown at 30 1C in 2-L baffled flasks with 200 mL medium for about 8 h. Cells from the pre-culture were harvested by centrifugation (8800g, 5 min, 4 1C), washed once with sterile 5% NaCl, re-suspended in 30 mL of 5% NaCl, and then used as inoculum. During the process, the temperature was maintained at 30 1C. The pH was monitored by an electrode (Mettler Toledo, Gießen, Germany) and kept constant by automatic addition of a 25% ammonia solution. Dissolved oxygen was determined using a pO2 electrode (Visiferm DO sensor, Hamilton, Bonaduz, Switzerland) and maintained above a saturation of 20% by increase of the stirrer speed from the initial 600 rpm up to 1600 rpm, and of the aeration rate from 9 sL h 1 up to 40 sL h 1. The composition of the exhaust gas was measured online by a quadrupole mass spectrometer (OmniStar, Pfeiffer Vacuum, Asslar, Germany). Feeding was started at a glucose concentration of 10 g L 1 and adjusted such that the glucose concentration in the fermenter remained above 10 g L 1. 2.7. Analysis of substrates and products Determination of cell concentration, as optical density, and cell dry mass was performed as described previously (Kiefer et al., 2004, Becker et al., 2009). The concentration of glucose was determined in 1:10 diluted culture supernatant using a glucose analyzer (2300 STAT Plus, Yellow Springs Instruments, Ohio, USA). For shake flask studies, HPLC was used to quantify trehalose and organic acids (Kind et al., 2010a), amino acids, as well as diamines in the culture supernatant (Kind et al., 2010a; Krö mer et al., 2005). For the quantification of substrates and products from fed-batch fermentation, the supernatant was obtained by centrifugation (10,000g, 5 min, 4 1C) and subsequent filtration (0.2-mm Minisart filter, Sartorius, Gö ttingen, Germany). The diaminopentane concentration was then determined by HPLC (LaChrome Elite, VWR Hitachi, West Chester, Pennsylvania, USA), employing an Ionospher 5C column (100 3.0 mm; Chrompack, Engstingen, Germany) as stationary phase at 40 1C and an eluent consisting of 0.6 g L 1 citrate, 4 g L 1 tartaric acid, and 1.4 g L 1 ethylenediamine in 5% methanol, and 95% water as mobile phase. Refraction index detection was used. The concentration of ammonium was determined by HPLC (Dionex ICS-2000; Dionex, Sunnyvale,

3

California, USA) with a Dionex IonPac CS16 column (250 3 mm; Dionex, Sunnyvale, California, USA) as stationary phase. Separation was performed at 0.5 mL min 1 and 40 1C, using 30 mM methansulfonic acid as mobile phase. Subsequent detection was performed using a conductivity detector. For analysis of its purity, GC measurements were performed employing a system, consisting of an HP 6890N gas chromatograph (Hewlett Packard, Paolo Alto, California, USA), a J & W scientific column (30 m 0.25 mm 0.25 mm; Agilent, Waldbronn, Germany) and an HP 6890N flame ionization detector (Agilent, Waldbronn, Germany) for detection. Helium (purity 499.999%) was used as carrier gas, at a flow rate of 1.7 mL min 1. The inlet temperature was 250 1C and the temperature at the detector was 300 1C. 2.8. Determination of diaminopentane distribution coefficients Experiments to extract diaminopentane from fermentation broth were performed in a heat-jacketed glass reactor with a turbine stirrer and four baffles. The reactor was thermostated and equipped with an internal thermometer. The alcohols for the extraction experiments, i.e. n-butanol, 2-butanol, 2-octanol, and cyclohexanol, were saturated with deionized water prior to use. For the later comparison of distribution coefficients, the fermentation broth always contained the same amount of diaminopentane. It was mixed with the corresponding alcohol, as specified below, and incubated overnight at 60 1C and 375 rpm, followed by phaseseparation and diaminopentane analysis. 2.9. Purification of diaminopentane by distillation Experiments were performed in a heat-jacketed glass reactor with a turbine stirrer and four baffles. The reactor was thermostated and equipped with an internal thermometer, a 30-cm distillation column filled with metal rings, and a Normag distillation bridge with a water-cooled condenser. Crude extracts were combined and concentrated to approximately 40 wt% diaminopentane during a first distillation step, at a vacuum of 100 mbar and an internal temperature of up to 95 1C. A second rectification step was then performed at reduced pressure of 40 mbar and an internal temperature up to 105 1C to finally yield a product purity of 4 98.5%. 2.10. Polymerization, polymer analysis and testing For poly-condensation, the obtained diaminopentane in water solution (40.1%) was mixed with sebacic acid (38.9%) and water (21.0%) in a heated reaction vessel upon stirring. The reactor was heated to an inner temperature of 240 1C and maintained at this temperature for 1 h, whereby the inner pressure was 16.4 bar. Then, the excess pressure was released within 1 h. The temperature was raised to 271 1C during this step. Subsequently, the reaction mixture was maintained under a nitrogen stream for another 70 min to support the post-condensation reaction and to increase the molecular weight. The polymer melt was then released through a die plate into water for pelletization. The viscosity number (VZ) was determined using an Ubbelohde viscosimeter with a DIN II column according to ISO1628-1 (Deutsches Institut für Normung, 2011). The measurements were performed at 25 1C with sulfuric acid (96%) as solvent, and with a concentration of the polymer of 0.5 g/100 g. Melting point (Tm) and glass transition (Tg) temperatures of polyamides were measured by differential scanning calorimetry (DSC) according to DIN53765 (Deutsches Institut für Normung, 1994). Water uptake of the polymer was measured according to ISO1110 (Deutsches Institut für Normung, 1998). Haze and transmission were measured according to ASTM D1003 (ASTM International, 1995).



4

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

For reinforcement of the bio-nylon, the PA5.10 pellets were compounded with 30%, glass-fibers (OCF1110DS, 10-mm, Owens Corning, Toledo, OH, USA) in a twin-screw extruder at 200 rpm (Coperion Wþ P, Stuttgart, Germany). The temperature of the extruder was kept at 260 1C. After pelletizing, the sample was dried at 80 1C under vacuum for 16 h. In order to measure mechanical properties of the obtained polymer, test specimens with a thickness of 1.5 mm were produced by injection molding at 280 1C and 100 rpm (Arburg, Loßburg, Germany). The temperature of the mold was 80 1C. Tensile testing and impact testing of the test specimen, produced by injection molding, were performed according to ISO527-2 (Deutsches Institut für Normung, 2012) and ISO179 (Deutsches Institut für Normung, 2010; Deutsches Institut für Normung, 1997), respectively. 2.11. Elementary flux mode analysis The stoichiometric network of C. glutamicum included the reactions for lysine production, which were taken from the genome scale model (Kjeldsen and Nielsen, 2009) and expanded by the lysine decarboxylase reaction, responsible for the generation of diaminopentane. The biomass composition was taken from previous reports in the literature (Marx et al., 1996). The reactions of the carbon core metabolism of C. glutamicum and their stoichiometry were described previously (Buschke et al., 2013).

3. Results 3.1. Expression of lysine decarboxylase in the lysine hyper-producer C. glutamicum LYS-12 provides diaminopentane as major product The recently created strain C. glutamicum LYS-12 fortuitously possesses an engineered core metabolism favoring production of lysine, the precursor of the desired diamine (Becker et al., 2011). Marked by twelve genomic traits (Table 1), LYS-12 cells secreted large amounts of lysine using glucose (Table 2). In the first step, this strain was converted into a producer of diaminopentane by genome-based overexpression of codon-optimized lysine decarboxylase from E. coli. The relevant gene, ldcC, which allows addition of the missing terminal reaction in the pathway of producing diaminopentane from lysine, was integrated into the bioD locus. The resulting strain, designated C. glutamicum DAP-13, was tested for growth and production performance with glucose as sole carbon source (Fig. 1A). The strain grew exponentially and accumulated diaminopentane from early on. On a molar basis, about 31% of the sugar was directed to product biosynthesis (Table 2). In addition, N-acetyl-diaminopentane was secreted as the major by-product. However, no lysine production was observed, indicating that conversion of lysine to diaminopentane by the heterologous lysine decarboxylase proceeded to completion under the tested cultivation conditions.

3.2. Deletion of the N-acetyltransferase NCgl1469 eliminates N-acetyl-diaminopentane as byproduct The undesired accumulation of the acetylated product by C. glutamicum DAP-13 indicated the need for deletion of the responsible N-acetyltransferase. Accordingly, the acetyltransferase gene NCgl1469 (614 bp), recently discovered as the responsible enzyme (Kind et al., 2010b), was replaced by a non-functional fragment of 250 bp. The deletion was verified by PCR. The analysis revealed a 1113-bp fragment, as expected for the inactivated gene, whereas the wild type gene would have involved an extended 1476-bp fragment. In the resulting strain, C. glutamicum DAP-14, N-acetyl-diaminopentane formation was completely abolished (Fig. 1B). Consequently, the molar product yield increased to 37% (Table 2). Comparison with the parent strain revealed that the additional diaminopentane generated after elimination of the acetyltransferase was fully secreted into the growth medium. The specific rates of growth and the glucose uptake rate, as well as the biomass yield, remained unaffected.

3.3. Metabolic engineering of the exporters for lysine and diaminopentane increases product yield We found that none of the producing strains secreted lysine when grown on glucose. However, this did not guarantee efficient product formation without lysine overflow, particularly not in the production process incorporating extended feed phases, changing nutrient conditions, and introduction of potential limitations (Kind et al., 2011). In order to prevent such potential loss of carbon, the lysine exporter gene (lysE) was deleted by replacing the native gene with a shortened DNA fragment, lacking 575 bp of the coding region. The deletion resulted in a PCR fragment of 893 bp, whereas the wild type gene was represented by a 1468 bp fragment. This modification did not affect the production characteristics, at least under the tested conditions involving minimal medium (Fig. 1C, Table 2). The maintecance of production capacity in DAP-15 confirmed that diaminopentane secretion in C. glutamicum was independent of lysE expression. Subsequently, cellular transport was enhanced by overexpression of the major facilitator permease Cg2893 (Kind et al., 2011). For this purpose, the natural promoter of the permease gene was by replaced with a strong sod promoter. The modification was verified by PCR, resulting in an expected 192-bp longer PCR product. This modification led to further improvement of the production properties (Fig. 1D). In shake flask culture, C. glutamicum DAP-16 reached a final diaminopentane yield of 17.5 mM, corresponding to a molar yield of 41% (Table 2). In line, the specific diaminopentane production rate was increased to 1.7 mmol g 1 h 1, whereas specific rates of growth and glucose uptake remained stable. N-acetyl-diaminopentane and lysine production were not observed.

Table 2 Growth and production characteristics of lysine and diaminopentane-producing Corynebacterium glutamicum strains, grown in minimal medium with glucose as sole carbon source. The given data, representing mean values from three biological replicates with corresponding standard deviations, comprise the specific growth rate (m), the yields for diaminopentane (YDap/Glc), N-acetyl-diaminopentane (YN-Ace-Dap/Glc), lysine (YLys/Glc), and biomass (YX/Glc), the specific glucose uptake rate (qGlc), and the specific production rates for diaminopentane (qDap) and lysine (qLys). The yields reflect the slopes of the linear relation between substrate consumption and diaminopentane/biomass formation. The data for LYS-12 were taken from Becker et al. (2011). Strain

m [h 1]

YDap/Glc [mmol mol 1]

YN-Ace-Dap/Glc [mmol mol 1]

YLys/Glc [mmol mol 1]

YX/Glc [g mol 1]

qGlc [mmol g 1 h 1]

qDap [mmol g 1 h 1]

qLys [mmol g 1 h 1]

LYS-12 DAP-13 DAP-14 DAP-15 DAP-16

0.32 0.22 7 0.00 0.23 7 0.00 0.217 0.00 0.217 0.01

o0.1 314 75 3687 6 3647 8 405 7 5

o 0.1 507 1 o 0.1 o 0.1 o 0.1

258 o 0.1 o 0.1 o 0.1 o 0.1

64.0 52.0 7 1.1 51.17 0.2 50.0 7 1.6 51.4 7 0.7

5.0 4.3 7 0.1 4.4 7 0.0 4.3 7 0.2 4.3 7 0.1

– 1.3 7 0.0 1.6 7 0.0 1.6 7 0.1 1.7 7 0.1

1.3 – – – –



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

5

Fig. 1. Cultivation profiles of diaminopentane-producing Corynebacterium glutamicum strains DAP-13 (A), DAP-14 (B), DAP-15 (C), and DAP-16 (D). The cultures were grown in shake flasks on glucose as sole carbon and energy source. In all cultures, the dissolved oxygen level remained above 20%, which ensured sufficient oxygen supply to the cells. The data represent mean values from three biological replicates with corresponding standard deviations. The cultivation time was 9 h for strains DAP-13, DAP-14, and DAP-15, and 12 h for strain DAP-16, respectively.

3.4. The designed producer C. glutamicum DAP-16 accumulates 88 g L 1 diaminopentane under industrial fermentation conditions The production performance of the final strain, C. glutamicum DAP-16, was evaluated in a fed-batch process on an industrial glucose medium. During the initial batch phase, the strain grew unlimited with a specific growth rate of 0.16 h 1 (Fig. 2A). Diaminopentane was secreted from the start of cultivation. Within 12 h, the cells consumed the initially supplied glucose (90 g L 1) and reached a high cell concentration of OD660 ¼ 120. Upon depletion of glucose, the feed phase was initiated. The feed rate was adjusted such that the glucose level in the medium remained at about 10 g L 1. The diaminopentane concentration continuously increased to a noteworthy final titer of 88 g L 1 within 50 h. While the greatest increase in product was achieved during the feed phase, growth showed a different behavior. The cell concentration peaked at the beginning of the feed phase. This coincided with the time point of maximum productivity (2.2 g L h 1). Subsequently, the concentration of cells slightly decreased until the end of the process. Trehalose accumulated as by-product to a final value of 10.0 g L 1. The levels of other by-products, such as lactate, acetate, pyruvate, succinate, and glycerol remained below 1 g L 1 throughout the process. The diaminopentane yield differed between the batch and the feed phase (Fig. 2B). In particular, during the feed phase, C. glutamicum DAP-16 was most efficient and exhibited a high molar yield of 50%, equal to a yields in grams of 0.29 g (diaminopentane) (g glucose) 1 (Fig. 2B). However, also during the initial batch phase, cells converted the substrate into substantial amounts of diaminopentane. The enhanced yield during the feed phase was provided by appropriate composition of the

feed solution. In comparison to the batch medium, the ratio of glucose to other nutrients was almost seven-fold higher in the feed medium. This intentionally reduced growth and increased product formation. There seemed no specific causal relationship between initial high glucose concentration and strain performance, indicated by the fact that the product yield was constant during the entire batch phase, despite the glucose level decreased. Taken together, the cultivation profile of the fed-batch process was as desired: an initial batch phase as major growth phase for accumulation of high biomass concentration and feeding phase with reduced growth and enhanced production. It should be added that efficient product formation relied on sufficient ammonium supply. The feed concentration for the given fermentation (275 g L 1 ammonium) provided a rather constant ammonium level between 10 and 13 g L 1 until the end of the fermentation. In contrast, preliminary studies with less concentrated feed (150 g L 1 ammonium) resulted in limited ammonium availability in the culture medium and reduced diaminopentane production of only 59 g L 1 (data not shown). The next steps focused on purification of diaminopentane from the culture broth to provide the monomer in sufficient quality for the polycondensation into bio-nylon. 3.5. Extraction with n-butanol and two-step distillation provides polymer-grade diaminopentane First tests aimed at the identification of a suitable solvent for extracting diaminopentane from the aqueous fermentation broth. The most important criteria for selection of an appropriate solvent were (i) a miscibility gap in the broth, i.e. low solubility in water, (ii) sufficient polarity to allow for high solubility of diaminopentane,


6

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66


Fig. 2. Fermentative production of the polyamide building block diaminopentane by engineered Corynebacterium glutamicum DAP-16. The fed-batch process involved glucose-based medium during the initial batch phase and a subsequent feeding phase. The data comprise the cultivation profile (A) and the resulting yield of diaminopentane for the batch and the feed-phase (B).

i.e. high relative polarity, and (iii) stability against alkaline conditions that were needed to obtain uncharged diaminopentane by extraction. The comparison of different potential solvents, based on their physic-chemical properties, indicated that e.g. chloroform, ethyl acetate, and benzene were too unpolar to be used, but longer chain alcohols are immiscible with water, but have high polarity and may be potential extraction solvents (Fig. 3A). Accordingly, the distribution coefficients for diaminopentane between solvent and broth were compared for n-butanol, 2-butanol, 2-octanol, and cyclohexanol in a series of five consecutive extractions at 60 1C (Fig. 3B). The distribution coefficients of 2-octanol were markedly low, indicating that this solvent was too unpolar to be practical for use in extraction. The other alcohols were equally suitable for extraction. An exception was the first extraction step with nbutanol. Here, only a small organic layer separated, which was found to contain high levels of the antifoam agent. The aqueous layer was actually diluted by absorbing some water from the presaturated alcohol. The subsequent extractions, however, exhibited a similar behavior to the other two alcohols. Closer inspection revealed that, although extraction performance was similar, nbutanol indeed was most promising. Among all solvents, 2butanol showed the highest water levels dissolved in the organic phase, e.g. as much as 71%, after the first extraction step, which was suboptimal with regard to subsequent distillation. Cyclohexanol

showed good extraction properties, but its boiling point of 161 1C is relatively close to that of the product (179 1C), which also would have complicated subsequent distillation of the product. Thus, nbutanol was selected and the pH-dependence of the distribution was investigated in order to assess the feasibility of extraction from a less alkalized fermentation broth to potentially reduce costs. Five consecutive extractions at 60 1C were compared for broths at different pH values (Fig. 3C). However, the phase transfer of the product into the organic layer was reduced at lower pH, due to increasing protonation of the diaminopentane molecule. At a pH10, a transfer of diaminopentane into the organic layer could no longer be measured. Subsequent purification of diaminopentane was performed by a two-step distillation process. A first crude distillation served to reduce volumes by removing most of the water and n-butanol to yield a diaminopentane concentration of about 40%. The second rectification resulted in material of 99.6% to 99.9% purity, as determined by GC. 3.6. Bio-nylon PA5.10 reveals excellent material properties as compared to those of the petrochemical polyamides PA6 and PA6.6 The isolated diaminopentane was successfully applied in synthesis of bio-based PA5.10 by polymerization with sebacic acid using a standard polycondensation procedure. Subsequently, the material



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

7

Fig. 3. Down-stream purification of diaminopentane from the fermentation broth by solvent extraction. Water solubility and relative polarity index, as selection criteria for appropriate solvents (A); the distribution coefficients for diaminopentane in a series of five consecutive extractions, comparing results when using n-butanol, 2-butanol, 2-octanol, and cyclohexanol as solvent (B); investigation of the pH-dependence of the distribution coefficient for n-butanol (C). The values for water solubility and polarity were extracted from previous work (Reichardt, 2003). Hereby, the values for relative polarity are normalized to water (relative polarity ¼ 1).

properties of the synthesized bio-nylon were examined and compared to the commercial polymers, Ultramid B27 (PA6) and Ultramid A27 (PA6.6). The bio-based material had a high melting point of 215 1C, which was similar to that of PA6 (Table 4). In terms of other important characteristics, the bio-nylon performed even better than did the conventional, petroleum-based nylons. This included lower water absorption and density. Polymers of PA5.10, with a thickness of 1.5 mm, produced by injection molding, were translucent due to their lower haze, which allows applications beyond those of the conventional polymers.

Table 3 Material properties of petroleum-based polyamide PA6 and PA6.6, and bio-based polyamide PA5.10. Gray color indicates superior material properties.

Bio-based content [%] Viscosity number [mL g 1] Melting point [1C] Glass transition temperature [1C] Density [g cm 3] Water absorption [%] Haze

PA6

PA66

PA5.10

0 150 220 54 1.14 3.0 102

0 150 260 60 1.14 2.8 102

100 141 215 50 1.07 1.8 64

3.7. Reinforcement of the bio-nylon PA5.10 with glass fibers increases mechanical stability

4. Discussion

For automotive or electronic applications, polyamides are typically used as glass fiber (GF) reinforced compounds. Therefore, the mechanical properties of GF-reinforced PA5.10, PA6, and PA6.6 were compared (Table 4). For each polymer, the GF content was 30%. The tensile strength, which measures the force required to pull material to the point where it breaks, and the E-modulus of PA5.10 were slightly lower than those of PA6 and PA6.6, indicating that the bio-based variant was softer than the more rigid PA6 and PA6.6. The lower viscosity number indicated a smaller molecular weight of the bio-based polymer (ISO1628-1, Deutsches Institut für Normung, 2011). The elongation at break was comparable for all materials. Subsequent tests investigated the ability of the fibers to absorb shock and impact energy without breaking, i.e., the impact strength. As this property is dependent upon the temperature and the shape of the test specimen, the analysis involved two different temperatures and different test specimens (Table 4). In general, all analyzed materials became more brittle at lower temperatures. PA6 was somewhat more sensitive to temperature changes, as judged from the higher variation in response to the temperature changes (Table 4) The impact strength of PA5.10 reached a value between those of PA6 and PA6.6. The heat distortion temperature, i.e., the temperature at which a polymer sample deforms under a specified load, was highest for PA6.6, due to the much higher melting point of this polymer (Table 3) and lowest for PA5.10. A better visual surface quality was observed for glass fiber reinforced PA5.10, as compared to that of PA6 and PA6.6.

In this study, we have demonstrated a novel route to produce bio-based nylon through integration of biological and chemical methodologies. The central aspect of the study led to fermentative production of the bio-nylon monomer, diaminopentane, in a tailor-made C. glutamicum cell factory. The present study is the first report of de novo generation of an industrially competitive producer for a bio-nylon building block, an achievement that has not been considered feasible previously. It underlines the power of systems metabolic engineering, not only in complementing traditional strategies of strain optimization, but as an autonomous strategy for the creation of tailor-made production strains. The diaminopentane titer reached (88 g L 1) is about ten-fold higher than previous levels attained in other glucose-grown strains of C. glutamicum and E. coli, and the molar yield is almost doubled to 50% (Mimitsuka et al., 2007; Tateno et al., 2009; Qian et al., 2010). The ability to attain formerly inaccessible phenotypes therefore distinguishes this study from previous promising, but more local, metabolic engineering approaches. Through systemswide pathway engineering (Fig. 4), diaminopentane is obtained almost without by-products, which streamlines down-stream processing to a high purity of 99.8% and, subsequently, straightforward standard poly-condensation. An additional unique feature of this work is the first successful integration of bio-based monomer production into the value chain from renewables to a truly novel industrial bio-nylon. Our PA5.10 exhibited excellent material properties and competes with the well-established polyamides PA6 and PA6.6, for instance, in terms of mechanical stability, indicating market. Future applications in energy-friendly transportation are



8

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

Table 4 Comparison of the mechanical properties of glass-fiber reinforced (30 wt%) petrochemical PA6 (Ultramid B27), PA6.6 (Ultramid A27), and bio-based PA5.10. Gray color indicates superior material properties. Property

Condition

Standard

Unit

PA6 GF30%

PA6.6 GF30%

PA5.10 GF30%

Tensile strength

23 1C dry 23 1C conditioned 50% RH 14 d 23 1C dry 23 1C conditioned 50% RH 14 d 23 1C dry 23 1C conditioned 50% RH 14 d 23 1C dry 23 1C conditioned 50% RH 14 d 30 1C dry 30 1C conditioned 50% RH 14 d 23 1C dry 23 1C conditioned 50% RH 14 d 30 1C dry 30 1C conditioned 50% RH 14 d 1.8 MPa dry

ISO ISO ISO ISO ISO ISO ISO ISO ISO ISO ISO ISO ISO ISO ISO

MPa MPa % % MPa MPa kJ m 2 kJ m 2 kJ m 2 kJ m 2 kJ m 2 kJ m 2 kJ m 2 kJ m 2 1C

179 154 3.8 4.0 9424 8667 15 18 9 9 99 100 86 87 202

188 172 3.7 3.6 9586 9499 10 11 8 8 83 86 69 69 250

155 140 3.9 4.0 8310 7927 12 13 8 8 86 87 82 82 186

Elongation at break E-modulus Impact strength, notched

Impact strength, unnotched

Heat distortion temperature

527-2 527-2 527-2 527-2 527-2 527-2 179-2/1eA(S) 179-2/1eA(S) 179-2/1eA(S) 179-2/1eA(S) 179-2/1eU 179-2/1eU 179-2/1eU 179-2/1eU 75-2:2004

even more promising, because the lower density of PA5.10 allows for a weight saving of about 6%.

4.1. The synthetic producer C. glutamicum DAP-16 enables good diaminopentane production from renewable resources The integration of production characteristics of the created production strains into the theoretical flux space, calculated by in silico elementary flux mode analysis, allowes a careful evaluation of the obtained strain performance (Fig. 5). Remarkably, the successive implementation of beneficial genetic traits enables stepwise improvement of production. Along the glucose-grown strain genealogy, the desired increase in diaminopentane yield is accompanied by reduced cellular growth (points 1–13), most likely due to a re-direction of carbon from anabolism to product formation. The best producer, C. glutamicum DAP-16, even reaches optimal production levels (Fig. 5, point 13). Clearly, these cells are tuned for maximum production at the given growth. This ideal channeling of carbon towards the target compound without loss into by-products is unique among all previous diaminopentane producers and represents a phenotype not achieved to date (Mimitsuka et al., 2007; Tateno et al., 2009; Qian et al., 2010). The performance is improved even further by targeted reduction of growth in an industrial fed-batch environment. The molar yield of 50%, achieved during the feed-phase of the production process, approaches the predicted theoretical optimum of 75%, predicted by elementary flux mode analysis. Taking into account that this upper boundary at zero growth is hardly achievable in a realistic process, the biosynthetic power of the created cell factory is remarkable. Even under conditions of high accumulation of the non-natural diaminopentane, the created producers exhibited fast growth and substrate uptake (Table 2). The overall glucose uptake rate of 4.3 mmol g 1 h 1 remained close to the level of the wild type of C. glutamicum (Krö mer et al. 2008). C. glutamicum DAP-16 maintained its high yield throughout the whole feed-phase of production (Fig. 2B). Even at the end of the process, C. glutamicum DAP-16 still accumulated diaminopentane, promising even higher titers if the fermentation were to be prolonged (Fig. 2A). Its maximum space–time yield of 2.2 g L 1 h 1 as at the level of C. glutamicum strains used in industry (Becker et al., 2011). The impact of the ammonia content in the feed solution underlines the great importance of using appropriate process settings to fully exploit the potential of the strain. It seems that the low ammonium affinity (Km ¼36 mM) of diaminopimelate dehydrogenase in the biosynthetic pathway (Misono et al., 1986) required high ammonium levels to support a high flux.

4.2. From the viewpoint of material performance, bio-based PA5.10 might replace conventional petrochemical nylons Fermentative diaminopentane was successfully polymerized with sebacic acid under standard conditions. This demonstrated that such bio-based polyamides are accessible through existing polyamide production facilities. Extensive comparison of pure and glass-fiber reinforced PA5.10 with petrochemical nylons PA6 and PA6.6 revealed attractive characteristics of the bio-based material (Tables 3 and 4). PA5.10 had a lower density, particularly relevant for the use in energy-friendly transportation. The lower water uptake of the bio-nylon, as compared to both petro-nylons, provides a better long-term dimensional stability, which is attractive in cases where the exact geometrical shapes of the polymer need to be maintained. Moreover, this feature leads to a higher glass transition temperature, as well as superior mechanical properties, visualized by the E-modulus and the tensile strength. The fact that polymers with a thickness of 1.5 mm were translucent also opened further applications, e.g. in the production of containers where the amount of liquid can be seen directly from the outside. Regarding the mechanical properties, the glass-fiber reinforced PA5.10 could compete with the conventional polyamides in all examined fields, except for the heat distortion temperature. This property, however, could be further tuned by adjustment of the glass fiber content. Therefore, from the perspective of material performance, fully bio-based PA5.10 is an alternative to petrochemicals PA6 and PA6.6. Given the fact that these polyamides are among the most important industrial polymers worldwide, this would open up a large global market for the novel bio-nylon, once PA5.10 becomes available on large-scale. Straightforward standard poly-condensation with the dicarboxylic acid sebacic acid from castor oil, the latter being a naturally-occurring, inexpensive, and environmentally friendly resource with a world production that exceeds 500,000 t per year (Ogunniyi, 2006), is likely to support this development. 4.3. Bio-based polyamide PA5.10 – a pioneering bio-based polymer with advanced material properties Savings in greenhouse gas (GHG) emissions and primary energy input and cost-efficiency are major drivers for bio-based production (Buschke et al., 2013; Meeusen, 2010,). For bio-based materials, life cycle assessment studies have suggested that one ton of bio-based material might save about 55 GJ of primary energy and 3 t CO2 equivalents of GHG, relative to conventional plastics (Weiss et al., 2012). In line with this, chemical companies



9

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 Q9 Fig. 4. Systems metabolic engineering of Corynebacterium glutamicum for diaminopentane production. The modifications reflect the sixteen genomic traits engineered into 54 C. glutamicum DAP-16. Green color indicates increased activity by overexpression, whereas red color (in combination with a dotted line and the symbol “X”) relates to gene 55 deletion, and orange color (in combination with a dashed line) denotes gene attenuation. The enzymes encoded by the corresponding genes are: zwf, glucose 6-phosphate dehydrogenase; pgl, 6-phosphogluconolactonase; tkt, transketolase; tal, transaldolase; fbp, fructose 1,6-bisphosphatase; pck, phosphoenolpyruvate carboxykinase; pycA, 56 pyruvate carboxylase; icd, isocitrate dehydrogenase; lysC, aspartokinase; hom, homoserine dehydrogenase; dapB dihydrodipicolinate reductase; ddh, diaminopimelate 57 dehydrogenase; lysA, diaminopimelate decarboxylase; lysE, lysine exporter; and ldcC, lysine decarboxylase from E. coli, NCgl1469, N-acetyltransferase; and cg2893, major 58 facilitator permease. More specifically, the genomic changes comprise overexpression of the tkt operon (Psod tkt), overexpression of fructose 1,6-bisphosphatase (Ptuf fbp), 59 modification and amplification of pycA (Psod pycAP458S), deletion of pck (Δpck), attenuation of icd (icdatt), modification, and amplification of lysC (Psod lysCT311I), attenuation of hom (homV59A), amplification of dapB (Psod dapB), duplication of ddh (2xddh), duplication of lysA (2xlysA), overexpression of a codon-optimized variant of E. coli ldcC (Ptuf 60 ldcCopt), deletion of NCgl1469 (ΔNCgl1469), deletion of lysE (ΔlysE), and overexpression of cg2893 (Psod cg2893). (For interpretation of the references to color in this figure 61 legend, the reader is referred to the web version of this article.) 62 63 64 have recently started joint ventures with agribusiness for ecoonly yielded low performance polyesters, such as polylactic 65 efficient production pipelines towards generation of bio-plastics acid (PLA) or polyhydroxy-alkanoates (PHA) (Liu et al., 2007; 66 (Meeusen, 2010). However, these previous developments have Matsumoto et al., 2011). To date, polyamides could not be derived Please cite this article as: Kind, S., et al., From zero to hero – Production of bio-based nylon from renewable resources using engineered Corynebacterium glutamicum. Metab. Eng. (2014), http://dx.doi.org/10.1016/j.ymben.2014.05.007i

10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Q2 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66


Fig. 5. Evaluation of the performance of the genealogy of tailor-made diaminopentane-producing C. glutamicum strains constructed in this study and previously (Kind et al., 2010a, Kind et al., 2010b, Kind et al., 2011) by integration into the in silico flux space created by elementary flux mode analysis (triangles). The strains displayed are C. glutamicum DAP-1 (1), DAP-2 (2), DAP-3a (3), DAP-3b (4), DAP-3c (5), DAP-4 (6), DAP-3c ΔlysE (7), DAP-3c Psod2893 (8), DAP-3c Δ2894 (9), DAP-13 (10), DAP-14 (11), DAP-15 (12), and DAP-16 (13), all investigated in shake flask cultivations in minimal medium with 10 g L 1 glucose as sole carbon source. In addition, production properties for C. glutamicum DAP-16 (14) refer to the feed phase of the fed-batch process. The underlying genetic changes for each strain are summarized in Table 1.

from renewables, despite the fact that their petrochemical counterparts have a severe impact on global climate, require high amounts of fossil energy, and inevitably generate large amounts of toxic wastes (Sato et al., 1998, Vink et al., 2003). In this regard, we have demonstrated a milestone for the bio-based industry, which will hopefully stimulate and strengthen further development towards a post-petroleum society.

Uncited references Carole et al. (2004), Frost (2005), Lee et al. (2002), McKinlay et al. (2005), Okano et al. (2010), Raemakers-Franken et al. (2007), Sahm et al. (2000), Scholten and Dägele (2008), Scott et al. (2010), Stellmacher et al. (2010) and Yokota and Lindley (2005).

Acknowledgments The study was supported by the BMBF grant “Bio-based Polyamides through Fermentation” (No. 0315239A) within the initiative Bioindustry21. References ASTM International, 1995. Standard test method for haze and luminous transmittance of transparent plastics. 83.140.99(ASTM D1003). Becker, J., Wittmann, C., 2012a. Bio-based production of chemicals, materials and fuels – Corynebacterium glutamicum as versatile cell factory. Curr. Opin. Biotechnol. 223, 631–640. Becker, J., Wittmann, C., 2012b. Systems and synthetic metabolic engineering for amino acid production – the heartbeat of industrial strain development. Curr. Opin. Biotechnol. 23, 718–726.

Becker, J., Klopprogge, C., Schrö der, H., Wittmann, C., 2009. Metabolic engineering of the tricarboxylic acid cycle for improved lysine production by Corynebacterium glutamicum. Appl. Environ. Microbiol. 75, 7866–7869. Becker, J., Klopprogge, C., Zelder, O., Heinzle, E., Wittmann, C., 2005. Amplified expression of fructose 1,6-bisphosphatase in Corynebacterium glutamicum increases in vivo flux through the pentose phosphate pathway and lysine production on different carbon sources. Appl. Environ. Microbiol. 71, 8587–8596. Becker, J., Zelder, O., Häfner, S., Schrö der, H., Wittmann, C., 2011. From zero to hero design-based systems metabolic engineering of Corynebacterium glutamicum for L-lysine production. Metab. Eng. 13 (2), 159–168. Buschke, N., Schrö der, H., Wittmann, C., 2011. Metabolic engineering of Corynebacterium glutamicum for production of 1,5-diaminopentane from hemicellulose. Biotechnol. J. 6, 306–317. Buschke, N., Schäfer, R., Becker, J., Wittmann, C., 2013. Metabolic engineering of industrial platform microorganisms for biorefinery applications – optimization of substrate spectrum and process robustness by rational and evolutive strategies. Bioresour. Technol. 135, 544–554. Carole, T.M., Pellegrino, J., Paster, M.D., 2004. Opportunities in the industrial biobased products industry. Appl. Biochem. Biotechnol. 113–116, 871–885. Carothers, W.H., 1938. Linear polyamides and their production. US Patent Office, Pat 2, 130. Cox, J.M., 2001. Applications of nylon membrane arrays to gene expression analysis. J. Immunol. Methods 250, 3–13. Deutsches Institut für Normung, 1994. Prüfung von Kunststoffen und Elastomeren; Thermische Analyse; Dynamische Differenzkalorimetrie (DDK)(DIN 53765). Deutsches Institut für Normung, 1997. Plastics – Determination of Charpy impact properties – Part 2: Instrumented impact test (DIN EN ISO 179-2). Deutsches Institut für Normung, 1998. Plastics – Polyamides – Accelerated conditioning of test specimens (DIN EN ISO 1110). Deutsches Institut für Normung, 2010. Plastics – Determination of Charpy impact properties – Part 1: Non-instrumented impact test (DIN EN ISO 179-1). Deutsches Institut für Normung, 2011. Plastics – Determination of the viscosity of polymers in dilute solution using capillary viscometer – Part 1: General principles (DIN EN ISO 1628-1). Deutsches Institut für Normung, 2012. Plastics – Determination of tensile properties – Part 2: Test conditions for moulding and extrusion plastics (DIN EN ISO 527-2). Frost J., 2005. Synthesis of caprolactam from lysine. US Patent Office, Pat 123669. Hong, S.H., Kim, J.S., Lee, S.Y., In, Y.H., Choi, S.S., Rih, J.K., Kim, C.H., Jeong, H., Hur, C. G., Kim, J.J., 2004. The genome sequence of the capnophilic rumen bacterium Mannheimia succiniciproducens. Nat. Biotechnol. 22, 1275–1281. Jäger, W., Schäfer, A., Pühler, A., Labes, G., Wohlleben, W., 1992. Expression of the Bacillus subtilis sacB gene leads to sucrose sensitivity in the gram-positive bacterium Corynebacterium glutamicum but not in Streptomyces lividans. J. Bacteriol. 174, 5462–5465. Jang, Y.-S., Kim, B., Shin, J.H., Choi, Y.J., Choi, S., Song, C.W., Lee, J., Park, H.G., Lee, S.Y., 2012. Bio-based production of C2–C6 platform chemicals. Biotechnol. Bioeng. 109, 2437–2459. Kiefer, P., Heinzle, E., Zelder, O., Wittmann, C., 2004. Comparative metabolic flux analysis of lysine-producing Corynebacterium glutamicum cultured on glucose or fructose. Appl. Environ. Microbiol. 70, 229–239. Kind, S., Wittmann, C., 2011. Bio-based production of the platform chemical 1,5diaminopentane. Appl. Microbiol. Biotechnol. 91, 1287–1296. Kind, S., Jeong, W.K., Schrö der, H., Wittmann, C., 2010a. Systems-wide metabolic pathway engineering in Corynebacterium glutamicum for bio-based production of diaminopentane. Metab. Eng. 12, 341–351. Kind, S., Jeong, W.K., Schrö der, H., Zelder, O., Wittmann, C., 2010b. Identification and elimination of the competing N-acetyldiaminopentane pathway for improved production of diaminopentane by Corynebacterium glutamicum. Appl. Environ. Microbiol. 76, 5175–5180. Kind, S., Kreye, S., Wittmann, C., 2011. Metabolic engineering of cellular transport for overproduction of the platform chemical 1,5-diaminopentane in Corynebacterium glutamicum. Metab. Eng. 13, 617–627. Kjeldsen, K.R., Nielsen, J., 2009. In silico genome-scale reconstruction and validation of the Corynebacterium glutamicum metabolic network. Biotechnol. Bioeng. 102, 583–597. Krö mer, J.O., Bolten, C.J., Heinzle, E., Schrö der, H., Wittmann, C., 2008. Physiological response of Corynebacterium glutamicum to oxidative stress induced by deletion of the transcriptional repressor McbR. Microbiology 154, 3917–3930. Krö mer, J.O., Fritz, M., Heinzle, E., Wittmann, C., 2005. in vivo quantification of intracellular amino acids and intermediates of the methionine pathway in Corynebacterium glutamicum. Anal. Biochem. 340, 171–173. Lee, P.C., Lee, S.Y., Hong, S.H., Chang, H.N., 2002. Isolation and characterization of a new succinic acid-producing bacterium, Mannheimia succiniciproducens MBEL55E, from bovine rumen. Appl. Microbiol. Biotechnol. 58, 663–668. Liu, Q., Ouyang, S.P., Kim, J., Chen, G.Q., 2007. The impact of PHB accumulation on L-glutamate production by recombinant Corynebacterium glutamicum. J. Biotechnol. 132, 273–279. Marx, A., Graaf, A., de, Wiechert, W., Eggeling, L., Sahm, H., 1996. Determination of the fluxes in the central metabolism of Corynebacterium glutamicum by Nuclear Magnetic Resonance Spectroscopy combined with metabolite balancing. Biotechnol. Bioeng. 49, 111–129. Matsumoto, K., Kitagawa, K., Jo, S.J., Song, Y., Taguchi, S., 2011. Production of poly(3hydroxybutyrate-co-3-hydroxyvalerate) in recombinant Corynebacterium glutamicum using propionate as a precursor. J. Biotechnol. 152, 144–146.



1 2 3 4 Q6 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

McKinlay, J.B., Zeikus, J.G., Vieille, C., 2005. Insights into Actinobacillus succinogenes fermentative metabolism in a chemically defined growth medium. Appl. Environ. Microbiol. 71, 6651–6656. Meeusen, M.J.G., 2010. Biobased production chainsThe Biobased Economy: Biofuels, Materials and Chemicals in the Post-Oil EraEarthscan, London, pp. 229–243. Mimitsuka, T., Sawai, H., Hatsu, M., Yamada, K., 2007. Metabolic engineering of Corynebacterium glutamicum for cadaverine fermentation. Biosci. Biotechnol. Biochem. 71, 2130–2135. Misono, H., Ogasawara, M., Nagasaki, S., 1986. Characterization of meso-diaminopimelate dehydrogenase from Corynebacterium glutamicum and its distribution in bacteria. Agric. Biol. Chem. 50, 2729–2734. Ogunniyi, D.S., 2006. Castor oil. A vital industrial raw material. Bioresour. Technol. 97 (9), 1086–1091. Okano, K., Tanaka, T., Ogino, C., Fukuda, H., Kondo, A., 2010. Biotechnological production of enantiomeric pure lactic acid from renewable resources: recent achievements, perspectives, and limits. Appl. Microbiol. Biotechnol. 85, 13–23. Qian, Z.G., Xia, X.X., Lee, S.Y., 2010. Metabolic engineering of Escherichia coli for the production of cadaverine. A five carbon diamine. Biotechnol. Bioeng. 108, 93–103. Raemakers-Franken P., Nossin P., Brandts P., Wubbolts M., Peeters W., Ernste S., Wildeman S. de, Schuermann M., 2007. Biochemical synthesis of 6-amino caproic acid US Patent Office, Pat 0254341. Reichardt, C., 2003. Solvents and Solvent Effects in Organic Chemistry, 3rd ed. Wiley-VCH Publishers, Weinheim. Sahm, H., Eggeling, L., Graaf, A.A., de, 2000. Pathway analysis and metabolic engineering in Corynebacterium glutamicum. Biol. Chem. 381, 899–910.

11

Schneider, K., Schütz, V., John, G.T., Heinzle, E., 2010. Optical device for parallel online measurement of dissolved oxygen and pH in shake flask cultures. Bioprocess. Biosyst. Eng. 33, 541–547. Scholten, E., Dägele, D., 2008. Succinic acid production by a newly isolated bacterium. Biotechnol. Lett. 30, 2143–2146. Scott, E.L., van Haveren, J., Sanders, J.P.M., 2010. The production of chemicals in a biobased economyThe Biobased Economy: Biofuels, Materials and Chemicals in Q7 the Post-Oil EraEarthscan, London, pp. 146–164. Stellmacher, R., Hangebrauk, J., Wittmann, C., Scholten, E., Abendroth, G.von, 2010. Fermentative manufacture of succinic acid with Basfia succiniciproducens DD1 in serum flasks. Chemie Ingenieur Technik 82, 12231229. Q8 Tateno, T., Okada, Y., Tsuchidate, T., Tanaka, T., Fukuda, H., Kondo, A., 2009. Direct production of cadaverine from soluble starch using Corynebacterium glutamicum coexpressing alpha-amylase and lysine decarboxylase. Appl. Microbiol. Biotechnol. 82, 115–121. Vink, E.T., Rábago, K.R., Glassner, D.A., Gruber, P.R., 2003. Applications of life cycle assessment to NatureWorks™ polylactide (PLA) production. Polym. Degrad. Stab 80, 403–419. Weiss, M., Haufe, J., Carus, M., Brandao, M., Bringezu, S., Hermann, B., Patel, M., 2012. A review of the environmental impacts of biobased materials. J. Indust. Ecol. 16, 169–181. Wittmann, C., Kim, H.M., John, G., Heinzle, E., 2003. Characterization and application of an optical sensor for quantification of dissolved O2 in shake-flasks. Biotechnol. Lett. 25, 377–380. Yokota, A., Lindley, N.D., 2005. Central metabolism. Sugar uptake and conversion. In: Eggeling, L., Bott, M. (Eds.), Handbook of Corynebacterium glutamicum. CRC Press, Boca Raton, pp. 215–240.


Biosynthesis of pinene from glucose using metabolically-engineered Corynebacterium glutamicum.

Recent advances in the metabolic engineering of Corynebacterium glutamicum for the production of lactate and succinate from renewable resources.

Aerobic production of succinate from arabinose by metabolically engineered Corynebacterium glutamicum.

Direct production of organic acids from starch by cell surface-engineered Corynebacterium glutamicum in anaerobic conditions.

L-citrulline production by metabolically engineered Corynebacterium glutamicum from glucose and alternative carbon sources.

Two-step production of gamma-aminobutyric acid from cassava powder using Corynebacterium glutamicum and Lactobacillus plantarum.

From ZeRO to HeRO: Saving lives one HeRO at a time.

Engineering Corynebacterium glutamicum to produce 5-aminolevulinic acid from glucose.

Production of the Marine Carotenoid Astaxanthin by Metabolically Engineered Corynebacterium glutamicum.

Ciprofloxacin triggered glutamate production by Corynebacterium glutamicum.

Engineering Corynebacterium glutamicum for violacein hyper production.

Synthesis of β-alanine from L-aspartate using L-aspartate-α-decarboxylase from Corynebacterium glutamicum.

A Facile Solid-Phase Route to Renewable Aromatic Chemicals from Biobased Furanics.

Production of hydrophobic amino acids from biobased resources: wheat gluten and rubber seed proteins.

Characterization of 3-phosphoglycerate kinase from Corynebacterium glutamicum and its impact on amino acid production.

Engineering of Corynebacterium glutamicum for growth and L-lysine and lycopene production from N-acetyl-glucosamine.

Functional and structural analyses of threonine dehydratase from Corynebacterium glutamicum.

Identification of essential tryptophan in amylomaltase from Corynebacterium glutamicum.

Fatty acid from the renewable sources: a promising feedstock for the production of biofuels and biobased chemicals.

Interaction sites of DivIVA and RodA from Corynebacterium glutamicum.

Characterization and partial purification of L-asparaginase from Corynebacterium glutamicum.

Excretion of glutamate from Corynebacterium glutamicum triggered by amine surfactants.

Nucleotide sequence of the dapA gene from Corynebacterium glutamicum.

Production of green biodegradable plastics of poly(3-hydroxybutyrate) from renewable resources of agricultural residues.