Production of glutaric acid from 5-aminovaleric acid using Escherichia coli whole cell bio-catalyst overexpressing GabTD from Bacillus subtilis

Accepted Manuscript Title: Production of glutaric acid from 5-aminovaleric acid using Escherichia coli whole cell bio-catalyst overexpressing GabTD from Bacillus subtilis Authors: Yun-Gi Hong, Yu-Mi Moon, Ju-Won Hong, So-Young No, Tae-Rim Choi, Hae-Rim Jung, Su-Yeon Yang, Shashi Kant Bhatia, Jung-Oh Ahn, Kyung-moon Park, Yung-Hun Yang PII: DOI: Reference:

S0141-0229(18)30125-X https://doi.org/10.1016/j.enzmictec.2018.07.002 EMT 9242

To appear in:

Enzyme and Microbial Technology

Received date: Revised date: Accepted date:

28-3-2018 24-6-2018 9-7-2018

Please cite this article as: Hong Y-Gi, Moon Y-Mi, Hong J-Won, No S-Young, Choi T-Rim, Jung H-Rim, Yang S-Yeon, Bhatia SK, Ahn J-Oh, Park K-moon, Yang Y-Hun, Production of glutaric acid from 5-aminovaleric acid using Escherichia coli whole cell bio-catalyst overexpressing GabTD from Bacillus subtilis, Enzyme and Microbial Technology (2018), https://doi.org/10.1016/j.enzmictec.2018.07.002 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Production of glutaric acid from 5-aminovaleric acid using Escherichia coli whole cell bio-catalyst overexpressing GabTD from Bacillus subtilis

IP T

Yun-Gi Honga, Yu-Mi Moona, Ju-Won Honga, So-Young Noa, Tae-Rim Choia, Hae-Rim Junga, Su-

a

SC R

Yeon Yanga, Shashi Kant Bhatia a,b, Jung-Oh Ahnc, Kyung-moon Parkd, Yung-Hun Yanga,b,*

Department of Biological Engineering, College of Engineering, Konkuk University, Seoul, 05029,

South Korea

Institute for Ubiquitous Information Technology and Applications (CBRU), Konkuk University,

U

b

Biotechnology Process Engineering Center, Korea Research Institute Bioscience Biotechnology

A

c

N

Seoul 05029, South Korea

d

M

(KRIBB), Gwahangno, Yuseong-Gu, Daejeon 305-806, Korea Department of Biological and Chemical Engineering, Hongik University, Sejong Ro 2639,

Author for correspondence (E-mail: [email protected])

A

CC

EP

*

TE D

Jochiwon, Sejong City, Republic of Korea

1

Highlights

The first E. coli whole cell bioconversion from 5-aminovalerate to glutaric acid

•

Finding of critical factors for GabTD reaction

•

Achievement of high bioconversion rate over 90% based on α-ketoglutarate concentration.

•

Repetitive use of whole cell biocatalyst to accumulate more glutaric acid

SC R

IP T

•

U

Abstract

N

Glutaric acid is one of the promising C5 platform compounds in the biochemical industry. It can be produced chemically, through the ring-opening of butyrolactone followed by hydrolysis. Alternatively,

A

glutaric acid can be produced via lysine degradation pathways by microorganisms. In

M

microorganisms, the overexpression of enzymes involved in this pathway from E. coli and C.

TE D

glutamicum has resulted in high accumulation of 5-aminovaleric acid. However, the conversion from 5-aminovaleric acid to glutaric acid has resulted in a relatively low conversion yield for unknown reasons. In this study, as a solution to improve the production of glutaric acid, we introduced gabTD

EP

genes from B. subtilis to E. coli for a whole cell biocatalytic approach. This approach enabled us to determine the effect of co-factors on reaction and to achieve a high conversion yield from 5-

CC

aminovaleric acid at the optimized reaction condition. Optimization of whole cell reaction by different plasmids, pH, temperature, substrate concentration, and cofactor concentration achieved full

A

conversion with 100mM of 5-aminovaleric acid to glutaric acid. Nicotinamide adenine dinucleotide phosphate (NAD(P)+) and α-ketoglutaric acid were found to be critical factors in the enhancement of

conversion in selected conditions. Whole cell reaction with a higher concentration of substrates gave 141mM of glutaric acid from 300mM 5-aminovaleric acid, 150mM α-ketoglutaric acid, and 60mM NAD+ at 30°C, with a pH of 8.5 within 24 hours (47.1% and 94.2% of conversion based on 52

aminovaleric acid and α-ketoglutaric acid, respectively). The whole cell biocatalyst was recycled 5 times with the addition of substrates; this enabled the accumulation of extra glutaric acid.

A

CC

EP

TE D

M

A

N

U

SC R

IP T

Key words: B. subtilis gabTD, whole cell conversion, optimization, Co-factor effect

3

Introduction As an alternative approach to producing platform chemicals to replace petroleum based production, biological techniques through fermentation or biotransformation from renewable sources have been extensively studied for several decades as a sustainable technology [1-5]. Consequently, many studies

IP T

have been carried out on the use of platform chemicals such as 3-hydroxypropionic acid [6], succinic acid [7], itaconic acid [8], putrescine [9], cadaverine [10], pipecolic acid [11], and 5-aminovaleric acid [12-16]. Among these, 5-aminovaleric acid (5-AVA) and glutaric acid are basic 5-carbon platform

SC R

chemicals used for the production of 5-nylon. Similar to cadaverine, they can be manufactured from L-lysine, which can be produced in quantities of up to 2.2 million tons per year [17]. 5-aminovaleric

U

acid is a precursor of 5-hydroxyvaleric acid, glutaric acid, and 1,5-pentanediol. Glutaric acid is a precursor of a plasticizer, 1,5-pentanediol, and glutaric acid itself can be involved in polymerization

N

reaction, generating polyols and polyamides [12]. The production of 5-aminovaleric acid and glutaric

A

acid from model organisms such as Corynebacterium glutamicum and Escherichia coli has been

M

reported in previous works [12, 14, 18-22]. Most of the metabolically engineered strains were

TE D

designed to utilize a 5-aminovaleric acid degradation pathway (AMV pathway), which naturally exists in Pseudomonas putida [13, 16, 23, 24].

Conversion of lysine to glutaric acid via the AMV pathway is catalyzed by four enzymes. Enzymes

EP

involved in the native glutaric acid production pathway of P. putida include lysine monooxygease (DavB), 5-aminovaleramidase (DavA), 5-aminovaleric acid aminotransferase (DavT), and glutarate

CC

semialdehyde dehydrogenase (DavD) [23, 24]. Previous studies demonstrated that 4-aminobutyrate aminotransferase (GabT) and succinate semialdehyde dehydrogenase (GabD) can also be utilized in

A

the conversion of 5-AVA and glutarate semialdehyde (GSA), respectively [14]. The conversion of lysine to 5-AVA is a relatively simple step compared to that of 5-AVA to glutaric acid. The two reactions of DavB and DavA yielding 5-AVA do not require costly co-factors, except for oxygen for oxygenation of L-lysine by DavB. Otherwise, the conversion of 5-AVA to glutaric acid by DavT and 4

DavD requires expensive co-factors such as PLP and NAD(P)+, respectively. In addition, DavT reaction requires another substrate, α-ketoglutaric acid as an amine acceptor which is conferred one amine group during the reaction from 5-AVA. (Fig. 1). The first two steps in this pathway involve davB and davA encoding of lysine monooxygenase and 5-

IP T

aminovaleramidase, respectively. This results in a high concentration of 5-AVA production of up to 773mM and 2134mM by fermentation and whole cell conversion, respectively [18, 25]. In contrast, the fermentative production of glutaric acid in engineered strains via the AMV pathway using various

SC R

aminotransferases (DavT or GabT) and diacid semialdehyde dehydrogenases (DavD or GabD)

catalyzing 5-AVA to glutaric acid have produced only low concentrations of glutaric acid, up to

U

103mM with 0.656mM/h of productivity [12, 14, 20, 26]. Although some genuine pathways have been studied to produce glutaric acid via the AMA (2-aminoadipate) pathway or a new synthetic

N

pathway using E. coli as the host strain, these pathways yielded even lower concentration of glutaric

M

AVA to glutaric acid is not clearly known.

A

acid than the AMV pathway (Table 1) [21, 22, 27]. The reason for the low conversion rate from 5-

TE D

Previous studies have suggested that the potent 5-aminovaleric acid transporter or permeases might export 5-aminovaleric acid to outside of the cell, yielding low glutaric acid production by fermentation [12, 13, 16]. Adkins et al. highlighted the importance of α-ketoglutaric acid to improve

EP

glutaric acid production, indicating that the regeneration of this co-substrate using glutamic acid dehydrogenase may lead to an improvement in productivity [12]. However, no previous studies have

CC

been carried out to determine the influence of co-factors such as pyridoxal 5′-phosphate (PLP) and NAD(P)+, which are co-factors of 5-AVA aminotransferase and glutarate semialdehyde

A

dehydrogenase, respectively. Additionally, previous studies have not identified which of the reaction

steps is a rate limiting step in 5-AVA to glutaric acid conversion reaction among the transamination of 5-AVA and dehydrogenation of glutaric acid semialdehyde. As a solution to determine the important nodes and to increase the production yield, whole cell 5

biotransformation can be considered [8, 10, 28-32]. A whole cell bioconversion system has advantages such as relatively easy control of each enzyme, supply of cofactors in a short period of time, and higher robustness to harsh environments than purified enzymes. In addition, our previous results showed that, with sufficient starting materials such as lysine and citric acid, the whole cell

IP T

system could improve yield and productivity after the optimization of reaction parameters [8, 10]. In this paper, gabTD from Bacillus subtilis was cloned into E. coli and applied to 5-AVA degradation for glutaric acid production. The E. coli whole cell biocatalyst was applied to determine

SC R

the detailed reaction conditions and optimum concentration of 5-AVA, α-ketoglutaric acid, PLP, and NAD(P)+, which are required for glutaric acid production (Fig. 1). Through this study, the effect of

U

PLP and NAD(P)+, which are co-factors of GabT and GabD, respectively, on reaction was observed. Although we could not clarify all the reasons for the low glutaric acid productivity, several ways to

M

A

N

improve glutaric acid conversion were revealed.

Chemicals

TE D

2. Materials and Methods

3-aminopropanoic acid (>99%), 5-aminovaleric acid (>97%), 6-aminocaproic acid (>99%), glutaric (>99%),

glutamic

EP

acid

acid

(>99%),

pyridoxal-5-phosphate

hydrate

(>98%),

and

diethylethoxymethylene malonate (>99%) were purchased from Sigma-Aldrich Co. (USA). 4-

CC

aminobutyric acid (>98%) and α-ketoglutaric acid (>99%) were obtained from Tokyo Chemical Industry Co. (Japan). β-nicotinamide adenine dinucleotide sodium salt (>99%), β-nicotinamide

A

adenine dinucleotide phosphate sodium salt (>99%), Tris hydrochloride (>99%), Tris-base(>99.9%), sodium borate decahydrate (>99%), iso-propyl-β-D-thiogalactopyranoside (>99%), and other medium components were obtained from Biosesang Co. (Korea). Sodium acetate trihydrate (>98.5%) and acetic acid (gracial) were purchased from Samchun Co. (Korea) and EMD Millipore Co. (Germany), 6

respectively.

Bacterial Strains and Plasmid Construction During plasmid construction, E. coli strains were grown at 37 °C in lysogeny broth (LB) (10 g l−1

IP T

tryptone, 5 g l−1 yeast extract, and 5 g l−1 sodium chloride). The medium was supplemented with 50 µg ml−1 kanamycin (Km) for selection when required. E. coli K12 MG1655 was used as a host strain for

SC R

cloning gabTD genes into expression vectors (Table. 2). PCR product of gabT and gabD genes was

amplified from Pseudomonas putida KT2440, Corynebacterium glutamicum ATCC13032, and Bacillus subtilis subsp, subtilis str. 168. Amplified genes were inserted into a pET24ma vector

U

(constructed by Dr. David Sourdive, Pasteur Institute, France) containing p15A replication origin

N

using conventional cloning technique with restriction enzymes and T4-ligase. To insert the amplified

A

BsgabTD gene into duet vectors, the Gibson assembly master mix (New England Biolabs) was used.

M

Detailed information about primers used in this study for vector construction is provided in the

TE D

supplementary table.

EP

Culture Condition and Whole Cell Preparation Constructed plasmids were transformed into E. coli BW25113 (DE3) competent cells for the

CC

preparation of whole cell biocatalyst. Cultivation was carried out at 37°C in a shaking incubator (HanBeak Science Co. Korea) at 200 rpm. Seed-cultures were prepared overnight in 5ml of LB medium

A

with kanamycin (50 µg/µl) in a 14ml round bottom tube. The precultures were inoculated into 50 ml of LB contained in a 250 ml baffled Erlenmeyer flask with kanamycin (50 µg/µl), followed by incubation at 37°C with shaking. Upon reaching an OD600 of 0.6~0.7, the cultures were induced by the addition of Isopropyl β-D-1-thiogalactopyranoside (IPTG) at 0.5mM. Cultures were then incubated at 25°C with shaking for 16 h. To prepare the cell concentrate for further whole cell reaction, cultures 7

were harvested by centrifugation at 5,000 rpm for 10 min at 4°C. The cell pellets were then washed with deionized (DI) water. After centrifuging again, the cell pellets were resuspended in DI water to produce a concentrate with an OD600 of 200. These cells were stored at -70°C.

IP T

Whole Cell Reaction Strain BW25113(DE3) pET24ma::BsgabTD whole cell concentrate was used to measure the

SC R

activity of γ-aminobutyrate aminotransferase (GabT) and succinate semialdehyde dehydrogenase (GabD). Assays were performed in a total volume of 500ul. Whole cell reaction was initiated by the

addition of proper substrates and co-factors. The cell OD600 and buffer concentration used in this study

U

were constant at 20 and 500mM, respectively. Whole cell activity tests were conducted with 100mM

N

substrates and 5mM NADP+. The prepared reaction mixture was incubated at 30°C for 24 h in a

A

shaking incubator at 200rpm. After incubation, the reaction was stalled by heating at 94°C for 3 min,

M

and the samples were then diluted to reach an appropriate concentration for use in high-

EP

Analytical Methods

TE D

performance liquid chromatography (HPLC) analysis.

Amine compounds (3-aminopropionic acid, 4-aminobutyric acid, 5-aminovaleric acid, and

CC

glutamic acid) were measured after diethyl ethoxymethylenemalonate (DEEMM) derivatization reaction by HPLC (YL-9100, Korea) at a UV-absorbance of 284nm. Separation of these derivatized

A

amine compounds was performed on a C18 column (Shimadzu, shimpack GIS, 250 4.6mm id). The column temperature was maintained at 35 °C. The mobile phase composed of acetonitrile (solvent A) and 25mM sodium acetate buffer pH 4.8 (solvent B) were supplied at 1 ml/min and the composition of solvent A to B (A:B, v/v) was changed with the following gradient program: 0 min (20:80), 2 min (25:75), 32 min (60:40), 37 min (20:80), 40 min (20:80). To analyze the organic acid compounds such 8

as α-ketoglutaric acid and glutaric acid, HPLC (Perkin Elmer) equipped with refractive index detector (RID) (Perkin Elmer) was used. Separation of organic acids was performed on an Aminex HPX-87H column (300 7.8mm id). The mobile phase in which the flow rate was constant at 0.6 ml/min was 0.008N H2SO4. The oven temperature was constant at 60°C during the operation.

IP T

Multiple sequence analysis was conducted with ClustalW and GeneDoc software. After aligning the protein sequence in FASTA format, sequence alignment was conducted by ClustalW and the result

was clarified using GeneDoc. In addition, for statistical analysis, Minitab 16 software was used. One-

SC R

way ANOVA (Tukey’s method) was used to determine the significant difference between data points

U

with 95% level of confidence.

A

N

3. Results and Discussion

M

Introduction of new gabTD from B. subtilis for whole cell reaction of glutaric acid production To determine the different sources of gabTD system other than the known strains such as P. putida

TE D

and C. glutamicum, we searched several strains through bibliographic and bioinformatic approaches and found gabTD derived from B. subtilis (BsgabTD). In nature, the gabTD operon plays a role in the utilization of glutamate as a nitrogen source correlated with glutamate decarboxylase (Gad) enzymes

EP

in many microorganisms [33, 34]. Especially for the Bacillus species, which forms spores at the end stage of its lifecycle, the GABA degradation pathway is important in forming spores. Sequence

CC

alignment results showed that GabT from B. subtilis (BsGabT) has 44% and 46% homology to DavT from P. putida KT2440 (PpDavT) and GabT from C. glutamicum ATCC 13032 (CgGabT),

A

respectively (Fig. S1). In addition, GabD from B. subtilis (BsGabD) has 52% and 48% homology to P. putida KT2440 (PpDavD) and C. glutamicum ATCC13032 (CgGabD), respectively (Fig. S2). To investigate the possibility of BsGabT/BsGabD in the production of glutaric acid (replacing previously used enzymes, PpDavTD or CgGabTD), pET24ma harboring each gene from B. subtillis, 9

P. putida, and C. glutamicum was constructed and expressed by E. coli BW25113(DE3) cells (Table. 2). Previously, the whole cell conversion of 5-AVA to glutaric acid has only been investigated with PpDavTD at low concentration (12.8mM 5-AVA) [12]. In order to more clearly confirm the capability of the BsGabT/BsGabD enzyme for glutaric acid production, a higher concentration of substrate (100mM substrates and 5mM NAD+) than that in the previous study reported about BsGabD was used

IP T

in this reaction [12]. PpDavT, PpDavD, CgGabT, CgGabD, BsGabT, and BsGabD were overexpressed separately in the host strain. Activity of the reaction was estimated by quantified

SC R

glutaric acid concentration, and the capability of BsGabT/BsGabD was accessed by comparing relative activity with PpDavT/PpDavD and CgGabT/CgGabD (Fig. 2A). Although PpDavT/PpDavD

showed the highest activity among the compared cells, CgGabT/CgGabD and BsGabT/BsGabD

U

exhibited similar activity with PpDavT/PpDavD. It is therefore believed that BsGabT/BsGabD can be

N

potent glutaric acid producing enzymes, besides PpDavTD or CgGabTD.

A

The ligand preference of the BsGabT whole cell was studied by comparing the decreased

M

concentrations of the amine substrates (Fig. 2B). Reaction was conducted with equal concentrations of

TE D

100mM of amine substrates and α-ketoglutaric acid. BsGabT showed the highest activity to 4aminobutyric acid (GABA), which is the original substrate of BsGabT. BsGabT showed poor activity on 3-aminopropionic acid and 6-aminohexanoic acid. On the other hand, BsGabT had high activity

EP

with 5-aminovaleric acid, similar to that of GABA. When 3-aminopropionic acid, 5-aminovaleric acid, and 6-aminohexanoic acid were used as an amine substrate, the relative activity based on the

CC

activity of GABA was 9.0%, 86.9%, and 28.0%, respectively. While BsGabT showed the highest

A

activity with GABA as the substrate, it showed similar activity with 5-aminovaleric acid.

Optimization of whole cell reaction condition with different copy numbers of BsgabTD, pH, and temperature

10

To improve the efficiency of whole cell reaction, BsgabTD was cloned into different vectors having different copy numbers and some variations (Table 2): pET24ma (copy numbers 10 to 12), pACYC duet-1 (10 to 12), pET duet-1 (~40), pCDF duet-1 (20 to 40) and pRSF duet-1 (more than 100) [35]. Among these vectors, the best glutaric acid production was shown when BW25113 harboring the pET24ma vector was used for whole cell reaction (Fig. 3A). pCDF duet-1 achieved 97.5% of relative

IP T

activity compared to the pET24ma vector. pRSF duet-1 showed the poorest results for the BsgabTD

whole cell reaction vector as it showed a relative activity of only 48.3%. It seems that the extremely

SC R

high copy number of the pRSF duet-1 vector negatively affected the activity of the enzyme expressed.

Although pET24ma and pACYC duet-1 have the same origin of replication and both have a low copy number, pET24ma showed a better result than the pACYC duet-1 because of its different T7 promotor

U

(constructed by Dr. David Sourdive, Pasteur Institute, France) [36]. pET24ma::BsgabTD was used for

N

further experiments.

A

The optimal pH range of purified GabT and GabD (EC 2.6.1.19 and EC 1.2.1.79, respectively) is

M

known to be in the alkaline range. However, the optimal pH in a whole cell conversion system can

TE D

differ to that of purified enzyme because enzymes are protected by the cell envelope in whole cell reaction [8, 10]. Thus, to examine the effect of pH on the BsGabTD whole cell system, various initial pH ranges were tested (Fig. 3B). pH was adjusted to the desired value using a Tris-HCl buffer and

EP

NaOH solution. The relative activity peaked at pH 7, but sharply decreased when the pH range was greater than pH 7. At pH 6 and 8, relative activity was 66.7% and 56.5%, respectively. In contrast,

CC

residual α-ketoglutaric acid decreased to 23.3mM up to pH 11, although the glutaric acid peak was not detected after pH 10. At an extreme alkalinity of pH 12, 69mM of α-ketoglutaric acid remained.

A

Residual α-ketoglutaric acid implies that the alkaline range has the optimal pH of GabT. In contrast, the greatest amount of glutaric acid production was shown at pH 7. These contradictory results imply that the optimal pH for the GabT and GabD enzymes differ. Also, the difference of residual αketoglutaric acid concentration among the pH points was not significant, except for pH 12. The fine activity of GabT at the broad pH range implies that the final conversion reaction, which coverts 11

glutarate semialdehyde to glutaric acid by GabD, determines the total conversion rate. Finally, the optimal pH was determined as pH 7, which yielded the highest conversion from α-ketoglutaric acid to glutaric acid. After selecting the optimal pH, the reaction temperature was examined. Previously, Zhong Li et al.

IP T

showed that the temperature of whole cell reaction of 30°C was optimal to produce 5-AVA using DavT from P. putida KT2440 overexpressed in E. coli BL21(DE3) [16]. This result concurs with the

whole cell reaction temperature of 30°C obtained in the current study (Fig. 3C). The highest activity

SC R

was observed at 30°C. At a temperature greater than 30°C, the decreasing rate of relative activity based on the result at 30°C was much less than that at less than 30°C. Compared to the relative

N

U

activity at 16°C, which decreased to 49.2%, the relative activity at 45°C only decreased to 92.6%.

A

Optimization of major substrates of whole cell reaction

M

Whole cell conversion rate can be affected by many factors such as inhibition by substrate,

TE D

intermediate metabolite, byproduct or product. To achieve accumulation of the final product with high concentration, determining the upper limit of the initial substrate concentration should be the first step, as other factors such as intermediates or products cannot be easily controlled at a simple batch

EP

reaction. The effect of the initial substrate concentration on the reaction was investigated by observing the scanning activity at a broad range of 5-aminovaleric acid and α-ketoglutaric acid concentration.

CC

Different concentrations of 5-aminovaleric acid from 100mM to 1000mM were added to each

reaction mixture, and α-ketoglutaric acid was fixed at 100mM (Fig. 4A). Although this reaction

A

theoretically requires two substrates with the same ratio, activity increased as the ratio of 5aminovaleric acid to α-ketoglutaric acid was increased. After 300mM of 5-aminovaleric acid, no

significant change in activity was observed. At this point, more than 98% of α-ketoglutaric acid was consumed. In addition, substrate inhibition by 5-aminovaleric acid was not observed, even with 1M 512

aminovaleric acid. However, unlike 5-aminovaleraic acid, α-ketoglutaric acid obviously showed substrate inhibition due to the considerably declining reaction pH. Reaction was conducted with a broad range of initial α-ketoglutaric acid (100-500mM) and 5-AVA was equally controlled at 500mM to all reaction mixtures (Fig. 4B). Until 200mM, most of the α-ketoglutaric acid was consumed within 24 h. However, residual α-ketoglutaric acid significantly increased when more than 250mM of initial

IP T

α-ketoglutaric acid was added. The acidic characteristic of α-ketoglutaric was so strong that the pH of the reaction mixture decreased to 6.46 at the initial 250mM α-ketoglutaric acid. The unconsumed α-

SC R

ketoglutaric acid concentration was 121mM at this point. Moreover, none of the supplied αketoglutaric acid was converted at 500mM point. Even though a high concentration of 500mM Tris-

buffer was used for the reaction, high α-ketoglutaric acid concentration in the mixture diminished the

U

pH to 3.88 at 500mM initial α-ketoglutaric acid. Further attempts were made in this study to produce

N

glutaric acid with high concentration with 300mM 5-aminovaleric acid and 150mM α-ketoglutaric

TE D

Influence of co-factor on reaction

M

A

acid.

It was expected that enzymes encoded by gabT and gabD would require co-factors such as pyridoxal-5’-phosphate (PLP) and nicotinamide adenine dinucleotide (phosphate) (NAD(P)+) because

EP

these enzymes are types of aminotransferase and oxidoreductase, respectively. To determine the effect

CC

of co-factors on GabTD reaction, various reactions with different concentrations of PLP and NAD(P)+ were performed on reaction mixtures of 100mM substrates.

A

Generally, PLP was expected to be regenerated after a ping-pong reaction that first converts 5-

aminovaleric acid to glutarate semialdehyde yielding GabT-PMP. This enzyme-PMP intermediate is transformed again to enzyme-PLP, converting α-ketoglutaric acid to glutamate. PLP had no significant effect on the reaction, and the conversion rate was not changed with the addition of the PLP range

13

from 0mM to 2mM (Fig. 5A). Although some enzymes such as inducible lysine decarboxylase (CadA) from E. coli were highly dependent on PLP concentration, GABA aminotransferase from B. subtilis seemed to efficiently bind and regenerate PLP [10]. Thus, it is concluded that innate PLP in the enzyme or cell is sufficient to perform the aminotransferase activity of GabT.

IP T

On the other hand, NAD(P)+ should be supplemented or regenerated from an outer source as it will not be regenerated automatically after catalyzing glutarate semialdehyde dehydrogenation. NAD(P)+ showed a surprising effect on improving BsGabTD conversion reaction (Fig. 5B). Interestingly,

SC R

similar to a study by Seong Ah Park et al., which reported that BsGabD has a similar Km/Kcat value to both substrates NAD+ and NADP+; nevertheless, this type of enzyme is reported as a NADP+dependent enzyme [37]. The activity of GabD was increased by 2.99- and 2.61-fold when 1mM of

U

NAD+ and NADP+ was added, respectively. An addition of 20mM of NAD(P)+ improved the

N

production by 5.00- and 4.88-fold. Glutaric acid concentration increased almost linearly as NAD(P)+

A

was added. Linear regression between glutaric acid and NAD(P)+ concentration range from 1mM to

M

20mM resulted in R square values of 0.98 and 0.96 for NAD+ and NADP+, respectively. The slopes in this range for NAD+ and NADP+ were 1.34 and 1.38, respectively, and these values did not

TE D

significantly differ (data is not shown here). We chose to use NAD+ for further reaction because NAD+ is more relatively cost effective than NADP+. Subsequently, saturation concentration of NAD+

EP

was investigated at a substrates concentration of 100mM (Fig. 5C). The effect of the additional supply of NAD+ was significant, such that glutaric acid production was saturated with only the addition of

CC

60mM NAD+. α-Ketoglutaric acid was fully consumed after 60mM of NAD+, and full conversion

A

appeared to have been achieved. Further experiments were conducted with 60mM NAD+.

Glutaric acid production with repetitive supply of reaction mixture To produce more glutaric acid using the whole cell conversion system, the use of highly

14

concentrated substrates is preferred because reaction does not need a large scale reactor and concentration steps [38]. However, it is not easy to increase substrate concentration in the GabTD whole cell conversion system because the excess addition of α-ketoglutaric acid considerably decreases the initial pH mixture too far from its optimal pH range. Although the regeneration of αketoglutaric acid using a regeneration system may be possible later, as an alternative strategy to

IP T

produce more glutaric acid, the continual supply of reaction mixture was applied. This involves

recollecting the used whole cells and refreshing the reaction substrates and co-factor to the recollected

SC R

whole cell [39]. The order of the reaction conditions were pH 7, 30°C, 300mM 5-aminovaleric acid, 150mM α-ketoglutaric acid, and 60mM NAD+ which were selected at the above experiments, the whole cell reaction was monitored at each cycle. The recycle frequency of the whole cell was

U

determined to be 24 h after time course monitoring of the reaction for 48 h (Fig. 6A). Although

N

reaction continued, even after 24 h, and full conversion based on α-ketoglutaric acid was achieved, the

A

reaction rate seemed to decrease steeply 24 h after already showing 93.2% conversion. Thus, samples

M

were harvested every 24 h, the cell was then centrifuged, and the reacted medium was replaced with fresh substrate medium. At the first cycle, 141mM of glutaric acid was recorded (Fig. 6B). Conversion

TE D

continued for 5 cycles. During the cycles, the glutaric acid concentration at each cycle and the accumulated glutaric acid were monitored. The result showed that the production of glutaric acid was stable until the 2nd cycle. The produced glutaric acid concentration was decreased by 0.40-, 0.24-, and

EP

0.10-fold at the third, fourth, and fifth cycles, respectively. At each cycle, 47.1%, 44.8%, 19.9%, 11.4%, and 4.94% of the conversion based on 5-AVA was achieved. The conversion based on α-

CC

ketoglutaric acid was 94.2%, 89.6%, 37.7%, 22.9%, and 9.89% for each reaction cycle. The total

A

amount of glutaric acid produced was 191µmol at 500µl scale of 5 repeated recycles.

Conclusion To date, 5-AVA and glutaric acid have been produced by fermentation, introducing davB, davA, 15

davT, and davD genes from P. putida KT2440. High conversion from lysine to 5-aminovaleric acid was achieved up to 773mM by fermentation and 2134mM by whole cells with high productivity of 8.05 mM/h and 76.2 mM/h, respectively. However, until now, low concentrations (103mM) with low productivity (0.656mM/h) have been recorded for the conversion of 5-aminovaleric acid to glutaric acid via transamination by DavT or GabT followed by dehydrogenation by DavD or GabD. Moreover,

IP T

a detailed description of the 5-AVA to glutaric acid pathway has not yet been presented, despite the complicated considerations on the reaction such as co-factors and co-substrate. Thus, the aim of this

SC R

study was to establish a whole cell reaction system to monitor the effect of several factors and to increase the production of glutaric acid by introducing gabTD from B. subtilis. We optimized the reaction conditions, substrate concentrations, and cofactor concentrations. We determined the effect of

U

co-factors such as PLP and NAD(P)+ on reaction. The addition of PLP was not necessary, as PLP did

N

not affect GabT activity. Meanwhile, NAD(P)+ was revealed to be an important factor in improving

A

the productivity of glutaric acid. Additionally, the result showed that NAD+ and NADP+ have an equal

M

effect on glutaric acid production reaction on the selected condition, and this result proved that NAD+, which is a comparatively more inexpensive co-factor than NADP+, can be used for glutaric acid

TE D

production with BsGabTD whole cell conversion reaction. Although further research is needed on the optimization process for a more concentrated reaction and solution to supply large amounts of expensive co-factor such as NAD+, 300mM of 5-AVA reaction gave 47% conversion under optimized

EP

reaction condition. Repeated use of the whole cell showed that the activity of the biocatalyst could be

CC

maintained for few cycles.

In conclusion, we reported a new system for glutaric acid production using E. coli whole cell

A

overexpressing GabTD from B. subtilis. Comprehensive reaction optimization for a new enzyme system led to the highest glutaric acid conversion and accumulation by the whole cell, giving new insight and direction to improve the glutaric acid bio-conversion. Although full conversion with 300mM 5-AVA could not be achieved, we found limiting factors such as α-ketoglutaric acid and NAD(P)+, which could be the next targets for increasing glutaric acid productivity. 16

Acknowledgement This study was supported by the National Research Foundation (NRF) of Korea (NRF2015M1A5A1037196, NRF-2016R1D1A1B03932301), the Research Program initiated to address social issues highlighted by the National Research Foundation of Korea (NRF) funded by the Ministry

IP T

of Science and ICT (2017M3A9E4077234), and the R & D Program of MOTIE/KEIT (10067772, 10049674). The consulting service of the Microbial Carbohydrate Resource Bank (MCRB, Seoul,

A

CC

EP

TE D

M

A

N

U

SC R

Korea) is greatly appreciated.

17

References

[1] M.G.A. Vieira, M.A. da Silva, L.O. dos Santos, M.M. Beppu, Natural-based plasticizers and biopolymer films: A review, European Polymer Journal 47(3) (2011) 254-263.

IP T

[2] I. Bechthold, K. Bretz, S. Kabasci, R. Kopitzky, A. Springer, Succinic acid: a new platform chemical for biobased polymers from renewable resources, Chemical engineering & technology 31(5)

SC R

(2008) 647-654.

[3] M. Dusselier, P. Van Wouwe, A. Dewaele, E. Makshina, B.F. Sels, Lactic acid as a platform chemical in the biobased economy: the role of chemocatalysis, Energy & Environmental Science 6(5)

U

(2013) 1415-1442.

N

[4] S. Kind, C. Wittmann, Bio-based production of the platform chemical 1, 5-diaminopentane,

A

Applied microbiology and biotechnology 91(5) (2011) 1287.

M

[5] Y. Su, H.M. Brown, X. Huang, X.-D. Zhou, J.E. Amonette, Z.C. Zhang, Single-step conversion of cellulose to 5-hydroxymethylfurfural (HMF), a versatile platform chemical, Applied Catalysis A:

TE D

General 361(1-2) (2009) 117-122.

[6] B. Andreeßen, A.B. Lange, H. Robenek, A. Steinbüchel, Conversion of glycerol to poly (3hydroxypropionate) in recombinant Escherichia coli, Applied and environmental microbiology 76(2)

EP

(2010) 622-626.

[7] S. Okino, R. Noburyu, M. Suda, T. Jojima, M. Inui, H. Yukawa, An efficient succinic acid

CC

production process in a metabolically engineered Corynebacterium glutamicum strain, Applied microbiology and biotechnology 81(3) (2008) 459-464.

A

[8] J. Kim, H.-M. Seo, S.K. Bhatia, H.-S. Song, J.-H. Kim, J.-M. Jeon, K.-Y. Choi, W. Kim, J.-J. Yoon, Y.-G. Kim, Production of itaconate by whole-cell bioconversion of citrate mediated by expression of multiple cis-aconitate decarboxylase (cadA) genes in Escherichia coli, Scientific reports

7 (2017). 18

[9] J. Schneider, V.F. Wendisch, Putrescine production by engineered Corynebacterium glutamicum, Applied microbiology and biotechnology 88(4) (2010) 859-868. [10] H.J. Kim, Y.H. Kim, J.H. Shin, S.K. Bhatia, G. Sathiyanarayanan, H.M. Seo, K.Y. Choi, Y.H. Yang, K. Park, Optimization of Direct Lysine Decarboxylase Biotransformation for Cadaverine Production with Whole-Cell Biocatalysts at High Lysine Concentration, Journal of microbiology and

IP T

biotechnology 25(7) (2015) 1108-13.

[11] F. Pérez-García, P. Peters-Wendisch, V.F. Wendisch, Engineering Corynebacterium glutamicum

SC R

for fast production of L-lysine and L-pipecolic acid, Applied microbiology and biotechnology 100(18) (2016) 8075-8090.

[12] J. Adkins, J. Jordan, D.R. Nielsen, Engineering Escherichia coli for renewable production of the

U

5 carbon polyamide building blocks 5 aminovalerate and glutarate, Biotechnology and

N

bioengineering 110(6) (2013) 1726-1734.

A

[13] S.J. Park, E.Y. Kim, W. Noh, H.M. Park, Y.H. Oh, S.H. Lee, B.K. Song, J. Jegal, S.Y. Lee,

M

Metabolic engineering of Escherichia coli for the production of 5-aminovalerate and glutarate as C5 platform chemicals, Metabolic engineering 16 (2013) 42-47.

TE D

[14] C.M. Rohles, G. Gießelmann, M. Kohlstedt, C. Wittmann, J. Becker, Systems metabolic engineering of Corynebacterium glutamicum for the production of the carbon-5 platform chemicals 5aminovalerate and glutarate, Microbial cell factories 15(1) (2016) 154.

EP

[15] J.M. Jorge, F. Pérez-García, V.F. Wendisch, A new metabolic route for the fermentative production of 5-aminovalerate from glucose and alternative carbon sources, Bioresource technology

CC

(2017).

[16] Z. Li, J. Xu, T. Jiang, Y. Ge, P. Liu, M. Zhang, Z. Su, C. Gao, C. Ma, P. Xu, Overexpression of

A

transport proteins improves the production of 5-aminovalerate from l-lysine in Escherichia coli, Scientific reports 6 (2016). [17] L. Eggeling, M. Bott, A giant market and a powerful metabolism: L-lysine provided by Corynebacterium glutamicum, Applied microbiology and biotechnology 99(8) (2015) 3387-3394. 19

[18] S.J. Park, Y.H. Oh, W. Noh, H.Y. Kim, J.H. Shin, E.G. Lee, S. Lee, Y. David, M.G. Baylon, B.K. Song, High level conversion of L lysine into 5 aminovalerate that can be used for nylon 6, 5 synthesis, Biotechnology journal 9(10) (2014) 1322-1328. [19] N. Buschke, H. Schroder, C. Wittmann, Metabolic engineering of Corynebacterium glutamicum for production of 1,5-diaminopentane from hemicellulose, Biotechnology Journal 6(3) (2011) 306-

IP T

317.

[20] J.H. Shin, S.H. Park, Y.H. Oh, J.W. Choi, M.H. Lee, J.S. Cho, K.J. Jeong, J.C. Joo, J. Yu, S.J.

SC R

Park, Metabolic engineering of Corynebacterium glutamicum for enhanced production of 5aminovaleric acid, Microbial cell factories 15(1) (2016) 174.

[21] J. Wang, Y. Wu, X. Sun, Q. Yuan, Y. Yan, De Novo Biosynthesis of Glutarate via α-Keto Acid

U

Carbon Chain Extension and Decarboxylation Pathway in Escherichia coli, ACS Synthetic Biology

N

(2017).

A

[22] J.-L. Yu, X.-X. Xia, J.-J. Zhong, Z.-G. Qian, A novel synthetic pathway for glutarate production

M

in recombinant Escherichia coli, Process Biochemistry (2017). [23] O. Revelles, M. Espinosa-Urgel, S. Molin, J.L. Ramos, The davDT operon of Pseudomonas

TE D

putida, involved in lysine catabolism, is induced in response to the pathway intermediate δaminovaleric acid, Journal of bacteriology 186(11) (2004) 3439-3446. [24] O. Revelles, M. Espinosa-Urgel, T. Fuhrer, U. Sauer, J.L. Ramos, Multiple and interconnected

EP

pathways for L-lysine catabolism in Pseudomonas putida KT2440, Journal of bacteriology 187(21) (2005) 7500-7510.

CC

[25] X. Wang, P. Cai, K. Chen, P. Ouyang, Efficient production of 5-aminovalerate from l-lysine by engineered Escherichia coli whole-cell biocatalysts, Journal of Molecular Catalysis B: Enzymatic 134

A

(2016) 115-121.

[26] J.C. Joo, Y.H. Oh, J.H. Yu, S.M. Hyun, T.U. Khang, K.H. Kang, B.K. Song, K. Park, M.-K. Oh, S.Y. Lee, Production of 5-aminovaleric acid in recombinant Corynebacterium glutamicum strains from a Miscanthus hydrolysate solution prepared by a newly developed Miscanthus hydrolysis 20

process, Bioresource Technology (2017). [27] J.-L. Yu, X.-X. Xia, J.-J. Zhong, Z.-G. Qian, Enhanced production of C5 dicarboxylic acids by aerobic-anaerobic shift in fermentation of engineered Escherichia coli, Process Biochemistry 62 (2017) 53-58. [28] T. Ishige, K. Honda, S. Shimizu, Whole organism biocatalysis, Current opinion in chemical

IP T

biology 9(2) (2005) 174-180.

[29] Z. Xiao, C. Lv, C. Gao, J. Qin, C. Ma, Z. Liu, P. Liu, L. Li, P. Xu, A novel whole-cell biocatalyst

SC R

with NAD+ regeneration for production of chiral chemicals, PloS one 5(1) (2010) e8860.

[30] T. Matsumoto, S. Takahashi, M. Kaieda, M. Ueda, A. Tanaka, H. Fukuda, A. Kondo, Yeast whole-cell biocatalyst constructed by intracellular overproduction of Rhizopus oryzae lipase is

U

applicable to biodiesel fuel production, Applied microbiology and biotechnology 57(4) (2001) 515-

N

520.

A

[31] K. Ban, M. Kaieda, T. Matsumoto, A. Kondo, H. Fukuda, Whole cell biocatalyst for biodiesel

M

fuel production utilizing Rhizopus oryzae cells immobilized within biomass support particles, Biochemical engineering journal 8(1) (2001) 39-43.

TE D

[32] C.C. Carvalho, Whole cell biocatalysts: essential workers from Nature to the industry, Microbial biotechnology 10(2) (2017) 250-263.

[33] L. Zhu, Q. Peng, F. Song, Y. Jiang, C. Sun, J. Zhang, D. Huang, Structure and regulation of the

EP

gab gene cluster, involved in the γ-aminobutyric acid shunt, are controlled by a σ54 factor in Bacillus thuringiensis, Journal of bacteriology 192(1) (2010) 346-355.

CC

[34] C. Feehily, K. Karatzas, Role of glutamate metabolism in bacterial responses towards acid and other stresses, Journal of applied microbiology 114(1) (2013) 11-24.

A

[35] N.H. Tolia, L. Joshua-Tor, Strategies for protein coexpression in Escherichia coli, Nature methods 3(1) (2006) 55. [36] S.G. Lee, J.O. Lee, J.K. Yi, B.G. Kim, Production of cytidine 5′ monophosphate N

acetylneuraminic acid using recombinant Escherichia coli as a biocatalyst, Biotechnology and 21

bioengineering 80(5) (2002) 516-524. [37] S.A. Park, Y.S. Park, K.S. Lee, Kinetic characterization and molecular modeling of NAD (P)+dependent succinic semialdehyde dehydrogenase from Bacillus subtilis as an ortholog YneI, J. Microbiol. Biotechnol 24(7) (2014) 954-958. [38] J.-H. Kim, H.-M. Seo, G. Sathiyanarayanan, S.K. Bhatia, H.-S. Song, J. Kim, J.-M. Jeon, J.-J.

IP T

Yoon, Y.-G. Kim, K. Park, Development of a continuous L-lysine bioconversion system for cadaverine production, Journal of Industrial and Engineering Chemistry 46 (2017) 44-48.

SC R

[39] S.K. Bhatia, Y.H. Kim, H.J. Kim, H.M. Seo, J.H. Kim, H.S. Song, G. Sathiyanarayanan, S.H. Park, K. Park, Y.H. Yang, Biotransformation of lysine into cadaverine using barium alginateimmobilized Escherichia coli overexpressing CadA, Bioproc Biosyst Eng 38(12) (2015) 2315-2322.

U

[40] P. Liu, H. Zhang, M. Lv, M. Hu, Z. Li, C. Gao, P. Xu, C. Ma, Enzymatic production of 5-

N

aminovalerate from L-lysine using L-lysine monooxygenase and 5-aminovaleramide amidohydrolase,

A

Scientific reports 4 (2014) 5657.

M

[41] R.G. Taylor, D.C. Walker, R. McInnes, E. coli host strains significantly affect the quality of small scale plasmid DNA preparations used for sequencing, Nucleic acids research 21(7) (1993) 1677.

TE D

[42] S. Landais, P. Gounon, C. Laurent-Winter, J.-C. Mazié, A. Danchin, O. Bârzu, H. Sakamoto, Immunochemical analysis of UMP kinase fromEscherichia coli, Journal of bacteriology 181(3) (1999)

A

CC

EP

833-840.

22

Figure legends

Figure 1. E. coli whole cell overexpressing BsgabTD system converting 5-aminovaleric acid to glutaric acid

IP T

Substrates and co-factors were supplied externally and utilized for catalytic reaction by BsGabT and

BsGabD. The supplied 5-aminovaleric acid was finally converted to glutaric acid. Glutamic acid and

SC R

NAD(P)H can be produced as byproducts after GabT and GabD reaction, respectively. Figure 2. Application of new GabTD and comparison with known systems

U

Comparison of capability of producing glutaric acid using 5-aminovaleric acid aminotransferase and

N

α-ketoglutaric acid from different sources of organisms (A). Relative activity of BsGabT/BsGabD

A

mixture to amine substrates of similar structure with different chain-lengths (B).

TE D

BsgabTD, pH, and Temperature.

M

Figure 3. Optimization of whole cell reaction conditions with different copy numbers of

Relative activity of BsGabTD expressed with different types of vectors having different copy numbers (A). Optimal reaction pH from 6 to 12, for BsGabTD co-expressed whole cell biocatalyst (B), and

EP

broad range of reaction points to determine the optimal reaction temperature (C).

CC

Figure 4. Optimization of major substrate concentration for whole cell reaction Reaction at different concentrations of 5-aminovaleric acid from 100mM to 1000mM with fixed

A

concentration of α-ketoglutaric acid (A). Effect of increasing α-ketoglutaric acid in reaction mixture was determined (B). Change in initial pH by α-ketoglutaric acid were monitored together. Figure 5. Influence of co-factor on reaction Effect of different concentrations of PLP on BsGabT reaction (A), different concentrations of NAD+ 23

and NADP+ from 0 to 20mM on glutaric acid production (B), and impact of increasing NAD+ concentration on glutaric acid production (C). Figure 6. Time course monitoring and repetitive supply of reaction mixture Glutaric acid production monitoring by time-dependent manner with selected concentration of

IP T

substrates and co-factor (A) and reusability test of BsGabTD whole cell (B). Reaction mixture was changed with fresh substrate mixture containing 300mM 5-aminovaleric acid, 150mM α-ketoglutaric

A

CC

EP

TE D

M

A

N

U

SC R

acid, and 60mM NAD+. At each cycle, the reaction was performed for 24 h.

24

EP

CC

A TE D

IP T

SC R

U

N

A

M

Figure 1.

25

EP

CC

A TE D

B

IP T

SC R

U

N

A

M

Figure 2.

A

26

Figure 3. A

80

60

IP T

Relative activity (% )

100

40

SC R

20

0 et 1 et 1 4ma F du C du pET2 pCD pACY

1 et 1 due t F du pET pRS

TE D

M

A

N

U

B

EP

C

80

CC

Relative activity (% )

100

60

A

40

20

0 15

20

25

30

35

40

45

Temperature ( )

27

EP

CC

A TE D

B

IP T

SC R

U

N

A

M

Figure 4.

A

28

Figure 5. A

80

60

IP T

Relative activity (% )

100

40

20

0 0.1

0.5

1

2

SC R

0

PLP concentration (mM)

TE D

M

A

N

U

B

A

CC

EP

C

29

EP

CC

A TE D

IP T

SC R

U

B

N

A

M

Figure 6.

A

30

Titer (Glutar ic acid) [mM]

3200mM glucose, fed at glucose depletion up to 55.5mM

4.27

0.119

N/D

Whole cell

17.1mM Lysine

15

0.833

N/D

N/D

Whole cell

12.8 mM 5AVA

N/D

N/D

10.8

0.6

Fermentati on

91.3 mM Glucose

0.153

6.21

0.129

Pathway

Conversio n process

AMV

Fermentati on

(P. putida DavTD)

(Fedbatch)

A

(P. putida DavTD)

7.34

N/D

Referen ce

[13]

[12]

M

E. coli

N

AMV

productivi ty (Glutaric acid) [mM/h]

SC R

E. coli

Substrate

Productivi ty (5-AVA) [mM/h]

U

Host

Tite r (5AVA ) [mM ]

IP T

Table 1. Comparison of 5-AVA and glutaric acid titers and productivities.

(P. putida DavTD)

773

8.05

N/D

N/D

[18]

Enzyme

205mM Lysine

178

14.8

N/D

N/D

[40]

Whole cell

821mM Lysine, Fed lysine to maintain 821~1026m M

2134

76.2

N/D

N/D

[25]

(Semiwhole cell)

EP

E. coli

Fermentati on

TE D

AMV

111mM glucose, pH-stat feeding, At OD 60, 821mM lysine was supplement ed gradually

CC

AMV (P. putida DavTD)

A

E. coli

E. coli

AMV (P. putida DavTD)

31

(endogeno us GabTD)

AMV

1830mM glucose, maintained at 111mM

239

4.78

53

1.06

[14]

555mM glucose, maintained at 55.5222mM

341

4.61

29.2

0.365

[26]

N/D

539

3.46

N/D

[16]

Fermentati on

N/D

43.5

1.02

N/D

N/D

[15]

300mM 5aminovaleri c acid, 150mM αketoglutaric acid, 60mM NAD+

N/D

N/D

141

5.88

In this study

(Semiwhole cell)

AMV (E. coli PatAD)

AMV (B. subtilis GabTD)

Fermentati on

Whole cell

CC

EP

E. coli

[20]

218mM glucose

M

C. glutamicu m

(P. putida DavTD)

(Fedbatch)

0.656

TE D

E. coli

Fermentati on

103

IP T

AMV C. glutamicu m

(Fedbatch)

1.07

SC R

(C. glutamicu m GabTD)

Fermentati on

168

U

AMV C. glutamicu m

(Fedbatch)

494mM glucose, maintained at 111mM

N

(endogeno us GabTD)

Fermentati on

A

AMV C. glutamicu m

AMA

Fermentati on

111mM glucose

N/D

N/D

3.18

0.0663

[21]

E. coli

Glutaconat e

Fermentati on

55.5mM glucose

N/D

N/D

0.0454

0.00568

[22]

E. coli

Glutaconat e

Fermentati on

55.5mM glucose

N/D

N/D

0.825

0.0344

[27]

A

E. coli

32

Table 2. List of bacterial strains and plasmids used in this study. Strain/plasmid

Genotype/Strain description

Source/reference

Bacterial strains F− ϕ 80lacZ M15 endA recA hsdR(rk−mk−) supE thi gyrA relA∆(lacZYA-argF)U169

[41]

E. coli BW25113

F' λ- ∆(araD-araB)567, ∆lacZ4787(::rrnB-3), lambda-, rph-1, ∆(rhaD_rhaB)568, hsdR514

CGSC

E. coli BW25113 (DE3)

λDE3 lysogen of BW25113

SC R

B. subtilis str. 168 C. glutamicum ATCC13032

M

A

Kmr. 1 MCS site with T7 promoter, lac operator, RBS. P15A replicon. 10 to 12 copy number. Cmr. 2 MCS site with T7 promoter, lac operator, RBS. P15A replicon. 10 to 12 copy number. Specr. 2 MCS site with T7 promoter, lac operator, RBS. CloDF13 replicon. 20 to 40 copy number. Ampr. 2 MCS site with T7 promoter, lac operator, RBS. colE1(PBR322) replicon. 40 copy number. Kmr. 2 MCS site with T7 promoter, lac operator, RBS. RSF1030 replicon(NTP1). Over 100 copy number.

TE D

pCDF-duet1

N

Plasmids

pACYC-duet1

pET-duet1

pRSF-duet1

ATCC

ATCC ATCC

[36, 42] Novagen Novagen

Novagen

Novagen

gabT of B. subtilis str. 168 inserted into pET24ma

In this study

pET24ma-BsgabD

gabD of B. subtilis str. 168 inserted into pET24ma

In this study

CC

EP

pET24ma-BsgabT

pET24ma-CggabT

pET24ma-CggabD

A

In this study

U

P. putida KT2440

pET24ma

IP T

E. coli DH5α

gabT of C. glutamicum ATCC13032 inserted into pET24ma gabD of C. glutamicum ATCC13032 inserted into pET24ma

In this study In this study

pET24ma-PpdavT

davT of P. putdia KT2440 inserted into pET24ma

In this study

pET24ma-PpdavD

davD of P. putdia KT2440 inserted into pET24ma

In this study

pET24ma-BsgabTD

Native gabTD operon of B. subtilis str. 168 inserted into pET24ma

In this study

33

pCDF duet1-BsgabTD pRSF duet1-BsgabTD

Native gabTD operon of B. subtilis str. 168 inserted into pCDF duet1 Native gabTD operon of B. subtilis str. 168 inserted into pRSF duet1

In this study In this study

Native gabTD operon of B. subtilis str. 168 inserted into pACYC duet1

In this study

pET duet1-BsgabTD

Native gabTD operon of B. subtilis str. 168 inserted into pET duet1

In this study

A

CC

EP

TE D

M

A

N

U

SC R

IP T

pACYC duet1-BsgabTD

34