Engineering Corynebacterium glutamicum for the de novo...

Journal Pre-proof

Engineering Corynebacterium glutamicum for the de novo biosynthesis of tailoredpoly-γ-glutamic acid

Guoqiang Xu, Jian Zha, Hui Cheng, Mohammad H.A. Ibrahim, Fan Yang, HunterDalton, Rong Cao, Yaxin Zhu, Jiahua Fang, Kaijun Chi, Pu Zheng, Xiaomei Zhang,Jinsong Shi, Zhenghong Xu, Richard A. Gross, Mattheos A.G. Koffas

PII: S1096-7176(19)30193-4

DOI: https://doi.org/10.1016/j.ymben.2019.08.011

Reference: YMBEN 1590

To appear in: Metabolic Engineering

Received Date: 3 May 2019

Revised Date: 15 August 2019

Accepted Date: 17 August 2019

Please cite this article as: Xu, G., Zha, J., Cheng, H., Ibrahim, M.H.A., Yang, F., Dalton, H., Cao, R.,Zhu, Y., Fang, J., Chi, K., Zheng, P., Zhang, X., Shi, J., Xu, Z., Gross, R.A., Koffas, M.A.G., EngineeringCorynebacterium glutamicum for the de novo biosynthesis of tailored poly-γ-glutamic acid, MetabolicEngineering (2019), doi: https://doi.org/10.1016/j.ymben.2019.08.011.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the additionof a cover page and metadata, and formatting for readability, but it is not yet the definitive version ofrecord. This version will undergo additional copyediting, typesetting and review before it is publishedin its final form, but we are providing this version to give early visibility of the article. Please note that,during the production process, errors may be discovered which could affect the content, and all legaldisclaimers that apply to the journal pertain.

© 2019 Published by Elsevier Inc. on behalf of International Metabolic Engineering Society.

https://doi.org/10.1016/j.ymben.2019.08.011

https://doi.org/10.1016/j.ymben.2019.08.011

Engineering Corynebacterium glutamicum for the de novo biosynthesis of tailored 1

poly-γ-glutamic acid 2

Guoqiang Xua,b,c,d ξ, Jian Zhac ξ, Hui Chenga,b,d, Mohammad H. A. Ibrahimc,f, Fan 3

Yangc, Hunter Daltonc, Rong Cao a,b,d, Yaxin Zhua,b,d, Jiahua Fanga,b,d, Kaijun Chia,b,d, 4

Pu Zhenga, Xiaomei Zhanga,e, Jinsong Shia,e, Zhenghong Xua,b,d,e* , Richard A. Grossc, 5

f,g*, Mattheos A. G. Koffasa,c, f,g* 6

aThe Key Laboratory of Industrial Biotechnology, Ministry of Education, Jiangnan 7

University, Wuxi, Jiangsu 214122, China; 8

bNational Engineering Laboratory for Cereal Fermentation Technology, Jiangnan 9

University, Wuxi, Jiangsu 214122, China; 10

cCenter for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, 11

Troy, New York 12180, United States; 12

dJiangsu Provincial Research Center for Bioactive Product Processing Technology, 13

Jiangnan University, Wuxi 214122, China; 14

eLaboratory of Pharmaceutical Engineering, School of Pharmaceutics, Jiangnan 15

University, Wuxi 214122, China 16

fChemistry of Natural and Microbial Products Department, National Research Centre, 17

Al-Bohoos St., Cairo12622, Egypt. 18

gDepartment of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, 19

Troy, NY USA. 20

hDepartment of Biological Sciences, Rensselaer Polytechnic Institute, Troy, NY USA. 21

ξThese authors contributed equally to this work. 22

*Corresponding authors: Mattheos A. G. Koffas ([email protected]), Richard A. Gross 23

([email protected]), or Zhenghong Xu ([email protected]). 24

25

Abstract 26

γ-Polyglutamic acid (γ-PGA) is a biodegradable polymer naturally produced by Bacillus 27

spp. that has wide applications. Fermentation of γ-PGA using Bacillus species often 28

requires the supplementation of L-glutamic acid, which greatly increases the overall cost. 29

Here, we report a metabolically engineered Corynebacterium glutamicum capable of 30

producing γ-PGA from glucose. The genes encoding γ-PGA synthase complex from B. 31

subtilis (pgsB, C, and A) or B. licheniformis (capB, C, and A) were expressed under 32

inducible promoter Ptac in a L-glutamic acid producer C. glutamicum ATCC 13032, which 33

led to low levels of γ-PGA production. Subsequently, C. glutamicum F343 with a strong 34

L-glutamic acid production capability was tested. C. glutamicum F343 carrying capBCA 35

produced γ-PGA up to 11.4 g/L, showing a higher titer compared with C. glutamicum 36

F343 expressing pgsBCA. By introducing B. subtilis glutamate racemase gene racE under 37

Ptac promoter mutants with different expression strength, the percentage of L-glutamic 38

acid units in γ-PGA could be adjusted from 97.1% to 36.9%, and stayed constant during 39

the fermentation process, while the γ-PGA titer reached 21.3 g/L under optimal initial 40

glucose concentrations. The molecular weight (Mw) of γ-PGA in the engineered strains 41

ranged from 2,000 to 4,000 kDa. This work provides a foundation for the development of 42

sustainable and cost-effective de novo production of γ-PGA from glucose with 43

customized ratios of L-glutamic acid in C. glutamicum. 44

45

Keywords: Corynebacterium glutamicum, poly-γ-glutamic acid, glutamic acid, PGA 46

synthase, glutamate racemase 47

48

1. Introduction 49

γ-Polyglutamic acid (γ-PGA) is a natural polyanionic polypeptide consisting of D- and/or 50

L-glutamate repeat units linked between α-amino and γ-carboxylic groups (γ-amide 51

linkages) with molecular weights of ~100 to 8,000 kDa (Luo et al., 2016). γ-PGA has 52

attracted much attention due to its unique properties of biocompatibility, biodegradability, 53

water-solubility and viscosity, promoting its wide-ranging applications in medicine, food 54

and cosmetic industries (Rodriguez-Carmona and Villaverde, 2010; Sha et al., 2019). 55

Interestingly, γ-PGA with different stereochemical composition can find different 56

applications. For example, γ-PGA with high content of L-glutamate can be used as 57

materials for cosmetics due to its skin compatible features (Wu et al., 2006); in some 58

other fields, γ-PGA with high content of D-glutamate is used for drug delivery owing to 59

its low immunogenicity and minimum side effects (Baumgartner et al., 2017; Khalil et al., 60

2018). As a result, it is necessary to tune γ-PGA stereochemical composition to meet the 61

demands of different applications. 62

γ-PGA is produced naturally and mostly by Gram-positive bacteria such as Bacillus 63

subtilis, B. licheniformis, and B. amyloliquefaciens when fed with glutamic acid or 64

directly from glucose/glycerol. Their natural function for these producing strains is as a 65

capsular protectant or for nutrition preservation under adverse conditions (Ogunleye et al., 66

2015). Thus, production of γ-PGA using natural producer strains has been investigated 67

extensively. Such investigations involve screening and selection of superior mutant 68

strains, optimization of media and fermentation process, as well as strain improvement by 69

metabolic engineering (Cai et al., 2018; Cai et al., 2017; Feng et al., 2017a; Feng et al., 70

2017b; Massaiu et al., 2019; Zhan et al., 2018). Recently, the enzymes and genes 71

involved in γ-PGA synthesis in Bacillus spp have been identified and studied. The 72

membrane-associated enzyme complex γ-PGA synthetase, encoded by the genes pgsB, C, 73

and A (denoted as capB, C and A in some other strains), can polymerize glutamate to 74

γ-PGA in an ATP-dependent manner (Buescher and Margaritis, 2007; Sawada et al., 2018; 75

Sung et al., 2005). As γ-PGA is synthesized from D- and L-glutamic acid, glutamate 76

racemase genes racE/glr and yrpC have been identified in B. subtilis and were shown to 77

be essential for converting L-glutamate into D-glutamate for γ-PGA synthesis (Ashiuchi et 78

al., 1999; Jiang et al., 2011). γ-PGA producing strains can be divided into two classes. 79

Strains belonging to the first class require an external supply of L-glutamate to make 80

γ-PGA (Cai et al., 2018) while strains belonging to the second do not need the 81

supplementation of L-glutamate (Cao et al., 2011; Kongklom et al., 2015). 82

Recently, the heterogenous production of γ-PGA using E. coli or Corynebacterium 83

glutamicum has been reported (Cao et al., 2010; Cao et al., 2018). C. glutamicum is of 84

special interest as it is a GRAS strain (generally recognized as safe) approved by the US 85

Food and Drug Administration (FDA) and can naturally produce glutamate, a precursor to 86

synthesize γ-PGA, in a high titer (Becker et al., 2018). It is genetically tractable and it is 87

easy to reconstitute and modulate the γ-PGA biosynthesis pathway in this organism for 88

improving γ-PGA production. Moreover, it is possible to alter the chemistry of the 89

recombinantly produced γ-PGA by varying the ratios of D- and L-glutamate units, as well 90

as to tailor its molecular weight, using techniques in protein design, metabolic 91

engineering and synthetic biology. However, the production titer accomplished was 92

modest and the major monomer of the produced γ-PGA was L-glutamate. Thus, the need 93

remains for a γ-PGA producing C. glutamicum strain that produces γ-PGA in 94

substantially higher titers in variable stereochemical compositions. 95

In this study, we employed C. glutamicum strain ATCC13032 as a host to express the 96

γ-PGA biosynthetic pathway (Figure 1). For producing γ-PGA, we tested different 97

pathway gene orthologues, derived from B. subtilis Ia1a (Ibrahim et al., 2016) and B. 98

licheniformis 9945a (Birrer et al., 1994). Furthermore, we modulated gene expression to 99

increase γ-PGA production. For this purpose, we tested the potent glutamate producing 100

strain C. glutamicum strain F343 as the host (Zheng et al., 2012). Moreover, we 101

expressed the racemase gene (racE) from B. subtilis Ia1a to facilitate the modification of 102

L-glutamate ratio in γ-PGA. To further modulate the percentage of L-glutamate in γ-PGA, 103

we regulated the gene expression level of racE using Ptac promoters with different 104

transcriptional strengths. Our efforts led to a production titer of 21.3 g/L γ-PGA with 105

varied percentage of L-glutamate (36.9%-97.1%) in the recombinant C. glutamicum, 106

paving the way for the efficient production of γ-PGA with the control of stereochemistry 107

from glucose. 108

109

Fig.1. The metabolic pathways of L-glutamic acid in C. glutamicum and an artificial synthetic 110

pathway as well as related genes for γ-PGA. Heterologous synthetic pathways for γ-PGA containing 111

pgsB, C, and A or capB, C and A were tested in the present study in C. glutamicum. 112

113

2. Materials and methods 114

2.1. Strains and plasmids 115

All strains and plasmids used in this study are listed in Table 1. E. coli DH5α was 116

used for plasmid construction, the L-glutamic acid producers C. glutamicum ATCC13032 117

and C. glutamicum F343 were used as chassis organisms. Two wild-type Bacillus strains 118

B. subtilis Ia1a (Ibrahim et al., 2016) and B. licheniformis ATCC 9945a (Birrer et al., 119

1994) were used as sources of the PGA biosynthetic genes and as controls for native PGA 120

production. The plasmid pZM1 was used for gene cloning and plasmid construction (Zha 121

et al., 2018). 122

2.2. Construction of plasmids and recombinant strains 123

The genes pgsB, pgsC, pgsA and racE from B. subtilis strain Ia1a and the genes capB, 124

capC and capA from B. licheniformis ATCC 9945a (strain WU635) were amplified 125

(ACCUZYME 2X mix, Bioline) using genomic DNA as template that was extracted with 126

PureLink Genomic DNA Kit (Invitrogen). All the primers used in this study are listed in 127

Supplementary Table 1, and were purchased from Integrated DNA technologies (IDT, 128

USA). The genes were then cloned into plasmid pZM1 and assembled in a monocistronic 129

form using a previously published method (Xu et al., 2012). The transformation of 130

plasmids into C. glutamicum was carried out following a previously published protocol 131

(van der Rest et al., 1999). 132

2.3. Construction of racE plasmid series 133

The modified pZM1-RacE plasmids containing two copies of lacO were constructed 134

using a protocol previously established (Xu et al., 2012). The fragment covering lacO, a 135

ribosome-binding site (RBS), and racE was amplified by PCR, digested by NheI/KpnI 136

and cloned into XbaI/KpnI digested pZM1, resulting in the pZM1-RacE plasmid with two 137

copies of lacO. To construct an inactive racemase BsRacE, the tripeptide C184T185H186 138

was chosen as the target to be deleted according to literature (Glavas and Tanner, 2001; 139

Puig et al., 2009). The construction of BsRacE∆CTH was carried out using ClonExpress 140

MultiS One Step Cloning Kit (Vazyme Biotech Co., Ltd). The primers used are listed in 141

Supplementary Table 1. 142

2.4 Medium and cultivation methods 143

B. subtilis Ia1a and B. licheniformis ATCC 9945a were grown overnight in a 144

modified medium E (Leonard et al., 1958) at 37 oC and 220 rpm. The modified medium E 145

contains (per liter): 12.0 g, citric acid; 7.0 g, NH4Cl; 0.5 g, MgSO4·7H2O; 0.5 g, K2HPO4; 146

0.15 g, CaCl2·2H2O; 0.104 g MnSO4·H2O; and 0.04 g, FeCl3·6H2O (pH 7.4). E. coli used 147

for plasmid propagation was grown in Luria-Bertani (LB) medium containing 5 g/L Yeast 148

extract, 10 g/L tryptone, and 10 g/L NaCl. The medium for preparation of recombinant C. 149

glutamicum seeds (seed medium) contained 25 g/L glucose, 1.5 g/L K2HPO4, 0.6 g/L 150

MgSO4, 0.005 g/L MnSO4, 0.005 g/L FeSO4, 25 g/L corn steep liquor (Sigma, USA), and 151

2.5 g/L urea. Fermentation medium for C. glutamicum ATCC13032 is composed of 100 152

g/L glucose, 30 g/L (NH4)2SO4, 3.0 g/L KH2PO4, 0.5 g/L MgSO4, 0.02 g/L FeSO4, 0.02 153

g/L MnSO4, 5.0×10-5 g/L biotin, 4.5×10-4 g/L thiamine-HCl, 3.0×10-5 g/L protocatechuic 154

acid and 60 g/L CaCO3. The medium for C. glutamicum F343 is composed of 40 g/L 155

glucose, 1.0 g/L K2HPO4, 0.6 g/L MgSO4, 0.002 g/L MnSO4, 0.002 g/L FeSO4, 5.0×10-5 156

g/L thiamine, 15 g/L corn steep liquor, and 3.0 g/L urea. 157

For C. glutamicum and its recombinant strains, colonies were inoculated into seed 158

medium with kanamycin (25 mg/L) and cultured for 12 h at 30 oC (for C. glutamicum 159

ATCC13032) or 38 oC (for C. glutamicum F343) and 220 rpm. Then, the preculture (10%, 160

v/v) was inoculated into 25 mL fermentation medium with kanamycin (25 mg/L) in 161

250-mL Erlenmeyer flasks and grown for 36 h at 30 oC (for C. glutamicum ATCC13032) 162

or at 38 oC (for C. glutamicum F343) and 220 rpm. IPTG (1 mM) was added 3 h after 163

inoculation to induce for gene expression. 164

For the glutamic acid fermentation of wildtype C. glutamicum F343 in 7-L 165

bioreactors (New Brunswick Scientific, Edison NJ, USA), one colony was inoculated into 166

seed medium, and cultured for 10-12 h at 36 °C and 220 rpm. Then 300 mL preculture 167

was inoculated into 3-L fermentation medium. Initially, 450 µL of 500 mg/L 168

thiamine-HCl and 4.5 µL of 200 mg/L biotin was supplemented. Another 1.3 µL of biotin 169

stock was further added at 12 h and at the same time the temperature was switched from 170

36 °C to 39 °C. The pH was controlled at 7.2 by automatically feeding 25% ammonia 171

hydroxide and 6 M HCl during the fermentation. The air ventilation was set to 1 vvm and 172

the dissolved oxygen was controlled at 30% saturation by automatic control of agitation. 173

2.5 Purification of γ-PGA 174

After the pH was adjusted to 3.0 by adding citric acid, the fermentation broth was 175

centrifuged for 30 min at 5,000 rpm. After collecting the supernatant, 4 volumes of pure 176

ethanol were added and the mixture was left to cool at 4 oC overnight. Subsequently, the 177

pellet containing γ-PGA was re-dissolved in water and was purified by using a 30 KDa 178

dialysis bag. Finally, γ-PGA solution was dried in a vacuum freeze dryer and the product 179

was used for analysis. 180

2.6 Analytical procedures 181

Sampling for the measurement of cell growth, glucose, glutamic acid and γ-PGA was 182

performed every 4 h during fermentation. Cell growth was determined by measuring the 183

optical density at 600 nm (OD600) on a Biotek Synergy plate reader. 184

Before the determination of OD600, the samples were diluted by distilled water to the 185

OD600 range of 0.2-0.8, and then the OD600 values in the diluted samples were measured. 186

The concentration of glucose was analyzed on an Agilent 1200 Series HPLC (Agilent 187

Technologies, Santa Clara, CA, USA), equipped with a ZORBAX column (4.6×250 mm) 188

and refractive index detector (RID). The mobile phase consisted of acetonitrile (75%, v/v) 189

and the flow rate was set at 2 mL/min. For L-glutamic acid analysis, glutamic acid was 190

derivatized by phenylisothiocyanate (PITC). The level of L-glutamic acid was analyzed 191

by HPLC according to the modified method in our lab (Xu et al., 2015). The mobile 192

phase A was prepared by adding 15.2 g sodium acetate into 1850 mL H2O followed by 193

adjusting pH to 6.5 with acetic acid, and addition of 140 mL acetonitrile. Mobile phase B 194

consisted of acetonitrile (80%, v/v). The flow rate was set at 1 mL/min. 195

Gel permeation chromatography (GPC) was performed to both determine the 196

molecular weight of purified γ-PGA samples and to determine the concentration of 197

γ-PGA in cultures. The HPLC system used was equipped with Shodex KB00 series 198

columns (two KB80 M, one KB802.5) and a refractive index detector (RI detector). The 199

mobile phase consisted of 0.3 M Na2SO4 and the flow rate was set at 0.6 mL/min. 200

Dextran standards of narrow polydispersity (Polyscience Corporation) were used to 201

construct a calibration curve from which molecular weights of γ-PGA were calculated 202

with no further correction. The method used to quantify γ-PGA in culture supernatants 203

was adapted from a previous publication (Birrer et al., 1994). In summary, aliquots from 204

cultures were withdrawn, cells were removed by filtration through a 0.45 µm cellulose 205

acetate syringe filter and 50 µL of the resulting solution was directly injected into the 206

GPC using the identical chromatographic conditions described above. The peak area 207

corresponding to γ-PGA was measured and a calibration curve was constructed using 208

purified γ-PGA samples of peak area vs. γ-PGA concentration. 209

One-dimensional (1-D) proton (1H) NMR spectra were recorded on a Bruker 210

Spectrometer (600 MHz) in deuterated dimethyl sulfoxide (DMSO) (Birrer et al., 1994). 211

For the qualitative assessment to determine whether γ-PGA is present in media 212

supernatants, cells were removed by filtering the supernatant aliquot through a 0.45 µm 213

cellulose acetate syringe filter unit. Subsequently, salt and media components were 214

removed using an Amicon Centrifugal filter unit with a 5,000 molecular weight cutoff. 215

The retained material was then lyophilized, dissolved in d6-DMSO, and analyzed by 1H 216

NMR (600 MHz) using at least 500 transients and a one-second delay time. Based on 217

standard γ-PGA samples, we determined whether ≥ 0.05 g/L γ-PGA was present in 218

culture supernatants, 1H NMR signals corresponding to γ-PGA were detected. 219

2.7 Transcriptional-level assay 220

Gene transcriptional levels were measured by real-time PCR. Specifically, the total 221

RNA at log phase of γ-PGA fermentation (8, 16, 24 h) was collected by liquid nitrogen 222

grinding and extracted with TRIzol® reagent; cDNA was amplified by the RevertAid 223

First Strand cDNA Synthesis Kit (Takala, China). The primers used for amplifying the 224

corresponding genes are listed in Supplementary Table 1, and 16S rRNA from C. 225

glutamicum ATCC13032 was used as the reference gene to normalize the data. The gene 226

transcriptional levels of racE were compared with the control strain (PGA010) after 227

normalization to the reference gene 16S rRNA. The reaction system of qPCR is 17 µL, 228

including 1.25 µL cDNA, 8.5 µL SYBR Green Mix, 0.425 µL forward primer and reverse 229

primer, as well as 6.4 µL RNase free steaming water. The reaction conditions are as 230

follows: 50 oC, 2 min; 95 oC, 10 min, 1 cycle; 95 oC, 15 s; 60 oC, 1 min, 40 cycles. All the 231

experiments were performed in triplicate. 232

2.8 Western blot and enzyme assays 233

Cells grown under γ-PGA production conditions were collected and washed twice 234

with 50 mL of 0.85% NaCl solution. Harvested cells were resuspended in 5 mL standard 235

buffer (0.1 M Tris-HCl, pH 8.0), and lysozyme was added to lyse the cells at 4 oC for two 236

hours. Finally, the cells were disrupted by sonication, and cell debris was removed by 237

centrifugation at 12,000 rpm for 15 min (4 oC) (Wu et al., 2006). 238

For western blot analysis, SDS-PAGE was performed using the SDS-PAGE gel 239

preparation kit (BBI) according to manufacturers’ instructions. SDS-containing sample 240

buffers (10 µL) were added to 30 µL of samples and heated for 10 min, followed by gel 241

loading. Gels were electrophoresed at 80 V until the samples fully entered into the 242

concentrate (Schagger, 2006). After 40 min, the voltage was changed to 140 V and 243

maintained for 50 min. Western blot was performed using the Western Blot Kit (Mouse) 244

with PVDF membrane (BBI) according to instructions. 245

For the enzyme assay of racemase with different copies of lacO, the reaction of 100 246

µL crude enzyme resolution with 100 µL substrate (0.5 g/L L-glutamte) was conducted at 247

37 oC for 30 min, and the reaction was terminated by boiling in a water bath. The content 248

of D-glutamte in the reaction liquid was determined by HPLC (Oh et al., 2015; Wu et al., 249

2006). One unit (U) was defined as the amount of D-glutamate produced by enzyme per 250

minute, and U/mg represents the enzyme activity unit contained in 1 mg total protein. 251

Total protein concentration was determined by Bradford assay. 252

2.9 Molecular weight determination 253

The weight-average molecular weights (Mw) of γ-PGA were measured by GPC using 254

a Waters Empower3 system equipped with a column (Ultrahydrogel™ Linear 300 255

mm×7.8 mmid×2) and a refractive index detector (waters 2414 RI detector). The purified 256

γ-PGA was dissolved with the mobile phase (0.1 M NaNO3) and then filtered through a 257

0.45 µm membrane. The flow rate was set at 0.8 mL/min under 35 oC. Dextran with 258

molecular weight ranging from 9.75 kDa to 5348 kDa (National Institutes for Food and 259

Drug Control) was used as standard to determine the molecular weight of γ-PGA. 260

2.10 Determination of the ratio of L- and D-glutamic acid in γ-PGA 261

The method used was adapted from a literature procedure (Cromwick and Gross, 262

1995). γ-PGA samples were dissolved in 2 M HCl and hydrolyzed at 121 oC for 50 min. 263

Thereafter, after cooling the reaction to 5 oC, 2 M NaOH was added dropwise with 264

stirring until the solution reached pH 7. Next, 100 µL of 265

1-fluoro-2-4-dinitrophenyl-5-L-alanine amide (FDAA, Marfey's Reagent) and 20 µL of a 266

NaHCO3 solution were added to 50 µL of a 2.5 µmol glutamic acid (L-/D-mixture) 267

solution from hydrolysis of γ-PGA. The mixed solutions were incubated at 40 oC and 400 268

rpm for 1 h. Samples were then cooled to room temperature and 10 µL of 20 µmol HCl 269

was added and mixed thoroughly. The samples were diluted with 1 mL acetonitrile for 270

HPLC analysis. 271

For HPLC analysis, 5/95(v/v) acetonitrile/ammonium acetate buffer (50 mM, pH 5.3) 272

was used as the eluent at a flow rate of 0.6 mL/min. The separation was performed on a 273

XBridge TM column C8 (Waters, 3.5 µm, 2.1×50 mm) at 30 oC. Detection of eluents was 274

with a diode array detector at 340 nm. 275

276

3. Results 277

3.1. Tuning the architecture of the biosynthetic pathway and selecting gene orthologues to 278

increase PGA production 279

C. glutamicum ATCC13032 is a commonly used strain to produce L-glutamic acid. 280

To direct the synthesis of γ-PGA from glucose, the genes which encode PGA synthase in 281

B. subtilis Ia1a and B. licheniformis ATCC9945a were cloned into plasmid pZM1 in a 282

monocistronic form, and then expressed constitutively under Peftu in C. glutamicum 283

ATCC13032. The pathway consisting of genes pgsB, pgsC and pgsA from B. subtilis Ia1a 284

was not functional based on the observation that no γ-PGA was detected in the 285

fermentation broth of the chassis strain C. glutamicum ATCC13032 harboring the plasmid 286

pZM1-eftu-pgsB*CA (PGA002) (Figure S1). This determination was made by analyzing 287

cell- and medium-free, lyophilized culture supernatants by 1H NMR. 1H NMR spectra 288

can suggest protons in γ-PGA structure. Using 1H NMR, we were able to confirm the 289

presence of γ-PGA in cultures. Additionally, we further applied GPC to reconfirm the 290

presence of and to quantify the content and molecular weight of γ-PGA in cell cultures. 291

Based on these two analytical methods, we also investigated the PGA pathway genes 292

capB, capC, and capA derived from B. licheniformis ATCC 9945a controlled by Peftu 293

(strain PGA003) but no γ-PGA production was observed (Figure S1). 294

We hypothesized that the inability to produce γ-PGA using the two constructed C. 295

glutamicum strains may be due to the high expression levels of the γ-PGA biosynthetic 296

genes controlled by the strong promoter Peftu. To prove this hypothesis, promoter Peftu was 297

replaced by the inducible promoter Ptac, which is weaker than Peftu. The pathway 298

consisting of pgsB, pgsC and pgsA was first introduced in C. glutamicum ATCC13032 299

(PGA004) and the γ-PGA production accomplished using this recombinant strain was 300

0.47 g/L (Fig. 2A). Similarly, the inducible expression of capB, capC and capA in C. 301

glutamicum ATCC13032 (PGA005) led to γ-PGA production of 0.60 g/L (Fig. 2A). The 302

production of L-glutamic acid by PGA004 reached 0.40 g/L while L-glutamic acid 303

achieved by PGA005 was 0.33 g/L (Fig. 2A). In addition, higher cell growth was 304

achieved by PGA005 than PGA004 under similar glucose consumption (Figure S2). In 305

both cases, we verified the chemical structure of the produced γ-PGA by using 1H NMR 306

(Fig. 2B) and confirmed the presence of γ-PGA in the fermentation samples by GPC 307

(Figure S1). Even though the γ-PGA titer was low, these results suggested that functional 308

expression of the γ-PGA recombinant pathway was feasible in C. glutamicum ATCC 309

13032. Considering L-glutamic acid is the precursor of γ-PGA synthesis, and the 310

observation that B. licheniformis ATCC 9945a can produce higher γ-PGA when more 311

L-glutamic acid was supplemented in our study (Data not shown), selection of a C. 312

glutamicum strain with higher production of L-glutamic acid as the host strain was 313

conducted. 314

315

Fig.2. γ-PGA production of strains carrying the pgsBCA or capBCA genes under the Ptac promoter, 316

and characterization of produced γ-PGA by 1H NMR. (A) γ-PGA production by strains PGA004, 317

PGA005, PGA008 and PGA009. (B) 1H NMR of the standard and samples. 318

3.2 Construction of a C. glutamicum strain with high γ-PGA production capability 319

The limited supply of L-glutamic acid in C. glutamicum ATCC 13032 may represent 320

a bottleneck in the production of γ-PGA. Thus, the industrial strain C. glutamicum F343, 321

which can grow well even at high temperature (38 °C) with high L-glutamic acid 322

production capability (Zheng et al., 2012), was evaluated as a host for γ-PGA production. 323

In this study, L-glutamic acid fermentation with C. glutamicum F343 was performed in a 324

7-L bioreactor; the maximum biomass of this strain was reached at 28 h (28.3, OD600), the 325

titer of L-glutamic acid reached 28.3 g/L at 28 h, and the residual glucose concentration 326

was about 36 g/L in 48 h (Figure S3). Firstly, the constitutive expression of genes 327

pgsB*CA(PGA006) or capB**CA(PGA007) was investigated in C. glutamicum F343 and, 328

as previously shown with C. glutamicum ATCC13032, no γ-PGA production was detected. 329

Subsequently, inducible expression of pgsB*CA or capB**CA in C. glutamicum F343 330

(PGA008 and PGA009) was investigated. As a result, the host strain C. glutamicum F343 331

expressing pgsB*CA under the control of the Ptac promoter produced 1.6 g/L γ-PGA, 332

while a slightly higher γ-PGA titer was achieved from the strain carrying the capB**CA 333

genes (1.8 g/L) (Fig. 2A). Thus, strain C. glutamicum F343, with inducibly expressed 334

capB**CA, was chosen for subsequent studies. 335

First, the effect of induction time (addition of inducer at 1, 2, 3, and 4 h after starting 336

the shake flask culture) on the cell growth and γ-PGA production was investigated. As 337

shown in Figure 3A, the biomass did not change significantly while the γ-PGA titer was 338

very sensitive to the induction time (≥ 2 h), an induction time of 2 h gave the highest 339

γ-PGA titer (10.5 g/L); however, earlier induction time (1 h) led to a dramatic decrease of 340

biomass and a lower γ-PGA titer (8.0 g/L) at 24 h. Next, the fermentation profile of the 341

engineered strain C. glutamicum F343 expressing capBCA (PGA009) was investigated. 342

As shown in Figure 3B, at approximately 24 h, cell growth entered the stationary phase 343

(maximum OD600 reached 15). γ-PGA production reached about 11.4 g/L at 48 h, after 344

which, the γ-PGA titer decreased slightly. Rapid consumption of glucose occurred in the 345

log phase, however, high levels of residual glucose (~60 g/L) was present at the end of 346

the fermentation. The medium concentration for glutamic acid increased to 8.0 g/L and, 347

thereafter, decreased. Presumably, this is due to utilization of produced glutamic acid for 348

γ-PGA synthesis. About 6 g/L L-glutamic acid was still present at the end of 349

fermentation. 350

351

Fig.3. Production of γ-PGA by strain PGA009. (A) Effect of induction times (1, 2, 3, and 4 h) on cell 352

growth and γ-PGA production by strain PGA009; (B) γ-PGA fermentation profile by strain PGA009. 353

3.3 Modification of γ-PGA composition by heterologous expression of RacE 354

In γ-PGA produced by the engineered C. glutamicum F343, the major monomer unit 355

was L-glutamate as demonstrated by the observation that 97.1% of the glutamate units 356

were L-glutamate. This is possibly due to the low intrinsic activity or complete absence of 357

glutamate racemase, the enzyme responsible for the conversion of L-glutamate to 358

D-glutamate. To change the ratio of L-glutamate units in the product γ-PGA and 359

investigate the effect of the L-glutamate supply with different ratios of L-glutamate and 360

D-glutamate on γ-PGA production, the glutamate racemase gene racE derived from B. 361

subtilis Ia1a was expressed with different transcriptional levels in the engineered strain 362

PGA009 (Fig. 4A). 363

364

Fig.4. Modification of γ-PGA composition by heterologous expression of BsracE. (A) Metabolic 365

strategies for regulating γ-PGA monomer percentage: PGA009, module for γ-PGA synthesis; PGA010, 366

module for γ-PGA synthesis and alteration of L-glutamic acid percentage by incorporating one lacO 367

operator upstream of racE; PGA011, module for γ-PGA synthesis and alteration of L-glutamic acid 368

percentage by incorporating two lacO operators upstream of racE; Ptac, inducible promoter; T7t, 369

terminator; RBS, ribosome binding site; lacO, operator. (B) Relative transcriptional level of racE in 370

PGA010 and PGA011. (C) Enzyme activity of glutamate racemase in PGA010 and PGA011. (D) 371

L-glutamic acid percentage in γ-PGA produced by PGA009, PGA010 and PGA011. 372

Real-time quantitative PCR was employed to measure the expression levels of 373

BsracE. The transcriptional levels of BsracE in C. glutamicum F343 374

pZM1-capBCA-(2lacO)BsracE (PGA011), where two lacO operators were incorporated, 375

were approximately 70.0%, 41.2% and 37.6% (P<0.05) of that in strain C. glutamicum 376

F343 pZM1-capBCA-(1lacO)BsracE (PGA010) at 8 h, 16 h and 24 h, respectively (Fig. 377

4B). Subsequently, western blot analysis and enzyme assay of glutamate racemase with 378

different copies of lacO introduced upstream of its gene were performed to characterize 379

the expression of BsracE. Western blot analysis showed that differentiated expression of 380

BsracE has been achieved successfully (Figure S4). Enzyme assay also showed that the 381

enzyme levels of BsracE in PGA011 were approximately 62.8%, 54.7% and 44.6% of 382

that in strain PGA010 at 8 h, 16 h and 24 h, respectively (Fig. 4C). Then the effect of 383

different expression of BsracE on the ratio of L-glutamate in γ-PGA was monitored. As 384

expected, 97.1% of the glutamate units is L-glutamate in the engineered strain PGA009, 385

and the percentage of L-glutamic acid in PGA decreased to 36.9% in the engineered strain 386

PGA010, while in engineered strain PGA011, the ratio of L-glutamate monomer was 387

decreased to almost half the PGA moieties (59.5%) (Fig. 4D). Thus, a large and even 388

range of L-glutamate (36.9-97.1%) was achieved in the γ-PGA-producing C. glutamicum 389

strains. 390

Next, the fermentation of these strains with different transcriptional levels of BsracE 391

were investigated. The introduction of BsracE in strain PGA010 led to dramatically 392

decreased cell growth and glucose consumption as well as γ-PGA titer relative to strain 393

PGA009; the maximum cell growth was achieved at 16 h (4.8, OD600), and was only 394

32.7% of that of strain PGA009 (Fig.5A). Moreover, the production by strain PGA010 of 395

γ-PGA reached only 2.2 g/L with a low residual L-glutamic acid concentration (0.36 g/L) 396

at 56 h (Fig. 5A). In contrast, enhanced growth was observed for strain PGA011 397

compared to the strain PGA009; the maximum biomass of the former strain was reached 398

at 48 h (20.5, OD600), which is 39.5% higher than that of the latter strain (14.7, OD600). 399

Even at 24 h, the OD600 of strain PGA011 was 17.4. Despite the higher residual 400

L-glutamic acid concentration detected at the end of the fermentation for strain PGA011 401

(8.5 g/L), γ-PGA production reached 15.4 g/L at 24 h, 35.1% higher than for PGA009 402

(Figs 3B and 5B). However, the residual glucose is 50 g/L at the end of fermentation of 403

PGA011 suggesting the initial glucose concentration is too high. In addition, an inactive 404

form of glutamate racemase was constructed as described in section 2.3, which resulted in 405

strain C. glutamicum F343 pZM1-capBCA-(1lacO)BsracE∆CTH (PGA012). Enzyme 406

assay showed that no glutamate racemase activity can be detected in PGA012. 407

Fermentation profile also showed that PGA012 exhibited medium level of cell growth, 408

glucose consumption, glutamic acid accumulation as well as γ-PGA production compared 409

to PGA009 and PGA010 (Figure S5). 410

411

Fig.5. Production of γ-PGA by engineered strains PGA010 and PGA011. (A) fermentation profile of 412

strain PGA010; (B) fermentation profile of strain PGA011. 413

3.4 γ-PGA stereochemistry and molecular weight 414

For strains, PGA009, PGA010 and PGA011, determination of γ-PGA 415

stereochemistry and molecular weight as a function of culture time is reported (Figs 6A 416

and 6B, respectively). The proportion of L-glutamic acid in γ-PGA was determined by 417

hydrolysis, derivatization with Marfey's Reagent and subsequent LC analysis (see 418

Experimental). For strains PGA009, PGA010 and PGA011, small variations in 419

stereochemistry as a function of culture time was observed and is given by the range of 420

L-glutamate ratio values reported. For strain PGA009, the percentage of L-glutamate 421

repeat units varies between 92-98%. Correspondingly, the percentage of L-glutamate 422

values for strains PGA010 and PGA011 are between 50-70% and 25-35%, respectively. 423

Hence, this set of recombinant engineered strains derived from C. glutamicum F343 424

provides the ability to produce γ-PGA with a wide range of stereochemical compositions. 425

This is most interesting for PGA009 and PGA011 since the titers of γ-PGA production 426

are ≥11 g/L. Further work on regulating the expression of the racE gene is expected to 427

allow an even wider range of γ-PGA stereochemical compositions. 428

The weight average molecular weight (Mw) of γ-PGA produced during fermentations 429

of strains PGA009, PGA010 and PGA011, that differ in expression levels of racE, was 430

determined (Fig.6B). For strain PGA009, Mw varied between 1000 kDa-4000 kDa during 431

the culture. The highest concentration of γ-PGA occurred at 48 h, this corresponded to 432

when the Mw of γ-PGA had its highest value (~4000 kDa). The Mw value remained at 433

about 4000 kDa to 80 h and then decreased to about 1000 kDa at 96 h. Relative to Mw 434

values for strains PGA010 and PGA011, the Mw of γ-PGA between 48 and 80 h, 435

produced by PGA090 that lacks RacE, are the highest values observed in this study. The 436

Mw of γ-PGA reached its highest value of about 8000 kDa at 24 h, which corresponded to 437

the time at which the culture reached its highest γ-PGA concentration and cell growth 438

was in the early stationary phase. While the γ-PGA concentration in the culture media of 439

PGA011 remained at about 15 g/L, Mw decreased rapidly between 24 and 32 h to about 440

2000 kDa and then slowly decreased to about 800 kDa as the culture time reached 72 h. 441

However, Mw of γ-PGA in the strain PGA010 kept stable during the fermentation. 442

443

Fig.6. Monomer ratio and molecular weight (Mw) of γ-PGA produced by strains PGA009, PGA010, 444

and PGA011 during the fermentation process. (A) monomer proportion of γ-PGA produced by strains 445

PGA009, PGA010, and PGA011 during the fermentation process; (B) Molecular weight (Mw) of γ-PGA 446

produced by strains PGA009, PGA010, and PGA011 during the fermentation process 447

3.5 Fermentation optimization for γ-PGA production in the engineered strain 448

Since there was still moderate amount of glucose at the end of the fermentation for 449

the strain PGA011, we optimized the initial concentration of glucose to relieve possible 450

inhibition on cell growth and γ-PGA production. Thus, the initial concentration of 451

glucose was varied to determine an optimal value. As shown in Fig. 7A, at 24 h, cell 452

growth increased as the initial concentration of glucose decreased from 120 g/L to 60 g/L, 453

then decreased when the concentration was further decreased to 40 g/L. However, at 80 454

g/L, cell growth plateaued at 72 h. From Fig. 7B, 80 g/L is the lowest initial glucose 455

concentration that remains available throughout the fermentation but is nearly completely 456

depleted at 72 h. Fig. 7C shows that, with 80 g/L initial glucose concentration, the highest 457

titer (21.3 g/L) was achieved by the strain PGA011 in 56 h (productivity 0.38 g/L/h). The 458

highest production for other conditions ranged from 12.3 g/L to 15.4 g/L. To the best of 459

our knowledge, this is the highest production of γ-PGA in recombinant C. glutamicum. 460

461

Fig.7. Effect of different initial glucose concentrations on the fermentation profile of strain PGA011. 462

(A) cell growth, (B) residual glucose, and (C) γ-PGA production during fermentation process. 463

464

4. Discussion 465

During the past several decades, C. glutamicum has been the platform organisms for 466

industrial production of amino acids. C. glutamicum is generally regarded as a safe, and 467

its metabolic engineering toolbox continues to develop making it an attractive host 468

system for the industrial production of various chemicals (Zha et al., 2018). In this work, 469

C. glutamicum was used as the chassis organism to synthesize γ-PGA from glucose at 470

high yields and tailored ratios of L- to D-glutamic acid. 471

Both constitutive and inducible expression of the γ-PGA biosynthetic genes pgsBCA 472

and capBCA were tested in C. glutamicum. Our results showed that γ-PGA can be 473

produced when the heterologous genes pgsBCA or capBCA were inducibly overexpressed, 474

while no PGA was detected when those genes were overexpressed constitutively. This 475

result is consistent with previous reports (Cao et al., 2010) that γ-PGA production was 476

achieved by inducibly overexpressing the heterologous genes pgsBCA, and consistent 477

with our previous study (Zha et al., 2018) that constitutive promoters reduced the 478

production of anthocyanins in C. glutamicum by 52-84% when compared with the 479

inducible promoter. The reason for this behavior may be due to substantial increase in 480

metabolic burden which is a longstanding problem in metabolic engineering (Wu et al., 481

2016), resulting from the constitutive expression of pgsBCA or capBCA, as indicated by 482

our optimization of induction time for the expression of capBCA (Figure S6). The effect 483

of induction time (1 h, 2 h, 3 h and 4 h for PGA009, 4 h, 8 h and 12 h for PGA010) on the 484

production of γ-PGA by the corresponding strains was investigated and the results show a 485

remarkable change in titer of γ-PGA as a function of induction time (Figure S6). 486

The chassis organism C. glutamicum F343 achieved higher γ-PGA production when 487

harboring genes capBCA instead of genes pgsBCA. This result is consistent with an 488

earlier observation by our group that B. licheniformis ATCC 9945a was a better γ-PGA 489

producer than B. subtilis Ia1a (Data not shown). Sequence alignment of amino acids of 490

γ-PGA synthetase complex from B. subtilis Ia1a and B. licheniformis ATCC 9945a 491

showed that PgsA shared the identity of 65.30% with CapA, while PgsB and PgsC 492

possessed greater similarity with 90.08% and 89.93%, respectively, when compared with 493

CapB and CapC. This result is consistent with the previous report that PgsA showed low 494

homology, while PgsB and PgsC showed great similarity based on the sequence 495

alignment of amino acids (Cao et al., 2013), and B. licheniformis-derived CapBCA 496

appeared to have greater catalytic activity than B. subtilis Ia1a-derived PgsBCA, different 497

function of PgsB and PgsC as well as PgsA (Buescher and Margaritis, 2007; Sung et al., 498

2005). However, the catalytic mechanisms of these enzymes should be explored through 499

three-dimensional modelling and crystal structure to elucidate the precise function of 500

each membrane-associated component in γ-PGA polymerization and transportation. 501

While little is known about how the properties of γ-PGA differ as a function of its 502

stereochemical composition, it is expected that this intrinsic structural characteristic of 503

γ-PGA can be important, especially in the context of interactions with biological systems. 504

Previous work showed that, for Bacillus licheniformis ATCC 9945a, the γ-PGA 505

stereochemical composition can be regulated by changing the concentration of 506

manganese II sulfate (MnSO4) ion in the culture medium (Cromwick and Gross, 1995). 507

Through this strategy, the percentage of L-glutamate units in γ-PGA varied from 59% to 508

10%. The methods described in our work enable the preparation of γ-PGA with very high 509

L-glutamate content (~97%) and more precise regulation could be achieved by using the 510

heterologous expression of BsracE in C. glutamicum. However, heterologous expression 511

of BsracE led to weaker cell growth. It may be that high expression of BsracE in C. 512

glutamicum led to metabolic imbalances. This hypothesis is supported by the observation 513

that moderate expression of BsracE in PGA011 not only restored cell growth; also, it 514

regulated the content of L-glutamate units in γ-PGA and elevated the titer of γ-PGA up to 515

15.4±0.4 g/L. This result is consistent with previous study in E. coli that claimed racE 516

integration enhanced γ-PGA yield by increasing its D-glutamate content, which is 517

preferred by the γ-PGA production system (Cao et al., 2013). Other researchers reported 518

that overexpression of glr (racE) gene by Bacillus spp both increased the production of 519

γ-PGA as well as the percentage of D-glutamate in γ-PGA (Ashiuchi et al., 1999; Jiang et 520

al., 2011). Thus, in future work, we plan to use racE to dynamically regulate the 521

stereochemical composition of D-/L-glutamic acid in γ-PGA, which might lead to further 522

production increase and a wider range of D-glutamic acid percentages in γ-PGA. 523

Furthermore, what was not been addressed herein, or in these other literature papers, is 524

how the distribution of D- and L-glutamic acid along chains is altered with changes in 525

racE expression. Such changes in repeat unit sequence distribution would be expected to 526

have profound effects on both the physical and biological properties of γ-PGA. 527

For native γ-PGA producing species such as those of the genus Bacillus, degradation 528

of γ-PGA over extended culture times is often observed due to the presence of γ-PGA 529

hydrolases in these organisms (Kimura et al., 2004). There are three types of γ-PGA 530

hydrolases, namely γ-PGA hydrolase (PgdS), D-/L-endopeptidase (Such as CwlO), and 531

γ-glutamyltransferase (GGT) (Feng et al., 2014; Liu et al., 2018; Sha et al., 2019). 532

Deletion of genes encoding for γ-PGA hydrolases led to a higher molecular weight in B. 533

subtilis (Kimura et al., 2004). In addition, disruption of the cell wall lytic enzyme CwlO 534

affects the amount and molecular weight of γ-PGA produced by B. sublitis (natto) (Mitsui 535

et al., 2011). In this study, high levels of residual glucose were present at the end of the 536

fermentation. C. glutamicum F343 could tolerate higher concentrations of glucose when it 537

was used to produce L-glutamic acid (Zheng et al., 2012); However, the engineered strain 538

grew slower than the original strain, and the residual glucose was much higher than that 539

in the original strain. The possible reason may be that the accumulation of γ-PGA made 540

the fermentation broth very viscous, preventing the diffusion and dissolution of oxygen 541

and slowing down cellular metabolism (Liu et al., 2017). Furthermore, decreases in 542

molecular weight of γ-PGA were observed over extended culture period, indicating that 543

endogenous degradation enzymes such as D-/L-endopeptidase might exist in C. 544

glutamicum. Based on the complete genome sequence of strain C. glutamicum ATCC 545

13032, the gene (ggtB) encoding a γ-glutamyl transpeptidase in C. glutamicum has been 546

identified (Walter et al., 2016), and can be a target in future metabolic engineering efforts. 547

548

5. Conclusions 549

In the present work, we chose C. glutamicum as the host organism to construct an 550

artificial synthetic pathway to produce γ-PGA by heterologous expression of the genes 551

pgsBCA or capBCA from Bacillus spp. We compared two ways of overexpression 552

(constitutive expression versus inducible expression), and found that inducible expression 553

can lead to accumulation of γ-PGA, while constitutive expression cannot. Subsequently, a 554

high glutamic acid producing C. glutamicum strain was used as the host organism, 555

resulting in a final γ-PGA titer of 11.4 g/L and the percentage of L-glutamic acid in chains 556

during cultures is between 92-98%. We also constructed a strain that can produce γ-PGA 557

with a lower percentage of L-glutamic acid by moderating the expression of a racemase 558

gene, racE, resulting in the production of γ-PGA with a wide range of L-to-D-glutamic 559

acid ratios (97.1-36.9%) as well as a higher titer (21.3 g/L). This work provides a new 560

strategy for the de novo biosynthesis of γ-PGA in C. glutamicum that produces high 561

molecular weight (typically between 2,000 and 4,000 kDa) and tailored L-glutamic acid 562

content. 563

Acknowledgment

We thank Iza Radecka, University of Wolverhampton, UK, for kindly providing

the strain B. licheniformis ATCC 9945a (Copy “WU635”). We also thank Katharina

Neufeld, Anthony Maiorana, and Robert Centorfrom Rensselaer Polytechnic Institute

(RPI) for their help during NMR characterization. The technical and financial support

from Center for Biotechnology and Interdisciplinary Studies at RPI, is also

acknowledged. This work was financially supported by the Program of Introducing

Talents of Discipline to Universities (No. 111-2-06). The National Key Research and

Development Program of China (2018YFA0900303). International Joint Research

Laboratory for Engineering synthetic biosystems for Intelligent Biomanufacturing at

Jiangnan University. The Six talent peaks project in Province (No.2015-SWYY-006),

and Top-notch Academic Programs Project of Jiangsu Higher Education Institutions

(TAPP).

References

Ashiuchi, M., Soda, K., Misono, H., 1999. Characterization of yrpC gene product of Bacillus subtilis

IFO 3336 as glutamate racemase isozyme. Biosci Biotech Bioch. 63, 792-798.

Baumgartner, R., Kuai, D., Cheng, J., 2017. Synthesis of controlled, high-molecular weight

poly(l-glutamic acid) brush polymers. Biomater Sci. 5, 1836-1844.

Becker, J., Rohles, C. M., Wittmann, C., 2018. Metabolically engineered Corynebacterium glutamicum

for bio-based production of chemicals, fuels, materials, and healthcare products. Metab Eng.

50, 122-141.

Birrer, G. A., Cromwick, A. M., Gross, R. A., 1994. Gamma-poly(glutamic acid) formation by Bacillus

licheniformis 9945a: physiological and biochemical studies. Int J Biol Macromol. 16, 265-75.

Buescher, J. M., Margaritis, A., 2007. Microbial biosynthesis of polyglutamic acid biopolymer and

applications in the biopharmaceutical, biomedical and food industries. Crit Rev Biotechnol. 27,

1-19.

Cai, D., Chen, Y., He, P., Wang, S., Mo, F., Li, X., Wang, Q., Nomura, C. T., Wen, Z., Ma, X., Chen, S.,

2018. Enhanced production of poly-gamma-glutamic acid by improving ATP supply in

metabolically engineered Bacillus licheniformis. Biotechnol Bioeng. 115, 2541-2553.

Cai, D., He, P., Lu, X., Zhu, C., Zhu, J., Zhan, Y., Wang, Q., Wen, Z., Chen, S., 2017. A novel approach

to improve poly-gamma-glutamic acid production by NADPH regeneration in Bacillus

licheniformis WX-02. Sci Rep. 7, 43404.

Cao, M., Feng, J., Sirisansaneeyakul, S., Song, C., Chisti, Y., 2018. Genetic and metabolic engineering

for microbial production of poly-gamma-glutamic acid. Biotechnol Adv. 36, 1424-1433.

Cao, M., Geng, W., Liu, L., Song, C., Xie, H., Guo, W., Jin, Y., Wang, S., 2011. Glutamic acid

independent production of poly-gamma-glutamic acid by Bacillus amyloliquefaciens LL3 and

cloning of pgsBCA genes. Bioresour Technol. 102, 4251-7.

Cao, M., Geng, W., Zhang, W., Sun, J., Wang, S., Feng, J., Zheng, P., Jiang, A., Song, C., 2013.

Engineering of recombinant Escherichia coli cells co-expressing poly-gamma-glutamic acid

(gamma-PGA) synthetase and glutamate racemase for differential yielding of gamma-PGA.

Microb Biotechnol. 6, 675-84.

Cao, M. F., Song, C. J., Jin, Y. H., Liu, L., Liu, J., Xie, H., Guo, W. B., Wang, S. F., 2010. Synthesis of

poly (gamma-glutamic acid) and heterologous expression of pgsBCA genes. J Mol Catal

B-Enzym. 67, 111-116.

Cromwick, A. M., Gross, R. A., 1995. Effects of manganese (II) on Bacillus licheniformis ATCC

9945A physiology and gamma-poly(glutamic acid) formation. Int J Biol Macromol. 17,

259-67.

Feng, J., Gao, W., Gu, Y., Zhang, W., Cao, M., Song, C., Zhang, P., Sun, M., Yang, C., Wang, S., 2014.

Functions of poly-gamma-glutamic acid (gamma-PGA) degradation genes in gamma-PGA

synthesis and cell morphology maintenance. Appl Microbiol Biotechnol. 98, 6397-407.

Feng, J., Gu, Y., Quan, Y., Gao, W., Dang, Y., Cao, M., Lu, X., Wang, Y., Song, C., Wang, S., 2017a.

Construction of energy-conserving sucrose utilization pathways for improving

poly-gamma-glutamic acid production in Bacillus amyloliquefaciens. Microb Cell Fact. 16,

98.

Feng, J., Quan, Y., Gu, Y., Liu, F., Huang, X., Shen, H., Dang, Y., Cao, M., Gao, W., Lu, X., Wang, Y.,

Song, C., Wang, S., 2017b. Enhancing poly-gamma-glutamic acid production in Bacillus

amyloliquefaciens by introducing the glutamate synthesis features from Corynebacterium

glutamicum. Microb Cell Fact. 16, 88.

Glavas, S., Tanner, M. E., 2001. Active site residues of glutamate racemase. Biochemistry. 40,

6199-6204.

Ibrahim, M. H. A., Cress, B. F., Linhardt, R. J., Koffas, M. A. G., Gross, R. A., 2016. Draft genome

sequence of Bacillus subtilis Ia1a, a new strain for poly-γ-glutamic acid and

exopolysaccharide production. Microbiology Resource Announcements.

Jiang, F., Qi, G., Ji, Z., Zhang, S., Liu, J., Ma, X., Chen, S., 2011. Expression of glr gene encoding

glutamate racemase in Bacillus licheniformis WX-02 and its regulatory effects on synthesis of

poly-gamma-glutamic acid. Biotechnol Lett. 33, 1837-40.

Khalil, I. R., Khechara, M. P., Kurusamy, S., Armesilla, A. L., Gupta, A., Mendrek, B., Khalaf, T.,

Scandola, M., Focarete, M. L., Kowalczuk, M., Radecka, I., 2018. Poly-gamma-glutamic acid

(gamma-PGA)-based encapsulation of adenovirus to evade neutralizing antibodies. Molecules.

23.

Kimura, K., Tran, L. S., Uchida, I., Itoh, Y., 2004. Characterization of Bacillus subtilis

gamma-glutamyltransferase and its involvement in the degradation of capsule

poly-gamma-glutamate. Microbiology. 150, 4115-23.

Kongklom, N., Luo, H. Z., Shi, Z. P., Pechyen, C., Chisti, Y., Sirisansaneeyakul, S., 2015. Production

of poly-gamma-glutamic acid by glutamic acid-independent Bacillus licheniformis TISTR

1010 using different feeding strategies. Biochem Eng J. 100, 67-75.

Leonard, C. G., Housewright, R. D., Thorne, C. B., 1958. Effects of some metallic ions on glutamyl

polypeptide synthesis by Bacillus subtilis. J Bacteriol. 76, 499-503.

Liu, T. Y., Chu, S. H., Shaw, G. C., 2018. Deletion of the cell wall peptidoglycan hydrolase gene cwlO

or lytE severely impairs transformation efficiency in Bacillus subtilis. J Gen Appl Microbiol.

64, 139-144.

Liu, X. X., Yang, S., Wang, F., Dai, X. F., Yang, Y. K., Bai, Z. H., 2017. Comparative analysis of the

Corynebacterium glutamicum transcriptome in response to changes in dissolved oxygen levels.

J Ind Microbiol Biot. 44, 181-195.

Luo, Z. T., Guo, Y., Liu, J. D., Qiu, H., Zhao, M. M., Zou, W., Li, S. B., 2016. Microbial synthesis of

poly-gamma-glutamic acid: current progress, challenges, and future perspectives. Biotechnol

Biofuels. 9.

Massaiu, I., Pasotti, L., Sonnenschein, N., Rama, E., Cavaletti, M., Magni, P., Calvio, C., Herrgard, M.

J., 2019. Integration of enzymatic data in Bacillus subtilis genome-scale metabolic model

improves phenotype predictions and enables in silico design of poly--glutamic acid production

strains. Microb Cell Fact. 18.

Mitsui, N., Murasawa, H., Sekiguchi, J., 2011. Disruption of the cell wall lytic enzyme CwlO affects

the amount and molecular size of poly-gamma-glutamic acid produced by Bacillus subtilis

(natto). J Gen Appl Microbiol. 57, 35-43.

Ogunleye, A., Bhat, A., Irorere, V. U., Hill, D., Williams, C., Radecka, I., 2015. Poly-gamma-glutamic

acid: production, properties and applications. Microbiology. 161, 1-17.

Oh, S. Y., Richter, S. G., Missiakas, D. M., Schneewind, O., 2015. Glutamate racemase mutants of

Bacillus anthracis. J Bacteriol. 197, 1854-61.

Puig, E., Mixcoha, E., Garcia-Viloca, M., Gonzalez-Lafont, A., Lluch, J. M., 2009. How the substrate

D-glutamate drives the catalytic action of Bacillus subtilis glutamate racemase. J Am Chem

Soc. 131, 3509-3521.

Rodriguez-Carmona, E., Villaverde, A., 2010. Nanostructured bacterial materials for innovative

medicines. Trends Microbiol. 18, 423-30.

Sawada, K., Araki, H., Takimura, Y., Masuda, K., Kageyama, Y., Ozaki, K., Hagihara, H., 2018.

Poly-L-gamma-glutamic acid production by recombinant Bacillus subtilis without pgsA gene.

Amb Express. 8.

Schagger, H., 2006. Tricine-SDS-PAGE. Nat Protoc. 1, 16-22.

Sha, Y. Y., Zhang, Y. T., Qiu, Y. B., Xu, Z. Q., Li, S., Feng, X. H., Wang, M. X., Xu, H., 2019. Efficient

biosynthesis of low-molecular-weight poly-gamma-glutamic acid by stable overexpression of

PgdS hydrolase in Bacillus amyloliquefaciens NB. J Agr Food Chem. 67, 282-290.

Sung, M. H., Park, C., Kim, C. J., Poo, H., Soda, K., Ashiuchi, M., 2005. Natural and edible

biopolymer poly-gamma-glutamic acid: synthesis, production, and applications. Chem Rec. 5,

352-66.

van der Rest, M. E., Lange, C., Molenaar, D., 1999. A heat shock following electroporation induces

highly efficient transformation of Corynebacterium glutamicum with xenogeneic plasmid

DNA. Appl Microbiol Biotechnol. 52, 541-5.

Walter, F., Grenz, S., Ortseifen, V., Persicke, M., Kalinowski, J., 2016. Corynebacterium glutamicum

ggtB encodes a functional gamma-glutamyl transpeptidase with gamma-glutamyl dipeptide

synthetic and hydrolytic activity. J Biotechnol. 232, 99-109.

Wu, G., Yan, Q., Jones, J. A., Tang, Y. J., Fong, S. S., Koffas, M. A. G., 2016. Metabolic burden:

cornerstones in synthetic biology and metabolic engineering applications. Trends Biotechnol.

34, 652-664.

Wu, Q., Xu, H., Xu, L., Ouyang, P., 2006. Biosynthesis of poly(gamma-glutamic acid) in Bacillus

subtilis NX-2: Regulation of stereochemical composition of poly(gamma-glutamic acid).

Process Biochem. 41, 1650-1655.

Xu, G. Q., Zhu, Q. J., Luo, Y. C., Zhang, X. J., Guo, W., Dou, W. F., Li, H., Xu, H. Y., Zhang, X. M.,

Xu, Z. H., 2015. Enhanced production of L-serine by deleting sdaA combined with modifying

and overexpressing serA in a mutant of Corynebacterium glutamicum SYPS-062 from sucrose.

Biochem Eng J. 103, 60-67.

Xu, P., Vansiri, A., Bhan, N., Koffas, M. A., 2012. ePathBrick: a synthetic biology platform for

engineering metabolic pathways in E. coli. ACS Synth Biol. 1, 256-66.

Zha, J., Zang, Y., Mattozzi, M., Plassmeier, J., Gupta, M., Wu, X., Clarkson, S., Koffas, M. A. G., 2018.

Metabolic engineering of Corynebacterium glutamicum for anthocyanin production. Microb

Cell Fact. 17, 143.

Zhan, Y. Y., Sheng, B. J., Wang, H., Shi, J., Cai, D. B., Yi, L., Yang, S. H., Wen, Z. Y., Ma, X., Chen, S.

W., 2018. Rewiring glycerol metabolism for enhanced production of poly-gamma-glutamic

acid in Bacillus licheniformis. Biotechnol Biofuels. 11.

Zheng, P., Liu, M., Liu, X. D., Du, Q. Y., Ni, Y., Sun, Z. H., 2012. Genome shuffling improves

thermotolerance and glutamic acid production of Corynebacteria glutamicum. World J

Microbiol Biotechnol. 28, 1035-43.

Table 1 Bacterial strains and plasmids used in this study.

Plasmids/Strains Description Sources

Plasmids

pZM1-eftu Constitutive expression plasmid, promoter Peftu Lab collection

pZ8-1 Used for point mutation Lab collection

pZM1-Ptac Inducible expression plasmid, promoter Ptac Lab collection

pZ8-1-capB pZ8-1 containing capB This study

pZ8-1-capB* The first NdeI in capB was removed This study

pZ8-1-capB** Both NdeI in capB were removed This study

pZM1-eftu-pgsB*CA Constitutive expression plasmid containing genes

pgsB*, pgsC and pgsA

This study

pZM1-eftu-capB**CA Constitutive expression plasmid containing genes

capB**, capC and capA

This study

pZM1-Ptac-pgsB*CA Inducible expression plasmid containing genes

pgsB*, pgsC and pgsA

This study

pZM1-Ptac-capB**CA Inducible expression plasmid containing genes

capB**, capC and capA

This study

pZM1-Ptac-capB**CA-B

sracE (1lacO)

Inducible expression plasmid containing genes

capB**, capC and capA, as well as gene BsracE

with one lacO

This study

pZM1-Ptac-capB**CA-B Inducible expression plasmid containing genes This study

sracE (2lacO) capB**, capC and capA, as well as gene BsracE

with two lacO

pZM1-Ptac-capB**CA-B

sracE (1lacO)∆CTH

Inducible expression plasmid containing genes

capB**, capC and capA, as well as gene

BsracE∆CTH with one lacO

This study

Strains

B. subtilis Ia1a Produce γ-PGA with high yield naturally (Ibrahim et al.,

2016)

B. licheniformis ATCC

9945a

“WU635”

Produce γ-PGA with high yield naturally (Birrer et al.,

1994)

C. glutamicum

ATCC13032

Can accumulate L-glutamic acid Lab collection

C. glutamicum F343 Industrial strain which can accumulate L-glutamic

acid with high titer

(Zheng et al.,

2012)

PGA001 C. glutamicum F343 harbors empty plasmid This study

PGA002 C. glutamicum ATCC13032 harbors the genes

pgsB*CA in constitutive expression plasmid

This study


capB**CA in constitutive expression plasmid

This study


pgsB*CA in inducible expression plasmid

This study


capB**CA in inducible expression plasmid

This study

PGA006 C. glutamicum F343 harbors the genes pgsB*CA in

constitutive expression plasmid

This study

PGA007 C. glutamicum F343 harbors the genes capB**CA

in constitutive expression plasmid

This study

PGA008 C. glutamicum F343 harbors the genes pgsB*CA in

inducible expression plasmid

This study


in inducible expression plasmid

This study


and the gene BsracE in inducible expression

plasmid

This study


and the gene BsracE under 2lacO in inducible

expression plasmid

This study


and the gene BsracE∆CTH in inducible expression

plasmid

This study

* represent one NdeI site was mutated in the gene.

** represent two NdeI sites were mutated in the gene.

SUPPLEMENTARY MATERIAL

Supplementary Figure 1. GPC chromatogram of fermentation broth produced by the

strain C. glutamicum ATCC13032 carrying the plasmid pZM1-eftu-pgsB*CA or

pZM1-Ptac-pgsB*CA. (A) blank; (B) the standard of γ-PGA; (C) sample produced by

C. glutamicum ATCC13032 pZM1-eftu-pgsB*CA (PGA002); (D) sample produced by

C. glutamicum ATCC13032 pZM1-Ptac-pgsB*CA (PGA004).

Figure S1

Supplementary Figure 2. Fermentation profiles (cell growth and glucose

consumption) of strain C. glutamicum ATCC13032 carrying the pgsBCA (PGA004) or

capBCA (PGA005) genes under the control of Ptac promoter at 48 h.

Figure S2

Supplementary Figure 3. L-glutamic acid fermentation by strain C. glutamicum 1

F343 in 7-L bioreactor. 2

3

Figure S3 4

5

1

Supplementary Figure 4. Western blot analysis of glutamate racemase with different 2

copies of lacO at different time (8, 16, 24 h). Number 1 represents C. 3

glutamicumF343 pZM1-capBCA-(lacO)BsracE(PGA011); while 2 represents C. 4

glutamicumF343 pZM1-capBCA-(2lacO)BsracE(PGA011). 5

6

7 Figure S4 8

Supplementary Figure 5. Fermentation profiles of strains PGA009 (no RacE) and 1

PGA010 (active RacE) and the strain PGA012 (inactive RacE). (A) Biomass; (B) 2

Consumed glucose; (C) L-glutamic acid; (D) γ-PGA. 3

4

Figure S5 5

Supplementary Figure 6. Effect of induction times on the cell growth and γ-PGA

production of engineered strains. (A) C. glutamicum F343 pZM1-Ptac-capB**CA

(PGA009) with induction times (1, 2, 3, and 4 h); (B) C. glutamicum F343

pZM1-Ptac-capB**CA-BsracE(1lacO) (PGA010) with induction times (4, 8, and 12

h).

Figure S6

Supplementary Table 1 Primers used in this study.

Primers Sequence (5’ to 3’) Notes

pgsB-Nde-F TATACATATGTGGTTACTCATTATAGCCTGTGC Clone the gene pgsB,

mutate the NdeI site in

pgsB and construction of

the recombinant plasmids

pgsB-BamH-R CGCGGATCCCTAGCTTACGAGCTGCTTAACCTTG

pgsB-R1 CCATGACATCCATGTGGTCTTCTAAAAC

pgsB-F2 GTTTTAGAAGACCACATGGATGTCATGG

pgsC-Nde-F TATACATATGTTCGGATCAGATTTATACATCG Clone the gene pgsC and

construction of the

recombinant plasmids

pgsC-BamH-R CGCGGATCCttaaattaagtagtaaacaaacatgatagc

pgsA-Nde-F TATACATATGAAAAAAGAACTGAGCTTTCAT Clone the gene pgsA and

construction of the


pgsA-BamH-R CGCGGATCCTTATTTAGATTTTAGTTTGTCGCTATG

CapB-Nde-F AAAACTGCAGCATATGTGGGTAATGCTATTAGCCTG Clone the gene capB,


capB and construction of

the recombinant plasmids

CapB-BamH-R CGCGGATCCCTAGCTAACGAGCTGCTTAATCTTG

mutation-1-F CGACTTTAAACATTTGGAAGC

mutation-1-R GCTTCCAAATGTTTAAAGTCG

mutation-2-F CGTCAAAGCGTATGAAGCAG

mutation-2-R CTGCTTCATACGCTTTGACG

CapC-Nde-F TATACATATGTTTGGATCAGATTTATATATCGC Clone the gene capC and

construction of the


CapC-BamH-R CGCGGATCCttagattagatagtaagcatacataatgacg

CapA-Nde-F TATACATATGAAAAAACAACTGAACTTTCAGG Clone the gene capA and

CapA-BamH-R CGCGGATCCTCATTTGTTCACCACTCCGT construction of the


Bs-RacE-F-NdeI TATACATATGTTGGAACAACCAATAGGAGTC Clone the gene BsracE and

construction of the


Bs-RacE-R-Bam

HI

CGCGGATCCCTATCTTCTAATCGGTTCTTGCAGTG

Bl-RacE-Nde-F1 TATACATATGGATCAACCGATAGGAGTCATC Clone the gene BlracE,


BlracE and construction

of the recombinant

plasmids

Bl-RacE-R1 GCTTCTTCATAGGCGCCGC

Bl-RacE-F2 GCGGCGCCTATGAAGAAGC

Bl-RacE-BamH-

R2

CGCGGATCCTTATTTTATAGCGGTTTCCTGAAGAG

16S rRNA-F ATATCAGGAGGAACACCAAT Be used as the reference

gene to normalize the data

16S rRNA-R ACTACCAGGGTATCTAATCC

RacE rRNA -F ATCGCATTGGAAGACATC Transcriptional-Level

Assay of racE from B.

subtilis RacE rRNA-R TGCTCTTAATCGTATTCTCTG

Insert1-F CTTGAGGGGTTTTTTGCTAGCTTGACAATTAATCATCG

GCTCG

Clone the gene BlracE

except for 9 base pairs that

encodes the tripeptide

C184T185H186

Insert1-R AGGATAGCCTAAAATCAGCGAATCAATCG

Insert2-F CGCTGATTTTAGGCTATCCTATTTTAAAAGAATCCATTC

AGAG

Insert2-R AGTTTGTAGAAACGCGTCGACAAAAAACCCCTCAAGA

CCCG

1. Inducible expression of the γ-PGA synthase can lead to de novo synthesis of γ-PGA.

2. Limited supply of L-glutamate in C. glutamicum restricts the synthesis of γ-PGA.

3. The stereochemistry of γ-PGA can be tailored by regulating the expression of racE.

Engineering Corynebacterium glutamicum for the de novo...

Documents

Transcript of Engineering Corynebacterium glutamicum for the de novo...