Stoichiometrically Controlled Immobilization of ... - ACS Publications

Nov 16, 2015 - Toru Honda, Tsuyoshi Tanaka, and Tomoko Yoshino*. Division of Biotechnology and Life Science, Institute of Engineering, Tokyo Universit...
0 downloads 0 Views 1MB Size
Subscriber access provided by CARLETON UNIVERSITY

Article

Stoichiometrically Controlled Immobilization of Multiple Enzymes on Magnetic Nanoparticles by the Magnetosome Display System for Efficient Cellulose Hydrolysis Toru Honda, Tsuyoshi Tanaka, and Tomoko Yoshino Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.5b01174 • Publication Date (Web): 16 Nov 2015 Downloaded from http://pubs.acs.org on November 21, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

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

Biomacromolecules

1

2

Stoichiometrically Controlled Immobilization of Multiple Enzymes on Magnetic

3

Nanoparticles by the Magnetosome Display System for Efficient Cellulose Hydrolysis

4

5

Toru Honda, Tsuyoshi Tanaka, and Tomoko Yoshino*

6 7

Division of Biotechnology and Life Science, Institute of Engineering, Tokyo University of

8

Agriculture and Technology, 2-24-16 Naka-cho, Koganei, Tokyo 184-8588, Japan

9 10

*Corresponding author: Tomoko Yoshino; Division of Biotechnology and Life Science,

11

Institute of Engineering, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho,

12

Koganei, Tokyo 184-8588, Japan

13

Telephone: +81-42-388-7021

14

Fax: +81-42-385-7713

15

E-mail: [email protected]

16 17

Keywords: magnetosome display system; cellulosome; protein complex; magnetic nanoparticle

18 19

ACS Paragon Plus Environment

1

Biomacromolecules

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

Page 2 of 23

20

Abstract

21

The immobilization of multiple cellulase complexes receiving attention for use in the efficient

22

hydrolysis of celluloses. In this study, the magnetosome display system was employed for the

23

preparation of systems mimicking natural multiple cellulase complexes (cellulosomes) on

24

magnetic nanoparticles (MNPs). Initially, two fluorescent proteins, namely green fluorescent

25

protein and mCherry, were immobilized on MNPs. Fluorescence analysis revealed the close

26

proximity of two different proteins on the MNPs. Enzyme-linked immunosorbent assay analysis

27

showed that stoichiometrically equivalent amounts of the proteins were immobilized on the

28

MNPs. Next, endoglucanase (EG) and β-glucosidase (BG) were immobilized on MNPs to give

29

EG/BG-MNPs. The resulting MNPs were applied for the hydrolysis of celluloses, with rapid

30

hydrolysis of carboxymethyl cellulose being observed. Furthermore, the fusion of the cellulose-

31

binding domain to EG/BG-MNPs promoted improved hydrolysis activity against the insoluble

32

cellulose. We could therefore conclude that the magnetosome display system can expand the

33

possibilities of mimicking natural cellulosome organization on MNPs.

34 35 36 37 38 39 40

ACS Paragon Plus Environment

2

Page 3 of 23

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

Biomacromolecules

41

1. Introduction

42

Cellulose is the most abundant and promising biomass for bioethanol production. Its hydrolysis

43

by cellulase is key for this process, however, the large consumption of cellulase makes it costly,

44

resulting in a bottleneck in cellulose-based bioethanol production 1. An alternative approach is

45

the utilization of immobilized cellulase on solid materials, as immobilization provides

46

advantages in terms of handling, reusability, and cellulase stability2-6.

47

Although a number of studies have been reported for the preparation of immobilized cellulase,

48

the immobilization of multi-enzyme complexes, known as cellulosomes7, has also attracted

49

considerable attention. Cellulosomes has been discovered in several cellulolytic bacteria, such as

50

Clostridium cellulolyticum, Clostridium thermocellum, and Ruminococcus flavefaciens8. It is

51

composed of a scaffold protein containing a cohesin domain, a cellulose-binding domain (CBD),

52

and several cellulases with the dockerin domain (i.e., endoglucanase, exoglucanase, and β-

53

glucosidase). The dockerin domain can specifically bind to the cohesin domain on the scaffold

54

protein, allowing the construction of highly organized multi-enzyme complexes in close

55

proximity to one another. This close proximity leads to synergistic effects in enzymatic activity,

56

resulting in efficient cellulose hydrolysis9. Recently, to mimic the cellulosome on solid materials,

57

the immobilization of multiple cellulases onto magnetic particles has been attempted10,

58

However, reproduction of the highly organized multiple enzyme assembly was difficult, as the

59

enzymes were randomly immobilized on the magnetic particle surfaces through chemical cross-

60

linking or Au-S bonding.

61

We previously proposed a novel process for the immobilization of recombinant proteins onto

62

magnetic nanoparticles (MNPs) using the magnetosome display system with a magnetotactic

11

.

ACS Paragon Plus Environment

3

Biomacromolecules

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

12

Page 4 of 23

63

bacterium, Magnetospirillum magneticum strain AMB-1

64

in magnetotactic bacteria, consisting of a magnetite core covered by a lipid bilayer membrane,

65

and an average diameter of 75 nm 12 (see also Figure 1(A)). In the magnetosome display system,

66

foreign proteins can be immobilized on magnetosomes using the mms13 gene as a fusion partner

67

by genetic engineering13. Mms13, a protein kwon to double-pass transmembrane protein and

68

tightly bind to the magnetite core. Therefore, M. magneticum AMB-1 was transformed by the

69

fusion gene of mms13 and target protein gene, and then desired target protein can be efficiently

70

expressed (or immobilized) on magnetosomes via Mms13 in M. magneticum. Until now, various

71

proteins, including single-chain variable fragments

72

complex

73

have been successfully immobilized on MNPs using the magnetosome display system. The

74

genetically engineered MNPs can be easily extracted and purified from the M. magneticum

75

AMB-1 transformant cell lysate by magnetic separation. The complexes of foreign proteins with

76

MNPs have been applied to immunoassays, ligand screening assays, and cell separation systems

77

as magnetic carriers18-23. The magnetosome display system can therefore enable us to construct

78

highly organized multi-enzyme complexes on MNPs.

79

We report the preparation of highly organized protein complexes on MNPs using the

80

magnetosome display system, for application in the hydrolysis of celluloses. Initially, two

81

fluorescent proteins, namely green fluorescent protein and mCherry, will be immobilized on

82

MNPs for the proof-of-concept study. Fluorescent microscopy and fluorescence resonance

83

energy transfer analysis will be used to determine the proximity of two different proteins on the

84

MNPs based on the interaction between the cohesin and dockerin domains. Furthermore, ELISA

85

analysis will be used to determine the quantities of proteins immobilized on the MNPs.

15

, a tropomyosin receptor kinase A

16

14

. Magnetosome is a unique organelle

, the class I major histocompatibility

, and a thyroid-stimulating hormone receptor

17

,

ACS Paragon Plus Environment

4

Page 5 of 23

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

Biomacromolecules

86

Subsequently, endoglucanase and β-glucosidase immobilized on MNPs will be prepared using

87

the magnetosome display system and tested for cellulose hydrolysis activity. We expect to

88

confirm synergistic effects when using two close proximity enzymes and the CBD on MNPs,

89

resulting in the efficient degradation of celluloses. We predict that the magnetosome display

90

system will expand the possibilities to mimic natural cellulosome organization.

91 92

2. Experimental Section

93

2.1 Bacterial Strains and Culture Conditions

94

Escherichia coli TOP10 (Life technologies, CA, US) and EPI300 (ARBROWN, Tokyo, Japan)

95

were used for gene cloning. E. coli transformant cells were cultured at 37 °C in Luria-Bertani

96

(LB) agar plates or in ampicillin-containing media (50 µg mL-1). The Magnetospirillum

97

magneticum AMB-1 (ATCC 700264) mms13 gene deletion mutant (strain AMB-1 ∆mms13) was

98

used for the production of magnetic nanoparticles (MNPs)24. Strain AMB-1 ∆mms13 was

99

microaerobically cultured in magnetic spirillum growth medium (MSGM) containing

100

gentamycin (2.5 µg mL-1) at 28 °C as described previously

101

constructed by electroporation

102

and ampicillin (5 µg mL-1). Furthermore, the expression of target proteins was induced by the

103

addition of anhydrotetracycline hydrochloride (500 ng mL-1, Kanto Chemical, Tokyo, Japan) at

104

the mid-log phase, and the mixture cultured for a further 12 h.

26

25

. AMB-1 transformants were

and cultured in MSGM containing gentamycin (2.5 µg mL-1)

105 106

2.2 Plasmid Construction

ACS Paragon Plus Environment

5

Biomacromolecules

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

Page 6 of 23

107

A schematic diagram representing the co-expression plasmids in Figure S1. Co-expression

108

plasmids were comprised of two fusion gene sets, namely “mms13-cohesin” genes and “target-

109

dockerin” genes. Cohesin domains from Clostridium thermocellum (CohC: CipA 364-702 amino

110

acids) and Ruminococcus flavefaciens (CohR: ScaB 29-186 amino acids) were fused to mms13.

111

Dockerin domains from C. thermocellum (DocC: CelS 673-741 amino acids) and R. flavefaciens

112

(DocR: ScaA 787-879 amino acids) were fused to target proteins (GFP and mCherry, or

113

endoglucanase celCCA (EG) and β-glucosidase Ccel_2454 (BG). EG and BG were removed

114

signal sequence and native dockerin. The gene sequences of two cellulases and the cellulose

115

binding domain (CBD) were also obtained from C. thermocellum. All genes were generated by

116

artificial gene synthesis (Integrated DNA Technologies, Coralville, IA, USA). Fusion genes were

117

cloned into pUMtOR which is a tetracycline-inducible vector for M. magneticum AMB-1 27. The

118

linker is a hydrophilic polypeptide linker containing repeating amino acid chains of 4 asparagine

119

residues and 1 serine residue (linker length: 10 amino acids)28. The thrombin cleavage site was

120

designed between the linker and CohC. All prepared transformants are listed in Table S1.

121 122

2.3 Preparation of MNPs

123

Cultured cells of M. magneticum AMB-1 transformants using MSGM were collected by

124

centrifugation at 9000 g for 10 min at 4 °C. Collected cells were re-suspended in HEPES-Ca

125

(40 mL, 10 mM, supplemented with 1 mM CaCl2, pH 7.4), and disrupted by passing through a

126

French press at 1500 kg cm-2 (Ohtake Works, Tokyo, Japan). MNPs were collected from the

127

disrupted cell suspension using a columnar neodymium-boron magnet. The MNPs were washed

128

10 times with HEPES-Ca (20 mL). The concentration of MNPs in suspension was determined by

ACS Paragon Plus Environment

6

Page 7 of 23

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

Biomacromolecules

129

measuring the optical density at 660 nm using a UV-2200 spectrophotometer (Shimadzu, Kyoto,

130

Japan). A value of 1.0 corresponded to 172 µg MNPs (dry weight) per mL.

131 132

2.4 Enzyme-Linked Immunosorbent Assay (ELISA)

133

Sandwich ELISA analysis was performed for the quantification of GFP and mCherry on the

134

MNPs. Mouse anti-GFP IgG (Takara Bio, Otsu, Japan) and alkaline phosphatase (ALP)-

135

conjugated anti-GFP IgG (Rockland Immunochemicals, PA, USA), or mouse anti-mCherry IgG

136

(Takara Bio, Otsu, Japan) and ALP-rat anti-mCherry IgG (Life Technologies, CA, USA) were

137

used as primary and secondary antibodies, respectively. ALP conjugation to rat anti-mCherry

138

IgG was performed using the Alkaline Phosphatase Labeling Kit–NH2 (Dojindo, Kumamoto,

139

Japan). Primary antibodies (2 µg mL-1, 100 µL) in bicarbonate buffer (pH 9.8) were added to a

140

96 well palate (Costar 3925, Corning Inc., NY, USA) and incubated at 4 °C overnight to allow

141

sufficient time for adsorption to take place. After washing three times with PBST (phosphate-

142

buffered saline, PBS, containing 0.1% Tween-20, at pH 7.4), the plates were blocked with 0.2%

143

bovine serum albumin (BSA) in PBS for 1.5 h at room temperature. After washing with PBST,

144

the protein samples (100 µL) were added, and the plates incubated for 1 h at 37 °C. The protein

145

fractions on the MNPs were extracted according to a modified literature procedure

146

washing with PBST, secondary antibodies (1 µg mL-1 for GFP, 2 µg mL-1 for mCherry) were

147

added and incubated at 37 °C for 1 h. After washing once more with PBST, Lumi-Phos 530

148

(100 µL, Wako Pure Chemical Industries) was added as a luminescence substrate. After

149

incubation for 5 min, the luminescence intensity was measured using a microplate reader SH-

150

9000 (Corona Electric, Ibaraki, Japan). The quantities of GFP and mCherry were estimated using

29

. After

ACS Paragon Plus Environment

7

Biomacromolecules

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

Page 8 of 23

151

luminescent curves of the GFP and mCherry standards (Cell Biolabs, Inc., CA, USA). Finally,

152

the quantities of GFP (27 kDa) and mCherry (29 kDa) on a single MNP were calculated as

153

described previously28.

154 155

2.5 FRET Analysis

156

Prior to fluorescence measurements, the protein complexes were purified from MNPs by

157

thrombin digestion to avoid interference from MNP light scattering30. The MNPs (1 mg) were

158

suspended in thrombin solution (100 µL, 100 U mL-1 in PBS) and incubated at 4 °C overnight.

159

Following centrifugation at 18000 g for 1 min, the supernatants were collected and used for the

160

measurements. Fluorescence emission spectra using an excitation wavelength of 483 nm were

161

acquired between 540 and 680 nm to detect the fluorescence resonance energy transfer (FRET)

162

between GFP and mCherry using a microplate reader SH-9000 (Corona Electric, Ibaraki, Japan).

163 164

2.6 Hydrolysis of Soluble and Insoluble Celluloses using MNP-Immobilized Cellulases

165

MNPs (50 µg) were added to a substrate solution (0.5% carboxymethyl cellulose (CMC) or 1%

166

Avicel in 100 mM Tris-HCl containing 1 mM CaCl2, at pH 5.5). Following incubation at 30 °C,

167

the MNPs were removed from the suspension by magnetic separation, and the supernatant was

168

collected. The supernatant (5 µL) was added to tetrazolium blue chloride assay buffer (195 µL,

169

composition = 1 mg mL-1 tetrazolium blue chloride, 0.5 M potassium sodium tartrate, 50 mM

170

NaOH)

171

absorbance of the solutions at 704 nm was measured using a microplate reader SH-9000

31

. After boiling at 100 °C for 3 min, the solutions were cooled using an ice bath. The

ACS Paragon Plus Environment

8

Page 9 of 23

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

Biomacromolecules

172

(Corona Electric, Ibaraki, Japan). The concentrations of reducing sugars present were calculated

173

from a standard curve using glucose.

174 175

3. Results and Discussion

176

3.1 Protein Proximity Analysis of Marker Proteins on MNPs

177

Prior to preparation of the cellulase-MNP complexes, two marker proteins, namely GFP and

178

mCherry, were immobilized onto MNPs for the evaluation of spatial protein-protein proximity.

179

Immobilization was achieved using the magnetosome display system. A schematic diagram of

180

the transformants used in this section is given in Figure 1(A). In this study, the dockerin domains

181

from C. thermocellum (DocC) and R. flavefaciens (DocR) were fused to GFP and mCherry,

182

respectively. Furthermore, the cohesin domains from C. thermocellum (CohC) and R.

183

flavefaciens (CohR) were fused to mms13. The CohC and CohR were linked with a peptide

184

linker of 10 amino acids. GFP and mCherry were expected to be co-localized on magnetosomes

185

arranged in a chain through interactions between cohesin and dockerin in AMB-1/CohCR-GmC

186

(details of the M. magneticum AMB-1 transformants are given in Table S1). As expected,

187

fluorescence from both GFP and mCherry was observed at the same positions, with

188

magnetosome chains present in AMB-1/CohCR-GmC (Figure 1(B), top), although no

189

fluorescence was detected in the wild type (Figure 1(B), bottom). Furthermore, fluorescence

190

from mCherry was widely distributed in the cytoplasm of CohR-lacking AMB-1/CohC-GmC

191

(Figure 1(B), middle), indicating that two different proteins are co-immobilized onto the MNPs

192

through cohesin-dockerin interactions.

ACS Paragon Plus Environment

9

Biomacromolecules

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

Page 10 of 23

193

The spatial protein-protein proximity was then investigated by FRET analysis. The MNPs were

194

extracted from cell lysates of AMB-1 transformants by magnetic separation, followed by

195

purification of the MNP protein fractions by enzymatic cleavage, as described in the

196

Experimental Section (Section 5). Figure 2 shows the emission spectra of fluorescent proteins

197

under an excitation wavelength of 483 nm. No FRET signal was detected from the MNP/CohC-

198

GmC protein complexes (Figure 2, dashed line), as these protein complexes include only GFP. A

199

slight increase in the signal at ~610 nm was observed, even when free mCherry was added to the

200

GFP-containing protein complexes (Figure 2, gray solid line). In contrast, mCherry fluorescence

201

was detected in protein complexes purified from MNP/CohCR-GmC (Figure 2, black solid line,

202

indicated by an arrow). mCherry may therefore act efficiently as a fluorescence acceptor towards

203

GFP, which acts as a fluorescence donor in the FRET system on MNP/CohCR-GmC. FRET

204

occurs when GFP and mCherry are co-localized within 1-10 nm of each other, with the highest

205

efficiency reached between 4-6 nm

206

CohR) were linked with a peptide linker consisting of 10 amino acids. If the DocC and DocR

207

fused to fluorescence proteins were respectively bound to each scaffold protein, two fluorescence

208

proteins were located within about 4 nm corresponding to a peptide unit’s length of 0.38 nm33.

209

Therefore, this closer position led to the generation of FRET.

210

The quantities of GFP (1.3 ± 0.2 µg mg-1-MNP) and mCherry (1.4 ± 0.1 µg mg-1-MNP) present

211

on the MNPs were estimated using enzyme-linked immunosorbent assay (ELISA) analysis.

212

Based on these results, the number of GFP and mCherry molecules per single MNP were

213

calculated according to literature methods28, and found to be 33 molecules/MNP for both

214

proteins. This value was in agreement with previously reported results, i.e., 22 molecules-Protein

215

A/MNP

13

32

. The distance between two scaffold proteins (CohC and

and 20-34 molecules-Protein G/MNP23. These results indicate that stoichiometrically

ACS Paragon Plus Environment

10

Page 11 of 23

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

Biomacromolecules

216

equivalent amounts of GFP and mCherry were successfully immobilized on MNPs using the

217

magnetosome display system. This demonstrates that if desired, the number of recombinant

218

proteins present on MNPs can be altered by genetic engineering.

219 220

3.2 Hydrolysis of Carboxymethyl Cellulose (CMC) using Endoglucanase (EG) and β-

221

Glucosidase (BG) Immobilized on MNPs (EG/BG-MNPs)

222

According to GFP and mCherry immobilization conditions, endoglucanase (EG) and β-

223

glucosidase (BG) were immobilized onto MNPs via cohesin-dockerin interactions (Figure 3(A),

224

EG/BG-MNPs) for application in the hydrolysis of the soluble cellulose, carboxymethyl

225

cellulose (CMC). A schematic diagram of the transformants used in this section is given in

226

Figure 3(A). To evaluate the proximity effect of the two enzymes on CMC hydrolysis, the

227

mixture of EG-MNPs and BG-MNPs was used as a control (Figure 3(A), EG-MNPs and BG-

228

MNPs). It should be noted that equal quantities of the two enzymes were used for CMC

229

hydrolysis. When the CMC solution was treated with EG/BG-MNPs, the quantity of reducing

230

sugar increased with time, reaching a plateau at >3 h (Figure 3(B), closed circles). In contrast,

231

the hydrolysis rate of the EG-MNPs and BG-MNPs mixture was significantly slower than that of

232

EG/BG-MNPs, with the quantity of reducing sugar reaching a plateau beyond 7 h (Figure 3(B),

233

open circles). This indicates that the close proximity of EG and BG may contribute to the higher

234

hydrolysis rate in EG/BG-MNPs. The previous report revealed the higher activity of

235

immobilized three enzymes (EG, BG and exoglucanase) against CMC than that of free

236

enzymes9. The resulting concentration of reducing sugars was increased from about 3.2 mM to

237

3.9 mM. These values (3.2 to 3.9 mM) are similar level with the obtained value (1.6 mM) in this

ACS Paragon Plus Environment

11

Biomacromolecules

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

Page 12 of 23

238

study. Based on these results, we have concluded that the co-immobilization by our proposed

239

method has an advantage over using free enzymes. In general, cellulose is degraded to

240

oligosaccharides by EG, and the oligosaccharides are subsequently degraded to glucose by BG.

241

Thus, the spatial proximity of these two enzymes leads to synergistic enzyme reactions for the

242

efficient degradation of cellulose, suggesting that the efficient degradation of CMC by EG/BG-

243

MNPs may be caused by the oligosaccharides being unable to diffuse away from the reaction

244

center due to the proximity of EG and BG.

245

The reusability of the EG/BG-MNPs was then investigated (Figure 3(C)). Following hydrolysis

246

of CMC, the MNPs were magnetically collected, and reused in further CMC hydrolysis cycles

247

under the same conditions. The hydrolysis activity in the first trial was defined as 100%, and the

248

activity of EG/BG-MNPs was maintained at >70% after five trials. Previous reports using

249

chemically conjugated cellulases and magnetic particles indicate that enzyme activity dropped to

250

20-50% after five trials, even with the use of thermostable cellulases and optimal conditions for

251

CMC hydrolysis

252

system show good performance in terms of cellulase reusability. The slight decrease in

253

enzymatic activity could be explained by two enzymes being released from or inactivated on

254

MNPs during the hydrolysis process. It is also possible that the amount of MNPs used gradually

255

decreased during the magnetic separation process.

2, 5

. This suggests that the MNPs prepared using the magnetosome display

256 257

3.3 Effect of the Cellulose-Binding Domain (CBD) on Hydrolysis of Microcrystalline

258

Cellulose

ACS Paragon Plus Environment

12

Page 13 of 23

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

Biomacromolecules

259

The EG/BG-MNPs were applied in the hydrolysis of the insoluble cellulose, Avicel, but little

260

hydrolysis was detected (Figure 4(B)). In order to improve the hydrolysis activity, the cellulose-

261

binding domain (CBD) was fused to the end of the scaffold protein attached to the MNPs (Figure

262

4(A), EG/BG-CBD-MNPs). The CBD plays an important role in the binding of cellulosome to

263

the crystalline cellulose34-36. When compared to the EG/BG-MNPs, the production of reduced

264

sugar was significantly enhanced when EG/BG-CBD-MNPs were used (Figure 4(B)). The

265

binding of the CBD to microcrystalline cellulose may promote the hydrolysis of EG and BG on

266

MNPs. Although the hydrolysis rate of Avicel was approximately 10 times lower than that of

267

CMC (Figure 3(B)), this tendency correlates with previous results.

268 269

4. Conclusions

270

We reported the construction of highly organized protein complexes on magnetic nanoparticles

271

(MNPs) using the magnetosome display system. A proof-of-concept study using green

272

fluorescent protein (GFP) and mCherry revealed the close proximity (