Engineering and Production of Glucooligosaccharide

Oxidases for Site-specific Activation of Cellulose and

Hemicellulose Substrates

by

Maryam Foumani Alhaeri

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy

Graduate Department of Chemical Engineering and Applied Chemistry

University of Toronto

© Copyright by Maryam Foumani Alhaeri 2015

ii

Engineering and Production of Glucooligosaccharide Oxidases for

Site-specific Activation of Cellulose and Hemicellulose Substrates

Maryam Foumani Alhaeri Doctor of Philosophy

Department of Chemical Engineering and Applied Chemistry

University of Toronto

2015

Abstract

Canada has an extensive supply of residual biomass, which comprises plant

polysaccharides that represent a renewable resource for the production of biochemicals,

polymers and fuels. Enzymatic oxidation of plant oligo- and poly-saccharides can alter

the characteristics of these compounds, enhancing their application in food products;

enzymatic oxidation could also facilitate site-specific chemical derivatizations of

carbohydrates, leading to new bio-based polymers.

A glucooligosaccharide oxidase from Sarocladium strictum (GOOX) with reported

activity on oligosaccharides comprising up to seven glucose units, was engineered in this

study using a genetic engineering approach to extend the activity of this enzyme on a

wider range of plant polysaccharides and oligosaccharides. For the first time, in addition

to the previously reported cello- and malto-oligosaccharides, activity of wild type GOOX

on xylooligosaccharides was reported. The catalytic efficiency of two mutant enzymes

(GOOX-Y300A, GOOX-Y300N) on cellooligosaccharides and xylooligosaccharides was

2-fold higher than the wild-type enzyme. Notably, the binding affinity of these mutants

iii

towards oligosaccharides had decreased, which was correlated to reduced substrate

inhibition by cellooligosaccharides. The GOOX-W351F mutant showed enhanced

activity and affinity on galactose compared to wild-type GOOX, suggesting that this

mutation reduces steric hindrance between galactose and substrate binding amino acids

within the substrate binding cleft of the enzyme.

Wild type GOOX exhibited low activity on polysaccharides including konjac

glucomannan, and barley β-glucan, and weak activities on carboxy-methyl cellulose,

regenerated amorphous cellulose, microcrystalline cellulose (Avicel), and xyloglucan.

The specific activity was improved by up to 56 %, 55 % and 30 % for crystalline

cellulose (Avicel), regenerated amorphous cellulose (RAC) and glucomannan,

respectively, by constructing fusions between GOOX and various carbohydrate binding

modules (CBMs). Binding capacities of the fusion proteins on crystalline and amorphous

cellulose, as well as glucomannan, β-glucan, and xyloglucan also increased by over 10-

fold as determined by SDS-polyacrylamide gel electrophoresis and affinity gel

electrophoresis. The immobilized fusion enzyme on a solid cellulose surface remained

stable and active. This finding is anticipated to broaden applications of GOOX as an

immobilized enzyme used in cellulose-based biosensing devices.

In addition to enzyme engineering and biochemical characterizations, three applications

of GOOX enzymes were evaluated: 1) production of oxidized oligosaccharides as sugar

standards, 2) synthesis of plant oligosaccharides and polysaccharides with enhanced

nutraceutical value, and 3) substitution of glucose oxidase used in baking applications.

iv

Acknowledgments

I would like to express my deepest and sincere gratitude to Prof. Emma R. Master for her

kind and firm supervision, motivations for taking unpredictable science routes, and her

continuous supports all through this exciting journey. Her dedication to provide an

energetic environment for students to explore and discover new findings and enthusiasm

towards research never cease to amaze me.

I am also very grateful to all the members of the Master lab for being excellent labmates,

making my graduate study a rewarding and enjoyable experience. In particular, Dr. Thu

Vuong for his kind assistance and guidance toward my research, and Dr. Dragica

Jeremic, Julie-Anne Gandier, Ruoyu Yan, and Mabel Wong for being great friends and

emotionally supportive throughout my study. I am also thankful to all the BioZone

members especially Endang Susilawati (Susie), Melanie Duhamel and Angelika Duffy

for their supports.

I am very thankful to my committee members Professor Bradley Saville and Professor

Ning Yan for their valuable feedback and insightful inputs to my research throughout the

course of my PhD.

I would like to extend my heartfelt gratitude to my parents, Nasrin Moeini and Mostafa

Foumani for their inspiration, support and unconditional love throughout my life, and to

my husband, Mohammad Reza, for his gracious patience, considerate guidance, and

never-ending kindness and support. I am especially thankful to my 5-year old daughter,

Parnian, who showed an exceptional independence towards the end of my work while

caring for her little sister, Parimah, in my absence at the family times. I am also grateful

for my siblings, Fatemeh, Javad, Mohammad, and Mahdi for being the kindest and most

supportive siblings I could ever know. Finally, I am thankful to my wonderfull friends

and neigbours, especially Dr. Fatemeh Akbarian, Lili Zahedi, Hoda Mofidi and Zahra

Choolaei for being by my side through the difficult times and energizing those moments.

v

Table of Contents

Table of Contents ............................................................................. v

List of Tables .................................................................................. ix

List of Figures .................................................................................. x

List of Appendices ......................................................................... xii

List of Abbreviations .................................................................... xiii

Chapter 1 : Overview ....................................................................... 1

Chapter 2 : Literature review ........................................................... 3 2.1 Oxidation of sugars and polysaccharides ...............................................................................3 2.2 Examples of oxidoreductases from selected Auxiliary Activity (AA) families .....................4 2.3 Oligosaccharide oxidases within AA7 ...................................................................................8

2.3.1 Oligosaccharide oxidase from Microdochium nivale, MnCO .........................................8 2.3.2 Cello-oligosaccharide oxidase from Paraconiothyrium sp., PCOX ................................9 2.3.3 Chitooligosaccharide oxidase from Fusarium graminearum, ChitO ............................10 2.3.4 Gluco-oligosaccharide oxidases from Sarocladium strictum T1, GOOX-T1 ...............10

2.4 Current strategies for enzyme engineering ...........................................................................13 2.4.1 Mutagenesis: Random versus rational design................................................................13 2.4.2 Gene shuffling ...............................................................................................................15 2.4.3 Fusion proteins ..............................................................................................................15

2.5 Carbohydrate binding modules .............................................................................................16 2.5.1 Contributions in cellulose/hemicellulose active enzymes .............................................17 2.5.2 Industrial applications of CBMs ....................................................................................18 2.5.3 Effects of CBMs on neighbouring modules ..................................................................18 2.5.4 Important factors for designing CBM fusion proteins ...................................................20

2.6 Applications of carbohydrate oxidase in food industry ........................................................21 2.6.1 Oxidoreductases in baking applications ........................................................................21 2.6.2 Enzyme utilization for production of prebiotics ............................................................24

2.7 Research Hypotheses and Specific Research Objectives .....................................................26

vi

Chapter 3 Improved GOOX activity on mono and oligosaccharides

using site-directed mutagenesis ..................................................... 29 3.1 Introduction ..........................................................................................................................30 3.2 Materials and Methods .........................................................................................................32

3.2.1 Fungal strain and materials ............................................................................................32 3.2.2 Cloning of the GOOX-encoding gene ...........................................................................32 3.2.3 Site-directed mutagenesis ..............................................................................................33 3.2.4 Recombinant protein expression ...................................................................................34 3.2.5 Enzyme purification ......................................................................................................34 3.2.6 Enzymatic assays and kinetics analyses ........................................................................35 3.2.7 Deglycosylation .............................................................................................................36 3.2.8 Substrate docking ..........................................................................................................36 3.2.8 Nucleotide sequence accession number.........................................................................37

3.3 Results and Discussion .........................................................................................................37 3.3.1 Variations of GOOX ......................................................................................................37 3.3.2 Production of recombinant protein ................................................................................39 3.3.3 Novel substrate specificity.............................................................................................40 3.3.4 Improvement of substrate specificity ............................................................................45

3.4 Conclusions ..........................................................................................................................47

Chapter 4 Enhanced binding and activity of GOOX towards

polysaccharides through CBM fusions .......................................... 50 4.1 Introduction ..........................................................................................................................51 4.2 Materials and Methods .........................................................................................................53

4.2.1 Materials ........................................................................................................................53 4.2.2 Construction of fusion enzymes ....................................................................................53 4.2.3 Recombinant expression of fusion proteins in Pichia pastoris .....................................54 4.2.4 Purification of recombinant enzymes ............................................................................55 4.2.5 Specific activity on oligosaccharides, soluble polysaccharides and insoluble cellulose

substrates ................................................................................................................................56 4.2.6 Cellulose binding ...........................................................................................................57 4.2.7 Quartz crystal microbalance with dissipation (QCM-D) ...............................................58 4.2.8 Affinity gel electrophoresis ...........................................................................................59

vii

4.2.9 Temperature stability .....................................................................................................59 4.2.10 Nucleotide sequence accession number.......................................................................60

4.3 Results and Discussion .........................................................................................................60 4.3.1 Recombinant protein production ...................................................................................60 4.3.2 Improved binding to polymeric substrates ....................................................................63 4.3.4 Specific activity on polymeric substrates ......................................................................67 4.3.5 Immobilization of GOOX through CtCBM3 .................................................................70 4.3.6 Effect of CBM on thermostability .................................................................................72

4.4 Conclusions ..........................................................................................................................74

Chapter 5 Application trials of wild-type and engineered GOOX 75 5.1 GOOX in the production of sugar standards ........................................................................76

5.2.1 Introduction ...................................................................................................................76 5.2.2 Materials and Methods ..................................................................................................77

5.2.2.1 NMR analysis of oxidized products ..................................................................................... 77 5.2.2.2 Mass spectrometric analysis of oxidized products ............................................................... 78

5.2.3 Results and Discussion ..................................................................................................79 5.2.3.1 Confirming the regioselectivity of gluco-oligosaccharide oxidases ..................................... 79 5.2.3.2 Efficient oxidation of cellooligos. and Impact of chain length on GOOX activity .............. 82

5.2 Application of GOOX CBM fusions in the synthesis of plant oligosaccharides with

enhanced nutraceutical value ......................................................................................................84 5.2.1 Introduction ...................................................................................................................84 5.2.2 Materials and Methods ..................................................................................................86

5.2.2.1 Oxidation of xylooligosaccharides ....................................................................................... 86 5.2.2.2 Prebiotic Assay ..................................................................................................................... 87

5.2.3 Results and Discussion ..................................................................................................88 5.2.3.1 Small-scale fermentation on xylooligosaccharides and aldouronic acid .............................. 88 5.2.3.2 Cultivation of B. longum on oxidized and non-oxidized xylooligosaccharides ................... 89

5.2.4 Conclusions ...................................................................................................................91 5.3 A Mutant gluco-oligosaccharide oxidase is suitable to replace glucose oxidase for baking

applications .................................................................................................................................92 5.3.1 Introduction ...................................................................................................................92 5.3.2 Materials and Methods ..................................................................................................96

5.3.2.1 Activity assays ...................................................................................................................... 96 5.3.2.2 Oxidation Reactions for H2O2 inactivation study ................................................................. 97

viii

5.3.2.3 Detection of gluconic acid by HPLC .................................................................................... 97 5.3.2.4 Monitoring the oxygen content using Oxygraph .................................................................. 98

5.3.3 Results and Discussion ..................................................................................................98 5.3.3.1 GOOX-Y300A shows higher oxidation of oligosaccharides ................................................ 98 5.3.3.2 H2O2 inactivation .................................................................................................................. 99

5.3.4 Conclusions .................................................................................................................102

Chapter 6 : Conclusions ............................................................... 103

Chapter 7 : Future directions ....................................................... 109 7.1 Incorportaion of GOOX-oxidized oligosaccharides in an LPMO standard assay. ............109 7.2 Effect of debranching enzyme on prebiotic activity of polysaccharides ............................109 7.3 Additional value of a CBM fusion GOOX-Y300A for baking application ........................110

References .................................................................................... 112

Appendix 1: Supplemental information for chapter 3 ................. 126

Appendix 2: Supplemental information for chapter 4 ................. 128

Appendix 3: Supplemental information for chapter 5 ................. 131

ix

List of Tables

Table 2.1 Brief reviews of common enzyme engineering strategies ..............................................16

Table 3.1 List of oligonucleotide primers used for gene amplification and site-directed

mutagenesis. ...........................................................................................................................33

Table 3.2 Amino acid substitutions in GOOX in comparison with GOOX-T1 .............................38

Table 3.3 The effect of deglycosylation with PNGaseF on enzyme activity .................................40

Table 3.4 Kinetics parameters of wild-type and mutant GOOX enzymes. ....................................41

Table 3.5 Docking parameters of oligosaccharides with GOOX enzymes. ...................................44

Table 4.1 Specific activity of wild-type and CBM fusion GOOX on oligosaccharides. ................62

Table 4.2 Binding of wild-type GOOX and CBM fusions to insoluble cellulose. .........................63

Table 4.3 Kinetics parameters of wild-type GOOX and its CBM fusions on cellotetraose. ..........66

Table 4.4 Specific activity of the wild-type GOOX and CBM fusions on polysaccharides. .........67

Table 4.5 The half life of fusion and wild-type GOOX at 45°C. ...................................................73

Table 5.1 Growth rate of B. longum cultivations. ..........................................................................90

Table 5.2 Specific activities of GOOX-Y300A and GO on selected mono and oligosaccharides. 99

Table 5.3 Amount of Gluconic acid produced by GOOX-Y300A or GO in the presence of various

concentrations of H2O2. ........................................................................................................101

x

List of Figures

Figure 2.1 Oxidation mechanism of GOOX ...................................................................................11

Figure 2.2 GOOX substrate binding and active sites .....................................................................12

Figure 3.1 The structural model of GOOX.....................................................................................31

Figure 3.2 Conformational changes of S388 upon substrate binding .............................................43

Figure 3.3 Docking of monosaccharides to GOOX .......................................................................47

Figure 3.4 The biding site for GOOX-T1. ......................................................................................49

Figure 4.1 Schematic representation of wild-type GOOX and GOOX fusions..............................62

Figure 4.2 Affinity gel electrophoresis (AGE) of wild-type GOOX and CBM fusions .................64

Figure 4.3 Specific activity of wild-type GOOX and CBM fusions on polysaccharides.. .............69

Figure 4.4 Frequency - dissipation plot of enzymes binding to cellulose ......................................71

Figure 5.1 NMR spectra of cellobiose and xylobiose oxidation ....................................................80

Figure 5.2 MS/MS spectra and fragmentation of GOOX oxidized cellotriose ..............................81

Figure 5.3 Positive ion ESI-MS spectra of four cello-oligosaccharide samples ............................83

Figure 5.4 Structure of compounds used in prebiotic assay ...........................................................85

Figure 5.5 Viable cell count of B. longum preliminary cultures ....................................................88

Figure 5.6 Viable cell count of B. longum cultures ........................................................................90

Figure 5.7 Prosed mechanisms for GOOX benefits in baking applications. ..................................95

Figure 5.8 Proposed mechanism for GOOX reinforcing the protein network in dough ................95

Figure 5.9 H2O2 tolerance of GOOX-Y300A and GO .................................................................101

Figure S3.1 Multiple sequence alignment of GOOX homologues…………………………….. 126

Figure S3.2 Stability of wild-type and mutant GOOX at 37°C………………………………....127

Figure S3.3 SDS-PAGE of deglycosylated GOOX-VN and mutant enzymes………………….127

Figure S4.1 Purified wild-type and fusion GOOX proteins on 10% SDS-PAGE……………....128

Figure S4.2 Binding of wild-type GOOX and CBM fusions to insoluble cellulose…….........…128

xi

Figure S4.3 Specific activity of wild-type GOOX and CBM fusions on konjac glucomannan...129

Figure S4.4 Adsorbed mass of GOOX and CtCBM3_GOOX on cellulose-coated sensors ……129

Figure S4.5 Cellobiose oxidation of enzyme-bound sensors……………………………………130

Figure S4.6 Thermostability of proteins at 45°C………………………………………………..130

Figure S5.1 Behaviour of GOOX-Y300A in the presence or absence of H2O2………….….......132

Figure S5.2 - Log of GluA concentration produced versus H2O2 concentration………………..132

xii

List of Appendices

Appendix 1: Supplemental information for chapter 3 Appendix 2: Supplemental information for chapter 4 Appendix 3: Supplemental information for chapter 5

xiii

List of Abbreviations

AA7/9- Carbohydrate active enzymes with Auxiliary Activity from family 7 or 9 BMGY- Buffered complex medium containing glycerol BMMH- Buffered minimal medium containing methanol and histidine BSA- Bovine serum albumin CAZy- Carbohydrate active enzyme database CBM- Carbohydrate binding module CDH- Cellobiose dehydrogenase CelK- Clostridium thermocellum cellobiohydrolase ChitO- Chitooligosaccharide oxidase from Fusarium graminearum CMC- Carboxymethyl cellulose DP- Degree of polymerization FAD- Flavin adenine dinucleotide GH61- Glycosyl hydrolase from family 61 (re-named to AA9) GluM- Glucomannan from Konjac GO- Glucose oxidase from Aspergillus niger HPAEC- High-performance anion-exchange chromatography HRP- Horseradish peroxidase GOOX- Glucooligosaccharide oxidase from Sarocladium strictum strain CBS 346.70 GOOX-T1- Glucooligosaccharide oxidase from Sarocladium strictum strain T1 LPMO- Lytic polysaccharide monooxygenase MnCO- Carbohydrate oxidase from Microdochium nivale MRS- Bacterial growth medium so-named by its inventors: de Man, Rogosa and Sharpe NAG- N-acetyl-glucosamine NMR- Nuclear magnetic resonance PCOX- Cello-oligosaccharide oxidase from Paraconiothyrium sp. PDB- protein database of The Research Collaboratory for Structural Bioinformatics PI- Prebiotic index POX- Pyranose oxidase QCM-D- Quartz crystal microbalance with dissipation RAC- Regenerated amorphous cellulose SDS-PAGE- Sodium dodecyl sulfate polyacrylamide gel electrophoresis TCAG- Center for Applied Genomics TEMPO- 2,2,6,6-tetramethylpiperidine-1-oxyl XOS- Xylooligosaccharide FAEXynZ- Feruloyl esterase domain of a xylanase from Clostridium thermocellum YNB- Yeast nitrogen base without amino acids 4-AA- 4-aminoantipyrine

1

Chapter 1 : Overview

Production of biochemicals and functionalized fibres for value-added bio-products from

biomass can help to offset the price of biofuel while providing sustainable, and

environmentally friendly replacements for petroleum based materials. Enzymatic

modification of the plant oligo- and polysaccharides of biomass can alter the

characteristics of these materials to meet the requirements for the downstream

applications (e.g. reactivity, solubility, compatibility with other biopolymers). By using

an enzymatic approach, it is also possible to catalyze regio-selective and stereo-specific

modifications, while retaining the degree of polymerization and/or crystallinity of the

substrate. Moreover, routine molecular biology techniques can be applied to fine-tune

enzyme activities to broaden substrate range and increase enzyme stability. Lastly, since

enzyme reactions are typically performed in aqueous solutions at intermediate pH and

temperatures below 80°C, they represent a class of “green” catalysts with minimal safety

considerations.

Following a summary of my scholarly contributions below, Chapter 2 will review the

main literature and concepts pertinent to my PhD thesis, and will end by stating my

specific research hypotheses and objectives.

Summary of Scholarly Contributions

Peer-reviewed Publications

2

Foumani M, Vuong TV, and Master ER. Altered substrate specificity of the gluco-

oligosaccharide oxidase from Acremonium strictum, Biotechnology and Bioengineering.

2011, 108(10): 2261-2269.

Foumani M, Vuong TV, and Master ER. Oligosaccharide oxidase derived from

Acremonium strictum and uses thereof, Patent WO 2012/116431 A1, 2012.

Vuong T, Vesterinen A, Foumani M, Juvonen M, Seppälä J, Tenkanen M, Master ER.

Xylo- and cello-oligosaccharide oxidation by gluco-oligosaccharide oxidase from

Sarocladium strictum and variants with reduced substrate inhibition. Biotechnology for

Biofuels. 2013, 6: 148.

Foumani M, Vuong TV, MacCormick B, Master ER. Enhanced polysaccharide binding

and activity on linear β-glucans through addition of carbohydrate-binding modules to

either terminus of a glucooligosaccharide oxidase. PLOS ONE J. Accepted.

Manuscripts in Preparation

Vuong TV, MacCormick B, Master ER, Foumani M. Gluco-oligosaccharide oxidase

variants as suitable substitutes to glucose oxidase for baking applications.

Anticipated Manuscript

Vuong TV, Foumani M, Gudmundsson M, Master ER. Assessing enzymatic oxidation of

cellulosic substrates through XPS and fluorescence detection of carboxylic functionality.

3

Chapter 2 : Literature review

2.1 Oxidation of sugars and polysaccharides

Oxidation of hydroxyl groups to carbonyls can enhance the gelation, antiflocculation,

adhesion, thickening, and metal sequestration potential of polysaccharides (de Nooy et al.

1997; da Silva Perez et al. 2003). It can also alter the rheology of corresponding

polymers, and be performed as an initial step to subsequent esterification or amination of

hydroxyl groups. Pursuant to these objectives, chemicals such as 2,2,6,6-

tetramethylpiperidine-1-oxyl (TEMPO) have been used to oxidize primary hydroxyl

groups to uronic acids (Isogai and Kato 1998; da Silva Perez et al. 2003; Ciriminna and

Pagliaro 2010). Sodium periodate has also been used to oxidize C2 and C3 positions of

cyclic sugars, thereby introducing dialdehydes into polysaccharides (Kristiansen et al.

2010), whereas halide ions including I- and Br- have been used to further oxidize

aldehydes at positions C1 and C6 to aldonic acids (Diehl et al. 1974; Parikka et al. 2012).

However, chemical methods can compromise the polymerization and/or crystallinity of

the starting material, which is problematic when derivatizing oligosaccharide and

nanocrystalline substrates (Isogai et al. 2009; Saito et al. 2010).

Alternatively, carbohydrate oxidases can facilitate regio-selective oxidation of highly

functionalized carbohydrates without arduous protection/deprotection steps. Mild

reaction requirements also mean that loss in the degree of polymerization and

crystallinity of oligo- and poly-saccharide substrates can be minimized. Carbohydrate

oxidases (EC 1.1.3) can catalyze the oxidation of the primary hydroxyl (C6), secondary

hydroxyls (C2, C3 or C4) or anomeric carbon hydroxyl (C1) to an aldehyde, ketone or a

4

lactone (then carboxylic acid), respectively, with concomitant reduction of molecular

oxygen to H2O2 (van Hellemond et al. 2006). These enzymes were recently categorized

as auxiliary activities in the carbohydrate-active enzyme database (CAZy; Levasseur et

al. 2013). In the following paragraphs some of the enzymes from AA families will be

briefly reviewed.

2.2 Examples of oxidoreductases from selected Auxiliary Activity (AA) families

The most versatile branch of AA enzymes is perhaps AA3, which contains enzymes

belonging to the glucose-methanol-choline (GMC) oxidoreductase superfamily. These are

FAD containing enzymes include cellobiose dehydrogenase (CDH, EC 1.1.99.18,

AA3_1), a hemoflavoenzyme, oxidizing cellobiose and higher cellodextrin at the

anomeric position. The occurrence of CDH in wood degrading fungi, as well as oxidation

mechanism and structure-functional relationships, have been thoroughly reviewed for

CDH enzymes (Henriksson et al. 2000; Zamocky et al. 2006). Recent evidence suggests

that CDHs are physiological partners for polysaccharide monooxygenases, playing

important roles in oxidative cellulose decomposition (Langston et al. 2011; Phillips et al.

2011). From an applied perspective, the electron transfer by the CDH cytochrome domain

as well as its electrochemical properties make it suitable in biosensors and enzymatic

biofuel cell applications (Ludwig et al 2013).

The model enzyme for the GMC superfamily, glucose-1-oxidase (GO, EC 1.1.3.4,

AA3_2), oxidizes the anomeric hydroxyl of glucose, yielding a lactone that can be

hydrolyzed to form the corresponding acid (van Hellemond et al. 2006). GO has been

studied in detail given its importance in diagnostic reagents, biosensors, baking, and other

5

applications (Bankar et al. 2009). Unlike GO, aryl alcohol oxidase (AaO, EC 1.1.3.7,

AA3_2) is an extracellular flavoenzyme which is thought to be involved in fungal

degradation of lignin, providing H2O2 required for ligninolytic peroxidases (Hernández-

Ortega et al. 2012). Phylogenetic analyses of GMC oxidoreducases reveal that close to

the AaO cluster are pyranose dehydrogenases. Pyranose dehydrogenases (PDH, EC

1.1.99.29, AA3_2) have broader substrate specificity and regioselectivity than GO,

catalyzing the oxidation of C1, C2, or C3 hydroxyls of mono di- and tri-saccharides to

form the corresponding lactone or keto sugars (Peterbauer and Volc, 2010).

Unlike PDH, pyranose 2-oxidases (POX, EC 1.1.3.10, AA3_4) target C2/3 hydroxyls

(Giffhorn 2000; Kujawa et al. 2006) and can utilize molecular oxygen as an electron

acceptor. POX shows a hydride transfer mechanism for FAD reduction with a conserved

histidine as the catalytic base for deprotonation of the substrate; a feature that is also

conserved among all the above-mentioned GMC oxidoreducases (Wongnate et al. 2013).

The crystal structure of POX reveals a size exclusion mechanism for substrate binding

(Hallberg et al. 2004) similar to that observed in GO (Wohlfahrt et al. 1999). As a result,

the application of these enzymes is likely limited to the oxidation of mono- and di-

saccharides.

Whereas AA3 enzymes are flavoproteins, carbohydrate oxidases in family AA5 are

copper radical oxidases. Family AA5 comprises two subfamilies, the glyoxal oxidases in

subfamily AA5_1 and galactose oxidases in subfamily AA5_2. Rather than oxidizing

hydroxyl groups at C1, C2 or C3 positions, galactose 6-oxidase (GaOx, EC 1.1.3.9,

AA5_2) oxidizes the primary C6 hydroxyl of galactose along with galactose containing

oligosaccharides and polysaccharides (Whittaker 2005). The biochemistry and production

6

of GaOx has been studied in detail (Whittaker 2005; Spaduit et al. 2010) and extensive

analyses of its structure have revealed a comparatively shallow active site, explaining the

activity of this enzyme on galactopyranosyl units of galactoglucomannans, in addition to

monosaccharides (D/L-galactose) and oligosaccharides with terminal galactopyranosyl

units (Firbank et al. 2001; Parikka and Tenkanen 2009; Parikka et al. 2010). Studies with

GaOx reveal that a key benefit to enzymatic oxidation of polysaccharides is regio-

selectivity along with reduced loss in the degree of polymerization of oligomeric and

polymeric substrates (Parikka et al. 2010; Parikka et al. 2012).

Comparatively new classes of AA enzymes that act on polysaccharides include the lytic

polysaccharide monooxygenases (LPMOs), which are now classified as AA families 9,

10, 11, and 13 in the CAZy database. LPMOs are copper dependent enzymes involved in

oxidative cleavage of polysaccharides resulting in corresponding oxidized

oligosaccharides. These enzymes play a key role in lignocellulose degradation as they are

found in the genome of most plant cell wall degrading fungi (Morgenstern et al. 2014). In

addition, they show a synergistic boosting effect with hydrolytic enzymes and CDH,

however these latter enzymes are not necessary for the action of LPMOs (Dimarogona et

al. 2013; Vaaje-Kolstad et al. 2013).

Whereas fungal LPMOs from family AA9 (formerly GH61) target crystalline cellulose as

well as hemicellulose (Agger et al. 2014), bacterial LPMOs from family AA10 (formerly

CBM33) that have been characterized to date are selective towards cellulose and chitin

(Book et al. 2014). LPMOs from AA9 and AA10 families share a conserved ß-sandwich

fold as well as conserved histidine residues involved in copper binding within the

substrate binding site (Hemsworth et al. 2013; Book et al. 2014). However, studies on

7

the surface electrostatic potential of enzymes from families AA9 and AA10 reveal

significant differences in charge distribution particularly within the substrate binding

region (Book et al. 2014). While the anomeric C1 is the most favourable oxidation site

for LPMOs in general, AA9 enzymes have been also reported to oxidize hydroxyl groups

at the C4 position (Isaken et al. 2014).

Most recently, two additional LPMO families were discovered, namely AA11

(Hemsworth et al. 2014) and AA13 (Vu et al. 2014), which are specific towards chitin

and starch, respectively. These new classes share the conserved histidine brace for copper

binding in the active site and both require an electron donor such as ascorbic acid or

CDH. Similar to previously discovered LPMOs, C1 hydroxyl groups are oxidized by

AA11 and AA13 enzymes, although action of AA11 enzymes at the non-reducing C4 has

also been reported (Hemsworth et al. 2014).

Because LPMOs cleave glycosidic linkages in targeted polysaccharide, their action

ultimately reduces the degree of polymerization of the starting oligo- and

polysaccharides. By contrast, oligosaccharide oxidases from family AA7 oxidize the

anomeric carbon of existing reducing ends in oligomeric and polymeric substrates.

Despite the significance of the AA7 enzymes for carbohydrate oxidation, this enzyme

family is comparatively less well characterized. Given the importance and relevance of

this enzyme family to the context and scope of the present study, these enzymes will be

reviewed in detail in the following section.

8

2.3 Oligosaccharide oxidases within AA7

Oligosaccharide oxidases from family AA7 belong to the growing vanillyl-alcohol

oxidase (VAO) flavoenzyme family (Leferink et al. 2008), targeting the C1 hydroxyl of a

broad range of oligosaccharides, including cello-, -xylo-, and malto-oligosaccharides.

Examples of characterized AA7 enzymes include a carbohydrate oxidase from

Microdochium nivale (MnCO) (Xu et al. 2001), a cello-oligosaccharide oxidase from

Paraconiothyrium sp. (PCOX) (Kiryu et al. 2008), a chito-oligosaccharide oxidase from

Fusarium graminearum (ChitO) (Heuts et al. 2007), and a gluco-oligosaccharide oxidase

(EC 1.1.3.-) from Sarocladium strictum T1 (GOOX-T1) (Lin et al. 1991; Lee et al. 2005).

The above-mentioned enzymes are flavoproteins, with unique bi-covalent linkages to the

flavin adenine dinucleotide (FAD) cofactor, providing a relatively high redox potential

for these enzymes. Like other flavin carbohydrate oxidases that target the anomeric

carbon hydroxyl (C1), oligosaccharide oxidases are thought to mediate oxidoreductase

activity through two half-reactions: 1) oxidation of the reducing sugar to the

corresponding lactone, then 2) spontaneous hydrolysis of the lactone product to the

corresponding acid (Huang et al. 2005; van Hellemond et al. 2006).

2.3.1 Oligosaccharide oxidase from Microdochium nivale, MnCO

The gene encoding MnCO contains one intron and the coding region shows low

similarity to other FAD-containing carbohydrate oxidases other than AA7 family

enzymes (Xu et al. 2001). MnCO was recombinantly expressed in Aspergillus oryzae and

was well characterized in terms of specificity and kinetics parameters (Kulys et al. 2001a;

Xu et al. 2001). The crystal structure of the enzyme was also solved by Duskova et al.

9

(2009). Among the many different oligosaccharides, MnCO prefers tetrameric dextrins,

revealing that four α-(1→4) linked glucose units make a favourable interaction with the

substrate-binding pocket. Notably, this enzyme was the first reported AA7 enzyme that

can oxidize polysaccharides, including starch and carboxymethyl cellulose with 0.8 %

and 9 % activity relative to cellobiose, respectively (Xu et al. 2001). Compared to glucose

oxidase, the reactivity of this enzyme is lower on glucose, however the substrate

specificity of MnCO is broader towards mono- and di-saccharides (Kulys et al. 2001a).

Thus, this enzyme has been tested to replace glucose oxidase in biosensors (Kulys et al.

2001b) and baking applications (Schneider et al. 2003).

2.3.2 Cello-oligosaccharide oxidase from Paraconiothyrium sp., PCOX

PCOX efficiently oxidizes β-(1→4) linked sugars, such as cellooligosaccharides,

xylobiose and lactose at an optimal pH of 5.5 (Kiryu et al. 2008). Like MnCO, PCOX is

active on a wide range of sugar types ranging from glucose and galactose to xylose and

arabinose. Moreover, the results of Kiryu et al. (2008) indicate that the enzyme is capable

of oxidizing xylooligosaccharides in addition to cellooligosaccharides. This enzyme has

been mainly used to produce lactobionate, a lactose derivative with potential

nutraceutical benefits including prebiotic activity (Murakami et al. 2008). Despite the

wide substrate specificity of this enzyme, very little information is available from the

genetic perspective. Neither the nucleotide sequence of the gene encoding PCOX, nor the

amino acid sequence of this protein has been published, which limits our ability to study

this protein based on sequence homology to other enzymes.

10

2.3.3 Chitooligosaccharide oxidase from Fusarium graminearum, ChitO

ChitO catalyzes the oxidation of hydroxyl moieties at the C1 position of

chitooligosaccharides. Mutagenesis studies using the ChitO encoding gene reveal that the

conversion of the active site residue Q268 to R268 affects the recognition of N-acetyl

groups present on chitooligosaccharide substrates (Heuts et al. 2007). Interestingly, the

corresponding amino acid position in most other AA7 oligosaccharide oxidases,

including GOOX-T1, is arginine. Accordingly, arginine at this position is predicted to

hinder the formation of favourable interactions between branched molecules such as

glucosamine with GOOX-T1 active sites (Heuts et al. 2007). More recently, Ferrari et al.

(2015) created a set of ChitO variants with completely different substrate tolerance than

the wild-type enzyme. By combining single mutants with altered substrate preference, a

variant with activity towards lactose, cellobiose and maltose was generated although with

lower oxidation efficiency than those of GOOX and MnCO on these substrates. Notably

the engineered ChitO retained 40% of its catalytic efficiency towards

chitooligosaccharides.

2.3.4 Gluco-oligosaccharide oxidases from Sarocladium strictum T1, GOOX-T1

The GOOX-T1 gene consists of 1500-bp of coding sequence and one short 53-bp intron.

The deduced protein sequence is 499 amino acids and the predicted protein has a

molecular mass of 55.2 kDa. The predicted molecular weight of GOOX-T1 is lower than

the reported 61 kDa that was obtained from size exclusion chromatography. This

discrepancy is consistent with the N-glycosylation detected at Asn341 and Asn305

(Huang et al. 2005). Indeed, post-translational modification of GOOX-T1 might be

11

required for functional protein folding since recombinant expression of this enzyme in

E.coli was not successful; it is also possible that E. coli does not support complete

incorporation of the required FAD cofactor. By contrast, the recombinant GOOX

expression has been demonstrated in Pichia pastoris with yields reported as high as 300

mg per liter of cultivation medium (Lee et al. 2005).

GOOX-T1 is shown to function best at 37 °C and pH 8, and remain stable up to 50 °C. It

also shows a wide range of pH stability from pH 4 to pH 12 (Lin et al. 1991; Fan et al.

2000). The impact of temperature and pH on GOOX-T1 activity was studied extensively

using cello- and malto-oligosaccharides (Fan et al. 2000). In their study, Fan et al (2000)

revealed that the oxidation of maltose was highest between pH 9 to 10.5 and the Km of

this reaction was also highest at pH 10. Fan et al. (2000) also demonstrated that the

activation energy of GOOX is similar at pH 7 and pH 10, suggesting that the catalytic

mechanism of GOOX is retained within this pH range.

Figure 2.1 Oxidation mechanism of GOOX. The tyrosine residue (Y429) is a catalytic base and along with aspartic acid (D355) initiates the hydride transfer from the substrate, e.g. glucose to reduce the FAD cofactor, which is bi-covalently linked to GOOX through a cysteine (C130) and a histidine (H70) residue. The image was generated using ChemSketch.

12

Among tested carbohydrates and derivatives, GOOX-T1 oxidizes both α-linked and β-

linked glucose substrates, including lactose, malto-oligosaccharides and cello-

oligosaccharides (Lin et al. 1991; Fan et al. 2000; Lee et al. 2005). The highest catalytic

efficiency of native GOOX-T1 is observed with cellotriose (Lee et al. 2005). Notably,

this GOOX did not oxidize xylose, galactose, or many other sugars (Lin et al. 1991).

The crystal structure of GOOX-T1 is resolved by Huang et al. (2005) and was proposed

that Tyr429 initiates sugar oxidation by proton abstraction from the C1 hydroxyl,

followed by H1 hydride transfer to the N5 position of the FAD cofactor (Figure 2.1)

(Huang et al. 2005). Notably, the FAD is covalently bound by two amino acids, His70

and Cys130; this unique configuration is predicted to modulate the oxidative potential of

the FAD cofactor. Residues predicted to be involved in substrate binding and catalytic

Figure 2.2 GOOX substrate binding and active sites. A) Residues predicted to be involved in GOOX substrate binding B) Residues involved in FAD bi-covalent linkages and oxidation of substrate through FAD reduction. The predicted hydrogen bonds are shown with gray dashed-lines. The figure was generated using PyMOL.

13

mechanism of GOOX are illustrated in Figure 2.2. The crystal structure of GOOX-T1

further reveals that the enzyme possesses an open carbohydrate-binding groove, allowing

the accommodation of oligosaccharide substrates (Lee et al. 2005; Huang et al. 2008).

2.4 Current strategies for enzyme engineering

Several strategies have been developed for enzyme engineering including random

mutagenesis, site-directed mutagenesis, recombination, and fusion proteins. The

following paragraphs briefly review common protocols for each method, along with

corresponding advantages and disadvantages; the information is also summarized in

Table 2.1.

2.4.1 Mutagenesis: Random versus rational design

Random mutagenesis is a technique to obtain an improved enzyme by introducing

random mutations into the corresponding gene and then screening the recombinant

enzymes for mutants with desirable characteristics. A key requirement for this approach

is easy assessment of enzyme activity, for example via a colorimetric detection of

reaction products, This approach also benefits from the ability to express the target

enzyme in E. coli. In this way, activity measurements are less confounded by differences

in levels of recombinant protein expression between transformants. This method does not

require any information about the protein sequence nor the crystal structure of the

protein. The technique is usually applied to enhance thermostability or other

characteristics that are not easily predicted from protein sequence or structural

information. Several reviews have been published on the techniques used for random

mutagenesis (Cadwell and Joyce 1992; Cadwell and Joyce 1994; Bloom et al. 2005) and

14

consistently report the advantages of error-prone PCR over other methods. The approach

typically generates mutants through error-prone PCR, which applies a low fidelity Taq

polymerase. Notably, the error rate of the Taq polymerase used for this purpose is the

highest of the known DNA polymerases, approximately 2×10-5 compared to the error rate

of 4×10-7 for commercially available high fidelity polymerases. Additional changes to the

PCR reaction conditions can further increase the error rate, for example, increasing or

unbalancing the concentration of deoxynucleosides (dNTPs), addition of MnCl2, and

increasing the concentration of MgCl2 or Taq polymerase (Cadwell and Joyce 1992).

In contrast to random mutagenesis, site-directed mutagenesis requires knowledge of the

sequence and ideally the crystal structure of the protein. This method is usually used

when active sites of the enzyme are known and the engineering goal is to alter substrate

specificity. For instance, site-directed mutagenesis can be used to change a few amino

acid residues that are thought to prevent productive interactions with a target compound.

Molecular modeling software is typically used to help reveal amino acid residues

involved in substrate-enzyme interactions, and to compare the active sites of homologous

enzymes. For instance, Visual Molecular Dynamics (VMD) is a freely accessible

software (Humphrey et al. 1996) that provides tools for structural superimposition of

homologous proteins and compares the 3-dimenstional positioning of substrate

interacting residues; this tool is commonly used for proposing mutants with altered

specificity. Among the reported methods for site directed mutagenesis, non-PCR based

methods are more reliable since the primers are designed such that only the parental

plasmids are amplified in a linear fashion, preventing error propagation during successive

rounds of thermal cycling. Moreover, high accuracy is maintained through the use of

15

PfuTurbo DNA polymerase, to reduce random errors. Currently, the most cited

commercial kit for site-directed mutagenesis is QuikChange from Stratagene.

2.4.2 Gene shuffling

Gene shuffling is another approach to improve the functionality of a protein.

Recombining structurally similar proteins generate even more changes to the protein

sequence than random mutagenesis (Drummond et al. 2005). Yet, like random

mutagenesis, this approach is particularly valuable when the change required for a

desirable function is not predictable from the protein sequence or structure; it also

depends on an easy screen for protein function. Gene shuffling can be combined with

site-directed mutagenesis to produce a consensus gene sequence that comprises the most

frequent nucleotide residues identified from an alignment of genes corresponding to

homologous proteins (Lehmann and Wyss 2001; Lehmann et al. 2002; Steipe 2004).

2.4.3 Fusion proteins

Fusion proteins can be generated using a combination of PCR and recombination to fuse

domains from different proteins. For instance, the Green florescence protein (GFP) is a

common tag, which is fused to various proteins for quantitative bio analysis and easy

detection and localization studies (Remington 2011). Similarly, carbohydrate binding

modules (CBM) from naturally occurring glycoside hydrolases have been fused to

catalytic domains of other enzymes to increase the specific activity of the enzyme

towards cellulose and other polymeric carbohydrates (Shoseyov et al. 2006). Examples of

the CBM fused proteins will be reviewed in the following section.

16

Table 2.1 Overview of common enzyme engineering strategies

Approach Typical uses Advantages Disadvantages References

Random mutagenesis

When the characteristic under study is not predictable from protein sequence or structure

No information about sequence or structure is required

An easy screening method is required Ideally, targeted protein can be functionally expressed in E.coli

Cadwell and Joyce 1992 Bloom et al. 2005 Cadwell and Joyce 1994

Site-directed mutagenesis

When rational design is possible by structure-function correlations

Any expression system can be applied

A few mutations usually results in moderate changes to the function

Bhat 1996 Costa et al. 1996 Braman et al. 1996

Gene-shuffling When the function under study is not predictable or conserved among the sequence or structure of homologous proteins

Results in more significant changes

Benefits from sequence information for homologous proteins that score high on desired function.

Drummond et al. 2005 Steipe 2004 Lehmann and Wyss 2001 Lehmann et al. 2002

Fusion proteins When a tag is required for detection, binding or improved specificity

Specific choices are available depending on the application

The fusion module might change the conformation or functionality of the protein

Remington 2011 Shoseyov et al. 2006

2.5 Carbohydrate binding modules

Carbohydrate active enzymes that act on high molecular weight polysaccharides

including insoluble cellulose fibrils often contain carbohydrate binding modules (CBMs)

that can promote functional association of the enzyme and targeted substrate (Shoseyov

et al. 2006). To date, CBMs have been classified into more than 70 families based on

amino acid sequence similarities (www.cazy.org; 2015). These modules have been

further grouped into three types based on folding and substrate specificity (Boraston et al.

2004; Gilbert et al. 2013). Type A CBMs possess a flat binding site, which is thought to

17

associate with surfaces presented by crystalline cellulose. The planar architecture of the

binding site shows no or little affinity towards soluble carbohydrates (Zhang et al. 2012).

By contrast, Type B CBMs contain a cleft or a groove architecture that is better suited to

the conformation of amorphous cellulose or oligosaccharide chains. The depth of the

binding groove varies among Type B CBMs; it can be shallow or deep enough to

accommodate the entire width of a pyranose ring, as in CBM4-2 from C. fimi (Boraston

et al. 2002; Christiansen et al. 2009). Finally, lectin-like Type C CBMs have binding sites

that form several hydrogen bonds with sugar molecules typically having less than three

monosaccharide units (Notenboom et al. 2002).

2.5.1 Contributions in cellulose/hemicellulose active enzymes

The varied contributions and significance of CBMs on cellulolytic enzymes was recently

reviewed (Várnai et al. 2014). Notably, recent genome sequences indicate that the

majority of predicted cellulolytic enzymes lack CBMs and that CBMs are somewhat

enriched among cellobiohydrolases (Palonen et al. 2004). Whereas certain CBMs can

increase non-productive binding to lignin present in lignocellulose substrates (Várnai et

al. 2014; Palonen et al. 2004), CBMs can also improve the performance of cellulolytic

enzymes on insoluble substrates, particularly when presented with low substrate

concentrations (Tomme et al. 1988; Boraston et al. 2003; Costaouëc et al. 2013; Várnai et

al. 2013). Similar positive effect of CBMs is observed in mananases (Hagglund et al.

2003; Mizutani et al. 2012) and in some cases for xylanases (Lehtiö et al. 2003; Meng et

al. 2015) as truncated versions of these enzymes lacking the corresponding CBMs show

significant reduction of binding and activity towards insoluble mannans and xylans,

respectively. Moreover, CBMs from thermophiles can increase the thermostability of

18

carbohydrate-active enzymes (Charnock et al. 2000; Jun et al. 2009). Further relevant

examples will be reviewed in section 4.1.

2.5.2 Industrial applications of CBMs

CBMs have been applied broadly, and new applications continue to be reported. In

medical science, the affinity of CBMs contained in grass and dust allergens towards

oxidized cellulose was used to treat allergies and asthma (Shani et al. 2011). In terms of

process design, a family 3 CBM has been developed as an easy and cost effective tag for

protein purification (Guerreiro et al. 2008). To provide a tag free purification system, a

formic acid recognition site for chemical cleavage (Ramos et al. 2010) or an intein region

to excise and re-join the remaining protein, was designed between CBMs and target

proteins; this method was successfully applied for protein purification in E. coli and P.

pastoris (Wan et al. 2011). In the detergent industry, the effect of CBMs on enzyme

affinities has been used to improve the performance of laundry powders, where chimeric

amylases, proteases, lipases, and oxidoreductases are employed (Osten et al. 2000 a;

Osten et al. 2000 b).

2.5.3 Effects of CBMs on neighbouring modules

Recent reports on CBMs indicate that these modules can affect both the activity and

thermostability of the cognate catalytic module. In most cases, the activity of the

neighbouring domain is improved, harnessing the affinity of CBMs towards soluble and

insoluble polysaccharides. The significant decrease in binding affinity, and often catalytic

activity, of glycoside hydrolases upon genetic truncation of CBMs from the catalytic

module also reveals the boosting function of CBMs on enzyme activity (Ali et al. 2001;

19

Ali et al. 2005). The enhancement of activity has been attributed to: 1) a proximity effect,

2) a targeting function and 3) a disruptive function (Boraston et al. 2004). Firstly, through

binding to the substrate, CBMs would increase the local concentration of the polymeric

substrate relative to the active site of the enzyme. Secondly, the selectivity of the CBM

towards specific polymers, allows targeted binding to the substrate of interest from a

mixture of various polymers; this would be particularly relevant for enzyme applications

on plant biomass given the heterogeneity and complexity of the polysaccharides present.

Lastly, the potentially disruptive function of CBMs on cellulose could increase

amorphous structures within an otherwise crystalline substrate, although this role of

CBMs does not appear to be universal.

Besides enhancing the activity, varying effects of CBM fusion on the temperature

stability of associated enzymes have been reported. For instance, Jun et al. (2009) show

that appending a xylan specific CBM from Thermotoga maritima to xylanase 2 from

Hypocrea jecorina improves the thermostability and substrate affinity of the enzyme. In

another study, fusion of a family 42 CBM from Aspergillus kawachii to a feruloyl

esterase from A. awamori was shown to increase enzyme stability and affinity towards

arabinoxylan (Koseki et al. 2010). However, while fusion of a family 6 CBM from

Clostridium stercorarium Xy1A to Bacillus halodurans Xy1A does not affect enzyme

stability (Mangala et al. 2003), fusion of a family 22 CBM to B. halodurans C-125 family

10 xylanase decreases the thermostability of the enzyme (Mamo et al. 2007). The range

in effects of CBMs on protein stability is further exemplified by Kataeva et al. (2001),

who show that the native CBM4 of CelK from C. thermocellum increases the

20

thermostability of the catalytic module, which is not retained when substituting CBM4

for a CBM6 encoded by the same organism.

2.5.4 Important factors for designing CBM fusion proteins

Construction of chimeric enzymes is simple in theory; however, many challenges arise

when they are generated in practice. Several factors need to be considered when

designing the CBM-fusion enzymes. First, the type of CBM should be selected in a way

that imparts substrate selectivity to the fused enzyme. Ye et al. (2011) report that several

CBMs from different families (3,4,6,9) and types (A, B and C) were fused to C.

thermocellum cellodextrin phosphorylase (CtCDP) and only one of them enhanced

CtCDP activity on amorphous cellulose; the other fusion enzymes either showed less or

similar activity to the wild-type enzyme.

Second, the positioning of the CBM at the N-terminus or C-terminus of the catalytic

module should be thoughtfully selected. Notably, while family 4 CBMs are often located

at the N-terminus of GH9 glycosyl hydrolases, family 3 CBMs are typically located at the

C-terminus of this enzyme family (Kataeva et al. 2001). Moreover, Kateava et al.

reported that the fusion of a family 4 cellulose specific CBM to the C-terminus of

feruloyl esterase domain of a xylanase from C. thermocellum (FAE XynZ) did not

significantly affect the domain function, whereas positioning this CBM at the N-terminus

increased FAE XynZ affinity towards acid swollen cellulose (Kataeva et al. 2001).

Therefore, it appears that the CBM should be fused to the terminus that places the module

in closer proximity to the enzyme active site while not obstructing substrate accessibility

to the catalytic module.

21

Third, the stability towards proteases, and flexibility of the linker sequence, which is used

to connect the CBM to the catalytic module, should be carefully considered. In some

cases, the linker sequence affects the performance of the chimeric enzyme. For example,

Dias et al. report that the thermostability of C. thermocellum xylanase Xyn10B was

retained after removing the CBM22 domain and leaving the linker sequence, while

removing the whole linker-CBM22 sequence resulted in reduced thermostability of the

enzyme (Dias et al. 2004).

Lastly, if the microbial origin of the CBM or catalytic domain of the recombinant protein

is different from the expression host, codon optimization could be beneficial (Daly and

Hearn 2005). Codon optimization can be performed commercially, as a means to enhance

the expression of the full-length protein. For instance, codon optimization for yeast

resulted in 10.6 fold increase in protein expression of human glucocerebrosidase in P.

pastoris (Sinclair and Choy 2002). Similar strategies also resulted in increased expression

of α-amylase in P. pastoris (Tull et al. 2001).

Having reviewed the protein elements and engineering strategies most pertinent to this

study of carbohydrate oxidases, section 2.6 will review existing applications of

carbohydrate oxidase activity.

2.6 Applications of carbohydrate oxidase in food industry

2.6.1 Oxidoreductases in baking applications

In baking applications, carbohydrate oxidases, such as glucose oxidase from Aspergillus

niger (GO) and oligosaccharide oxidase from M. nivale (MnCO) have been used to

enhance the quality of the dough. Several authors have demonstrated the use of GO in

22

flour (Vemulapalli et al. 1998; Rasiah et al. 2005; Bonet et al. 2006; Hanft and Koehler

2006; Dagdelen and Gocmen 2007; Decamps et al. 2013) this enzyme converts glucose

constituents of the dough into gluconic acid, while reducing molecular oxygen to H2O2.

The H2O2 is shown to be the active reagent of a GO treatment, as it oxidizes different

positions of wheat proteins, i.e. gluten, including free thiols in cysteine residues to form

disulfide bonds, and phenolic tyrosine residues to generate di-tyrosines through oxidative

coupling (Hanft and Koehler 2006). These cross-linkages reinforce the protein network of

the dough (Rasiah et al. 2005), thereby reducing stickiness and enhancing machinability

of the dough. H2O2 also tends to generate a dry surface on the dough by oxidation and

gelation of water-soluble pentosans (Hanft and Koehler 2006). The water intake by

pentosans also improves freshness and softness of the baked products upon long storage

(Bonet et al. 2006).

Despite the advantages of GO to baking applications, its effectiveness is limited by the

selectivity of the enzyme towards glucose and susceptibility to inactivation by H2O2. An

early study by Kleppie (1966) suggests that in acidic pH, less than 0.01 M H2O2 oxidizes

methionine residues that are close to or in the active cleft of reduced GO. More

specifically, the methionine residues were shown to oxidize to methionine sulfoxide,

where the amount of methionine sulfoxide measured increased as the enzyme was

exposed to higher concentrations of H2O2 e.g. 0.1- 0.2 M, which in turn resulted in a

dramatic drop in enzyme performance by over 80%. Kleppie (1966) further highlighted

that GO in its reduced form is inhibited at least 100 times more readily than when in its

oxidized form. Hachimori et al. (1964) also studies the effect of H2O2 on several proteins

23

at alkaline pH, and report that in all cases, a tryptophan residue is oxidized by H2O2,

albeit to different extents.

The mechanism and kinetics of GO inhibition by H2O2 was studied in more detail by Bao

et al. (2003). Based on corresponding kinetics profiles in the presence of increasing H2O2

concentration, a competitive inhibition mechanism was suggested where the inhibition

constant was reported to be equal to the Michaelis constant KM. This result implies that

the affinity of GO in the reduced form to H2O2 is almost similar to that of oxygen. This

finding is consistent with their earlier report on competitive inhibition of immobilized

GO by H2O2 where the inhibition constant was also similar to the apparent Michaelis

constant (Bao et al. 2001).

Greenfield et al. already highlighted the importance of GO inactivation by H2O2 in 1975.

In that study, they demonstrate that the stability of immobilized GO in the presence of

H2O2 is inferior in a continuous operation than in storage tests. Therefore, they concluded

that for industrial usage of GO, it will be necessary to either directly reduce H2O2

inactivation or else indirectly minimize enzyme inactivation through GO immobilization

and/or addition of catalase. Much later, Yoshimoto et al. (2004) introduced a liposomal

capsulated GO system to decompose the H2O2 and protect GO from inactivation by H2O2.

The fluorescence properties of tryptophan residues and the FAD cofactor of GO at high

conversion of glucose revealed that the tertiary structure of the free enzyme is disordered,

while the encapsulated GO is protected (Yoshimoto et al. 2006).

Alternatives to GO are oligosaccharide oxidases, showing broader substrate specificity

and higher activity on oligosaccharides. Schneider et al. (2003) studied the rheology of

24

gluten content of the dough after addition of an oligosaccharide oxidase from M. nivale

(MnCO) and found that the enzyme increases the elastic modulus of the gluten, thereby

increasing the dough elasticity in a dose-dependent manner. They also studied the dough

consistency in terms of dough stickiness and firmness. The results indicated that the M.

nivale carbohydrate oxidase provides excellent dough consistency at 200-300 units/kg, as

evaluated using a scoring system by a skilled baker (Schneider et al. 2003). Despite the

better performance of MnCO compared to GO as a result of its wider substrate specificity

(Kulys et al. 2001a), this enzyme was reported to be also inactivated in the presence of

H2O2 (Nordkvist et al. 2007); however, the mechanism and extent level of H2O2

inactivation of MnCO was not reported.

2.6.2 Enzyme utilization for production of prebiotics

Carbohydrate active enzymes have been harnessed to produce prebiotic compounds.

Prebiotics were first described as “nondigestible food ingredients that beneficially affect

the host by selectively stimulating the growth and/or activity of one or a limited number

of bacteria in the colon, thus improving host health” (Gibson 1997). This definition was

later refined to include benefits from selective growth of microorganisms in other

sections of the gastrointestinal system: “a selectively fermented ingredient that allows

specific changes, both in the composition and/or activity in the gastrointestinal

microbiota that confers benefits.” (Gibson et al. 2004)

Carbohydrates with known prebiotic activity include but are not limited to

fructooligosaccharides, xylooligosaccharides (Kondepudi et al. 2012), konjac

glucomannan (Al-Ghazzewi et al. 2007), β-glucan, and arabinoxylan (Crittenden et al.

25

2002). As reviewed by Panesar et al. (2013) most enzymes used in prebiotic production

are glycoside hydrolases, which hydrolyze starting polysaccharides into a series of

oligosaccharides (Panesar et al. 2013). Very few studies have been done to harness other

enzyme types for prebiotic production, or to investigate the effect of those enzyme

treatments on the prebiotic activity of the product. For instance, oligosaccharide oxidases

such as GOOX-T1, MnCO and PCOX have been used to oxidize lactose into lactobionic

acid, which is a lactose derivative and potential prebiotic (Lin et al. 1996; Nordkvist et al.

2007; Murakami et al. 2008). However, the prebiotic potential of products prepared

using oligosaccharide oxidases has not been directly evaluated.

The prebiotic activity of a substrate can be studied both in vivo and in vitro. In the former

approach, often mice are fed with the potential prebiotic substrate and fecal samples are

collected for microbial analysis and to document fecal frequency as well as water content

(Wu et al. 2011). For in vitro studies, potential prebiotics are tested for their ability to

selectively stimulate the growth of bacteria correlated with healthy digestion, such as

species of Bifidobacteria and Lactobacilli, as compared to bacteria correlated with

unhealthy digestion including species of Clostridia and Bacteroides. In this case,

bacterial growth is typically measured in terms of viable cell counts or optical density

(Crittenden et al. 2002).

The prebiotic index (PI) was developed as a quantitative score to better compare the

prebiotic activity of different carbohydrates. Specifically, PI is defined as the portion of

healthy microflora subtracted by that of the unhealthy bacteria (Palframan et al. 2003).

For instance, when Bifidobacteria (Bif) and Lactobacilli (Lac) are considered as healthy

26

bacteria and Bacteroides (Bac) and Clostridia (Clos) constitute unhealthy bacteria, the

following equation describes PI correspondingly.

𝑃𝐼 = �𝐵𝑖𝑓𝑇𝑜𝑡𝑎𝑙

� + �𝐿𝑎𝑐𝑇𝑜𝑡𝑎𝑙

� − �𝐵𝑎𝑐𝑇𝑜𝑡𝑎𝑙

� − (𝐶𝑙𝑜𝑠𝑇𝑜𝑡𝑎𝑙

)

,where Bif, Lac, Bac and Clos are numbers of corresponding bacteria at sample time

divided by their numbers at inoculation; and Total refers to total bacteria numbers at

sample time divided by numbers at inoculation. Although PI has been used for systematic

analysis of commercial and novel prebiotics, a recent review by Bindels et al. (2015)

argues the selectivity criteria for prebiotics, and challenges our previous understanding of

healthy over unhealthy gut microorganisms. Thus, it is likely that new measures will be

developed in the coming years to better define and regulate prebiotic products.

In light of studies to date concerning carbohydrate oxidases, opportunities for protein

engineering, and existing applications of these enzymes, section 2.7 describes the

research hypotheses and specific research objectives addressed through my PhD thesis.

2.7 Research Hypotheses and Specific Research Objectives

The main objective of my PhD thesis was to study the enzymatic oxidation of plant oligo-

and polysaccharides by a glucooligosaccharide oxidase from Sarocladium strictum

(GOOX). This enzyme has reported activity on glucose and its oligomers with up to

seven sugar units. Thus, a specific aim of my PhD thesis was to engineer the GOOX

enzyme to expand its substrate specificity towards a broader range of plant

polysaccharides and oligosaccharides. My research hypotheses were:

1. Site-directed mutagenesis of the non-conserved residues in the active site of

27

GOOX will alter the sugar specificity of this enzyme so as to act not only on

glucose but also on xylose and galactose.

2. Fusion of selected carbohydrate binding modules to the GOOX catalytic domain

will extend its activity and binding capacity towards longer chain substrates e.g.

polysaccharides.

3. The GOOX-oxidized oligosaccharides can be directly used as sugar standards for

lytic polysaccharide monooxygenases (LPMOs), such as enzymes from family

AA9 (formerly GH61). Also, the enzymatic oxidation of oligo- and

polysaccharides will improve their nutraceutical value in various food

applications. In addition, the preferred oxidation of oligosaccharides versus

monosaccharides by GOOX makes this enzyme advantageous over commonly

used glucose oxidase for baking applications.

To test the above mentioned hypotheses, the following approaches were used:

1. Site-directed mutagenesis of the GOOX

a) The gene encoding GOOX protein was recovered from S. strictum RNA and

the non-conserved residues predicted to play an important role for substrate

specificity were mutated using the QuikChange site directed mutagenesis

technique. The wild type and mutant GOOXs were expressed in Pichia

pastoris and purified using an affinity chromatography method.

b) Enzyme kinetics study was performed using selected mono-, di- and tri-

saccharides to assess the effect of mutations on kinetics parameters.

28

2. CBM fusion to the GOOX

a) Selected CBM genes from Clostridium thermocellum, CBM3, CBM11, and

CBM44, were appended to both N- and C-terminus of the GOOX via natural

and synthetic linkers generating six fusion proteins that were expressed and

purified in P. pastoris.

b) The effect of the CBM on the binding capacity of protein fusions was assessed

using affinity gel electrophoresis; specific activities and thermostabilities were

compared using the standard GOOX assay, which detects H2O2 production.

3. Application of the GOOX-oxidized products

a) To confirm whether GOOX-oxidized oligosaccharides can be used as sugar

standards relevant to LPMO characterizations, the oxidized products were

characterized using NMR and mass spectroscopy to assess the oxidation

position and percent conversion.

b) To assess the prebiotic activity of plant oligosaccharides and polysaccharides

before and after GOOX treatment, Bifidobacterium. longum was grown on

oxidized and non-oxidized oligo- and poly-saccharides and the growth was

measured using viable cell count and optical density. The growth rates were

calculated and compared for each carbohydrate source.

c) To assess the advantage of GOOX over glucose oxidase in baking applications,

selected mono- and oligosaccharides were treated with both enzymes and

specific activities were measured. The inactivation of each enzyme by H2O2

was also compared.

29

Chapter 3 Improved GOOX activity on mono and oligosaccharides using

site-directed mutagenesis

Parts of this chapter are published in:

Foumani M, Vuong T.V, and Master E.R. 2011. Altered substrate specificity of the

gluco-oligosaccharide oxidase from Acremonium strictum, Biotechnology and

Bioengineering, 108(10): 2261-2269.

Contributions: Design of the study; performing the experiments, data collection and

analyses corresponding to strain cultivation, gene cloning, site-directed mutagenesis,

recombinant protein expression and purification, deglycosylation assay, specific activity

measurements, and kinetics studies; and manuscript preparation.

30

3.1 Introduction

Carbohydrate oxidases (EC 1.1.3) can catalyze the oxidation of hydroxyl groups to an

aldehyde, ketone or a lactone (then carboxylic acid), with concomitant reduction of

molecular oxygen to H2O2 (van Hellemond et al. 2006). Given the ease of detecting

H2O2, several carbohydrate oxidases, including glucose oxidase (GO) and pyranose

oxidase (POX), have been widely applied in clinical biosensors. Galactose oxidase

(GaOX) is also used to oxidize the primary C6 hydroxyl groups of polysaccharides

containing terminal galactose units (e.g. galactoglucomannan, galactomannan, and

xyloglucans) to alter the rheology of these compounds (Parikka et al. 2010).

By contrast, oligosaccharide oxidases that oxidize C1 hydroxyl groups of β-1,4-linked

sugars could be used to derivatize xylan and cellulosic substrates. Gluco-oligosaccharide

oxidase from Sarocladium strictum, previously Acremonium strictum, strain T1 (GOOX-

T1) (Lin et al. 1991) is an example of such enzyme. The crystal structure of GOOX-T1

reveals that similar to CDH and GaOX (Firbank et al. 2001; Hallberg et al. 2003), the

enzyme possesses an open carbohydrate-binding groove, allowing the accommodation of

oligosaccharide substrates (Figure 3.1A).

A screen of more than 50 carbohydrates and derivatives show that GOOX-T1 oxidizes

both α-linked and β-linked glucose substrates, including lactose, malto-oligosaccharides

and cello-oligosaccharides (Lin et al. 1991; Fan et al. 2000; Lee et al. 2005). The catalytic

efficiency of native GOOX-T1 is highest with cellotriose (Lee et al. 2005); however, this

GOOX did not oxidize either xylose, and galactose, or many other sugars (Lin et al.

1991).

31

Figure 3.1: The structural model of GOOX built by the Swiss-Model Workspace using the X-ray structure of GOOX-T1 (PDB ID: 2AXR). (A) The gross structure of GOOX with the active site containing the intermediate analogue 5-amino-5-deoxy-cellobiono-1,5-lactam (ABL) and the FAD cofactor. (B) The location of key residues Y300, W351, and N388 in relation to the ABL and the FAD cofactor (for neatness, the side chain of W300, N388 and a part of FAD are shown). Hydrogen bonds are shown as dashed lines.

While the catalytic mechanism of GOOX has been characterized, residues that affect the

substrate preference of this enzyme are still unknown. Given the limited arsenal of

biocatalysts that can be applied for oxidative modification of plant oligosaccharides,

GOOX variants with gained activity on xylose, galactose, and/or mannose containing

substrates would constitute a valuable set of new industrial enzymes. Accordingly, this

article reports the purification and substrate specificity of a gluco-oligosaccharide oxidase

from an S. strictum strain, and the improvement of its substrate specificity through site-

directed mutagenesis.

32

3.2 Materials and Methods

3.2.1 Fungal strain and materials

Sarocladium strictum (previously Acremonium strictum) type strain CBS 346.70 was

obtained from the American Type Culture Collection (ATCC) No.34717. 4-

aminoantipyrine (4-AA), 5-bromo-4-chloro-3-indolyl phosphate (BCIP), biotin, histidine,

imidazole, nitroblue tetrazolium (NBT) and phenol, as well as alkaline phosphatase-

linked anti-Rabbit IgG conjugates, anti-Myc antibodies, and horseradish peroxidase were

purchased from Sigma. Glucose, galactose, mannose, arabinose, N-acetyl-glucosamine,

xylose, maltose, and cellobiose were purchased from Sigma, while cellotriose,

mannobiose, mannotriose, xylobiose, and xylotriose were purchased from Megazyme. All

carbohydrates obtained from Sigma and Megazyme were between 95-99% pure.

3.2.2 Cloning of the GOOX-encoding gene

S. strictum was grown on 1 g mL-1 food-grade wheat bran at 27°C for 5 days, harvested

by filtration through Miracloth (Calbiochem), and then flash-frozen using liquid nitrogen.

Total RNA was extracted from the ground sample using the RNeasy Plant Mini Kit

(Qiagen). The full-length cDNA encoding the GOOX protein was isolated using the Long

Range 2Step RT-PCR Kit (Qiagen). Briefly, reverse transcription at 42°C for 90 min was

followed by PCR using Pfu DNA polymerase (Agilent Technologies), gene-specific

primers (Lee et al. 2005), and 35 cycles of 93°C for 30 s, annealing at 56°C for 40 s, and

extension at 72°C for 2 min. The PCR product was purified using the QIAquick PCR

Purification Kit (Qiagen), and then sequenced at the Centre for Applied Genomics

(TCAG, the Hospital for Sick Children). The GOOX encoding gene was cloned into the

33

pPICZαA expression vector (Invitrogen), downstream of an α-factor secretion signal,

using EcoRI and XbaI and T4 DNA ligase (Invitrogen).

3.2.3 Site-directed mutagenesis

Chito-oligosaccharide oxidase, ChitO, (accession no.: XP_391174) from Fusarium

graminearum and a carbohydrate oxidase from Microdochium nivale, MnCO, (accession

no.: CAI94231-2) were aligned to GOOX (accession no.: ADI58761) using the Megalign

program (DNASTAR-Lasergene) (Figure S3.1). Amino acids that were predicted to

participate in substrate binding, and that varied between the enzymes analyzed, were

selected for site-directed mutagenesis. Mutations Y300A, Y300N and W351F were

introduced using mutagenic primers (Table 3.1). PCR was performed for 14 cycles of

95°C for 30 s; 55°C for 1 min; and 68°C for 5 min, using the QuikChange method

(Agilent Technologies). The mutations were confirmed by sequencing (TCAG, the

Hospital for Sick Children).

Table 3.1 List of oligonucleotide primers used for gene amplification and site-directed mutagenesis.

Primer Sequence

EX1* GCTTCATGGATCCAGGAATTCAACTCAATCAACGCCTG

EX2* TTCAAGTCTAAATCATCTAGATAGGCAATGGGCTCAAC

Y300A-F CAACACCTACTTGGCCGGTGCTGACC

Y300A-R GGTCAGCACCGGCCAAGTAGGTGTTG

Y300N-F CAACACCTACTTGAACGGTGCTGACC

Y300N-R GGTCAGCACCGTTCAAGTAGGTGTTG

W351F-F GCGGCTGGTTCATCCAATGGGACTTC

W351F-R GAAGTCCCATTGGATGAACCAGCCGC

*From Lee et al. (2005) for gene amplification; others for site-directed mutagenesis.

34

3.2.4 Recombinant protein expression

Mutated plasmids were transformed into Pichia pastoris GS115 according to the

manufacturer’s instructions (Invitrogen, Pichia Expression version G). Transformants

were selected on buffered minimal methanol medium containing histidine (BMMH, 100

mM potassium phosphate, pH 6.0; 1.34 % yeast nitrogen base without amino acids

(YNB); 4 x 10-5 % biotin; 0.5 % methanol, 0.004% histidine), and then screened for

protein expression by immuno-colony blot using nitrocellulose membranes (0.45 µm,

Bio-Rad), anti-Myc antibodies, alkaline phosphatase-linked anti-Rabbit IgG conjugates,

and BCIP/NBT solution. Positive transformants were grown overnight in 100 mL of

buffered minimal glycerol medium containing histidine (BMGH, 100 mM potassium

phosphate, pH 6.0; 1.34 % YNB; 4 x 10-5 % biotin; 1 % glycerol, 0.004% histidine) at

30°C with continuous shaking at 300 rpm. The cells were harvested by centrifugation at

1,500 × g for 10 min and suspended in 300 mL of BMMH medium in 1 L- flasks to

OD600 ~ 1. Cultures were grown at 30°C and 300 rpm for 3 days and 0.5 % methanol was

added every 24 h to induce recombinant protein expression. Levels of recombinant

extracellular protein expressions were monitored every 24 h by activity and SDS-PAGE.

3.2.5 Enzyme purification

Supernatants from methanol-induced cultures of P. pastoris expressing the secreted

recombinant proteins were harvested by centrifugation at 6,000 × g for 10 min and

filtration through 0.22 μm Sterivex filter units (Millipore). Cleared supernatants were

concentrated approximately 150 times using Centricon concentration units (Millipore).

Each recombinant protein was purified using a new Ni-NTA resin (Qiagen). Fractions

35

were eluted with 250 mM imidazole and the buffer was replaced by 40 mM Tris-HCl (pH

8.0) using Vivaspin6 concentration units (GE Healthcare). Protein concentration

measurements were performed using the Pierce BCA assay (Thermo Scientific) and

enzyme purity was verified by SDS-PAGE.

In-gel trypsin digestion with sequencing-grade trypsin (Promega), followed by tandem

mass spectrometry was performed to confirm the identity of each protein sample. Tryptic

fragments were analyzed using the Applied Biosystems/MDS Sciex API QSTAR XL

Pulsar System coupled with an Agilent nano HPLC (1100 series) (The Advanced Protein

Technology Centre, the Hospital for Sick Children). Proteomic data were analyzed using

Scaffold Viewer (www.proteomesoftware.com).

3.2.6 Enzymatic assays and kinetics analyses

A chromogenic assay was used to measure H2O2 production (Lin et al. 1991). Reactions

contained 0.1 mM 4-AA, 1 mM phenol, 0.5 U horseradish peroxidase, 40 mM Tris-HCl

(pH 8.0), and different substrates were initiated by adding 0.2 µg of enzymes to the 250

μL reaction mixture. The production of H2O2 was coupled to the oxidation of 4-AA by

horseradish peroxidase and detected at 500 nm. Reactions were incubated at 37°C for 15

min to measure the specific activity of GOOX on 10 mM of monosaccharide or 1 mM of

oligosaccharide.

Kinetics parameters were determined with a wide range of substrate concentrations: 0.1

mM to 300 mM glucose, 1 mM to 1500 mM xylose, 1 mM to 600 mM galactose, 1 mM

to 600 mM N-acetyl-glucosamine (NAG), 0.1 mM to 300 mM maltose, 5 µM to 1.5 mM

cellobiose, 10 µM to 3.5 mM cellotriose, 20 µM to 40 mM xylobiose, and 20 µM to 50

36

mM xylotriose. At least 12 substrate concentrations were included to obtain kinetics

parameters for each substrate. Initial rates were obtained by measuring reaction products

every 30 s for 15 min at 37°C and pH 8.0, and kinetics parameters were calculated using

the Michaelis-Menten equation (GraphPad Prism5 Software).

The enzyme stability under activity assay conditions was evaluated in triplicate by

incubating 0.6 µg of each enzyme preparation in 40 mM Tris-HCl buffer (pH 8.0) for 0,

5, 15, 25, 35, and 60 min at 37°C. Residual enzyme activity was measured at 37oC for 15

min at pH 8.0 using 10mM maltose and 0.2 µg of protein.

3.2.7 Deglycosylation

Approximately 2 µg of purified enzyme was treated with Peptide -N-Glycosidase F from

Flavobacterium meningosepticum, also known as PNGaseF (New England Biolabs) using

denaturing and native conditions. Samples that were deglycosylated using denaturing

conditions were analysed by SDS PAGE, while samples deglycosylated using native

conditions were used to evaluate the impact of glycosylation on enzyme activity. The

activity of enzymes was measured on 10 mM maltose at 37oC for 15 min. N-

glycosylation was predicted by NetNGlyc (http://www.cbs.dtu.dk/services/NetNGlyc/)

while O-glycosylation was predicted by OGPET (http://ogpet.utep.edu/OGPET/).

3.2.8 Substrate docking

The structural model of GOOX from S. strictum type strain CBS 346.70 expressed in

Pichia was built by our post doctoral fellow co-authored in this work Dr. Vuong based on

the X-ray structure of GOOX from S. strictum strain T1, GOOX-T1 (PDB ID: 2AXR)

using the Swiss-Model Workspace (Arnold et al. 2006). The structures of glucose,

37

cellobiose, cellotriose, xylose, xylobiose, xylotriose and galactose were obtained from the

protein database of The Research Collaboratory for Structural Bioinformatics (PDB ID:

2FVY, 3ENG, 1UYY, X, 1B3W, 1UX7 and 2J1A, respectively). The program

AutodockTools 1.5.2 ran on Python 2.5 (http://autodock.scripps.edu/) was used to prepare

the oligosaccharides and the enzyme for docking. All hydrogen atoms were added and the

non-polar hydrogens were merged for all ligands and protein. A number of degrees of

torsions of each oligosaccharide were set up to evaluate different thermodynamic

properties. A Lamarckian genetic algorithm (Morris et al. 1998) with different number of

energy evaluations and a population size of 150 individuals were applied for docking.

The program, Autogrid 4, which pre-calculates grip maps of interaction energies, was

used to prepare the grid files, and then docking simulation was performed by Autodock 4

(http://autodock.scripps.edu/). After docking, free energies of binding ∆Gb and

dissociation constants Kd were reported.

3.2.8 Nucleotide sequence accession number

The cloned gene encoding GOOX from Sarocladium strictum type strain CBS 346.70 has

been deposited in the GenBank database under accession number GU369974.

3.3 Results and Discussion

3.3.1 Variations of GOOX

The GOOX gene cloned from Sarocladium strictum type strain CBS 346.70 encoded a

mature protein containing 474 amino acids, which is the same length as a previously

reported GOOX isolated from S. strictum strain T1 (hereafter GOOX-T1) (Lin et al.

1991; Lee et al. 2005). However, there were 15 amino acid substitutions between the two

38

proteins, 13 resulting from differences in corresponding wild-type gene sequences, and 2

(A38V and S388N) resulting from random mutations introduced during the construction

of the expression system (Table 3.2). The GOOX obtained in this study shares 97%

sequence identity with the reported GOOX-T1 (Lin et al. 1991), and it has a similar fold

to GOOX-T1.

Table 3.2 Amino acid substitutions in GOOX in comparison with GOOX-T1

No. Amino acid position

Amino acid in On protein surface*

Distance to sugar O1 (Å)

GOOX GOOX-T1 a

1 23 E D Yes 28.0

2 38 V A No 29.7

3 99 D N Yes 33.7

4 126 T S No 14.3

5 135 I V No 15.0

6 159 V I Yes 24.4

7 175 K E Yes 25.1

8 235 E Q No 22.2

9 259 Y F No 18.3

10 269 V I No 23.2

11 332 S Q No 26.7

12 366 S A Yes 21.2

13 367 H V Yes 20.5

14 388 N S No 11.3

15 435 D T Yes 24.1

*Determined by 20% accessible surface area a reported by Lin et al. (1991)

39

3.3.2 Production of recombinant protein

The recombinant expression of GOOX in P. pastoris GS115 was highest after three days

of incubation with 0.5 % methanol. Proteins were purified to more than 95 %

homogeneity by affinity chromatography; similar to previous reports of recombinant

GOOX-T1 expression by P. pastoris (Huang et al. 2008). Approximately 1.5 mg L-1 of

purified GOOX was recovered, and after confirming that one freeze-thaw cycle did not

affect enzyme activity, the purified enzyme was stored as 20 µL aliquots (~ 4 µg) at -

80°C. The enzyme remained fully active following pre-incubation at 37°C for 60 min

(Figure S3.2).

The deduced molecular mass of the mature protein with a c-myc epitope and a

polyhistidine tag is approximately 56 kDa (Protean, DNASTAR-Lasergene), which is

less than the electrophoretic molecular weight of purified GOOX (~70 kDa) (Figure

S3.3). By comparison, the reported molecular weight of GOOX-T1 determined by size

exclusion chromatography is approximately 61 kDa (Lee et al. 2005). Recombinant

proteins expressed in P. pastoris GS115 can be N-glycosylated with high-mannose-type

structures containing 8 to 14 mannose residues (Hirose et al. 2002; Blanchard et al.

2007). And NetNGlyc predicted three N-glycosylation sites in GOOX, including N305,

N341, and N394, which are all located in exposed loop regions. Still, the molecular

weight of deglycosylated GOOX was ~60 kDa, (Figure S3.3), suggesting that other post-

translational modifications, including O-glycosylation and/or phosphorylation, probably

occurred (Cereghino and Cregg 2000; Letourneur et al. 2001; Boraston et al. 2003).

Notably, deglycosylation of GOOX under native conditions did not cause a detectable

loss in enzyme activity (Table 3.3).

40

Table 3.3 The effect of deglycosylation with PNGaseF on enzyme activity

Enzyme Activity (nmol min-1)*

Glycosylated Deglycosylated

GOOX-VN 3.5 3.4

W351F 3.1 3.2

Y300A 4.1 3.9

Y300N 2.6 2.8

*Enzyme activity was measured in duplicate with 10 mM maltose following 15 min at 37oC.

3.3.3 Novel substrate specificity

GOOX oxidase activity was evaluated using glucose, xylose, galactose, NAG, mannose,

and arabinose. Glucose, xylose, galactose, and NAG were oxidized by the recombinant

GOOX, and the highest catalytic efficiency was observed using glucose (Table 3.4).

Previous analyses of GOOX-T1 did not detect activity on xylose, galactose or NAG, and

activity was limited to glucose and oligosaccharides with reducing end-glucosyl residues

(Lin et al. 1991; Fan et al. 2000). To check whether GOOX can oxidize oligomers of C5

sugars, the enzyme was then tested for oxidation of xylo-oligosaccharides. GOOX

oxidized xylo-oligosaccharides as efficiently as cello-oligosaccharides (Table 3.4), and

the catalytic efficiency of GOOX on these oligosaccharides was over two orders of

magnitude higher than that of the corresponding monomers. These findings show that

GOOX has broader substrate specificity than GOOX-T1, and GOOX oxidizes C6 and C5

mono- and oligomeric sugars.

41

Table 3.4 Kinetics parameters of wild-type and mutant GOOX enzymes.

Substrate Enzyme kcat

(min-1) Km

(mM) kcat/Km

d (mM-1min-1)

Specific activitya (µmol mg-1min-1)

From Vmax Defined substrate

concentrationb Glucose GOOX 449 ± 6c 17.4 ± 0.7 25.9 ± 1.1 7.4 2.6 Glucose W351F 337 ± 5 31.0 ± 1.3 10.9 ± 0.5 5.6 1.3 Glucose Y300A 793 ± 14 8.1 ± 0.4 98 ± 5 13.1 7.2 Glucose Y300N 649 ± 6 3.11 ± 0.12 209 ± 8 10.7 8.3 Xylose GOOX 315 ± 9 105 ± 10 3.0 ± 0.3 5.2 0.4 Xylose W351F 277 ± 4 288 ± 10 0.96 ± 0.04 4.6 0.2 Xylose Y300A 680 ± 8 51.8 ± 1.9 13.1 ± 0.5 11.2 1.6 Xylose Y300N 700 ± 10 32.0 ± 1.7 21.7 ± 1.2 11.5 2.3

Galactose GOOX 429 ± 8 132 ± 7 3.3 ± 0.2 7.1 0.5 Galactose W351F 394 ± 4 36.1 ± 1.3 10.9 ± 0.4 6.5 1.3 Galactose Y300A 798 ± 15 96 ± 5 8.3 ± 0.5 13.2 1.2 Galactose Y300N 705 ± 5 110 ± 2 6.4 ± 0.1 11.6 1.0

NAGe GOOX 485 ± 12 340 ± 17 1.4 ± 0.1 8.0 0.2 NAG W351F 470 ± 40 950 ± 120 0.5 ± 0.1 7.7 0.1 NAG Y300A 768 ± 13 92 ± 5 8.3 ± 0.4 12.7 1.1 NAG Y300N 680 ± 7 56.0 ± 1.9 12.2 ± 0.4 11.2 1.5

Mannose GOOX ND ND ND ND ND Mannose W351F ND ND ND ND ND Mannose Y300A ND ND ND ND 0.1 Mannose Y300N ND ND ND ND 0.2 Maltose GOOX 360 ± 5 2.81 ± 0.16 128 ± 8 5.9 1.5 Maltose W351F 323 ± 4 5.0 ± 0.2 65 ± 3 5.3 0.8 Maltose Y300A 625 ± 7 11.0 ± 0.5 57 ± 3 10.3 0.8 Maltose Y300N 624 ± 6 19.6 ± 0.7 31.8 ± 1.1 10.3 0.5

Cellobiose GOOX 375 ± 11 0.07 ± 0.01 6000 ± 800 6.2 5.4 Cellobiose W351F 344 ± 5 0.083 ± 0.005 4100 ± 60 5.7 5.1 Cellobiose Y300A 823 ± 16 0.25 ± 0.02 3400 ± 300 13.6 10.7 Cellobiose Y300N 684 ± 11 0.38 ± 0.02 1800 ± 100 11.3 8.2 Cellotriose GOOX 361 ± 12 0.085 ± 0.010 4200 ± 500 6.0 5.1 Cellotriose W351F 315 ± 6 0.11 ± 0.01 2900 ± 300 5.2 4.6 Cellotriose Y300A 670 ± 10 0.25 ± 0.02 2700 ± 200 11.0 9.0 Cellotriose Y300N 599 ± 5 0.44 ± 0.01 1400 ± 30 9.9 6.9 Xylobiose GOOX 529 ± 5 0.098 ± 0.004 5400 ± 50 8.7 7.9

42

Table 3.4 (Continued) Kinetics parameters of wild-type and mutant GOOX enzymes.

Substrate Enzyme kcat

(min-1) Km

(mM) kcat/Km

d (mM-1min-1)

Specific activitya (µmol mg-1min-1)

From Vmax Defined substrate

concentrationb Xylobiose W351F 478 ± 4 0.35 ± 0.01 1400 ± 40 7.9 5.9 Xylobiose Y300A 797 ± 7 5.11 ± 0.15 156 ± 5 13.2 2.1 Xylobiose Y300N 718 ± 8 4.83 ± 0.17 149 ± 6 11.8 1.8 Xylotriose GOOX 498 ± 7 0.10 ± 0.01 5100 ± 500 8.2 7.5 Xylotriose W351F 473 ± 4 0.31 ± 0.01 1500 ± 50 7.8 6.0 Xylotriose Y300A 832 ± 7 3.15 ± 0.11 260 ± 10 13.7 3.2 Xylotriose Y300N 718 ± 11 4.3 ± 0.2 170 ± 10 11.8 2.0

a 0.2 µg of enzyme was used in each reaction. b Reactions contained either 10 mM of monosaccharide or 1 mM of oligosaccharide substrate; ND- not detected. c Standard deviations (n=3). d Standard deviations (SD) for kcat /Km values were calculated using following formula: SD (kcat /Km) = kcat / Km *(SQRT((SD(Km)/Km)^2 + (SD(kcat)/kcat)^2)). e NAG, N-acetylglucosamine.

The broader substrate range of GOOX detected in the current study compared to previous

reports using GOOX-T1 could be the result of different assay conditions, in particular the

enzyme concentration. While reactions for kinetics analyses of GOOX proceeded for up

to 15 min and included substrate concentrations over 500 mM, oxidation of xylose,

galactose and NAG by GOOX was detected after 3 min using 10 mM of each sugar,

which were the reaction conditions previously used to screen GOOX-T1 activity (Lin et

al. 1991). However, the enzyme concentration was not specified in Lin et al. (1991), and

so it is possible that lower enzyme dose was used in that study. Furthermore, earlier

substrate screens were performed using the native GOOX-T1 (Lin et al. 1991) while in

the current study the recombinant GOOX was used to measure specific activities. This is

important in light of the fact that different kcat values were obtained by native and the

43

recombinant GOOX-T1 on maltose (531 min-1 and 361 min-1, respectively) (Lin et al.

1991; Lee et al. 2005), and the discrapency was explained by the various processing

methods for proteins in P. pastoris and S. strictum (Lee et al. 2005).

Alternatively, novel substrate specificity of GOOX could be due to amino acid

substitutions in this enzyme. Most substitutions are located on the protein surface or far

from the oxidation site (Table 3.2); however, N388 is positioned on the same β16-sheet

as conserved residues Q384 and Y386, which are predicted to participate in substrate

binding (Huang et al. 2005). The side chain of N388 is located near the predicted -2

subsite, within 6.2 Å from the substrate. When comparing the X-ray structures of

precursor and mature galactose oxidase from Fusarium spp., Firbank et al. (2001) showed

that the Cα of Tyr290 moved by 6.3 Å and the loop containing this residue could shift up

to 8 Å (Firbank et al. 2001).

Figure 3.2 Conformational changes of S388 upon substrate binding. (A) X-ray structure of GOOX-T1 (PDB ID: 1ZR6); the backbone of S388 formed two H-bonds (dashed lines) to that of G349. (B) X-ray structure of GOOX-T1 in the presence of a substrate analog (PDB ID: 2AXR); the side chain of S388 formed a weak H-bond to the G349 backbone while the side chain of D387 H-bonded to the backbone of S388. (C) The position of N388 in the structural model of GOOX.

44

While general loop movement was not observed when comparing GOOX-T1 structures

before and after inhibitor binding, the side chain of S388 in GOOX-T1 turned

significantly upon substrate binding to form a weak H-bond with the G349 backbone of

the β15-sheet (Figure 3.2) (Huang et al. 2005). Accordingly, the beneficial effect of the

S388N substitution on GOOX activity might be due to the potential of Asn to stabilize

substrates that contain fewer hydroxyl groups and/or to stabilize the β16-sheet for

substrate binding.

Docking analysis determined that the computational Kd for xylose was two times higher

than that for glucose, suggesting that low activity on xylose, which does not possess an

exocyclic CH2OH, might be due to weak binding of this substrate by GOOX (Table 3.5).

Table 3.5 Docking parameters of oligosaccharides with GOOX enzymes.

GOOX Y300A Y300N

Docked

energy ∆Gb

(kcal/mol)

Kd

(µM)

Distance to

Y429

Oη(Å)*

Docked

energy ∆Gb

(kcal/mol)

Kd

(µM)

Distance to

Y429 Oη(Å)

Docked

energy ∆Gb

(kcal/mol)

Kd

(µM)

Distance to

Y429

Oη(Å)

Glucose -5.2 150 2.8 -5.0 210 2.9 -5.1 190 2.8

Cellobiose -6.8 10 2.9 -6.1 35 2.7 -6.4 20 2.9

Cellotriose -6.8 10 2.8 -6.6 14 2.8 -6.7 12 2.7

Xylose -4.7 360 3.1 -4.6 430 2.9 -4.6 410 3.2

Xylobiose -6.2 27 2.6 -6.0 39 2.8 -5.9 45 3.2

Xylotriose -7.5 3.2 3.0 -5.7 63 2.9 -5.9 50 3.2

Galactose -5.1 190 2.6 -4.9 250 2.8 -5.0 230 2.9

45

*Distance between the Oη atom of Y429 and the O1 atom of oligosaccharides. This distance is 2.8 Å in the crystal structure of GOOX-T1 and an inhibitor (PDB ID: 2AXR).

The Km values for di- and tri-saccharides obtained experimentally, as well as the

corresponding Kd values derived from the docking models, are an order of magnitude

lower than the Km and Kd values for monosaccharides (Table 3.4, Table 3.5). These

results support the presence of two glycosyl-binding subsites in the carbohydrate-binding

groove of GOOX, which was also predicted by the X-ray structure of GOOX-T1 (Huang

et al. 2005).

3.3.4 Improvement of substrate specificity

The catalytic activity of GOOX on monosaccharides and oligosaccharides was further

improved through site-directed mutagenesis. Amino acids targeted for this analysis were

chosen by 1) referencing the published structure of GOOX-T1 (Huang et al. 2005), and 2)

identifying amino acids in GOOX that participate in substrate-binding, which consistently

differ from corresponding residues in ChitO from F. graminearum and MnCO from M.

nivale.

Y300 and W351 are located at the -2 glucosyl-binding subsite (Figure 3.1B), and likely

stabilize oligosaccharide binding through stacking interactions. Y300 is substituted by

alanine in ChitO and asparagine in MnCO while W351 is substituted by phenylalanine in

MnCO. Since MnCO is distinguished by its activity on galactose, xylose and to some

extent on mannose (Xu et al. 2001), altering the polarity and/or size of Y300 and W351

could increase the activity of GOOX on sugars with an axial OH4 group or that lack an

exocyclic CH2OH group. Accordingly, Y300N, Y300A and W351F substitutions were

46

generated in GOOX, and 3 mg L-1, 4 mg L-1 and 1.3 mg L-1 of each purified protein was

recovered, respectively. The mutant enzymes remained active after a one hour- pre-

incubation at 37°C (Figure S3.2).

The catalytic activity (kcat) of Y300A and Y300N mutant enzymes on all tested

monosaccharides and oligosaccharides was approximately two times higher than that of

GOOX (Table 3.4). These two mutant enzymes also gained low activity on mannose

(Table 3.4). However, the loss in hydrophobic interactions at the -2 subsite also increased

the Km and Kd values for oligosaccharides, reducing overall catalytic efficiency. These

results suggest that Y300 affects substrate positioning relative to the catalytic Y429

residue and the FAD cofactor, and that Y300 contributes to stacking interactions with

substrates containing more than two units.

The W351F mutation slightly reduced the catalytic activity of GOOX on all substrates.

Like Y300A and Y300N mutations, the W351F mutation also increased the Km values of

GOOX with oligomeric substrates (Table 3.4). These results are consistent with both

Y300 and W351 participating in stabilizing stacking interactions with penultimate

reducing sugars of oligomeric substrates, which also explains why the impact of these

mutations on Km is similar with di- and tri-saccharides (Table 3.4). Notably, the W351F

mutation also increased the Km values of GOOX with glucose and xylose, but decreased

the Km of GOOX with galactose, resulting in higher catalytic efficiency with this

substrate (Table 3.4).

47

Figure 3.3 Docking of monosaccharides to GOOX. Docking positions of glucose (A), xylose (B) and galactose (C); and the side chains of Y300 and W351 were shown. The O4 atom of galactose (circled) pointed to the benzene ring of W351, and their distance was 3.1Å.

Docking studies showed that while glucose and xylose binding at the active-site was not

restricted, the axial OH4 group of galactose points directly towards the benzene ring of

tryptophan (Figure 3.3), suggesting that the indole structure hinders GOOX binding of

sugars with axial OH4 groups.

3.4 Conclusions

This study demonstrates that GOOX with different substrate specificity were produced by

different strains of S. strictum, widening the application of GOOX from S. strictum for

the oxidation of mono- and oligo-saccharides. In addition to glucose, maltose and cello-

oligosaccharides, the new GOOX oxidized xylo-oligosaccharides, galactose, and N-

acetylglucosamine, which were not detected in GOOX-T1 from previous studies. Y300A

and Y300N substitutions increased the catalytic activity of GOOX on all substrates, and

gained low activity on mannose. Rational engineering approaches are now being applied

to decrease the Km of GOOX and its mutant enzymes on oligomeric substrates. In

particular, given the consistency between computational docking analyses and

experimental data reported in the current study, docking analyses will be used to predict

48

the effect of selected amino acid substitutions on the binding affinity, conformation, and

orientation of substrates bound by GOOX and variant enzymes. It is anticipated that

resulting carbohydrate oxidases will constitute important tools for the production of new

materials from plant fibre.

3.5 A followed-up study on the structure-function analysis of GOOX

Given the interesting outcomes of the site-directed mutagenesis work presented in this

chapter, a followed up study was designed to further unravel the role of amino acid

residues within the GOOX substrate binding site (Vuong et al. 2013). Since this analysis

was not a main part of my PhD thesis, I will only briefly summarize the main highlights

here.

The additional amino acids targeted for single mutation included Y72 and Q384 located

near the -1 substrate binding site, E247 and W351 located near the -2 substrate binding

site, and Q353, which is positioned between these two subsites (Figure 3.4). While most

substitutions were to alanine, some were mutated to related residues.

All mutations resulted in higher Km values than the wild-type GOOX, confirming the

important role of the targeted residues for substrate binding by GOOX. Notably, Y72 and

Q353 mutants lost oxidation activity on monosaccharides and the latter also lost activity

on xylooligosaccharides, suggesting the role of Q353 for substrate preference of this

enzyme.

49

Figure 3.4 The biding site for GOOX-T1. Residues for mutation in relation to the substrate analog, 5-amino-5-deoxy-cellobiono-1,5-lactam (ABL); the movement of residues in the absence of ABL (PDB ID: 1ZR6, green) compared with the presence of ABL (PDB ID: 2AXR, cyan) are indicated by arrows.

Moreover, it was confirmed that the two random mutations on S388 and A38 do not

explain the differences in substrate preference of GOOX and GOOX-T1. Yet, the

destabilizing effect of S388N proposed in this study was validated, as the back mutation

of N388S resulted in a more thermostable mutant consistent with GOOX-T1

thermostability. In addition, it was shown that the Y300A mutants generated by the

current study exhibits reduced substrate inhibition; this characteristic was also shown for

alanine substitution of W351 residue, suggesting these mutants as ideal candidates for

oxidizing oligosaccharides when present at high substrate concentrations (Vuong et al.

2013).

50

Chapter 4 Enhanced binding and activity of GOOX towards polysaccharides

through CBM fusions

Parts of this chapter are submitted in:

Foumani M, Vuong T.V, MacCormick B, Master E.R. 2015. Enhanced polysaccharide

binding and activity on linear β-glucans through addition of carbohydrate-binding

modules to either terminus of a glucooligosaccharide oxidase. PLOS ONE J.

Contributions: Design of the study; performing the experiments, data collection and

analyses corresponding to construction of CBM fusion genes, recombinant protein

expression and purifications, cellulose binding and affinity gel electrophoresis

experiments, specific activity measurements, kinetics studies, and thermal stability

assays; as well as manuscript preparation.

51

4.1 Introduction

Carbohydrate oxidases can facilitate regio-selective oxidation of sugars and

polysaccharides, and were recently categorized as auxiliary activities (AA) in the CAZy

(Levasseur et al. 2013). Among these enzymes, oligosaccharide oxidases from family

AA7 are comparatively less well characterized. The corresponding carbohydrate oxidases

target the C1 hydroxyl of a broad range of oligosaccharides, including cello-, -xylo-,

chito- and malto-oligosaccharides. Examples of characterized AA7 enzyme include a

gluco-oligosaccharide oxidase (EC 1.1.3) from Sarocladium strictum T1 (GOOX-T1)

(Lee et al. 2005) and a gluco-oligosaccharide oxidase from Sarocladium strictum CBS

346.70 (GOOX), which was discussed in previous chapter (Foumani et al. 2011). More

examples have been reviewed in section 2.2. These flavoenzymes are likely to share a

conserved FAD-binding domain, while having different substrate binding domains.

GOOX-T1 and GOOX were previously shown to function best at 37 °C and pH 8, and

remain stable up to 50 °C and 45 °C, respectively (Lin et al. 1991; Fan et al. 2000;

Foumani et al. 2011). Both enzymes oxidize maltose, lactose and cello-oligosaccharides

(Lee et al. 2005; Foumani et al. 2011); GOOX was later shown to also oxidize xylo-

oligosaccharides and xylan (Vuong et al. 2013; Vuong and Master 2014). Given the

precedence for GOOX activity on oligomers of glucose, the aim of the current study was

to evaluate the potential of selected CBMs to increase GOOX activity on plant

polysaccharides, particularly β-glucans.

In a recent study, Telke et al. (2012) found that fusing CBM3, CBM4 or CBM30 to a

family GH9 endoglucanase increased enzyme activity on different cellulose preparations

52

by roughly 10-fold. Similarly, Voutilainen et al. (2014) attached various cellulose

binding modules from both microbial and fungal origins to a fungal family GH7

cellobiohydrolase. In all cases, the CBM addition increased enzyme binding to cellulose

as well as thermostability, where the addition of a bacterial CBM3 from Clostridium

thermocellum (CipA) led to highest activity gains. Fusions of a xylan specific CBM from

family 22 to a xylanase from Bacillus halodurans also showed increased activity on

insoluble xylan while thermostability of the fusion enzyme was reduced (Mamo et al.

2007). Moreover, a xylan-binding module from family CBM22A was recently appended

to the N terminus of GOOX, and the resulting fusion protein retained activity after

immobilization to xylan-coated surfaces (Vuong and Master 2014).

In this study, we evaluate the potential of selected CBMs to increase the binding capacity

and activity of GOOX towards insoluble celluloses and β-glucans presented at low

concentrations. Specifically, three characterized CBMs were fused to either the N-

terminus or C-terminus of GOOX, namely 1) the Type-A CtCBM3 from C. thermocellum

CipA, which can bind cellulose (Lehtiö et al. 2003; Wan et al. 2011), 2) the Type-B

CtCBM11 from C. thermocellum Lic26A-Cel5E, which can bind β-(1→3), (1→4)-glucan

(Carvalho et al. 2004), and 3) the Type-B CtCBM44 from C. thermocellum Cel9D-

Cel44A, which can bind xyloglucan, glucomannan and β-(1→3), (1→4)-glucan

(Najmudin et al. 2006). These CBMs were chosen for this analysis since corresponding

binding affinities have been characterized; being sourced from a thermophilic organism

also presented possibility to confer thermostability to GOOX as previously observed for

cellulolytic enzymes (Telke et al. 2012; Voutilainen 2014). Although the number of CBM

fusion studies has been increased in recent years, we believe this work is unique in terms

53

of wide selection of CBMs appended to both terminus of a single domain non-hydrolytic

enzyme via several types of linkers. Notably, to our knowledge this work is among the

first studies examining the potential of CBMs from family 11 and 44 in fusion constructs

to enhance the behaviour of a catalytic module other than their native cognates.

4.2 Materials and Methods

4.2.1 Materials

All chemicals were reagent grade with high purity, and purchased from Sigma (Canada)

unless otherwise specified. Cellobiose (99% pure) was purchased from BioShop Inc.

(Canada). Cello-oligosaccharides, barley β-glucan, konjac glucomannan, xyloglucan from

tamarind seed, carboxymethyl cellulose, and hydroxyethyl cellulose, all with min 95%

purity, were purchased from Megazyme (Ireland). Mixed xylo-oligosaccharides (DP 2-7)

was obtained from Cascade Biochemicals (USA). Nanocrystalline cellulose was a kind

gift from Dr. Y. Boluk (University of Alberta). Regenerated amorphous cellulose was

produced from Sigmacell cellulose Type 20 after the cellulose was wetted with water,

dissolved in phosphoric acid, and regenerated in water as described previously (Zhang et

al. 2006). The CBM encoding genes from Clostridium thermocellum, CtCBM3,

CtCBM11, and CtCBM44 were purchased from Nzytech (Portugal). The same genes

were codon optimized for expression in Pichia pastoris and were synthesized by DNA

2.0 (USA).

4.2.2 Construction of fusion enzymes

To construct C-terminal CBM fusion proteins, genes encoding CtCBM3, CtCBM11 and

CtCBM44 with native N-terminal linkers (CtCBM3: PTNTPTNTPTNTP, CtCBM44:

54

PPPY) or no linker sequence (CtCBM11) were separately cloned into the XbaI restriction

site of pPICZαA-GOOX, a plasmid for recombinant expression of GOOX in P. pastoris

(Foumani et al. 2011). Three genes were synthesized by DNA 2.0 to produce

corresponding N-terminal CBM fusion proteins. In this case, a DNA sequence encoding

a TP linker (SRGGGTATPTPTPTPTP) was inserted between genes encoding CtCBM3,

CtCBM11 or CtCBM44, and the gene encoding GOOX (Figure 4.1). All plasmid

constructs were sequenced at the Center for Applied Genomics (the Hospital for Sick

Children, Toronto, Canada) before being transferred to P. pastoris for recombinant

protein production.

4.2.3 Recombinant expression of fusion proteins in Pichia pastoris

All plasmids were transformed into P. pastoris GS115 according to the manufacturer's

instructions (Invitrogen, Pichia Expression version G). P. pastoris transformants were

selected on buffered minimal methanol medium containing histidine (BMMH, 100 mM

potassium phosphate pH 6.0; 1.34 % YNB; 4×10−5 % biotin; 0.5 % methanol, 0.004%

histidine), and then screened for protein expression using an overlay activity assay.

Briefly, the assay mixture (0.3% agarose, 2% cellobiose, 50 mM Tris-HCl pH 8.0, 2 mM

phenol, 0.4 mM 4-AA, and 15 U/mL horseradish peroxidase) was maintained in a 40°C

water bath, and 10 mL of the solution was gently poured on top of each BMMH plate

containing the transformant colonies. After solidifying at room temperature for 15 min,

plates were transferred to 37°C for 60 min to induce the chromogenic reaction between 4-

AA and H2O2. Transformants with highest activity were then selected for liquid

cultivation.

55

Selected P. pastoris transformants were grown overnight in 100 mL of buffered minimal

glycerol medium (BMGY, 1% yeast extract; 2% peptone; 100 mM potassium phosphate

pH 6.0; 1.34% YNB; 4×10−5 % biotin; 1% glycerol) at 30 °C with continuous shaking at

250 rpm. The cells were harvested by centrifugation at 1,500 × g for 10 min and

suspended in 200 mL of BMMH medium supplemented with 1% casamino acid in 1-L

flasks to OD600 ∼2. Cultures were grown at 27 °C and 250 rpm for 4 days and 0.5%

methanol was added every 24 h to induce recombinant protein expression. To minimize

proteolysis of the secreted recombinant protein, 2 µM leupeptin was added to the culture

medium every 24 h.

4.2.4 Purification of recombinant enzymes

Culture supernatants containing the recombinant protein were harvested by centrifugation

at 6,000 × g for 10 min and filtered through 0.22 µm PES filter membrane (GE water and

process technologies, USA). The culture supernatants were concentrated and buffer

exchanged into binding buffer (100 mM potassium phosphate pH 8, 300 mM NaCl, and 5

mM imidazole) using a Jumbosep centrifugal device (Pall Corp, USA). Resulting

concentrates were incubated separately with Ni-NTA resin (Qiagen, Germany) and eluted

with 250 mM imidazole; the protein solution was exchanged to 40 mM Tris-HCl pH 8

using Vivaspin 20 concentration units (GE healthcare, UK). Protein concentrations were

measured using the Pierce BCA assay (Thermo Scientific, Canada) and confirmed using

SDS-PAGE densitometry, where the band density of purified protein and a dilution series

of bovine serum albumin (BSA) were determined using ImageJ (http://rsbweb.nih.gov/ij/)

(Schneider et al. 2012).

56

4.2.5 Specific activity on oligosaccharides, soluble polysaccharides and insoluble

cellulose substrates

A chromogenic assay was used to detect and measure H2O2 production (Lin et al. 1991).

To measure activity on oligosaccharides, reactions contained 3 pmol of enzyme, 0.1 mM

4-AA, 1 mM phenol, 0.5  U horseradish peroxidase, 50  mM Tris–HCl (pH 8.0), and 0.1

mM of each substrate; 20 mM CaCl2 was also included in reaction mixtures as CtCBMs

from family 3, 11 and 44 show conserved calcium binding sites (Najmudin et al. 2006;

Viegas et al. 2008; Yaniv et al. 2011). Reactions without enzyme served as negative

controls. In all cases, the final reaction volume was 250 µL. The production of H2O2 was

coupled to the reduction of 4-AA by horseradish peroxidase and detected at 500  nm.

One unit of the GOOX activity corresponded to the formation of 1 µmol of the product

per min. Reactions were incubated at 37°C for 10 min and absorbance was measured

every 1 min. H2O2 (0-0.04 mM) was used to generate a standard curve. All experiments

were performed as triplicates.

To determine kinetics parameters on cellotetraose, initial rates of reactions were

measured using the above assay with eight substrates concentrations from 0.01-1 mM.

Kinetics parameters were then calculated using the Michaelis–Menten equation

(GraphPad Prism5 Software).

The standard reaction described above was also used to measure enzyme activity on

soluble polysaccharides. Since GOOX activity significantly differed on each test

polysaccharide, enzyme and substrate concentrations were adjusted so that activity data

could be collected within the linear range of the standard curve for the assay.

57

Accordingly, between 7.7 pmol and 30 pmol (∼ 0.5-2 µg) of enzyme was added to the

assay mixture, and reactions contained 0.1% to 0.5 % (w/v) of test polysaccharides.

Specifically, reactions containing 7.7 pmol of enzyme were with 0.5% xyloglucan,

whereas 30 pmol of enzyme were used with 0.1% konjac glucomannan, 0.3% barley β-

glucan, 0.3% carboxymethyl cellulose, and 0.5% hydroxyethyl cellulose. All reactions

were incubated at 37°C for 30 min and the absorbance was measured every 5 min.

Similarly, to measure enzyme activity on insoluble substrates, between 7.7 pmol and 30

pmol of enzyme was added to the standard assay mixture, and reactions contained 0.2%

and 0.5 % of test polysaccharides. Specifically, reactions containing 7.7 pmol of enzyme

were with 0.5% of microcrystalline cellulose (Avicel pH-101), 0.5% of nanocrystalline

cellulose, and 0.5% of oat spelt xylan, whereas 30 pmol of enzyme were used with 0.2 %

of regenerated amorphous cellulose. All reactions were incubated with mixing (500 rpm)

at 37°C for up to 24 h using an Eppendorf thermomixer equipped with an adaptor for 96-

well plates. To avoid evaporation, microplates were sealed using an adhesive sheet. For

each time point (0, 2, 4, 6, and 24 h) the entire 250 µL reaction was collected, centrifuged

to precipitate the insoluble fraction, and then 150 µL of the supernatant was used to read

absorbance at 500 nm.

4.2.6 Cellulose binding

The binding of wild-type GOOX and CBM fusions to microcrystalline cellulose (Avicel

pH-101) and regenerated amorphous cellulose was determined by mixing 10 µg of

enzyme with 0.5 mg of the cellulose sample in 250 µL of a buffer solution (20 mM CaCl2,

0.05% Tween 20, 50 mM Tris-HCl pH 8). Following incubation for 2 h at 4˚C with

58

vigorous shaking at 1,400 rpm, mixtures were centrifuged to recover supernatants

containing the unbound protein fraction and pellets containing the bound protein fraction.

Supernatant samples were concentrated to 20 µL by vacuum centrifugation while pellets

were washed three times with the buffer solution before being extracted for 10 min at

100˚C with 20 µL of a denaturing solution (10% SDS and 10% β-mercaptoethanol). The

bound and unbound fractions were then analyzed by SDS-PAGE and the proteins were

visualized by Coomassie blue staining.

4.2.7 Quartz crystal microbalance with dissipation (QCM-D)

QCM-D experiments were performed by the co-authors Dr. Vuong and Ben MacCromick

with cellulose-coated sensors (QSX 334, Q-Sense, Sweden) using the Q-Sense E4

instrument (Q-Sense, Sweden). Briefly, this instrument detects adsorption of materials to

the sensor surface by measuring changes in oscillation frequency and dissipation values

dictated by the mass and viscosity of the bound material (Vuong and Master 2014). The

flow rate was kept constant at 0.05 mL/min and the temperature was maintained at 25 °C.

The changes in areal mass (ng/cm2) were obtained using the Voigt model of the Q-Tools

software (Q-sense, Sweden). All enzyme and substrate solutions were prepared in a

reaction buffer of 50 mM Tris-HCl pH 8.0. The sensors were equilibrated with 50 mM

Tris-HCl pH 8.0 for approximately 16 h, and then 1.5 µg/mL of CtCBM3_GOOX or the

wild-type GOOX was flowed over the coated sensors until the frequency and dissipation

values stabilized. The protein solutions were then replaced by the equilibration buffer to

rinse away unbound materials, and after washing, the equilibration buffer was replaced

by 0.5 mM cellobiose and the flow-through was collected. Following 40 min of reaction

between bound GOOX enzymes and cellobiose, corresponding quartz sensors were

59

removed from the QCM-D and washed 3 times with the equilibration buffer before being

incubated for 24 h with 800 µL of 0.5 mM cellobiose. In this way, reaction products

were allowed to accumulate in the reaction mixture, which facilitated product detection.

The cycle wash and batch incubation of GOOX-immobilized sensors with 0.5 mM

cellobiose was repeated 3 times. The presence of H2O2 in the flow-through from the

QCM-D, as well as in the sensor incubation solution, was detected using the standard

chromogenic assay.

4.2.8 Affinity gel electrophoresis

Binding of GOOX and fusion constructs to konjac glucomannan, barley β-glucan,

xyloglucan from tamarind seed, and carboxymethyl cellulose was examined by native

affinity gel electrophoresis as described by Freelove et al. (2001) with minor

modifications. Briefly, the native polyacrylamide gels prepared for these analyses

contained 7.5 % (w/v) acrylamide in 25 mM Tris, 250 mM glycine buffer (pH 8.3), and

0.01 % of the test polysaccharide. Approximately 5 µg of GOOX and each fusion

construct were loaded onto the gels and then run at 90 V for 2 h at room temperature.

Relative binding affinities were inferred from the migration distance of the fusion

proteins and wild-type GOOX on gels with and without the test polysaccharides. BSA (5

µg) was also used as a reference protein for these analyses.

4.2.9 Temperature stability

To investigate the potential of each CBM to increase the temperature stability of GOOX,

0.5 µg of each fusion protein or wild-type GOOX was incubated for up to 4 h at 45°C,

before being cooled to room temperature to measure residual enzyme activity using the

60

GOOX standard assay and 1 mM cellobiose as substrate. All experiments were performed

in triplicate. The half-life was measured by plotting the logarithm of percent remaining

activity versus the incubation time using Microsoft Excel (v 14.1.4).

4.2.10 Nucleotide sequence accession number

The genes encoding N-terminal CtCBM3_GOOX, CtCBM11_GOOX,

CtCBM44_GOOX and C-terminal GOOX_CtCBM3, GOOX_CtCBM11, and

GOOX_CtCBM44 have been deposited in the GenBank database under accession

numbers: JX181765, JX181766, JX181767, JX181768, JX181769, and JX181770,

respectively.

4.3 Results and Discussion

4.3.1 Recombinant protein production

When using the standard Pichia expression protocol to produce N-terminal CBM fusions,

degradation products that corresponded to the size of the wild-type GOOX and CBMs

separately were observed by SDS-PAGE. This observation implied proteolysis of the N-

terminal fusion proteins at the linker site. P. pastoris is expected to release extracellular

serine, cysteine and asparatic proteases (Shi et al. 2003). The activity of these proteases

depends on the pH, where serine protease activity is highest above pH 7, cysteine-type

protease activity is highest between pH 5-7, and aspartic protease activity is highest

below pH 5 (Shi et al. 2003).

In cases where proteolysis of recombinant proteins is observed, optimization of

expression parameters can significantly improve protein expression. For example,

61

lowering the induction temperature can reduce cell lysis and thereby minimize the release

of extracellular proteases; casamino acids can compete with the recombinant protein for

extracellular protease action and thereby protect the expressed protein from degradation

(Shi et al. 2003); daily addition of protease inhibitors can also increase protein expression

in P. pastoris (Kurokawa et al. 2002). Accordingly, in the present work the above

parameters were evaluated as a means of improving the functional, recombinant

expression of intact GOOX fusions.

Lowering the induction temperature to 15°C led to protein aggregation. However, slight

reduction of temperature to 27°C along with supplementation of 1% casamino acid and

daily addition of 2 µM leupeptin, significantly improved the expression of soluble, intact

fusion proteins. When using the improved cultivation conditions, the yield of the C- and

N- terminal constructs were similar, suggesting that the codon optimization used for N-

terminal constructs did not significantly increase recombinant protein expression in P.

pastoris. Moreover, the susceptibility of N-terminal fusion proteins to proteolysis at the

linker site suggests that in this case, the natural linkers used in the C-terminal constructs

were more resistant than the synthetic TP linker to secreted proteases.

The average yield of fusion and wild-type enzymes was approximately 5 mg/L and the

Ni-NTA purification system recovered wild-type and fusion proteins to over 95 % purity,

as judged by SDS-PAGE (Figure S4.1). The specific activities of the fusion proteins on

cellobiose were comparable or higher than that of wild-type GOOX with the exception of

GOOX_CtCBM3, which was approximately 80% of the wild-type activity (Table 4.1).

62

Figure 4.1 Schematic representation of wild-type GOOX and GOOX fusions. Schematic representation of carbohydrate binding modules fused to the amino and carboxyl terminal end of glucooligosaccharide oxidase from Sarocladium strictum. Natural linkers were used in C-terminal fusions while TP linkers were used in N-terminal fusions. The linker sequences are shown in the connecting lines.

Table 4.1 Specific activity of wild-type and CBM fusion GOOX on oligosaccharides.

Enzymes a Cellobiose Cellotetraose Cellohexaose Mixed XOS

GOOX wild-type 244 ± 7 b 197 ± 8 150 ± 2 229 ± 5 CtCBM3_GOOX 304 ± 4 253 ± 7 194 ± 3 289 ± 1 CtCBM11_GOOX 287 ± 1 235 ± 6 178 ± 6 274 ± 4 CtCBM44_GOOX 325 ± 6 256 ± 12 201 ± 4 317 ± 4 GOOX_CtCBM3 200 ± 8 166 ± 12 126 ± 0 185 ± 2 GOOX_CtCBM11 344 ± 24 279 ± 12 213 ± 4 327 ± 2 GOOX_CtCBM44 291 ± 3 239 ± 11 183 ± 4 271 ± 3

a All CBM fusions led to statistically significant differences in specific activities compared to wild-type GOOX as determined using a two-tailed t-test (P < 0.05) b The unit for specific activity is U/ µmol measured on 0.1 mM oligosaccharides. Standard deviations represent three biological replicates.

63

4.3.2 Improved binding to polymeric substrates

GOOX fusion to selected CBMs significantly improved GOOX binding to tested

polysaccharides, and binding selectivity was observed according to the appended CBM.

For instance, CtCBM3 fusion increased GOOX binding to crystalline cellulose (Avicel)

and regenerated amorphous cellulose by more than 10-fold (Table 4.2, Figure S4.2). As

expected, fusion to CtCBM11 and CtCBM44 increased the affinity of GOOX towards

soluble polysaccharides, including β-glucan, glucomannan, and xyloglucan (Figure 4.2);

fusion of GOOX to CtCBM11 or CtCBM44 also promoted enzyme binding to

regenerated amorphous cellulose (Table 4.2). Overall, binding results were consistent

with previous studies that confirmed CtCBM11 and CtCBM44 affinity towards β-glucan,

lichenan, hydroxyethyl cellulose, glucomannan and oat spelt xylan, along with CtCBM44

binding to xyloglucan (Carvalho et al. 2004; Najmudin et al. 2006; Viegas et al. 2008).

However, to our knowledge, binding of CtCBM11 and CtCBM44 to regenerated

amorphous cellulose has not been previously reported.

Table 4.2 Binding of wild-type GOOX and CBM fusions to insoluble cellulose.

Enzyme Portion bound (%) Avicel RAC

GOOX wild-type 10 10 CtCBM3_GOOX 60 95 CtCBM11_GOOX 5 65 CtCBM44_GOOX 10 75 GOOX_CtCBM3 55 95 GOOX_CtCBM11 5 70 GOOX_CtCBM44 10 95

The bound and unbound fractions were analyzed by SDS-PAGE (Figure S4.2), and the percentage of bound protein was calculated by measuring protein band density using ImageJ. Approximately, 5-10% uncertainty level is expected for the numbers given above.

64

In most cases, binding was not affected by CBM positioning at the N-terminus or C-

terminus of GOOX (Figure 4.2, Table 4.2). The exception was CtCBM44 fusions, where

C-terminal constructs showed slightly better binding to regenerated amorphous cellulose

than corresponding N-terminal constructs. Notably, the N-terminus of CtCBM44

contains a PKD domain, which would extend the linker region when CtCBM44 is fused

to the C-terminus of GOOX.

Figure 4.2 Affinity gel electrophoresis (AGE) of wild-type GOOX and CBM fusions. Purified proteins were subjected to AGE using a 7.5 % (w/v) polyacrylamide gel containing A: no polysaccharides, B: β-glucan, C: xyloglucan, D: glucomannan, E: carboxymethyl cellulose. The final concentration of each polysaccharide was 0.01 %. Bovine serum albumin (BSA) was used as a reference.

65

4.3.3 Activity on oligosaccharides

With the exception of C-terminal CtCBM3 fusion, all CBM fusions generated herein

increased GOOX activity on 0.1 mM cellobiose, cellotetraose and cellohexaose, as well

as mixed xylooligosaccharides with a DP of 2-7 (XOS) (Table 4.1). Similar results were

previously reported for GOOX activity on cello-oligosaccharides and xylo-

oligosaccharides after fusion to xylan-binding CtCBM22A (Vuong and Master 2014).

Subsequent kinetics analyses performed herein revealed that increased activities were

best explained by slight but statistically significant increases in kcat for all N-terminal

fusions as well as the C-terminal CtCBM11 construct (Table 4.3). Likewise, the reduced

activity of GOOX-CtCBM3 on all cellooligosaccharides could be explained by a

decrease in kcat.

The increase in kcat values on oligosaccharides observed for the N-terminal fusions

constructed herein was even greater in earlier studies where CtCBM22A was fused to the

N-terminus of GOOX (Vuong and Master 2014). In all cases, the same artificial TP

linker was encoded between the N-terminal CBM and GOOX. However in the case of

the earlier CtCBM22A fusion, the TP linker was extended towards the CBM by a 9-

amino acid loop (AVAGTVIEG), whereas the TP linker was extended by only 1 to 3

amino acids in the GOOX fusions generated herein. As previously proposed, CBM

addition (Vuong and Master 2014) or mutation (Vuong et al. 2013) at the N-terminus of

GOOX could cause a conformational change in the proximal FAD-binding domain,

thereby affecting the redox potential of the enzyme. We reasoned that a more flexible

linker would likely have a greater impact on the conformation of the FAD-binding

domain, which could explain the nearly 2-fold increase in kcat values upon CtCBM22A

66

fusion (Vuong and Master 2014) compared with approximately 0.5-fold increase in kcat

values observed in the current study (Table 4.3). By contrast, the slight increase in kcat

values for the C-terminal fusion of CtCBM11 construct was correlated to the specificity

of CBMs from family 11 towards short oligosaccharides including cellotetraose (Viegas

et al. 2008), which is anticipated to promote functional associations between the substrate

and substrate binding site that is positioned at the C-terminal end of GOOX. The

decrease in kcat observed for GOOX-CtCBM3 is more difficult to rationalize. However,

since kcat is a function of both the catalytic rate constant and rate constant for dissociation

of the product, it is conceivable that product release is reduced upon C-terminal

positioning of the cellulose-binding CtCBM3 and corresponding linker sequence.

The addition of CtCBM22A to GOOX did not affect the Km value for oligosaccharides

(Vuong and Master 2014). Likewise in this study, no statistically significant changes to

the Km values were observed for cellotetraose upon CBM fusions, suggesting that the

CBMs used herein do not compete with -1 or -2 substrate binding subsites of GOOX

(Foumani et al. 2011).

Table 4.3 Kinetics parameters of wild-type GOOX and its CBM fusions on cellotetraose.

Enzymes kcat (min -1) Km (mM) kcat/Km b (min -1. mM-1) GOOX wild-type 250 ± 20 a 0.07 ± 0.02 3400 ± 700 CtCBM3_GOOX 300 ± 20 * 0.10 ± 0.02 3100 ± 700 CtCBM11_GOOX 310 ± 20 * 0.10 ± 0.02 3100 ± 700 CtCBM44_GOOX 380 ± 30 * 0.12 ± 0.03 3100 ± 700 GOOX_CtCBM3 150 ± 10 0.05 ± 0.01 2900 ± 600 GOOX_CtCBM11 360 ± 30 * 0.11 ± 0.02 3200 ± 700 GOOX_CtCBM44 290 ± 20 0.09 ± 0.02 3300 ± 700 a Standard deviations (n=3), b Standard deviations (SD) for kcat/Km values were calculated using following formula: SD kcat/Km= kcat/Km × [SQRT((SD(Km)/Km)2 + (SD(kcat)/kcat)2)]. * Statistically significant increase compared to the wild type as evaluated by t-test (p < 0.05).

67

4.3.4 Specific activity on polymeric substrates

Specific activities were then tested using both soluble and insoluble polysaccharides.

Oxidation by wild-type GOOX or GOOX fusions was not detected on nanocrystalline

cellulose, total oat spelt xylan, or hydroxyethyl cellulose. However, activity of wild-type

GOOX was detected for the first time on konjac glucomannan, barley β-glucan,

carboxymethyl cellulose, regenerated amorphous cellulose, xyloglucan from tamarind

seed, and Avicel (Table 4.4). In the case of glucommanan, it is likely that the reducing

end glucose is mainly oxidized, given the low activity of GOOX on mannose (Foumani et

al. 2011).

Table 4.4 Specific activity of the wild-type GOOX and CBM fusions on polysaccharides.

Specific activitya (U/mmol)

GluM b(0.1%) RAC c (0.2%) β-Glu d (0.3%) CMC e (0.3%) XG f (0.5%) Avicel (0.5%)

GOOX wild-type 4,100 ± 300 g 290 ± 30 1,520 ± 90 600 ± 100 118 ± 3 79.2 ± 0.3

CtCBM3_GOOX 4,800 ± 100 * 420 ± 10 * 1,700 ± 100 630 ± 60 120 ± 20 123± 7 *

CtCBM11_GOOX 5,100 ± 100 * 430 ± 40 * 1,300 ± 200 560 ± 50 118 ± 6 86 ± 4

CtCBM44_GOOX 5,400 ± 200* 410 ± 50 * 1,700 ± 200 500 ± 100 112 ± 5 97 ± 8

GOOX_CtCBM3 4,100 ± 300 320 ± 20 1,500 ± 200 620 ± 50 110 ± 10 100 ± 4 *

GOOX_CtCBM11 5,000 ± 400 * 400 ± 30 * 1,500 ± 400 500 ± 90 120 ± 20 84 ± 7

GOOX_CtCBM44 5,400 ± 200 * 460 ± 60* 1,600 ± 200 500 ± 200 113 ± 3 99± 4 *

a Substrate concentrations were optimized to measure initial rates of reaction and are indicated in parentheses. bglucomannan from konjac, c regenerated amorphous cellulose, d β-glucan from barley, e carboxymethyl cellulose, f xyloglucan from tamarind seed.g Standard deviations (n=3) * Statistically significant improvements compared to wild-type GOOX as determined using a two-tailed t-test for two samples with unequal variance (p < 0.05)

68

In experiments using polysaccharides, substrate concentrations were optimized to

measure initial rates of reaction. While different substrate concentrations complicate

comparisons between substrates, this analysis permitted comparisons relative to wild-type

GOOX for a given substrate, which was the main objective of the study. Relative to

wild-type GOOX, CBM fusions led to statistically significant improvements in GOOX

activity on 0.5% Avicel, 0.2% regenerated amorphous cellulose, and 0.1% konjac

glucomannan (Figure 4.3). For example, CtCBM3 and CtCBM44 fusions increased

GOOX activity on Avicel by up to 56%, where the highest enhancement was observed

with the N-terminal CtCBM3 fusion. Consistent with this trend, Telke et al. (2012) and

Voutilainen et al. (2014) also report greatest enhancement of activity on Avicel upon

fusion of CtCBM3 to an endoglucanase from Alicyclobacillus acidocaldrious (Cel9A),

and cellobiohydrolase from Talaromyces emersonii (Cel7A), respectively. Comparable to

specific activities reported for oligosaccharides (Table 4.1), all CBM fusions with the

exception of GOOX-CtCBM3 increased GOOX activity on regenerated amorphous

cellulose and glucomannan by up to 55% and 30%, respectively (Figure 4.3).

The improvement to GOOX activity was comparable to previous studies reporting 20-

50% increased performance of a cellodextrin phosphorylase on regenerated amorphous

cellulase upon addition of a CBM9 (Ye et al. 2011), and the 10-80% improved activity of

Cel9A Alicyclobacillus acidocaldrious on β-glucan upon fusion of a CBM from families

3,4 and 30 (Telke et al. 2012). Highest improvement was observed herein with fusion to

CtCBM44, where impacts were approximately two times higher when using 0.1%

glucomannan compared to 0.3% (Figure S4.3). This result, along with highest activity

improvement on Avicel and regenerated amorphous cellulose, is consistent with previous

69

reports showing greatest impact of CBMs on glycoside hydrolase activity when using

insoluble substrates or relatively low concentrations of soluble polysaccharides (Bolam et

al. 1998; Ali et al. 2001; Boraston et al. 2003; Várnai et al. 2013).

Figure 4.3 Specific activity of wild-type GOOX and CBM fusions on polysaccharides. A: crystalline cellulose (Avicel, 0.5%), B: regenerated amorphous cellulose (RAC, 0.2%), and C: glucomannan from konjac (0.1%). Substrate concentrations were optimized to measure initial rates of reaction. One unit corresponds to 1 µmol of product per min. Error bars represents standard deviations; n=3. The dotted line represents the specific activity of wild-type GOOX.

Changes in specific activity values were generally consistent with the selectivity of the

respective CBM. For instance, the increased specific activity of CtCBM3_GOOX on

Avicel compared to wild-type GOOX was consistent with improved binding of this

fusion protein to Avicel (Table 4.2), as well as the previously reported affinity of

CtCBM3 towards microcrystalline cellulose (Lehtiö et al. 2003). Similarly, the highest

specific activity of GOOX_CtCBM44 on glucomannan was correlated to the relatively

high binding of this construct on glucomannan (Figure 4.2D) and the previously reported

affinity of CtCBM44 towards this polysaccharide (Najmudin et al. 2006).

Even though CBM fusion improved GOOX binding to all tested polysaccharides, it did

not increase GOOX oxidation of barley β-glucan, carboxymethyl cellulose or xyloglucan.

70

Whereas barley β-glucan comprises mixed β-1,3- and β-1,4-linkages and is unbranched,

xyloglucan and carboxymethyl cellulose have β-1,4-linked glucose backbones that are

substituted with branching sugars or carboxymethyl groups, respectively. An early study

of gluco-oligosaccharide oxidases did not detect GOOX-T1 activity on β-1,3-linked

glucose of laminaribiose (Lin et al. 1993), and a more recent study showed reduced

GOOX activity on branched xylo-oligosaccharides and anionic xylo-oligosaccharides

(Vuong et al. 2013). It is therefore likely that the mixed-linkage backbone structure of

barley β-glucan, and branching groups in carboxymethyl cellulose and xyloglucan,

restrict functional interactions between these substrates and the -1 and -2 subsites of

GOOX (Huang et al. 2005; Vuong et al. 2013).

4.3.5 Immobilization of GOOX through CtCBM3

In addition to enhancing activity on cellulose, we postulated that CtCBM3 fusion could

promote GOOX immobilization to cellulosic surfaces, which is relevant to several

applications including the use of enzymes in biosensing materials. Given the

comparatively high activity of CtCBM3-GOOX, additional comparative analyses were

restricted to CtCBM3-GOOX and wild-type GOOX, this time using a QCM-D equipped

with cellulose-coated piezoelectric crystal sensors.

In this analysis, protein adsorption is observed as a decrease in oscillation frequency of

the sensor, whereas an increase in dissipation reflects a more viscoelastic surface layer

(Hook et al. 1998). Accordingly, the higher change in frequency (∆f) observed using

CtCBM3-GOOX confirms enhanced binding of this enzyme to cellulose compared to the

wild-type GOOX (Figure 4.4; Figure S4.4). Considering corresponding differences in

71

molecular weight, ∆f values corresponded to a molar adsorption ratio of the fusion and

wild-type enzymes of approximately 1.6. Notably, the slight increase in slope (∆D/∆f)

of the D-f plot depicting CtCBM3-GOOX binding suggests a more viscoelastic surface

layer was formed by the fusion enzyme (Figure 4.4).

Figure 4.4 Frequency - dissipation plot of enzymes binding to cellulose. Changes in frequency (Δf) and dissipation (ΔD) during 1.5 µg/mL enzyme addition (1), 50 mM Tris-HCl pH 8 buffer washing (2) and 0.5 mM cellobiose addition (3) in the experiments with CtCBM3_GOOX (green, triangle) and wild-type GOOX (red, square). The total running time is 210 min. Linear fitting for cellulose binding of the CBM fusion (dashed arrow) and the wild-type (solid arrow) were analyzed by GraphPad Prism 5.

After extensive washing to remove loosely bound enzyme, cellobiose was passed over

the sensors coated with immobilized CtCBM3-GOOX or GOOX. The addition of

cellobiose resulted in negligible changes to frequency and dissipation (Figure 4.4),

indicating that the enzymes remained bound to cellulose in the presence of the soluble

substrate. Moreover, the activity of immobilized CtCBM3-GOOX and GOOX was

confirmed by measuring H2O2 after batch incubation of recovered sensors with 0.5 mM

72

cellobiose. Given that measured activities remained stable after repeated cycles of

washing and incubation of the sensors with cellobiose (Figure S4.5), this analysis

demonstrated that fusion of GOOX to CtCBM3 could have advantages beyond increased

activity on cellulose, including one-step purification of active enzyme from culture media

based on cellulose adsorption, as well as biosensing and biofuel cell applications

involving mixed sugars.

4.3.6 Effect of CBM on thermostability

Earlier studies of wild-type GOOX confirmed its stability at 40 °C, but loss in over 30%

and 90% activity after 1 h at 45°C and 50°C, respectively (Foumani et al. 2012).

Therefore, in the current study, the half life of wild-type GOOX and CBM fusions were

compared at 45 °C. Despite selecting CBMs from a thermophilic bacterium, their fusion

to GOOX had a moderate impact on the temperature stability of corresponding fusion

proteins (Table 4.5, Figure S4.6). In particular, C-terminal fusion of CtCBM44 and N-

terminal fusion of CtCBM3 and CtCBM11 increased the half-life of GOOX at 45 °C by

10-40% and the C-terminal fusion of CBM3 decreased the half-life by 15%.

From an applied perspective, the half-lives under the reaction conditions (i.e. 37 °C), or at

room temperature, are expected to be much longer and the impact of CBM fusion could

be different in those situations.

It is interesting to note that the impact of CBM fusion on GOOX stability was more

readily explained by the nature of the linker region rather than CBM family. For

example, the rigidity of the short linker and connecting loops discussed earlier might

73

restrict heat-induced conformational shifts in the N-terminal catalytic domain of GOOX,

which could explain the relative thermostabilities of N-terminal fusions (Table 4.5).

Table 4.5 The half-life of fusion and wild-type GOOX at 45°C.

Enzymes Half-life (min)a GOOX wild-type 119 ± 4 b CtCBM3_GOOX 132 ± 2 * CtCBM11_GOOX 132 ± 2 * CtCBM44_GOOX 127 ± 4 GOOX_CtCBM3 104 ± 9 GOOX_CtCBM11 111 ± 10 GOOX_CtCBM44 170 ± 20 *

a. Values were obtained by plotting the log (% residual activity) versus incubation time (min). b Standard deviations (n=3) * Statistically significant improvements compared to wild-type GOOX as determined using a two-tailed t-test for two samples with unequal variance (p < 0.05)

This notion was further evident when comparing linker sequences of C-terminal fusions.

Specifically, the higher half-life of GOOX-CtCBM44 compared to other C-terminal

fusions could be explained by the PPPY linker sequence between GOOX and CtCBM44,

which likely adopts a more rigid conformation compared to the longer linker of CtCBM3

(PTNTPTNTPTNTP) and the connecting loop of CtCBM11 (SRAVGE). The potential

impact of linker sequences on thermostability is interesting in the context of an earlier

report by Dias et al. (2004), who showed that the thermostability of C. thermocellum

xylanase Xyn10B was retained after removing the CBM22 domain and leaving the linker

sequence, whereas removing the whole linker-CBM22 sequence reduced the

thermostability of the enzyme. Moreover, cellulase fusions with either a flexible

polyglycine linker or a rigid alpha-helix linker showed that the rigid linker significantly

74

enhanced enzyme activities and thermostability (Zou et al., 2012). Further examples

showing variable effects of CBM fusion on the temperature stability of associated

enzymes have been reviewed in section 2.4.3.

4.4 Conclusions

A selection of CBMs with affinity towards different β-glucans were appended to either

the C-terminus or N-terminus of GOOX, and fusion proteins with similar or higher yield

than the wild-type enzyme were successfully expressed and purified from P. pastoris. All

N-terminal fusion proteins as well as the C-terminal CtCBM11 fusion showed higher

catalytic activity on tested oligosaccharides than wild-type GOOX, suggesting a

beneficial conformational change to the FAD binding domain. In addition, unchanged Km

values confirmed that the fused CBMs did not compete with the GOOX subsites for

oligosaccharide binding. Similar to activity studies, thermostability of fusion constructs

was dictated by the nature of the linker sequence rather than CBM type, where more rigid

linkers resulted in more stable fusion proteins, underscoring the relevance of linker

selection to fusion protein design. Finally, regardless of positioning, CBM fusion

promoted GOOX binding to cellulosic and hemicellulosic polysaccharides, and GOOX

remained active when immobilized to cellulose through CtCBM3. This result highlights

that CBM fusion, in particular the N-terminal CtCBM3, could facilitate applications of

GOOX in cellulose-based biosensing devices.

75

Chapter 5 Application trials of wild-type and engineered GOOX

Section 5.1 of this chapter are published in: Vuong T, Vesterinen A, Foumani M, Juvonen M, Seppälä J, Tenkanen M, Master E.R,

Xylo- and cello-oligosaccharide oxidation by gluco-oligosaccharide oxidase from

Sarocladium strictum and variants with reduced substrate inhibition. Biotechnology for

Biofuels, June 2013.

Contributions to the above article: Expression and purification a GOOX variant enzyme

used in this study, performing preliminary mass spectrometry experiments to validate the

method and to design sample preparation; Preparation of GOOX-oxidized

cellooligosaccharides (DP: 2-5) for the NMR and mass spectrometry analysis; and

preparing the corresponding sections in the manuscript.

76

The wild-type and engineered GOOX have been evaluated for the production of oxidized

carbohydrate standards required to better characterize the action of enzymes relevant to

cellulose conversion to sugars (e.g. LPMO enzymes); GOOX and GOOX variants were

also evaluated for their utility in two applications relevant to food industries. While the

results are promising, further experiments are needed to validate and implement these

applications. Accordingly, the following sections will describe analyses completed

through the course of my PhD, and Chapter 6 (Future Directions) will summarize our

next steps.

5.1 GOOX in the production of sugar standards

5.1.1 Introduction

Recent characterizations of lytic polysaccharide monooxygenases (LPMOs) from

auxiliary activity family 9 (AA9), has revealed new routes to enzymatic biomass

conversion (Horn et al. 2012). LPMO activity involves the oxidative cleavage of

polysaccharide chains, and while the mechanism of LPMO action is not fully resolved, it

is clear that these enzymes cleave glycosidic bonds while oxidizing the adjacent carbon.

Thus, the products of LPMOs activity are mainly oxidized at the anomeric carbon,

although oxidization of the non-reducing end has also been reported (Quinlan et al. 2011;

Beeson et al. 2012; Westereng et al. 2013). C1 oxidation produces oligosaccharides that

lack a reducing end. Instead, under physiological conditions, the lactone spontaneously

converts to an aldonic acid. As a result, cleavage of glycosidic linkages by LPMOs

cannot be monitored using regular glucose-detecting assays.

77

New assays have been developed to monitor LPMOs activity, and chemical routes were

applied to generate oxidized-oligosaccharides as standards. For instance, Westereng et al.

(2013) used iodine oxidation in a methanol:potassium hydroxide mixture to selectively

oxidize the C1 carbon of the cellooligosaccharides to serve as standards in an LPMO

assay (Westereng et al. 2013) and other reports applied similar techniques using bromine

for oxidation of the anomeric carbon in oligosaccharides (Diehl et al. 1974). However,

since the oxidation is not complete, these chemical techniques usually involve costly

purification steps; moreover they require handling of toxic regents such as methanol.

In this study, we hypothesize that GOOX can efficiently oxidize oligosaccharides at the

anomeric position, thus providing a cleaner method to generate oxidized sugar standards

for LPMO activity assays. Accordingly, the present project investigates the structure of

the GOOX-oxidized cellooligosaccharides with degree of polymerization from two to

five and studies the efficiency of oxidation by detecting the non-oxidized compounds.

5.1.2 Materials and Methods

5.1.2.1 NMR analysis of oxidized products

Reaction mixtures containing 10 mM cellobiose or 10 mM xylobiose, and 160 nM

GOOX or GOOX-Y300A, in 50 mM Tris HCl (pH 8.0) were incubated overnight at 37oC.

Oxidized products were analyzed at Aalto University, Prof. Seppälä Lab, by proton

nuclear magnetic resonance (1H NMR) using a Bruker 400 MHz NMR Spectrometer

(Bruker Ultrashield 400 Plus, USA). Samples were measured by our collaborator, Arja-

Helena Vesterinen, directly in the reaction solvent with water suppression using 10%

78

deuterium oxide as a co-solvent for deuterium lock. The peaks were identified using the

estimation program of ChemBioDrawUltra 12.0 (CambridgeSoft).

5.1.2.2 Mass spectrometric analysis of oxidized products

Reaction mixtures containing 1 mM of cello-oligosaccharides, from cellobiose to

cellohexaose, and 160 nM GOOX or GOOX-Y300A, in 50 mM Tris HCl (pH 8.0) were

incubated overnight at 37oC. To characterize oxidized products, 100 µL of each reaction

mixture was diluted in 900 µL of MilliQ-water, and diluted samples were purified at

University of Helsinki, by Minna Juvonen in Prof. Tenkanen’s lab and fractionated using

a Hypersep porous graphitized carbon column (Thermo Scientific, MA, USA), following

the protocols of Packer et al. (Packer et al. 1998) and Chong et al. (Chong et al. 2011) but

with modifications. Neutral sugars were eluted using 40 % acetonitrile, and a mixture of

50% acetonitrile and 0.05 % TFA was used to elute acidic sugars. Collected fractions

were dried with nitrogen gas for 20 min and then freeze-dried overnight.

Positive ion mass spectrometric analyses were performed using an Agilent XCT Plus

model ion trap mass spectrometer (Agilent Technologies, Waldbronn, Germany)

equipped with an electrospray source. For ESI-MS-analysis, freeze dried samples were

dissolved in 20 µL of MilliQ-water, and 6 µL of each sample was diluted in 100 µL of

methanol-water-formic acid solvent (50:49:1 (v:v:v)). Sample solutions were introduced

into the ES source at a flow rate of 5 µL/min via a syringe pump. The drying gas

temperature was set to 325 °C; drying gas flow was 4 L/min; the nebulizer pressure was

79

15 psi, and the ES capillary voltage was set to 3164 V. Ions were collected in the m/z

range of 50 to 1000.

5.1.3 Results and Discussion

5.1.3.1 Confirming the regioselectivity of gluco-oligosaccharide oxidases

To date, very few studies have confirmed the position of hydroxyl groups oxidized by

family AA7 gluco-oligosaccharide oxidases. Lee et al. used 13C and 1H NMR to confirm

that GOOX-T1 targets the hydroxyl group of the anomeric carbon; however, only

maltose was used in their analysis (Lee et al. 2005). Since gluco-oligosaccharide oxidase

activity is higher on cello-oligosaccharides and xylo-oligosaccharides than malto-

oligosaccharides (Foumani et al. 2011), 1H NMR was used here to evaluate the effect of

sugar type and chain length on the regio-selectivity and catalytic efficiency of GOOX

enzymes.

The disappearance of the H1 doublet signals from the reducing end of α- and β-glucose

units of cellobiose is consistent with oxidation at the anomeric C1 position (Figure 5.1A)

(Nouaille et al. 2009). Similarly, the peak height for the H1 signals from the reducing end

of α- and β-xylose units of xylobiose was decreased in oxidized xylobiose samples

(Figure 5.1B). Ring opening at the anomeric position was also revealed by the detection

of H2 and H3 signals at 4.05 ppm and 3.95 ppm in case of oxidized cellobiose, and at

4.01 ppm and 3.81 ppm, respectively in case of oxidized xylobiose (Higham et al. 1994;

Nouaille et al. 2009). The signals for the corresponding lactone were not observed

probably due to the relatively long oxidation reaction (24 h); similar observations were

80

reported after overnight incubation of Phanerochaete chrysosporium CHD with

cellobiose (Higham et al. 1994).

Figure 5.1 NMR spectra of cellobiose (A) and xylobiose (B) oxidation. (A): From top to bottom are the spectra of cellobiose, cellobiose that was oxidized by wild type GOOX, and cellobiose oxidized by GOOX_Y300A; CB red. alpha and CB red. beta: H1 signals due to reducing α-glucose and reducing β-glucose units of cellobiose, correspondingly; CBA-H2 and CBA-H3: H2 and H3 signals of the cellobionate molecule. (B): From top to bottom are the spectra of untreated xylobiose and GOOX oxidized xylobiose; XB red. alpha and XB red. beta: H1 signals due to reducing α-xylose and reducing β-xylose units of xylobiose, correspondingly; XBA: Overlapped signals of the xylobionate molecule. 10 mM cellobiose and 10 mM xylobiose were used in oxidation reactions.

ESI-MS/MS analyses also indicated the enzymatic oxidation of cellotriose at the

anomeric carbon. In the positive ionization mode, the acidic fraction of oxidized

81

cellotriose only produced glycosidic bond cleavage fragments, generating B- and Y-ions

(Figure 5.2A); cross ring cleavage fragmentation was not observed. Since neutral reducing

oligosaccharides usually form cross ring cleavage fragments from reducing ends if a

sodium cation is present (Hofmeister 1991; Asam and Glish 1997), oxidation of the

anomeric carbon seemed to change the fragmentation behaviour of sodium cationized

cellotriose. In the negative mode, B and C-ions from glycosidic bond cleavage were the

most abundant fragment ions (Figure 5.2B).

Figure 5.2 MS/MS spectra and fragmentation of GOOX oxidized cellotriose. (A): ESI MS/MS in the positive ionization; the m/z ratio of the precursor, [M+Na]+, is 543. (B): ESI MS/MS in the negative ionization; the m/z ratio of the precursor, [M-H]-, is 519. Fragment ions were named according to Domon and Costello (Domon and Costello 1988).

The molecular masses of Y- and Z-ions increased by 16 Da, compared to the un-oxidized

control sample (data not shown), supporting that the oxidation reaction occurred in the

reducing glucose. Cross ring cleavage fragmentation was also observed in the negative

82

mode. For instance, a peak at the m/z ratio of 383 was generated from oxidized

cellotriose (m/z 519) by the loss of 136 Da from cross ring cleavage of the oxidized

monosaccharide unit, leading to the formation of a 2,4A3-ion (Fig. 5.2B).

Additional, indirect evidence, from colorimetric assays, for the oxidation at C1 is that no

activity was detected on D-glucose derivatives lacking a C1 hydroxyl group, including

1,5-anhydroglucitol (D-glucose with -H instead of -OH at C1) and methyl-β -D-

glucopyranoside (D-glucose with -OCH3 instead of -OH at C1)

5.1.3.2 Efficient oxidation of cellooligos. and Impact of chain length on GOOX activity

Mass spectrometric analysis of oxidized cello-oligosaccharides from cellobiose to

cellopentaose revealed a 16 Da increase in m/z values of the acidic fraction (Figure

5.3M-P) compared to the control, un-oxidized samples (Figure 5.3A-D), confirming that

in all cases, the oxidation by GOOX introduced a single oxygen atom to all the

oligomeric substrates. The oxidation of different cello oligosaccharides was efficient, but

not complete at the tested concentrations, as can be seen from small amount of un-

oxidized oligosaccharides detected in the neutral fraction (Figure 5.3I-L).

The current analyses confirmed that the GOOX oxidize mono- and oligo-saccharides

only at the reducing anomeric position suggesting that GOOX production of oxidized

cello oligosaccharides would be an efficient way to generate oxidized carbohydrate

standards to facilitate the characterization of the C1-oxidizing enzymes of family AA9.

83

Figure 5.3 Positive ion ESI-MS spectra of four cello-oligosaccharide samples. G2: Cellobiose; G3: Cellotriose; G4: Cellotetraose; G5: Cellopentaose. (A)-(H): Un-oxidized cello-oligosaccharide samples; (I)-(P): GOOX oxidized cello-oligosaccharide samples; (A)-(D) and (I)-(L): Neutral fractions: (E)-(H) and (M)-(P): Acidic fractions.

84

5.2 Application of GOOX CBM fusions in the synthesis of plant oligosaccharides

with enhanced nutraceutical value

5.2.1 Introduction

Plant oligo- and poly-saccharides have been used to enhance the nutritional value of

different food products (Broekaert et al. 2011). For instance, certain oligosaccharides

have immunostimulatory activity, serving as a potent stimulator of the immune system;

or prebiotic activity, promoting a healthy digestive system (Barreteau 2006). Panesar et al.

(2013) review a comprehensive list of prebiotics currently used in the food industry, and

summarize those enzymes used in prebiotic production. Notably, most enzyme treatments

used to date hydrolyze starting polysaccharides to a series of oligosaccharides (Panesar et

al. 2013).

Previous studies have shown that among prebiotic carbohydrates, acidic oligosaccharides

such as those from pectin are correlated to comparatively high prebiotic activity

(Mandalari et al. 2007). Specific nutritional benefits include metabolic resistance and

promotion of mineral absorption, which have been attributed to the acidic functionality of

these oligosaccharides (Schaafsma 2008). These benefits are exemplified in lactobionic

acid, produced by oxidation of lactose using oligosaccharide oxidases such as GOOX-T1,

MnCO and PCOX (Lin et al. 1996; Nordkvist et al. 2007; Murakami et al. 2008). This

compound is a potential prebiotic that also enhances intestinal absorption of calcium

(Brommage et al. 1993) without contributing to calcification risk; it is also resistant to

digestive enzymes, which makes it available to be fermented by the intestinal flora

(Schaafsma 2008). In particular, the increase in the mineral uptake is believed to occur in

85

the large intestine by the passive absorption. Unlike the active mineral absorption that

depends mainly on vitamin D to transport the minerals across the cells and takes place in

the upper gastrointestinal tract, the passive mineral uptake occurs along the length of

small and large intestine by paracellular diffusion as a result of gradient in mineral

concentrations (Padma Ishwarya and Parbhasankar 2014). Thus, the potential of the acid

to make a strong complex with minerals leaves the acidic prebiotic compound in complex

with mineral in the large intestine where the prebiotic can be fermented to fatty acids,

which is predicted to lower the pH of contents in the large intestine and thereby increase

the solubility of minerals, promoting their passive absorption.

Aldouronic acids from xylan, which are acidic xylooligosaccharides (XOS), have also

been shown to relieve iron deficiency (Kobayashi et al. 2011). However these acidic

XOSs, which comprise xylooligosaccharides substituted with glucuronic acid by α-1,2

bonds (Figure 5.4), do not show prebiotic activity (Ohbuchi et al. 2009) even though

neutral XOSs are known to have prebiotic function (Okazaki et al. 1990; Aachary and

Prapulla 2011; Kondepudi et al. 2012).

Figure 5.4 Structure of compounds used in prebiotic assay. A) xylooligosaccharides (XOS), B) GOOX-CBM44 oxidized XOS, and C) aldouronic acid used in this study.

86

In this study, we hypothesized that the loss in prebiotic activity observed for aldouronic

acids from xylan reflect the branching structure of corresponding oligosaccharides, rather

than the acidity of the molecule. To test this hypothesis, we oxidized neutral XOSs using

a glucooligosaccharide oxidase (GOOX), which selectively produces an aldonic acid at

the reducing end of the molecule (Figure 5.4); both aldouronic acids from xylan and the

GOOX-treated XOSs were then tested using an in vitro prebiotic activity assay. In short,

GOOX treatment of XOS did not reduce the prebiotic potential of XOSs, as evaluated by

cultivation of B. longum. By contrast, the aldouronic acids from xylan were not utilized

by this strain of bifidobacteria. The growth rate of B. longum on GOOX-treated XOSs

and neutral XOSs was similar, and yields of B. longum were slightly higher after 24h

cultivation on GOOX-treated XOSs than neutral XOSs. These findings suggest that

GOOX oxidation of XOS can support bifidobacteria proliferation characteristic of

prebiotic activity, while also providing additional benefits previously correlated to the

acidic functionality.

5.2.2 Materials and Methods

5.2.2.1 Oxidation of xylooligosaccharides

Mixed xylooligosaccharide (DP 2-7, 95% pure) was purchased from Cascade

biochemical (USA), while aldouronic acid was obtained from Megazyme (Ireland). The

mixed xylooligosaccharide was oxidized using the GOOX-CBM fusion that was shown

in Section 4.3.3 to be most active towards XOS, namely GOOX-CBM44. Approximately

4 µg/ml of the enzyme was added to 1% w/v of the filter-sterilized substrate in final

volume of 10 mL, and the reaction was incubated overnight at 37°C. To confirm the

87

oxidation, small-scale reactions were performed in parallel and characterized using the

standard chromogenic GOOX assay described in section 3.2.6 to detect the production of

H2O2. Following enzymatic oxidation, GOOX-CBM44 was de-activated by incubation at

50 °C for 1 h.

5.2.2.2 Prebiotic Assay

Prebiotic assays were performed in Professor Elena Comelli’s lab in the Department of

Nutritional Sciences at the University of Toronto; all prebiotic experiments were

performed under the supervision of Dr. Amel Taibi, a post-doctoral fellow in Professor

Cornelli’s laboratory. Bifidobacterium longum NCC2705 was pre-cultured anaerobically

in a glove box (Coy Laboratories, Midland, MI, USA) containing an atmosphere of 90%

N2, 5% CO2, and 5% H2 (v/v). The pre-cultures were grown on MRS broth (1% Casein

peptone, 1% Meat extract, 0.5% Yeast extract, 0.1%g Tween-80, 0.2% K2HPO4, 0.5%

Na-acetate, 0.2% (NH4)2 citrate, 0.02% MgSO4-7H2O, and 0.005% MnSO4-H2O) at pH

6.5 supplemented with 0.5% glucose. Prior to fermentation, the cells from the pre-culture

were counted under the microscope using a Hemocytometer (Fisher scientific). A volume

corresponding to 5×105 CFU was centrifuged, and the cells were re-suspended in MRS

broth lacking glucose, and supplemented with the 0.5% of defined carbohydrates, i.e.

oxidized and non-oxidized XOS. Amended carbohydrates were the sole carbon source,

and 0.5 % glucose was used as the positive control. A final volume of 1-2 mL was used

for preliminary small-scale cultivations whereas 5 mL cultures were prepared for

studying the effect of enzyme treatment. The cultures were grown anaerobically at 37°C

under stationary conditions in either a gas-pack system (AnaeroGen TM, Oxoid) for

preliminary cultivation, or in a glove box for 5-mL cultures. An aliquot of each culture

88

was harvested at each time point to determine cell growth using viable colony counts.

Viable cell counts were determined by spreading 100 µL of 10-fold serially diluted

cultures onto solid MRS agar plates containing 0.5% glucose, and plates were incubated

in the glove box at 37°C for 24-48 h.

5.2.3 Results and Discussion

5.2.3.1 Small-scale fermentation on xylooligosaccharides and aldouronic acid

Neutral xylooligosaccharides promoted the growth of B. longum almost to the same level

as glucose, while aldoronic acid was not digested by this strain (Figure 5.5). The inability

of B.longum to grow on this compound was expected and consistent with the analysis of

Ohbuchi et al (2009). Based on this result, only xylooligosaccharides before and after

oxidation with GOOX-CBM44 were used for the subsequent cultivations.

Figure 5.5 Viable cell count of B. longum preliminary cultures. The growth media was MRS supplemented with xylooligosaccharides or aldouronic acid. Glucose was served as positive control and MRS media without a sugar was severed as the blank. Data points represent cultivation of one culture per compound.

89

5.2.3.2 Cultivation of B. longum on oxidized and non-oxidized xylooligosaccharides

The viable cell count results indicate slightly higher population of B.longum on GOOX-

CBM44 oxidized XOS vs the non-oxidized substrate at 24 h measurements (Figure 5.6).

However, as shown in Figure 5.6, the growth profile is different throughout the

cultivation time. For instance, the cultures grown on XOS shows higher population than

those grown on enzyme treated XOS after 2 h of incubation while this correlation is not

maintaed at 5 h or 24 h time points.Thus, to quantitavely compare this profile the growth

rate was calculated in the exponential phase using the following equation:

𝐾 =(log𝑁2 − log𝑁1) × 2.3

𝑇2 − 𝑇1

, where N1 is the population (colony forming unit, CFU) at T2 (h), and N1 is the

population at T1 (h).

As shown in Table 5.1, the growth rate of B. longum in the exponential phase from T1 =

2.5 h to T2 = 7.5 h is higher on glucose and is slightly higher on GOOX-CBM44 treated

XOS vs non-treated XOS. However, a paired t-test with p value of 0.05 revealed that the

difference between growth rates on treated and non-treated XOS was not statistically

significant. These data confirm that the acidic functional group does not reduce the

prebiotic activity of XOS as evaluated using B. longum . In the case of aldouronic acid,

also know as acidic XOS, glucoronic acid substituents are at terminal or branching

positions of the xylose backbone. It is possible then, that the main reason for the loss of

prebiotic value could be limited ability to remove glucuronic acids from the

xylooligosaccharide, rather than inability to metabolize acidic xylooligosacccharides in

90

general. In this case, GOOX-CBM44 oxidized XOS could be a viable route to enhancing

the nutritional value of XOSs, for example through promoting mineral absorption.

Figure 5.6 Viable cell count of B. longum cultures. The growth media was MRS supplemented with xylooligosaccharides (XOS) with and without GOOX-CBM44 treatment. Glucose was served as positive control. The error bars indicate standard deviation of three biological replicates.

Table 5.1 Growth rate of B. longum cultivations.

Substrates Growth rate, K (h-1)

XOS 0.51 ± 0.02 a

GOOX-CBM44 treated XOS 0.53 ± 0.09

Glucose 0.60 ± 0.05

a The values are measured at exponential phase between 2.5 to 7.5 h. standard deviations represents the growth rates of the three biological replicates.

1.00E+07

1.00E+08

1.00E+09

1.00E+10

0 5 10 15 20 25 30

Log

(cfu

/ml)

Time (h)

XOS XOS+GOOX Glucose

91

5.2.4 Conclusions

This study compared the growth profile of B. longum on XOS before and after oxidation

by GOOX-CBM44, which introduces a carboxyl groups at the reducing end of the

oligosaccharide. This analysis did not reveal statistically significant differences between

the growth rates of B. longum on the two XOS samples implying that the introduction of

the acidic groups at the reducing end of the xylooligosaccharide does not reduce the

digestibility of XOS by B. longum.

The acidic XOS prepared in this work by GOOX-CBM44 treatment maintains the

bifidobacteria proliferation characteristic of XOS while potentially increasing intestinal

mineral absorption potential. This finding presents a new opportunity for GOOX

application in the preparation of value added prebiotics.

92

5.3 A Mutant gluco-oligosaccharide oxidase is suitable to replace glucose oxidase for

baking applications

5.3.1 Introduction

In baking applications, additives have been widely used to improve the texture, volume,

flavor, and shelf life of the packed goods. Most additives contain enzymes, including

amylase, hemicellulase, cellulase, and carbohydrate oxidases, the action of which results

in improved rheological and handling properties of the dough (Sharma and Singh 2010).

Unlike amylase, xylanase and cellulase, which reduce the molecular weight of

polysaccharides present in the dough, oxidases target hydroxyl groups of corresponding

sugars, thereby producing aldonic acids.

Glucose oxidase (GO) has been widely used as a carbohydrate oxidase in baking

applications (Vemulapalli et al. 1998; Rasiah et al. 2005; Bonet et al. 2006; Hanft and

Koehler 2006; Dagdelen and Gocmen 2007; Decamps et al. 2013) to reduce stickiness

and enhance machinability and stability of the dough while retaining the softness of

baked products after long storage (Bonet et al. 2006). However, the effectiveness of GO

is limited to the selectivity of this enzyme towards glucose, and the generally low

concentration of glucose in cereal flour (Schneider et al. 2003).

Briefly, cereal flour from various grains generally comprise carbohydrates, dietary fibre,

and protein as their main ingredients. The primary constituent of the carbohydrate

fraction is starch, which accounts for 60-70% of the flour (Fišteš et al. 2014); free sugars

including glucose, maltose, and sucrose account for less than 2% (MacArthur and

Dappolonia 1979). The composition of dietary fibres such as cellulose, arabinoxylan, and

93

ß-(1→3), (1→4)-glucan originated mainly from brans and germs (Koehler and Wieser

2013) is different among the grains ranging from 12-20% of whole grain flour and 1-8%

of the white flour from wheat, rye, barley, corn, sorghum and rice (Nyman et al. 1984).

While these percent compositions reflect the polysaccharide content in cereal flour, the

actual fibre content in the baking ingredients is anticipated to be higher because prebiotic

dietary fibres are increasingly used in bakery for their nutritional and technical benefits to

the backed products (Padma Ishwarya and Parbhasankar 2014). In addition, the use of

hydrolytic enzymes in baking processes reflects the presence of corresponding short

oligosaccharides in the dough mixtures. This wide range of carbohydrates present in

cereal flour suggests that an oxidative enzyme with wider carbohydrate specificity is

advantages over GO for dough enhancement.

Accordingly, oxidoreductases that act on oligosaccharides have been shown to perform

better than GO in baking applications in terms of introducing aldonic acids and H2O2 to

the dough mixture (Schneider et al. 2003). More specifically, MnCO used in a baking

trial improved the machinability of the dough in a dose-dependent manner. As well, a

better consistency of the dough was exhibited upon addition of this enzyme (Schneider et

al. 2003). Despite the promising characteristics of MnCO, the performance of this

enzyme in terms of H2O2 inactivation has not been studied in detail.

GO is inactivated in the presence of low concentrations of H2O2, i.e. less than 0.01 M, as

described in section 2.5.1. This compound is the co-product of most enzymatic sugar

oxidations and is an antimicrobial factor, maintaining dough freshness. However, it has

also been shown to oxidize amino acids such as methionine and tryptophan that reside

94

near the active site of carbohydrate oxidases, thereby inactivating these enzymes

(Hachimori et al. 1964; Kleppie 1966). Earlier reports show that immobilized GO under

continuous operation is still inactivated by H2O2 (Greenfield et al. 1975) whereas

liposomal encapsulation of GO (Yoshimoto et al. 2004) can protect the enzyme from

H2O2 inactivation by decomposition of this compound at the lipid membrane, keeping the

concentration of H2O2 low inside and outside of the liposome capsules.

In this work, a mutant gluco-oligosaccharide oxidase, GOOX-Y300A, with higher

specific activity on glucose (as discussed in section 3.3.4) and reduced substrate

inhibition than the wild-type GOOX (Foumani et al., 2011), was investigated for its

potential to replace GO, benefiting the baking applications with a similar mechanism

(Figure 5.7, Figure 5.8) but likely with higher efficiency. GOOX-Y300A is capable of

oxidizing oligosaccharides including maltose-based oligosaccharides, which are the

components of starch, as well as glucose-based and xylose-based oligosaccharides

(Foumani et al., 2011), which are constituents of fibre in flour. GOOX-Y300A in fact,

shows higher rates of activity on oligosaccharides than the corresponding

monosaccharides. In particular, this enzyme is expected to benefit the properties of dough

commonly prepared from flour with the aid of hydrolytic enzymes (e.g. cellulases,

xylanases or amylases), which is expected to have relatively high oligosaccharide content.

This study compares GO and GOOX-Y300A for their specific activities on

oligosaccharides and their inactivation by H2O2.

95

Figure 5.7 Prosed mechanisms for GOOX benefits in baking applications. Proposed effects are similar to those of other oxidative enzymes used in baking (Rasiah et al. 2005; Bonet et al. 2006). The figures were taken from www.cornishpasties.com/breads and www.feastsforallseasons.com/english-muffins.

Figure 5.8 Proposed mechanism for GOOX reinforcing the protein network in dough; similar to that proposed for GO (Rasiah et al. 2005). The figure was generated using ChemSketch.

96

5.3.2 Materials and Methods

Cellobiose, maltose, xylose, glucose, and gluconic acid were reagent grade with over

95% purity and purchased from BioShop Inc. (Canada); cellotetraose (>95% pure) was

obtained from Megazyme (Ireland), whereras mixed xylooligosaccharides (DP-2-7, 95%

pure) were obtained from Cascade Biochemicals (United State). Glucose oxidase from A.

niger (GO) was purchased from Sigma (United state); this product exhibits less than 10

U/mg catalase activity. GOOX-Y300A was prepared as previously described in section

3.2.3 (Foumani et al., 2011).

5.3.2.1 Activity assays

Since different preparation methods were used for GOOX-Y300A and GO, the

concentration of both enzymes was re-measured in parallel using Pierce BCA Protein

Assay (Thermo Scientific). Specific activities of enzymes were then tested on mono- and

oligo-saccharides.

To determine the GOOX-Y300A activity, the standard chromogenic GOOX assay

described in section 3.2.6 was used to measure the production of H2O2. To initiate

oxidation by GOOX, 6 pmol of enzyme was added to the 250 µL reaction mixture (0.1

mM 4-AA, 1 mM phenol, 0.5 U horseradish peroxidase) in 50 mM Tris–HCl buffer (pH

8). Oxidations by GO were performed in the same reaction mixture as above except with

the buffer being 50 mM sodium acetate (pH 5) and were initiated by adding 6 pmol of

enzyme to the same reaction mixture as above. In both cases, reactions were incubated at

37 °C for 15 min and specific activities were separately measured on 10 mM xylose,

maltose and cellobiose as well as 1 mM of glucose, cellotetraose and mixed

97

xylooligosaccharides. One unit of the enzyme activity was correlated to the production of

1 µmol of the product per minute.

5.3.2.2 Oxidation Reactions for H2O2 inactivation study

To measure the effect of H2O2 on the enzyme activities, 0.1 μg of GOOX-Y300A or GO

were added to 250 μL mixtures of 1 mM glucose in either 50 mM Tris-HCl buffer (pH 8)

for GOOX-Y300A, or 50 mM sodium acetate buffer (pH 5) for GO reactions, containing

various concentrations from 0-0.2 M of H2O2. The reactions were incubated at 25 °C for

5h, and then were stopped by removing the enzyme using Nanosep centrifugal device

(Pall Corp). The extent of the oxidation was evaluated by analyzing the amount of

gluconic acid produced using HPLC.

5.3.2.3 Detection of gluconic acid by HPLC

The oxidized product (i.e. gluconic acid) was diluted 5-10 times and then 5 μL were

analyzed using an ICS5000 HPAEC equipped with a pulsed amperometric detector

(Dionex, USA) and CarboPac PA1 column coupled to a guard column. The eluents were

0.1 M NaOH (eluent A) and 1 M NaOAc (eluent B), and elution was performed at 30 °C

with a constant flow rate of 0.25 mL/min using a linear gradient of 0% to 10% eluent B

over 10 min, followed by a linear gradient of 10% to 30% eluent B over 15 min, and

finally from 30% to 100% eluent B over 5 min. Reconditioning of the column was

achieved by running the initial conditions (i.e. eluent A) for 10 min. This resulted in a

total runtime of 30 min.

98

Chromatograms were recorded and the peak area corresponding to gluconic acid was

calculated using Chromeleon 7.2 (Dionex, USA); gluconic acid solutions (0.01-0.5 mM)

were used as standards to correlate the peak area with the gluconic acid concentration.

5.3.2.4 Monitoring the oxygen content using Oxygraph

Oxygen concentrations in reaction mixtures were measured using an Oxygraph

(Hansatech, USA) equipped with a 10 mL chamber and attached to a circulating water

bath to maintain reaction temperatures at 25 °C. Briefly, 500 μL of the reaction buffer

containing 10 mM glucose with 0-2 mM H2O2 were equilibrated to 25 °C, and then the

reaction was initiated by adding 10 μL of either GO or GOOX containing approximately

16 pmol of the enzymes to the reaction chamber using a Hamilton syringe. The oxygen

content was measured continuously and enzyme action was evaluated by determining

oxygen consumption rates using the Oxygraph Plus software (Hansatech, USA).

5.3.3 Results and Discussion

5.3.3.1 GOOX-Y300A shows higher oxidation of oligosaccharides

Specific activity comparison shows wider substrate specificity for GOOX-Y300A

compared to GO. The oxidation activity of GO on maltose, cellotetraose, and

xylooligosaccharides were over two orders of magnitude less than that measured using

similar molar quantities of GOOX-Y300A. The activity on xylose and cellobiose was

more than 30 times lower with GO whereas on glucose, the oxidation activity of GO was

roughly 3 times higher than that of GOOX-Y300A (Table 5.2). The different substrate

preferences of GOOX-Y300A and GO is likely attributed to the open active site of

GOOX (Huang et al. 2005; Foumani et al. 2011) compared with the deep binding pocket

99

of GO which is specifically suited for glucose (Hecht et al. 1993). The activity of GO

reported herein is consistent with an earlier report by Kulys et al. (2001a), which show

weak activity of GO on xylose, maltose, and cellobiose compared to MnCO. Moreover,

due to the reduced substrate inhibition by GOOX-Y300A (Vuong et al. 2013), the current

study suggests that GOOX-Y300A may be particularly advantageous for oxidation of

maltose, as well as xylo- and cello-oligosaccharides generated by cellulases and

xylanases, which are also commonly used enzyme additives in baking applications.

Table 5.2 Specific activities of GOOX-Y300A and GO on selected mono- and oligosaccharides.

a errors represent standard deviations (n=3). b specific activities were calculated in mole basis considering the MW of a monomeric GO (80 KD)

5.3.3.2 H2O2 inactivation

Inactivation of GO by H2O2 has been a bottleneck for industrial applications of GO and

so GO is typically used in combination with catalase, which continuously degrades H2O2.

The requirement for a second enzyme has implications to enzyme cost as well as

environmental benefits of the enzyme approach (Pourbarfani et al. 2004). Thus, an

alternative enzyme with lower H2O2 inactivation is desirable.

Substrates Specific activity (U/µmol)

GOOX-Y300A GO

Glucose 200 ± 20 a 558 ± 7 b

Xylose 270 ± 10 5 ± 2

Maltose 640 ± 30 3.2 ± 0.4

Cellobiose 1000 ± 60 28 ± 3

Cellotetraose 1100 ± 50 0.5 ± 0.2

Mixed Xylooligos 420 ± 10 1.3 ± 0.1

100

Studying H2O2 inactivation of oxidases is challenging since most activity assays for this

class of enzymes detect the production of H2O2 as a by-product of the reaction. Using an

oxygen probe to detect oxygen as the co-substrate of GO, Bao et al. (2003) measured

oxygen consumption as a function of GO activity, and in this way, evaluated GO activity

in the presence of increasing H2O2 concentrations.

With a similar approach, an oxygraph was used to measure rates of glucose oxidation by

GO and GOOX-Y300A in the presence of H2O2. As reported in Appendix 3, the analysis

of GO in the presence of H2O2 was consistent with that reported by Bao et al (2003).

Interestingly however, using the oxygraph to measure GOOX-Y300A activity showed an

increase in oxygen concentrations when 2 mM H2O2 was added to the reaction mixture.

This result is presently difficult to explain, but will be further evaluated in future studies.

For the current investigation, high-performance anion-exchange chromatography

(HPAEC) was subsequently used to further compare the impact of H2O2 of GO and

GOOX-Y300A activity.

Liquid chromatography enables the quantitative analysis of oxidized reaction products.

Accordingly, HPAEC was used to measure the concentration of gluconic acid in

reactions containing 1 mM glucose, 0 to 0.2 M H2O2, and either GO or GOOX. Although

the rate of glucose oxidation by GOOX-Y300A is lower than that of GO, the current

analysis revealed that GOOX-Y300A might be slightly more stable than GO in the

presence of high concentration of H2O2 i.e. 0.2 M (Figure 5.9; Table 5.3).

101

Figure 5.9 H2O2 tolerance of GOOX-Y300A and GO. Percent activity of the enzymes in the presence of various amount of H2O2.

Table 5.3 Amount of gluconic acid produced by GOOX-Y300A or GO in the presence of various concentrations of H2O2.

a errors represent standard deviations (n=3).

The difference in reduction of the activity of GOOX-Y300A and GO with increased

concentration of H2O2 was statistically significant as evaluated using the SlopesTest

function in Excel (Real Statistics Resource Pack) where the log of the data is taken to

H2O2 (mM) Gluconic acid production (mM) Percent activity

GOOX-Y300A GO GOOX-Y300A GO

0 0.16 ± 0.01 a 0.81 ± 0.1 100 100

25 0.14 ± 0.01 0.63 ± 0.11 83 78

50 0.12 ± 0.02 0.59 ± 0.05 74 73

100 0.1 ± 0.01 0.39 ± 0.11

59 48

150 0.09 ± 0.02 0.31 ± 0.09 52 39

200 0.09 ± 0.02 0.26 ± 0.07 54 32

102

generate a linear trend (Figure S5.2) and the corresponding slopes are statistically

compared. This difference is notable when considering the fact that GOOX-Y300A does

not contain catalase while the GO sample used herein is reported to contain

approximately 10 U/mg of catalase activity. Our analysis of GO is consistent with those

report by Kleppe (1996), who show approximately 20% residual GO activity after

exposure to 0.2 M H2O2. Notably, the GO used by Kleppe contained less than 5 U/mg

catalase activity.

It is worth mentioning that for GOOX-Y300A, reduction in activity upon exposure to

H2O2 was not observed in reactions that proceed for only 30 min, or when 0.4 µg of

GOOX-Y300A was used in reactions rather than 0.1 µg. This could be explained by

incomplete oxidation of glucose after 30 min, and perhaps increased stability of GOOX-

Y300A enzyme at higher concentrations. Overall, it is anticipated that the relative

stability of GOOX-Y300A along with its wider substrate specificity will make this

enzyme a suitable substitute to GO in baking applications.

5.3.4 Conclusions

The findings of the current study demonstrate that in dough formulations where

oligosaccharides are in high abundance, GOOX-Y300A may be a better alternative to GO

for the production of aldonic acids and H2O2. Moreover, lower inactivation of GOOX-

Y300A by H2O2 compared to GO is anticipated to reduce requirements for catalase or

protective encapsulation, which could lead to cost savings.

103

Chapter 6 : Conclusions

Enzymes are excellent “green” catalysts in that they are site-specific, regio-selective, and

more importantly can be fine-tuned through protein engineering. The main objective of

my PhD study was built upon this latter property of enzymes: tunable catalysts through

molecular engineering. Accordingly, my aim was to use molecular biology techniques to

engineer enzymes from the AA7 family of oligosaccharide oxidase, and in particular the

glucooligosaccharide oxidase from Sarocladium strictum (GOOX), to enhance its activity

on insoluble and high molecular weight polysaccharides.

6.1 Rational design altered the substrate preference of the GOOX

The results presented in chapter 3 validated the hypothesis that site-directed mutagenesis

of residues predicted to determine substrate preference of GOOX could broaden the

specificity of this enzyme. In particular, the characterization of Y300A and Y300N

variants showed over 2-fold increase in catalytic performance of the enzyme on all tested

mono and oligo-saccharides, as well as gain of activity on mannose. Interestingly, the

Y300 mutation was later shown to result in reduced substrate inhibition compared to the

wild-type GOOX (Vuong et al. 2013). The W351F variant also displayed increased

enzyme efficiency on galactose proving the hypothesis that the low activity of GOOX on

substrates with axial OH group is due to the steric hindrance of this hydroxyl with the

tryptophan at 351position (Lee et al. 2005).

Despite the interesting trend in catalytic performance of the GOOX variants, and

improved binding behaviour on tested monosaccharides, the binding capacity of these

104

mutants was reduced on di- and tri- saccharides. Although this is a limitation for

applications of these variants on oligosaccharides, from a scientific point of view this

finding validated the hypothesis that the tyrosine and tryptophan at 300 and 351positions

are indeed involved in a stacking interaction with the second ring of disaccharide and

oligosaccharide substrates, being important residues for substrate binding by the GOOX

enzyme. Thus, the most significant contributions from this part of the study to the

scientific community were:

• To define GOOX residues that affect the substrate binding and preference of this

enzyme, giving clues for engineering of other AA7 enzymes.

• Production of valuable set of industrial enzymes, GOOX-Y300A and GOOX-

Y300N, with increased efficiency on glucose, xylose, galactose and a gain of

activity on mannose, as well as enhanced catalytic performance on

oligosaccharides and reduced substrate inhibition for applications where oxidation

of mono- and/or oligo-saccharides is required and where the oligosaccharide

concentration is not limiting.

6.2 CBM fusion promoted GOOX binding and activity towards polysaccharides

The results presented in chapter 4 demonstrated that fusion of selected CBMs with

affinity towards plant polysaccharides to GOOX could increase the binding and activity

of this enzyme on the corresponding polysaccharides. Specifically, GOOX fusion to

CtCBM3, CtCBM11 and CtCBM44, resulted in a significant enhancement in binding

capacity of GOOX towards Avicel, RAC, β-glucan and xyloglucan depending on the

selectivity of the appended modules. The activity of the fusion enzymes however, was

105

moderately enhanced on certain polysaccharides that contain no branching substitutions

and are linked through β-1->4 glycosidic bonds; namely glucomannan, RAC and Avicel.

This observation suggests that although the appended CBM increases the proximity of the

catalytic module and targeted substrate, favourable interaction with the enzymes active

site remain a bottleneck. Overall, the important breakthroughs and new insights from this

part of the study are as follows:

• Creation of a stable and active cellulose-immobilized CtCBM3-GOOX system

with significant potential in cellulose-biosensor applications and for cost-effective

and easy purification of GOOX enzyme using CtCBM3 as a purification tag.

• The first report on GOOX action towards polysaccharides including glucomannan,

β-glucan, xyloglucan, CMC, RAC, and Avicel broadens our concept of substrate

range among family AA7 oligosaccharide oxidases.

• The effect of linker on catalytic activity and thermostability was a surprise,

underscoring the importance of careful consideration of linker sequences when

constructing CBM fusions to AA7 enzymes as well as other carbohydrate-active

enzymes.

• The correlation between CBM type on binding of GOOX to particular

polysaccharides confirms that the substrate selectivity of the CBM is retained

even when it is linked to a catalytic module with broad substrate selectivity.

6.3 Engineered and wild-type GOOX have potential application in biofuel and food

industries

106

The results demonstrated in chapter 5 showed that the engineered enzymes as well as the

oxidized products presented in this study can be applied in following applications:

1. The oxidized cellooligosaccharides can be used as sugar standards for the

characterization of LPMO enzymes.

2. The fusion enzyme with comparatively high oxidation activity on

oligosaccharides can enhance the nutraceutical value of those compounds.

3. Mutant GOOX with enhanced activity on mono- and oligosaccharides and

reduced substrate inhibition can potentially replace GO in baking applications.

In particular, the application of GOOX products as sugar standards was demonstrated by

NMR and mass spectrometry analysis of reaction products, which confirmed nearly

complete, regio-selective oxidation of hydroxyls at the anomeric carbon of

oligosaccharides regardless of sugar type or chain length of these compounds.

The GOOX oxidation however, did not enhance the prebiotic activity of compounds as

the GOOX-CBM44 oxidized XOS were shown to promote Bifidobacterium longum

proliferation to the same extents of XOS. However, this result in itself was interesting

when considered in light of the negative impacts of aldouronic acid from xylan with

respect to B. longum utilization. It was concluded that unlike aldouronic acids which

obtain the acidic functionality through substitution of gluconic acid to the XOS backbone,

the GOOX-CBM44 oxidized XOS with an aldonic acid group at the anomeric carbon can

still support the growth of B. longum while potentially promoting intestinal mineral

absorption as a result of its acidic functionality.

107

The GOOX variants was shown to be indeed a very good replacing candidate for GO in

baking applications as GOOX-Y300A showed broader substrate specificity as well as

exhibiting lower level of H2O2 inactivation than the commercial GO.

Accordingly the key findings and significant contributions are as follows:

• GOOX-oxidized oligosaccharides as sugar standards would benefit the

bioconversion research community by enabling detailed biochemical analysis of

LPMO enzymes, which enhance cellulose hydrolysis for biofuel and biochemical

production.

• GOOX treatment of xylooligosaccharides was shown to be an important means to

fine-tune the performance of prebiotics and introduce additional nutraceutical

value.

• The GOOX-Y300A variant with wide substrate range is a better alternative to GO

in baking applications, particularly for dough with high fibre content.

Taking one step back from the specific context of this dissertation, the findings from this

study are aligned with the ultimate goal of our research group to discover, characterize

and engineer carbohydrate active enzymes, which could be then applied as industrial

biocatalysts in the production of high-value bioproducts. In the present study the GOOX

enzyme from a fungal origin was recombinantly expressed in P. pastoris and was

successfully engineered with enhanced activity, binding and stability performance.

Moreover, applications has been sought and studied which led to potential industrial

collaborations. Encouragingly, the outcomes from the present study have shown

108

opportunities for continuation or being branched to new projects. Accordingly the next

section will discuss the future recommendations.

109

Chapter 7 : Future directions

The following paragraphs propose future experiments, which could enable the transfer of

GOOX research to industrial practice.

7.1 Incorportaion of GOOX-oxidized oligosaccharides in an LPMO standard assay.

Studies presented in section 5.1 suggest that GOOX oxidized cellooligosaccharides can

be applied as sugar standards for LPMO assays. In particular, the spectrometry results

confirmed that the structure of oxidized oligosaccharides generated by GOOX is the same

as the most abundant product of LPMO action on cellulose. In order to implement this

concept, the GOOX oxidized cellooligosaccharides need to be purified from the oxidation

reaction mixture, and incorporated in a currently used LPMO assay. The LMPO activity

should be then measured using the GOOX products as standards as well as chemically

synthesized oxidized cellooligosaccharides. A direct comparison will advise the practical

feasibility of this application.

7.2 Effect of debranching enzyme on prebiotic activity of polysaccharides

In section 5.2 it was shown that introducing aldonic acid to the reducing end of XOS

using CBM44-GOOX supports proliferation of B. longum, whereas substiution of

gluconic acid on XOS, as presented in aldouronic acids, prevents digestion by B.

longum. It was suggested that the gluconic acid side-chains in aldouronic acid are

responsible for inability of B. longum to utilize this compound. Accordingly, it would be

interesting to remove the side-chains of aldouronic acid and its polysaccharide,

glucuronoxylan, with the aid of debranching enzymes, specifically α-glucoronidases, and

110

to test the effect of these enzyme treatments on the growth of Bifidobacterium cultures.

With that in mind, an α-glucoronidase have been already expressed and purified by a PhD

candidate in our lab, Ruoyu Yan, and following prebiotic experiments is to be conducted

as part of her PhD project. In addition, a study on the effect of oxidized XOS as well as

other GOOX-oxidized oligosaccharides on mineral absoption ought to be performed to

confirm the nutritional benefit of these products.

7.3 Additional value of a CBM fusion GOOX-Y300A for baking application

The results presented in section 5.3 suggest that GOOX-Y300A is suitable to replace GO

in baking applications due to its wider substrate specificity and lower H2O2 inactivation.

However, in this study the free enzyme has been studied whereas industrial applications

typically use immobilized enzyme for easier recycling and higher stability under

operational conditions. Lin et al. (1996) show that immoblization of the wild-type

GOOX-T1 led to significant enhancement in thermal stability and overall performance of

the enzyme. Accordingly, it is suggested to investigate the immobliziation of GOOX-

Y300A which is fused to a CBM specific to a diatery fibre. This way, the fibre -

immobilized enzyme would be added as a baking ingredient.

So far, the fusion enzyme CtCBM22_GOOX-Y300 containing the xylan specific

CtCBM22 has been constructed, expressed and purified by the post doctoral fellow in our

lab, Dr. Vuong; Ben MacCormick is studying the immobilization of this enzyme to oat

spelt xylan, and is assessing the coresponding effects on thermal stability and H2O2

tolerance of this enzyme. In addition, he is investigating the unique behavior of GOOX in

the presence of H2O2 (Figure S5.2) using HPAEC method as descriebed in section 5.3.2.3

with both free and immobilied enzymes to confirm this activation behavior. Ben

111

MacCormick is interested to also test the possibility of purifiying the enzyme with oat

spelt xylan due to the strong affinity of CtCBM22 towards this polysaccharide (Vuong

and Master 2014). This will be an additonal value for intorducing the GOOX variants in

baking applications.

112

References

Aachary AA, Prapulla SG. 2011. Xylooligosaccharides (XOS) as an Emerging prebiotic: Microbial Synthesis, Utilizatin Structural Characterization, Bioactive Properties and Application. Compr Rev Food Sci F. 10(1).

Agger JW, Isaksen T, Várnai A, Vidal-Melgosa S, Willats WG, et al. 2014. Discovery of LPMO activity on hemicelluloses shows the importance of oxidative processes in plant cell wall degradation. Proc Natl Acad Sci U S A 111: 6287-6292.

Al-Ghazzewi FH, Khanna S, Tester RF, Piggott J. 2007. The potential use of hydrolysed konjac glucomannan as a prebiotic. J Sci Food Agric 87:1758-1766.

Ali E, Zhao G, Sakka M, Kimura T, Ohmiya K, Sakka K. 2005. Functions of family-22 carbohydrate-binding module in Clostridium thermocellum Xyn10C. Biosci Biotechnol Biochem. 69:160-5.

Ali M, Hayashi H, Karita S, Goto M, Kimura T, Sakka K et al. 2001. Importance of the carbohydrate-binding module of Clostridium stercorarium Xyn10B to xylan hydrolysis. Biosci Biotechnol Biochem. 65:41-47.

Arnold K, Bordoli L, Kopp J, Schwede T. 2006. The SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. Bioinformatics 22(2):195-201.

Asam MR, Glish GL. 1997. Tandem mass spectrometry of alkali cationized polysaccharides in a quadrupole ion trap. J Am Soc Mass Spectrom 8:987-995.

Bankar SB, Bule MV, Singhal RS, Ananthanarayan L. 2009. Glucose oxidase-an overview. Biotechnol Adv 27:489-501.

Bao J, Furumoto K, Fukunaga K, Nakao K. 2001. A kinetic study on air oxidation of glucose catalyzed by immobilized glucose oxidase for production of calcium gluconate. Biochem Eng J. 8:91-102.

Bao J, Furumoto K, Yoshimoto M, Fukunaga K, Nakao K. 2003. Competitive inhibition by hydrogen peroxide produced in glucose oxidation catalyzed by glucose oxidase. Biochem Eng J. 13(1):69-72.

Barreteau H. 2006. Production of oligosaccharides as promising new food additive generation. Food Technol Biotechnol 44:323.

Beeson WT, Phillips CM, Cate JHD, Marletta MA. 2012. Oxidative cleavage of cellulose by fungal copper-dependent polysaccharide monooxygenases. J Am Chem Soc 134:890-392.

Bhat KS. 1996. Multiple site-directed mutagenesis. Methods Mol Biol 57:269-277.

113

Bindels LB, Delzenne NM, Cani PD, Walter J. 2015. Towards a more comprehensive concept for prebiotics. Nat Rev Gastroenterol Hepatol. Epub ahead of print.

Blanchard V, Gadkari R, Gerwig G, Leeflang B, Dighe R, Kamerling J. 2007. Characterization of the N-linked oligosaccharides from human chorionic gonadotropin expressed in the methylotrophic yeast Pichia pastoris. Glycoconj J 24(1):33-47.

Bloom JD, Meyer MM, Meinhold P, Otey CR, MacMillan D, Arnold FH. 2005. Evolving strategies for enzyme engineering. Curr Opin Struct Biol., 15:447-452.

Bolam D, Ciruela A, McQueen-Mason S, Simpson P, Williamson M, Rixon J et al. 1998. Pseudomonas cellulose-binding domains mediate their effects by increasing enzyme substrate proximity. Biochem J. :775-781.

Bonet A, Rosell C, Caballero P, Gó, mez M, Pé et al. 2006. Glucose oxidase effect on dough rheology and bread quality: A study from macroscopic to molecular level. Food Chem 99

Book AJ, Yennamalli RM, Takasuka TE, Currie CR, Phillips GN Jr, et al. 2014. Evolution of substrate specificity in bacterial AA10 lytic polysaccharide monooxygenases. Biotechnol Biofuels. 7:109.

Boraston A, Bolam D, Gilbert H, Davies G. 2004. Carbohydrate-binding modules: fine-tuning polysaccharide recognition. Biochem J 382:769-781.

Boraston A, Kwan E, Chiu P, Warren R, Kilburn D. 2003. Recognition and hydrolysis of noncrystalline cellulose. J Biol Chem. 278:6120-6127.

Boraston A, Nurizzo D, Notenboom V, Ducros V, Rose DR, Kilburn DG et al. 2002. Differential Oligosaccharide Recognition by Evolutionarily-related β-1,4 and β-1,3 Glucan-binding Modules. J Mol Biol 319:1143-1156.

Braman J, Papworth C, Greener A. 1996. Site-directed mutagenesis using double-stranded plasmid DNA templates. Methods Mol Biol 57:31-44.

Broekaert WF, Courtin CM, Verbeke K, Van de Wiele T, Verstraete W, Delcour JA. 2011. Prebiotic and other health-related effects of cereal-derived arabinoxylans, arabinoxylan-oligosaccharides, and xylooligosaccharides. Crit Rev Food Sci Nutr 51:178-194.

Brommage R, Binacua C, Antille S, Carrie A. 1993. Intestinal calcium absorption in rats is stimulated by dietary lac-tulose and other resistant sugars. J Nutr 123: 2186-2194.

Cadwell RC, Joyce GF. 1992. Randomization of Genes by PCR mutagenesis. PCR Methods Appl 2:28-33.

Cadwell RC, Joyce GF. 1994. Mutagenic PCR. PCR Methods Appl 3:136-140.

114

Carvalho AL. 2004. The family 11 carbohydrate-binding module of Clostridium thermocellum Lic26A-Cel5E accommodates beta-1,4- and beta-1,3-1,4-mixed linked glucans at a single binding site. J Biol Chem 279:34785-34793.

Cereghino JL, Cregg JM. 2000. Heterologous protein expression in the methylotrophic yeast Pichia pastoris. FEMS Microbiol Rev 24(1): 45-66.

Charnock S, Bolam D, Turkenburg J, Gilbert H, Ferreira L, Davies G et al. 2000. The X6 "thermostabilizing" domains of xylanases are carbohydrate-binding modules: structure and biochemistry of the Clostridium thermocellum X6b domain. Biochemistry. 39:5013-5021.

Chong S, Nissilä T, Ketola R, Koutaniemi S, Derba-Maceluch M, Mellerowicz E et al. 2011. Feasibility of using atmospheric pressure matrix-assisted laser desorption/ionization with ion trap mass spectrometry in the analysis of acetylated xylooligosaccharides derived from hardwoods and Arabidopsis thaliana. Anal Bioanal Chem. 401:2995-3009.

Christiansen C, Abou Hachem M, Janecek S, Viksø-Nielsen A, Blennow A, et al. 2009. The carbohydrate-binding module family 20-diversity, structure, and function. FEBS J 276: 5006-5029.

Ciriminna R, Pagliaro M. 2010. Industrial oxidations with organocatalyst TEMPO and its derivatives. Org Process Res Dev 14(1):245-251.

Costa GL, Bauer JC, McGowan B, Angert M, Weiner MP. 1996. Site-directed mutagenesis using a rapid PCR-based method. Methods Mol Biol 57:239-248.

Costaouëc TL, Pakarinen A, Várnai A, Puranen T, Viikari L. 2013. The role of carbohydrate binding module (CBM) at high substrate consistency: Comparison of Trichoderma reesei and Thermoascus aurantiacus Cel7A (CBHI) and Cel5A (EGII). Bioresour Technol 143: 196-203.

Crittenden R, Karppinen S, Ojanen S, Tenkanen M, Fagerstrom R, Matt J et al. 2002. In vitro fermentation of cereal dietary fibre carbohydrates by probiotic and intestinal bacteria. J Sci Food Agric 82:781-789.

da Silva Perez D, Montanari S, Vignon MR. 2003. TEMPO-mediated oxidation of cellulose III. Biomacromolecules 4:1417-1425.

Dagdelen AF, Gocmen D. 2007. Effect of glucose oxiase, hemicellulose and ascorbic acid on dough and bread quality. J Food Quality 30:1009-1022

Daly R, Hearn MTW. 2005. Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production. J Mol Recognit 18:119-38.

de Nooy A, Pagliaro M, Van Bekkum H, Besemer A. 1997. Autocatalytic oxidation of primary hydroxyl functions in glucans with nitrogen oxides. Carbohydr Res 304:117-123.

115

Decamps K, Joye IJ, Rakotozafy L, Nicolas J, Courtin CM, et al. 2013. The bread dough stability improving effect of pyranose oxidase from trametes multicolor and glucose oxidase from Aspergillus niger: unraveling the molecular mechanism. J Agric Food Chem 61: 7848-7854

Dias FMV, Goyal A, Gilbert HJ, José A M Prates, Ferreira LMA, Fontes CMGA. 2004. The N-terminal family 22 carbohydrate-binding module of xylanase 10B of Clostridium themocellum is not a thermostabilizing domain. FEMS Microbiol Lett 238:71-78.

Diehl HW, Pokorny M, Zissis E, Ness RK, Fletcher HG. 1974. Crystalline cellobiono-1, 5-lactone: its preparation and growth-inhibitory activity in the Avena coleoptile section test. Carbohydr Res 38:364-368.

Dimarogona M, Topakas E, Christakopoulos P. 2013. Recalcitrant polysaccharide degradation by novel oxidative biocatalysts. Appl Microbiol Biotechnol 97:8455-8465.

Domon B, Costello C. 1988. A systematic nomenclature for carbohydrate fragmentations in FAB-MS/MS spectra of glycoconjugates. Glycoconj J 5:397-409.

Drummond DA, Silberg JJ, Meyer MM, Wilke CO, Arnold FH. 2005. On the conservative nature of intragenic recombination. Proc Natl Acad Sci USA 102:5380–5385.

Dusková J, Dohnálek J, Skálová T, Østergaard LH, Fuglsang CC, Kolenko P, Stepánková A, Hasek J. 2009 .Crystallization of carbohydrate oxidase from Microdochium nivale. Acta Crystallogr Sect F Struct Biol Cryst Commun. 65(6):638-640.

Fan Z, Oguntimein GB, Reilly PJ. 2000. Characterization of kinetics and thermostability of Acremonium strictum glucooligosaccharide oxidase. Biotechnol Bioeng 68:231-237.

Ferrari AR, Lee M, Fraaije MW. 2015. Expanding the substrate scope of chitooligosaccharide oxidase from Fusarium graminearum by structure-inspired mutagenesis. Biotechnol Bioeng. Epub ahead of print

Firbank SJ, Rogers MS, Wilmot CM, Dooley DM, Halcrow MA, Knowles PF, McPherson MJ, Phillips SE. 2001. Crystal structure of the precursor of galactose oxidase: an unusual self-processing enzyme. Proc Natl Acad Sci USA 98(23):12932-12937.

Fišteš A, Došenović T, Rakić D, Pajin B, Šereš Z, et al. 2014. Statistical analysis of the basic chemical composition of whole grain flour of different cereal grains. Acta Universitatis Sapientiae, Alimentaria 7:45−53

Foumani M, Vuong TV, Master ER. 2011. Altered substrate specificity of the gluco-oligosaccharide oxidase from Acremonium strictum. Biotechnol Bioeng 108:2261-2269.

Foumani M, Vuong TV, Master ER. 2012. Glucooligosaccharide oxidases from Acremonium strictum and uses thereof. World intellectual property organization. WO 2012/116431.

116

Freelove AC, Bolam DN, White P, Hazlewood GP, Gilbert HJ. 2001. A novel carbohydrate-binding protein is a component of the plant cell wall-degrading complex of Piromyces equi. J Biol Chem 276: 43010-43017.

Gibson GR, Probert HM, Loo JV, Rastall RA, Roberfroid MB. 2004. Dietary modulation of the human colonic microbiota: updating the concept of prebiotics. Nutr Res Rev 17:259.

Gibson MS. 1997. More evidence on the link between bank health and investment in Japan. J Jpn Int Economies 11:296-310.

Giffhorn F (2000) Fungal pyranose oxidases: occurrence, properties and biotechnical applications in carbohydrate chemistry. Appl Microbiol Biotechnol 54: 727-740.

Gilbert HJ, Knox JP, Boraston AB. 2013. Advances in understanding the molecular basis of plant cell wall polysaccharide recognition by carbohydrate-binding modules. Curr Opin Struct Biol 23: 669-677.

Greenfield PF, Kittrell JR, Laurence RL. 1975. Inactivation of immobilized glucose oxidase by hydrogen peroxide. Anal Biochem. 65(1-2):109-124.

Guerreiro CIPD, Fontes CMGA, Gama M, Domingues L. 2008. Escherichia coli expression and purification of four antimicrobial peptides fused to a family 3 carbohydrate-binding module (CBM) from Clostridium thermocellum. Protein Expr Purif 59:161-8.

Hachimori Y, Horinishi H, Kurihara K, Shibata K. (1964). States of amino acid residues in proteins: V. Different reactivities with H2O2 of tryptophan residues in lysozyme, proteinases and zymogens. Biochim. Biophys. Acta 93: 346.

Hägglund P, Eriksson T, Collén A, Nerinckx W, Claeyssens M, Stålbrand H. 2003. A cellulose-binding module of the Trichoderma reesei beta-mannanase Man5A increases the mannan-hydrolysis of complex substrates. J Biotechnol 101:37-48.

Hallberg BM, Henriksson G, Pettersson G, Vasella A, Divne C. 2003. Mechanism of the reductive half-reaction in cellobiose dehydrogenase. J Biol Chem 278(9): 7160-7166.

Hallberg BM, Leitner C, Haltrich D, Divne C. 2004. Crystal structure of the 270 kDa homotetrameric lignin-degrading enzyme pyranose 2-oxidase. J Mol Biol 341(3): 781-796.

Hanft F, Koehler P. 2006. Studies on the effect of glucose oxidase in bread making. J Sci Food Agr 86:1699-1704.

Hecht HJ, Kalisz HM, Hendle J, Schmid RD, Schomburg D. 1993. Crystal structure of glucose oxidase from Aspergillus niger refined at 2.3 A resolution. J Mol Biol. 229(1): 153-172.

117

Hemsworth GR, Davies GJ, Walton PH. 2013. Recent insights into copper-containing lytic polysaccharide mono-oxygenases. Curr Opin Struct Biol 23:660-668.

Hemsworth GR, Henrissat B, Davies GJ, Walton PH. 2014. Discovery and characterization of a new family of lytic polysaccharide monooxygenases. Nat Chem Biol 10:122-126.

Henriksson G, Johansson G, Pettersson G. 2000. A critical review of cellobiose dehydrogenases. J Biotechnol 78: 93-113.

Hernández-Ortega A, Ferreira P, Martínez AT. 2012. Fungal aryl-alcohol oxidase: a peroxide-producing flavoenzyme involved in lignin degradation. Appl Microbiol Biotechnol 93:1395-1410.

Heuts DPHM, Janssen DB, Fraaije MW. 2007. Changing the substrate specificity of a chitooligosaccharide oxidase from Fusarium graminearum by model-inspired site-directed mutagenesis. FEBS Lett 581:4905-4909.

Higham C, Gordon-Smith D, Dempsey C, Wood P. 1994. Direct 1H NMR evidence for conversion of beta-D-cellobiose to cellobionolactone by cellobiose dehydrogenase from Phanerochaete chrysosporium. FEBS Lett. 351:128-132.

Hirose M, Kameyama S, Ohi H. 2002. Characterization of N-linked oligosaccharides attached to recombinant human antithrombin expressed in the yeast Pichia pastoris. Yeast 19(14):1191-1202.

Hofmeister GE. 1991. Linkage position determination in lithium-cationized disaccharides: tandem mass spectrometry and semiempirical calculations. J. Am. Chem. Soc. 113:5964-5970.

Hook F, Rodahl M, Brzezinski P, Kasemo B. 1998. Energy dissipation kinetics for protein and antibody-antigen adsorption under shear oscillation on a quartz crystal microbalance. Langmuir 14: 729-734.

Horn SJ, Vaaje-Kolstad G, Westereng B. 2012. Novel enzymes for the degradation of cellulose. Biotechnol Biofuels 5:1-3.

Huang CH, Lai WL, Lee MH, Chen CJ, Vasella A, Tsai YC, Liaw SH. 2005. Crystal structure of glucooligosaccharide oxidase from Acremonium strictum: a novel flavinylation of 6-S-cysteinyl, 8alpha-N1-histidyl FAD. J Biol Chem 280(46):38831-38838.

Huang CH, Winkler A, Chen CL, Lai WL, Tsai YC, Macheroux P, Liaw SH. 2008. Functional roles of the 6-S-cysteinyl, 8alpha-N1-histidyl FAD in glucooligosaccharide oxidase from Acremonium strictum. J Biol Chem 283(45):30990-30996.

Humphrey W, Dalke A, Schulten K. 1996. VMD - Visual Molecular Dynamics, J. Molec. Graphics 14:33-38

118

Isaksen T, Westereng B, Aachmann FL, Agger JW, Kracher D, Kittl R, Ludwig R, Haltrich D, Eijsink VG, Horn SJ. 2014. A C4-oxidizing lytic polysaccharide monooxygenase cleaving both cellulose and cello-oligosaccharides. J Biol Chem 289:2632-2642.

Isogai A, Kato Y. 1998. Preparation of Polyuronic Acid from Cellulose by TEMPO-mediated Oxidation. Cellulose 5:153-164.

Isogai T, Yanagisawa M, Isogai A. 2009. Degrees of polymerization (DP) and DP distribution of cellouronic acids prepared from alkali-treated celluloses and ball-milled native celluloses by TEMPO-mediated oxidation. Cellulose 16(1):117-127.

Jun H, Bing Y, Keying Z, Xuemei D, Daiwen C. 2009. Thermostable carbohydrate binding module increases the thermostability and substrate-binding capacity of Trichoderma reesei xylanase 2. New Biotechnol 26:53-59.

Kataeva I, Blum D, Li X, Ljungdahl L. 2001. Do domain interactions of glycosyl hydrolases from Clostridium thermocellum contribute to protein thermostability? Protein Eng. 14:167-172.

Kiryu T, Nakano H, Kiso T, Murakami H. 2008. Purification and characterization of a carbohydrate: acceptor oxidoreductase from Paraconiothyrium sp. that produces lactobionic acid efficiently. Biosci Biotechnol Biochem 72(3): 833-841.

Kleppie K. 1966. The effect of hydrogen peroxide on glucose oxidase from Aspergillus niger. Biochemistry, .5(1):139–143

Kobayashi Y, Wakasugi E, Ohbuchi T, Yokoyama M, Yasui R, Kuwahata M, Nakabou Y, Kido Y. 2011. Acidic Xylooligosaccharide promotes recovery from iron deficiency anemia by enhancing serum iron level in rats. Biomed Res. 22(4): 417-423

Koehler P, Wieser H. 2013. Chemistry of Cereal Grains. HandB Sourdough Biotechnol. Chapter 2:11-45.

Kondepudi KK, Ambalam P, Nilsson I, Wadström T, Ljungh Å. 2012. Prebiotic-non-digestible oligosaccharides preference of probiotic bifidobacteria and antimicrobial activity against Clostridium difficile. Anaerobe 18:489-497.

Koseki T, Mochizuki K, Kisara H, Miyanaga A, Fushinobu S, Murayama T et al. 2010. Characterization of a chimeric enzyme comprising feruloyl esterase and family 42 carbohydrate-binding module. Appl Microbiol Biotechnol 86:155-161.

Kristiansen KA, Potthast A, Christensen BE. 2010. Periodate oxidation of polysaccharides for modification of chemical and physical properties. Carbohydr Res 345:1264-1271.

Kujawa M, Ebner H, Leitner C, Hallberg BM, Prongjit M, et al. 2006. Structural basis for substrate binding and regioselective oxidation of monosaccharides at C3 by pyranose 2-oxidase. J Biol Chem 281: 35104-35115.

119

Kulys J, Tetianec L, Schneider P. 2001a. Specificity and kinetic parameters of recombinant Microdochium nivale carbohydrate oxidase. J Mol Catal B-Enzym. 13:95-101

Kulys J, Tetianec L, Schneider P. 2001b. Recombinant Microdochium nivale carbohydrate oxidase and its application in an amperometric glucose sensor. Biosens Bioelectron. 16(4-5):319-324.

Kurokawa T, Lee C, Shiomi N, Nakano A, Katoh S. 2002. Secretion of α-Amylase from Saccharomyces cerevisiae and Pichia pastoris and Characterization of Its C-Terminus with an Anti-Peptide Antibody. J Chem Eng Japan 35:1277-1281.

Langston JA, Shaghasi T, Abbate E, Xu F, Vlasenko E, et al. 2011. Oxidoreductive cellulose depolymerization by the enzymes cellobiose dehydrogenase and glycoside hydrolase 61. Appl Environ Microbiol 77: 7007-7015.

Lee MH, Lai WL, Lin SF, Hsu CS, Liaw SH, Tsai YC. 2005. Structural characterization of glucooligosaccharide oxidase from Acremonium strictum. Appl Environ Microbiol 71:8881-8887.

Leferink NG, Heuts DP, Fraaije MW, van Berkel WJ. 2008. The growing VAO flavoprotein family. Arch Biochem Biophys 474:292-301.

Lehmann M, and Wyss M. 2001. Engineering proteins for thermostability: the use of sequence alignments versus rational design and directed evolution. Curr Opin Biotechnol 12:371-375.

Lehmann M, Loch C, Middendorf A, Studer D, Lassen SF, Pasamontes L, Loon A, Wyss M. 2002. The consensus concept for thermostability engineering of proteins: further proof of concept. Prot Eng 15(2):403-411.

Lehtiö J, Sugiyama J, Gustavsson M, Fransson L, Linder M, et al. 2003. The binding specificity and affinity determinants of family 1 and family 3 cellulose binding modules. Proc Natl Acad Sci U S A. 100: 484-489.

Letourneur O, Gervasi G, Gaïa S, Pagès J, Watelet B, Jolivet M. 2001. Characterization of Toxoplasma gondii surface antigen 1 (SAG1) secreted from Pichia pastoris: evidence of hyper O-glycosylation. Biotechnol Appl Biochem 33(1):35-45.

Levasseur A, Drula E, Lombard V, Coutinho PM, Henrissat B. 2013. Expansion of the enzymatic repertoire of the CAZy database to integrate auxiliary redox enzymes. Biotechnol Biofuels 6:41.

Lin SF, Hu HM, Inukal T, Tsai YC. 1993. Production of novel oligosaccharide oxidase by wheat bran solid-state fermentation. Biotechnol Adv 11: 417-427.

Lin SF, Hwang YL, Tsai YC. 1996. Immobilization of glucooligosaccharide oxidase of Acremonium strictum for oligosaccharic acid production. Biotechnol Tech 10:63-68.

120

Lin SF, Yang TY, Inukai T, Yamasaki M, Tsai YC. 1991. Purification and characterization of a novel glucooligosaccharide oxidase from Acremonium strictum T1. Seizure 1118:41-47.

Ludwig R, Ortiz R, Schulz C, Harreither W, Sygmund C, Gorton L. 2013. Cellobiose dehydrogenase modified electrodes: advances by materials science and biochemical engineering. Anal Bioanal Chem 405:3637-3658.

MacArthur LA, D’appolonia. 1979. Comparison of oat and wheat carbohydrates. I. Sugars. Cereal Chem 56:455-457.

Mamo G, Hatti-Kaul R, Mattiasson B. 2007. Fusion of carbohydrate binding modules from Thermotoga neapolitana with a family 10 xylanase from Bacillus halodurans S7. Extremophiles 11:169-177.

Mandalari G, Nueno Palop C, Tuohy K, Gibson GR, Bennett RN, Waldron KW et al. 2007. In vitro evaluation of the prebiotic activity of a pectic oligosaccharide-rich extract enzymatically derived from bergamot peel. Appl Microbiol Biotechnol 73:1173-1179.

Mangala S, Kittur F, Nishimoto M, Sakka K, Ohmiya K, Kitaoka M et al. 2003. Fusion of family VI cellulose binding domains to Bacillus halodurans xylanase increases its catalytic activity and substrate-binding capacity to insoluble xylan. J Mol Catalysis B Enz 21:221-230.

Meng DD, Ying Y, Chen XH, Lu M, Ning K, Wang LS, Li FL. 2015. Distinct Roles for Carbohydrate-Binding Modules of GH10 and GH11 Xylanases from Caldicellulosiruptor sp. F32 in Thermostability and Catalytic Efficiency. Appl Environ Microbiol. Epub ahead of print.

Mizutani K, Fernandes VO, Karita S, Luís AS, Sakka M, Kimura T, Jackson A, Zhang X, Fontes CM, Gilbert HJ, Sakka K. 2012. Influence of a mannan binding family 32 carbohydrate binding module on the activity of the appended mannanase. Appl Environ Microbiol 78:4781-4787.

Morgenstern I, Powlowski J, Tsang A. 2014. Fungal cellulose degradation by oxidative enzymes: from dysfunctional GH61 family to powerful lytic polysaccharide monooxygenase family. Brief Funct Genomics 13:471-481.

Morris GM, Goodsell DS, Halliday RS, Huey R, Hart WE, Belew RK, Olson AJ. 1998. Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function. J Comp Chem 19(14):1639-1662.

Murakami H, Kiryu T, Kiso T, Nakano H. 2008. Production of calcium lactobionate by a lactose-oxidizing enzyme from Paraconiothyrium sp. KD-3. J Appl Glycosci 55:127-132.

Najmudin S, Guerreiro CI, Carvalho AL, Prates JA, Correia MA, et al. 2006. Xyloglucan is recognized by carbohydrate-binding modules that interact with beta-glucan chains. J Biol Chem 281: 8815-8828.

Nordkvist M, Nielsen PM, Villadsen J. 2007 .Oxidation of lactose to lactobionic acid by a

121

Microdochium nivale carbohydrate oxidase: kinetics and operational stability. Biotechnol Bioeng. 97(4):694-707.

Notenboom V, Boraston AB, Williams SJ, Kilburn DG, Rose DR. 2002. High-resolution crystal structures of the lectin-like xylan binding domain from Streptomyces lividans xylanase 10A with bound substrates reveal a novel mode of xylan binding. Biochemistry 41: 4246-4254.

Nouaille R, Matulova M, Pätoprstý V, Delort A, Forano E. 2009. Production of oligosaccharides and cellobionic acid by Fibrobacter succinogenes S85 growing on sugars, cellulose and wheat straw. Appl Microbiol Biotechnol. 83:425-433.

Nyman M, Siljestrom M, Pedersen B, Bachknudsen KE, ASP N-G et al. 1984. Dietary fiber content and composition in six cereals at different extraction rates. Cereal Chem 61:14-19.

Ohbuchi T, Takahashi T, Azumi N, Sakaino M. 2009. Structual analysis of neutral and acidic xylooligosaccharides from hardwood kraft pulp, and their utilization by intestinal bacteria in vitro. Biosci Biotechnol Biochem. 73(9):2070-2076

Okazaki M, Fujikawa S, Matsumoto N. 1990. Effect of xylo-oligosaccharide on the growth of bifidobacteria. Bifidobacteria Microflora 9:77–86.

Osten CVD, Bjornvad ME, Vind J, Rasmussen MD, inventor; 2000 a. Process and composition for desizing cellulosic fabric with an enzyme hybrid. Patent US6017751.

Osten CVD, Cherry JR, Bjornvad ME, Vind J, Rasmussen MD, inventor; 2000 b. Process for removal or bleaching of soiling or stains from cellulosic fabric. Patent US6015783.

Packer N, Lawson M, Jardine D, Redmond J. 1998. A general approach to desalting oligosaccharides released from glycoproteins. Glycoconj J. 15:737-747.

Padma Ishwarya S, Prabhasankar P. 2014. Prebiotics: application in bakery and pasta products. Crit Rev Food Sci Nutr 54:511-22.

Palframan R, Gibson G, Rastall R. 2003. Development of a quantitative tool for the comparison of the prebiotic effect of dietary oligosaccharides. Lett Appl Microbiol 37:281-284.

Palonen H, Tjerneld F, Zacchi G, Tenkanen M. 2004. Adsorption of Trichoderma reesei CBH I and EG II and their catalytic domains on steam pretreated softwood and isolated lignin. J Biotechnol 107: 65–72.

Panesar PS, Kumari S, Panesar R. 2013. Biotechnological approaches for the production of prebiotics and their potential applications. Crit Rev Biotechnol. 33:345–364.

Parikka K, Leppänen A, Pitkänen L, Reunanen M, Willför S, Tenkanen M. 2010. Oxidation of polysaccharides by galactose oxidase. J Agric Food Chem. 58:262-271.

122

Parikka K, Leppänen AS, Xu C, Pitkänen L, Eronen P, Österberg M et al. 2012. Functional and anionic cellulose-interacting polymers by selective chemo-enzymatic carboxylation of galactose-containing polysaccharides. Biomacromolecules 13:2418-2428.

Parikka K, Tenkanen M. 2009. Oxidation of methyl alpha-D-galactopyranoside by galactose oxidase: products formed and optimization of reaction conditions for production of aldehyde. Carbohydr Res 344(1):14-20.

Peterbauer CK, Volc J. 2010. Pyranose dehydrogenases: biochemical features and perspectives of technological applications. Appl Microbiol Biotechnol 85:837-848.

Peterbauer CK, Volc J. 2010. Pyranose dehydrogenases: biochemical features and perspectives of technological applications. Appl. Microbiol. Biotechnol. 85, 837-848.

Phillips CM, Beeson WT, Cate JH, Marletta MA. 2011. Cellobiose dehydrogenase and a copper-dependent polysaccharide monooxygenase potentiate cellulose degradation by Neurospora crassa. ACS Chem Biol 6:1399-1406.

Pourbarfani M, Jon McKechnie J, Shen T, Saville BA, MacLean HL. 2014. Impacts of pre-treatment technologies and co-products on greenhouse gas emissions and energy use of lignocellulosic ethanol production. J Clean Prod 78:104-111.

Quinlan RJ, Sweeney MD, Lo Leggio L, Otten H, Poulsen JCN, Johansen KS et al. 2011. Insights into the oxidative degradation of cellulose by a copper metalloenzyme that exploits biomass components. Proc Natl Acad Sci 108:15079-15084.

Ramos R, Domingues L, Gama M. 2010. Escherichia coli expression and purification of LL37 fused to a family III carbohydrate-binding module from Clostridium thermocellum. Protein Expr Purif 71:1-7.

Rasiaha I A, Suttonb KH, Lowa FL, Lina H-M, Gerrard JA. 2005. Crosslinking of wheat dough proteins by glucose oxidase and the resulting effects on bread and croissants. Food Chem 89: 325–332

Remington SJ. 2011. Green fluorescent protein: a perspective. Protein Sci. 20(9):1509-1519

Saito T, Hirota M, Tamura N, Isogai A. 2010. Oxidation of bleached wood pulp by TEMPO/NaClO/NaClO2 system: effect of the oxidation conditions on carboxylate content and degree of polymerization. J Wood Sci 56(3):227-232.

Schaafsma G. 2008. Lactose and lactose derivatives as bioactive ingredients in human nutrition. Int Dairy J. 18:458-465.

Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years of image analysis. Nat Methods 9:671-675.

123

Schneider P, Christensen S, Dybdal L, Fuglsang C, Xu F, Golightly E. 2003. Carbohydrate oxidase and use thereof in baking. Patent US 2003/0180416 Al.

Shani N, Shani Z, Shoseyov O, Mruwat R, Shoseyov D. 2011. Oxidized cellulose binding to allergens with a carbohydrate-binding module attenuates allergic reactions. J Immunol 186:1240-7.

Sharma K, Singh J. 2010. Enzymes in baking industry, in Panesar PS, Marwaha SS, Chopra HK (Eds.), Enzymes in Food Processing: Fundamentals and Potential Applications. I. K. International Publishing House Pvt. Ltd., New Delhi, pp. 75-96.

Shi X, Karkut T, Chamankhah M, Alting-Mees M, Hemmingsen SM, Hegedus D. 2003. Optimal conditions for the expression of a single-chain antibody (scFv) gene in Pichia pastoris. Protein Expr Purif 28:321-330.

Shoseyov O, Shani Z, Levy I. 2006. Carbohydrate binding modules: biochemical properties and novel applications. Microbiol Mol Biol Rev 70:283-295.

Sinclair G, Choy F. 2002. Synonymous codon usage bias and the expression of human glucocerebrosidase in the methylotrophic yeast, Pichia pastoris. Protein Expr Purif. 26:96-105.

Spaduit O, Olsson L, Brumer H. 2010. A comparative summary of expression systems for the recombinant production of galactose oxidase. Microb Cell Fact 9: 68.

Steipe B. 2004. Consensus based engineering of protein stability: from intrabodies to thermostable enzymes. Methods Enzym 388.

Telke AA, Ghatge SS, Kang SH, Thangapandian S, Lee KW, et al. 2012 Construction and characterization of chimeric cellulases with enhanced catalytic activity towards insoluble cellulosic substrates. Bioresour Technol 112: 10-17.

Tomme P, Van Tilbeurgh H, Pettersson G, Van Damme J, Vandekerckhove J, et al. 1988. Studies of the cellulolytic system of Trichoderma reesei QM 9414. Analysis of domain function in two cellobiohydrolases by limited proteolysis. Eur J Biochem 170: 575-581.

Tull D, Gottschalk T, Svendsen I, Kramhøft B, Phillipson B, Bisgård-Frantzen H et al. 2001. Extensive N-glycosylation reduces the thermal stability of a recombinant alkalophilic bacillus alpha-amylase produced in Pichia pastoris. Protein Expr Purif. 21:13-23.

Vaaje-Kolstad G, Horn SJ, Sørlie M, Eijsink VG. 2013. The chitinolytic machinery of Serratia marcescens--a model system for enzymatic degradation of recalcitrant polysaccharides. FEBS J 280:3028-3049.

van Hellemond EW, Leferink NG, Heuts DP, Fraaije MW, van Berkel WJ. 2006. Occurrence and biocatalytic potential of carbohydrate oxidases. Adv Appl Microbiol 60:17-54.

124

Várnai A, Mäkelä MR, Djajadi DT, Rahikainen J, Hatakka, A, et al. 2014. Carbohydrate-binding modules of fungal cellulases: occurrence in nature, function, and relevance in industrial biomass conversion. Adv Appl Microbiol 88: 103-165.

Várnai A, Siika-Aho M, Viikari L. 2013. Carbohydrate-binding modules (CBMs) revisited: reduced amount of water counterbalances the need for CBMs. Biotechnol Biofuels 6: 30.

Vemulapalli V, Miller KA, Hoseney RC. 1998. Glucose oxidase in breadmaking systems. Cereal Chem 75: 439–442.

Viegas A, Brás NF, Cerqueira NMFSA, Fernandes PA, Prates JAM, Fontes CMGA et al. 2008. Molecular determinants of ligand specificity in family 11 carbohydrate binding modules: an NMR, X-ray crystallography and computational chemistry approach. FEBS J 275:2524-2535.

Voutilainen SP, Nurmi-Rantala S, Penttilä M, Koivula A. 2014. Engineering chimeric thermostable GH7 cellobiohydrolases in Saccharomyces cerevisiae. Appl Microbiol Biotechnol 98: 2291-3001.

Vu VV, Beeson WT, Span EA, Farquhar ER, Marletta MA. 2014. A family of starch-active polysaccharide monooxygenases. Proc Natl Acad Sci U S A 111:13822-13827.

Vuong TV, Master ER. 2014. Fusion of a xylan-binding module to gluco-oligosaccharide oxidase increases activity and promotes stable immobilization. PLoS ONE. 9: e95170.

Vuong TV, Vesterinen AH, Foumani M, Juvonen M, Seppälä J, et al. 2013. Xylo- and cello-oligosaccharide oxidation by gluco-oligosaccharide oxidase from Sarocladium strictum and variants with reduced substrate inhibition. Biotechnol Biofuels 6: 148.

Wan W, Wang D, Gao X, Hong J. 2011. Expression of family 3 cellulose-binding module (CBM3) as an affinity tag for recombinant proteins in yeast. Appl Microbiol Biotechnol 91:789-798.

Westereng B, Agger JW, Horn SJ, Vaaje-Kolstad G, Aachmann FL, Stenstrøm YH et al. 2013. Efficient separation of oxidized cello-oligosaccharides generated by cellulose degrading lytic polysaccharide monooxygenases. (1). J Chromatogr A 1271:144-152.

Whittaker JW. 2005. The radical chemistry of galactose oxidase. Arch Biochem Biophys 433(1):227-239.

Wohlfahrt G, Witt S, Hendle J, Schomburg D, Kalisz HM, Hecht HJ. 1999. 1.8 and 1.9Å resolution structures of the Penicillium amagasakiense and Aspergillus niger glucose oxidases as a basis for modelling substrate complexes. Acta Crystallogr D Biol Crystallogr 55(Pt 5):969-977.

Wongnate T, Chaiyen P. 2013. The substrate oxidation mechanism of pyranose 2-oxidase and other related enzymes in the glucose-methanol-choline superfamily. FEBS J 280:3009-3027.

125

Wu WT, Cheng HC, Chen HL. 2011. Ameliorative effects of konjac glucomannan on human faecal β-glucuronidase activity, secondary bile acid levels and faecal water toxicity towards Caco-2 cells. British J Nutri 105:593-600.

Xu F, Golightly EJ, Fuglsang CC, Schneider P, Duke KR, Lam L et al. 2001. A novel carbohydrate:acceptor oxidoreductase from Microdochium nivale. Eur J Biochem 268:1136-1142.

Ye X, Zhu Z, Zhang C, Zhang YH. 2011. Fusion of a family 9 cellulose-binding module improves catalytic potential of Clostridium thermocellum cellodextrin phosphorylase on insoluble cellulose. Appl Microbiol Biotechnol 92:551-560.

Yoshimoto M, Miyazaki Y, Sato M, Fukunaga K, Kuboi R, Nakao K. 2004. Mechanism for high stability of liposomal glucose oxidase to inhibitor hydrogen peroxide produced in prolonged glucose oxidation. Bioconjug Chem. 15(5): 1055-1061.

Yoshimoto M, Sato M, Wang S, Fukunaga K, Nakao K. 2006. Structural stability of glucose oxidase encapsulated in liposomes to inhibition by hydrogen peroxide produced during glucose oxidation. Biochem Eng J. 30(2): 158–163

Zamocky M, Ludwig R, Peterbauer C, Hallberg BM, Divne C, Nicholls P, Haltrich D. 2006. Cellobiose dehydrogenase--a flavocytochrome from wood-degrading, phytopathogenic and saprotropic fungi. Curr Protein Pept Sci 7:255-280.

Zhang M, Wu SC, Zhou W, Xu B. 2012. Imaging and measuring single-molecule interaction between a carbohydrate-binding module and natural plant cell wall cellulose. J Phys Chem B 116: 9949-9956.

Zhang Y, Cui J, Lynd L, Kuang L. 2006. A transition from cellulose swelling to cellulose dissolution by o-phosphoric acid: evidence from enzymatic hydrolysis and supramolecular structure. Biomacromolecules 7:644-648.

Zou G, Shi S, Jiang Y, van den Brink J, de Vries RP, Chen L, Zhang J, Ma L, Wang C, Zhou Z. 2012. Construction of a cellulase hyper-expression system in Trichoderma reesei by promoter and enzyme engineering. Microb Cell Fact 11:21.

126

Appendix 1: Supplemental information for chapter 3

Figure S3.1 Multiple sequence alignment of GOOX homologues. The alignment between MnCO (CAI94231-2) from Microdochium nivale, ChitO (XP_391174) from Fusarium graminearum, and GOOX (noted as GOOX-VN) was generated by Megalign (DNASTAR-Lasergene). Amino acids, which were mutated, are highlighted with asterisks.

127

Figure S3.2 Stability of wild-type and mutant GOOX at 37°C. Residual activity of GOOX (circle), W351F (square), Y300A (cross) and Y300N (triangle) enzymes on 10 mM maltose after incubation at 37°C

in triplicate for up to 1 h.

Figure S3.3 SDS-PAGE of deglycosylated GOOX and mutant enzymes. SDS-PAGE was performed using a 12 % polyacrylamide gel and proteins were stained with Coomasie Blue. Deglycosylated samples were indicated with asterisks. Supernatant from P. pastoris containing pPICZαA (V) was also included. PageRulerTM Plus prestained protein ladder (Fermentas) was used.

128

Appendix 2: Supplemental information for chapter 4

Figure S4.1 Purified wild-type and fusion GOOX proteins on 10% SDS-PAGE. Lane1: PageRuler protein ladder; 2: CtCBM3_GOOX; 3: CtCBM11_GOOX; 4: CtCBM44_GOOX; 5: wild-type GOOX, 6: GOOX_CtCBM3; 7: GOOX_CtCBM11; 8: GOOX_CtCBM44.

Figure S4.2 Binding of wild-type GOOX and CBM fusions to insoluble cellulose as analyzed by SDS-PAGE. Purified proteins were incubated with crystalline cellulose (Avicel) or regenerated amorphous cellulose (RAC) for 2 h on ice with continuous shaking. Unbound (U) and bound (B) protein fractions are shown.

129

Figure S4.3 Specific activity of wild-type GOOX and CBM fusions on konjac glucomannan A. 0.1 %, B. 0.3%. All reactions contained 0.5 µg of enzyme. Error bars represents standard deviations; n=3. The doted line represents the specific activity of wild-type GOOX.

Figure S4.4 Adsorbed mass of wild-type GOOX and CtCBM3_GOOX on cellulose-coated sensors. Changes in adsorbed mass during 1.5 µg/mL enzyme addition (1), 50 mM Tris-HCl pH 8 buffer washing (2) and 0.5 mM cellobiose addition (3) in the experiments with CtCBM3_GOOX (green, solid line) and wild-type GOOX (red, dashed line). Mass values were obtained using the Voigt model.

130

Figure S4.5 Cellobiose oxidation of enzyme-bound sensors. QCM-D sensors that were previously bound with CtCBM3_GOOX were repeatedly washed and incubated with 0.5 mM cellobiose, and the regeneration of oxidized products was measured by the chromogenic assay.

Figure S4.6 Thermostability of proteins at 45°C. A: wild-type GOOX, B: CtCBM3_GOOX, C: CtCBM11_GOOX, D: CtCBM44_GOOX, E: GOOX_CtCBM3, F: GOOX_CtCBM11, G: GOOX_CtCBM44, H: logarithmic presentation of the graphs used to determine the half-lives.

131

Appendix 3: Supplemental information for chapter 5

Observations in experiments using Oxygraph

As mentioned in section 5.3.3.2 the H2O2 inactivation of GO and GOOX-Y300A was

first attempted to be measured by monitoring the oxygen concentration using Oxygraph.

Depending on the susceptibility of the enzyme to H2O2, the rate of oxygen depletion

would decrease with increased addition of H2O2. Notably, the oxygen concentration was

increased when GOOX-Y300A was added in the presence of 0.2 mM H2O2 (Figure S5.1)

resulting in oxygen production instead of oxygen consumption. This phenomenon was

not observed when GO, or solely the buffer was added to the reaction containing H2O2.

To the same reaction when catalase was added, a spontaneous increase of oxygen level

was observed as a result of decomposition of H2O2. Although the increase level of

oxygen with GOOX-Y300A was not as sharp as that with catalase, the similar trend

observed is interesting. These results suggest that GO and GOOX-Y300A will show

different behavior at the presence of H2O2. Further experiments are being designed to

propose a mechanism for explaining this behavior, which is described in the future

direction section.

132

Figure S5.1 Behavior of GOOX-Y300A in the presence (A) or absence (B) of H2O2. Reactions contain 10mM glucose, 16 pmole enzyme in Tris-HCl pH 8. The arrows show the time point where enzyme is added.

Figure S5.2 - Log of GluA concentration produced by GOOX-Y300A or GO versus H2O2 concentration.