Interactions between dietary polyphenols and plant cell wall models

This project addresses important aspects related to polyphenol bioaccessibility through investigations of: i the interactions between diverse polyphenols and cellulose, a main dietary fi

INTERACTIONS BETWEEN DIETARY POLYPHENOLS

AND PLANT CELL WALL MODELS

Anh Dao Thi Phan

B.Eng Food Technology (Can Tho University, Vietnam)

MSc Food Science and Technology (University of Copenhagen, Denmark)

A thesis submitted for the degree of Doctor of Philosophy at

The University of Queensland in 2016

ARC Centre of Excellence in Plant Cell Walls, Centre for Nutrition and Food Sciences, Queensland Alliance for Agriculture and Food Innovation

i

Abstract

The nutritional and health benefits of polyphenols in protecting humans from the risk of chronic diseases (e.g cardiovascular disease, cancers) are now well established from both epidemiological and intervention studies Recently, there has been emerging interest in the interactions between polyphenols and food components, particularly dietary fibres derived from plant cell walls (PCW),

in determining the health-promoting effects of polyphenols These interactions may have a major role in controlling the bioaccessibility of polyphenols by modulating the release of polyphenols in the upper digestive tract, or delivering bound polyphenols to the colon, for further release and metabolism by the resident microbiota

This project addresses important aspects related to polyphenol bioaccessibility through

investigations of: (i) the interactions between diverse polyphenols and cellulose, a main dietary fibre component, and determination of the mechanism behind such interactions; (ii) the effects of

environmental factors, that are relevant to food systems and digestion conditions, on the interactions

between polyphenols and cellulose; (iii) the binding selectivity of diverse polyphenols to different

PCW components (cellulose, hemicellulose, pectin), and identification of factors that influence such

binding selectivity; and (iv) the release and metabolism of polyphenols associated with PCW during

in vitro gastrointestinal digestion and colonic fermentation

A systematic understanding of the interactions between polyphenols and PCWs and the mechanisms behind such interactions has been developed Binding kinetics and binding isotherm studies of diverse polyphenols to cellulose and other PCW components permit quantification of the binding extents/rates, and estimation of the apparent maximum binding capacities and binding affinities of different polyphenols to various PCW components The binding capacities achieved were in the range of 20–60% of the PCW mass, and the apparent maximum binding capacity was predicted to

be 30–150% PCW mass, depending on the polyphenol molecules and the PCW components This implies that a high amount of ingested polyphenols might be present in the human digestive tract in non-bioaccessible forms, if the interactions with PCWs are relatively stable under gastric and small intestinal conditions

By employing a model system of bacterial cellulose-based composites which have previously been shown to represent important features of the chemical organization and assembly of primary PCWs, this project overcomes the difficulties of working with the complexity of heterogeneous cell wall materials and is able to define the differences in binding behaviours of polyphenols to various PCW

ii

constituents The binding selectivity of diverse polyphenols to different PCW components was determined and quantified, with the binding capacities following the order of Cellulose>Cellulose-arabinoxylan>Cellulose-xyloglucan>Cellulose-pectin>Apple cell walls, for the binding of (+/-)-catechin and ferulic acid, while Cellulose-pectin composites showed the greatest binding for cyanidin-3-glucoside (Cya-3-glc) In addition, both intrinsic factors (e.g the chemical characteristics of polyphenols, the chemical organization and local microstructure of cell wall components, and the charged nature of cell wall polymers) and extrinsic factors (e.g pH, temperature and salt) are major factors in modulating the binding process in a food system and/or under human digestion conditions, suggesting the simultaneous and complex effects of multiple factors on the binding rates and extents of polyphenols

This project observed the incomplete release of polyphenols associated with cellulose and apple cell walls under gastrointestinal digestion conditions, with the most release occurring in the gastric phase as a result of the effects of the acidic environment on polyphenol/PCW interactions There was limited release of bound (+/-)-catechin (2035%) and bound Cya-3-glc (3060%), whereas greater release was observed for the bound ferulic acid (4070%) This indicates that a large proportion of polyphenols are likely to be delivered to the colon for further release and metabolism during the fermentation by colonic microbiota

In order to investigate the potential release and metabolism of polyphenols in the large intestine, a

72h in vitro colonic fermentation was conducted for the complexes of polyphenols with apple cell

walls (ACW) or bacterial cellulose (BC), using a pig faecal inoculum All ACW substrates were

fermented faster than the BC substrates, as evidenced by the significantly (p<0.05) higher

production of gas and short chain fatty acids Bound polyphenols were partly released (22–60%) by 2h, followed by a quick decrease (2–6h), and total disappearance after 9h of fermentation, with the relative release rates being ferulic acid > Cya-3-glc > (+/-)-catechin The released polyphenols were completely metabolized by pig feacal microbiota via various proposed pathways, leading to the formation of a small number of possible metabolites Although the presence of different PCWs did not show an alteration in the metabolic pathways mediated by pig faecal bacteria for the three polyphenols studied, polyphenols associated with PCWs underwentfaster microbial metabolism compared to the control sample without any PCW The results suggest the essential role of PCW in the transportation of polyphenol precursors to the colon, and subsequent enhancement of the microbial metabolism,leading to the production of polyphenol metabolites that might exert health benefits

iii

Declaration by author

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis The content of my thesis

is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the policy and procedures of The University of Queensland, the thesis be made available for research and study in accordance with the Copyright Act 1968 unless a period of embargo has been approved by the Dean of the Graduate School

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis

iv

Publications during candidature:

Peer-reviewed papers:

 PHAN, A D T., NETZEL, G., WANG, D., FLANAGAN, B M., D’ARCY, B R & GIDLEY,

M J 2015 Binding of dietary polyphenols to cellulose: Structural and nutritional aspects Food Chemistry, 171, 388-396 (Incorporated in Chapter 3)

 PHAN, A D T., D'ARCY, B R & GIDLEY, M J 2016 Polyphenol-cellulose interactions:

effects of pH, temperature and salt International Journal of Food Science and Technology, 51,

203-211 (Incorporated in Chapter 4)

Conference abstracts and presentations:

 PHAN, A D T., NETZEL, G., D’ARCY, B R & GIDLEY, M J 2014 Adsorption of dietary polyphenols to cellulose: Quantification and mechanisms Proceeding “Polyphenol

Communication 2014”, 365-366, 27 th

International Conference on Polyphenols and 8 th Tannin Joint Conference, Nagoya University, Japan, 2-6th September 2014 (Poster presentation)

 PHAN, A D T., NETZEL, G., D’ARCY, B R & GIDLEY, M J 2013 The roles of cellulose

in controlling the bioaccessibility of dietary polyphenols Nutrition Society of Australia & Nutrition Society of New Zealand Joint Annual Scientific Meeting, Brisbane, 4-6th December

2013 (Oral presentation)

 PHAN, A D T., NETZEL, G., D’ARCY, B R & GIDLEY, M J 2013 Interactions between

food phenolic compounds and cellulose: mechanism and characterization The 46 th AIFST Annual Conference, Brisbane Convention & Exhibition Centre, Brisbane, 14-16th July 2013 (Poster presentation)

 PHAN, A D T., NETZEL, G., D’ARCY, B R & GIDLEY, M J 2013 Interactions between

food phenolic compounds and cellulose AIFST Food Science Summer School, University of

New South Wales, Sydney, 6-8th Feb 2013 (Poster presentation)

v

Publications included in this thesis

PHAN, A D T., NETZEL, G., WANG, D., FLANAGAN, B M., D’ARCY, B R & GIDLEY, M

J 2015 Binding of dietary polyphenols to cellulose: Structural and nutritional aspects Food Chemistry, 171, 388-396 (Incorporated as Chapter 3)

Performed binding experiments (100%) Statistical analysis of data (100%) Wrote paper (90%)

Edited paper (10%)

Analysed NMR data (100%) Edited paper (5%)

Edited paper (10%)

Wrote paper (10%) Edited paper (75%)

vi

PHAN, A D T., D'ARCY, B R & GIDLEY, M J 2016 Polyphenol-cellulose interactions: effects

of pH, temperature and salt International Journal of Food Science and Technology, 51, 203-211

(Incorporated as Chapter 4)

Performed experiments (100%) Analysed data (90%)

vii

Contributions by others to the thesis

Advice on project conception, experimental design, and the interpretation of research data as well as critical revision of thesis chapters were contributed by my advisory team: Professor Mike Gidley,

Dr Bruce D’Arcy, and Dr Gabi Netzel In addition, Dr Barbara Williams was involved during these processes for Chapter 6 Non-routine technical work involving various chemical analyses, SEM microscopy, and NMR study were contributed by other peoples as detailed in the following table:

Dr Barbara Williams Consulted for experimental design in Chapter 6

Consulted for statistical analysis in Chapter 6 Proof-reading and correction for Chapter 6

Dr Dongjie Wang Obtained SEM images: Figure 3.1 (Chapter 3) and Figure 5.1

(Chapter 5)

Dr Bernadine Flanagan Conducted NMR studies: Figure 3.2, Table 3.1 (Chapter 3), and

Figure 5.5 (Chapter 5) Barbara Gorham Assisted in preparation and lab work for the in vitro colonic

fermentation experiments (Chapter 6)

experiments (Chapter 6) Dagong Zhang

Dr David Appleton Determined protein content in apple cell wall extracts: incorporated

in Table 5.1 (Chapter 5)

Statement of parts of the thesis submitted to qualify for the award of another degree

None

viii

Acknowledgements

Up to now, this PhD project is the most challenging and exciting journey of my life I have achieved much during the 4 year-PhD study, not only in knowledge of food science and nutrition, but also in understanding myself in face of great challenges I would like to take this opportunity to thank all the people who have given me invaluable support, substantial encouragement, and for cheering me

up

First of all, I am grateful to my supervisors Professor Mike Gidley, Dr Bruce D’Arcy, and Dr Gabi Netzel for their excellent supervision, invaluable support and discussion throughout the research and thesis writing I have sincerest appreciation for you always having time to discuss and encourage me, not only during the study I had to deal with, but also with difficulties in my student life Your critical comments on my reports, your questions about research data, and your suggestions have resulted in significant contributions to my thesis Many thanks for your inspiration, and for developing my interests in plant cell wall and polyphenol research

I specially thank Dr Gabi Netzel and Dr Barbara Williams for your invaluable support during the in vitro colonic fermentation and polyphenol analysis This experiment could not have been carried

out without your help and useful advice

I would like to express my deep thanks to Dr Bernadine Flanagan and Dr Dongjie Wang for your help and explanation about NMR, SEM and bacterial cellulose production Many thanks to Dr Deirdre Mikkelsen and Dr Nima Gunness for your help and encouragements during my PhD study

I wish to thank Mr John Gorham for patiently teaching me confocal imagingtechniques, and Mrs Barbara Gorham for your great assistance in the fermentation experiment

I would like to extend my thanks to all CoE - UQ staff, past and present PhD students for the fortnightly meeting/discussion, your help and friendship Many thanks go to all the CNAFS/SAFS staff and students for your cooperation and contribution to a pleasant working environment

I sincerely thank all my Vietnamese friends, who are always giving me helps, and spending time with me and sharing with me happiness and sadness during my study life in Australia Special thanks to my colleagues and ex-teachers at Food Technology Department, Cantho University, Vietnam for your support during my PhD journey

Without the financial support provided by The Vietnamese Government Scholarship MOET), The University of Queensland, and The ARC Centre of Excellent in Plant Cell Walls, my PhD study would never have started

(VIED-ix

I am very grateful to my parents, my parents in law for their endless love and unlimited support for the whole of my life I would like to thank my older brother, my sister in law, and my nephews for their taking care our parents, creating a warm and happy atmosphere in our family during my time

of being far away from home

Last but not least, I would like to thank my beloved husband, Xuan Minh, for your endless love, understanding, encouragement, and support during my life I really appreciate all that you have done for me Thanks to my little daughter for forgiving me for leaving you behind for three years, and for giving me a strong motivation for completion of the PhD

x

Keywords

Polyphenols, bacterial cellulose composite, apple cell wall, interactions, in vitro digestion, in vitro

colonic fermentation, Cya-3-glc, (+/-)-catechin, ferulic acid, microbial metabolism

Australian and New Zealand Standard Research Classifications (ANZSRC)

ANZSRC code: 090801, Food Chemistry and Molecular Gastronomy (excl Wine), 75%

ANZSRC code: 090803, Food Nutritional Balance, 25%

Fields of Research (FoR) Classification

FoR code: 0908, Food Sciences, 100%

xi

Table of Contents

Abstract i

Declaration by author iii

Publications during candidature: iv

Publications included in this thesis v

Contributions by others to the thesis vii

Acknowledgements viii

Keywords x

List of Figures xvi

List of Tables xx

List of Abbreviations xxi

CHAPTER 1: INTRODUCTION 1

1.1 Project background 1

1.2 Hypotheses 3

1.3 Objectives 3

CHAPTER 2: LITERATURE REVIEW 5

2.1 Primary plant cell walls: molecular components, structure and functions 5

2.1.1 Plant cell wall structure and functions 5

2.1.2 Molecular components of primary plant cell walls 7

2.2 Bacterial cellulose – a plant cell wall model 11

2.2.1 Bacterial cellulose 11

2.2.2 Cellulose synthesis using Gluconacetobacter xylinus 12

2.2.3 The role of Gluconacetobacter xylinus in the synthesis of plant cell wall analogues 14

2.3 Phenolic compounds in plant foods 16

2.3.1 General structure and classification 17

2.3.2 Health benefits associated with polyphenol intake 22

2.4 Interactions between polyphenols and plant cell walls 23

2.4.1 The role of plant cell walls in controlling the bioaccessibility of phenolic compounds 24

2.4.2 Mechanism of binding of polyphenols to plant cell wall components 27

xii

2.4.3 Review of recent publications on the interactions between polyphenols and polysaccharides28

2.5 Bioaccessibility and bioavailability of polyphenols 30

2.5.1 Digestion and absorption of phenolic compounds 30

2.5.2 Limitation in the bioaccessibility of polyphenols bound to plant cell walls in the upper gastrointestinal digestion 32

2.5.3 The roles of colonic fermentation in release and metabolism of bioactive phenolic compounds from plant cell wall-polyphenol complexes 34

2.6 Conclusions and the gaps in knowledge 40

CHAPTER 3 42

BINDING OF DIETARY POLYPHENOLS TO CELLULOSE: STRUCTURAL AND NUTRITIONAL ASPECTS 42

Abstract 42

3.1 Introduction 43

3.2 Materials and methods 44

3.2.1 Standard phenolic compounds 44

3.2.2 Bacterial cellulose composites 45

3.2.3 Adsorption experiments 45

3.2.4 Solid State NMR analysis 46

3.2.5 Scanning electron microscopy 46

3.2.6 Confocal laser scanning microscopy (CLSM) 47

3.2.7 Data analysis 47

3.3 Results and discussion 47

3.3.1 Effects of alkali treatment on the purity and structural properties of BC 47

3.3.2 Adsorption kinetics study 50

3.3.3 Adsorption isotherms study 54

3.3.4 Confocal laser scanning microscopic observations 58

3.4 Conclusions 60

CHAPTER 4 61

xiii

POLYPHENOLCELLULOSE INTERACTIONS: EFFECTS OF pH, TEMPERATURE AND SALT

61

Abstract 61

4.1 Introduction 62

4.2 Materials and methods 63

4.2.1 Standards 63

4.2.2 Cellulose model 63

4.2.3 Adsorption experiments 64

4.2.4 Experimental design 64

4.2.5 Statistical analysis 66

4.3 Results and discussion 66

4.3.1 Fitting the response surface model 66

4.3.2 The effects of solution conditions on polyphenol adsorption 68

4.3.3 Interaction effects of multiple factors on polyphenol adsorption 72

4.4.4 Optimisation and model validation 74

4.4 Conclusions 75

CHAPTER 5 76

BINDING SELECTIVITY OF DIETARY POLYPHENOLS TO DIFFERENT PLANT CELL WALL COMPONENTS: QUANTIFICATION AND MECHANISM 76

Abstract 76

5.1 Introduction 77

5.2 Materials and methods 79

5.2.1 Materials 79

5.2.2 Production of cellulose-based composites 79

5.2.3 Preparation of apple cell walls 80

5.2.4 Adsorption experiments 81

5.2.5 Analytical 81

5.2.6 Scanning electron microscopy (SEM) 82

5.2.7 NMR study 82

xiv

5.2.8 Confocal Laser Scanning Microscopy (CLSM) 83

5.2.9 Statistics 83

5.3 Results and discussion 83

5.3.1 Characterisation of different plant cell wall models 84

5.3.2 Adsorption kinetics study 87

5.3.3 Adsorption isotherm study 91

5.3.4 Confocal microscopy images 97

5.3.5 Solid state 13C NMR Spectroscopy 98

5.3.6 Possible influences of the interactions on the bioaccessibility and nutritional/health aspects of polyphenols 100

5.4 Conclusions 101

CHAPTER 6 103

INDEPENDENT FERMENTATION/METABOLISM OF DIETARY POLYPHENOLS ASSOCIATED WITH DIFFERENT PLANT CELL WALL MODELS BY PIG COLONIC MICROBIOTA 103

Abstract 103

6.1 Introduction 104

6.2 Materials and methods 106

6.2.1 Materials 106

6.2.2 Production of bacterial cellulose 106

6.2.3 Preparation of apple cell walls 106

6.2.4 Adsorption experiments 106

6.2.5 In vitro digestion model 107

6.2.6 In vitro colonic fermentation using pig faecal inoculum 108

6.2.7 Apple cell wall characterisation 110

6.2.8 Quantification and identification of polyphenols release and possible metabolites 110

6.2.9 Statistical analysis 111

6.3 Results and discussion 111

6.3.1 Substrate characterisation 111

xv

6.3.2 Polyphenol release after two-stage in vitro digestion 114

6.3.3 Fermentation kinetics and end-product parameters 117

6.3.4 Polyphenol release and metabolism during in vitro fermentation 121

6.3.5 Microbial metabolism of polyphenols during in vitro colonic fermentation 124

6.3.6 The roles of dietary fibres in the microbial metabolism of bound polyphenols and implications for health benefits 132

6.4 Conclusions 134

CHAPTER 7: GENERAL CONCLUSIONS AND FUTURE DIRECTIONS 135

References Cited 138

Appendices 157

et al (2009)] 10 Figure 2.5: illustrates the egg-box model and the arrangement of major components in the cell wall [adapted from Tho et al (2005) and Cosgrove (1997)] 11 Figure 2.6: Scanning electron microscopy images of micro-architecture of bacterial cellulose (A) [scale bar = 100nm, (Mikkelsen et al., 2009)] and extracted plant-derived cellulose (B) [scale bar = 200 nm, (Eichhorn et al., 2010)] 12

Figure 2.7: Proposed model for cellulose synthesis from Ga xylinus (Mikkelsen and Gidley, 2011) 14

Figure 2.8: illustrates the visual changes in the formation of cellulose-xyloglucan composites as well as cellulose-pectin composites when adding xyloglucan and pectin, respectively, at different times of cultivation under an agitated condition [adapted from (Gu and Catchmark, 2012)] 15 Figure 2.9: Classification and chemical structures of major classes of polyphenols (Spencer et al., 2008) 17 Figure 2.10: General structure of six common anthocyanidins 18 Figure 2.11: illustrates predominant anthocyanin structure forms exist at different pH ranges (Lee et al., 2005) 19 Figure 2.12: Chemical structures of major catechins 20 Figure 2.13: Chemical structures of common phenolic acids 21 Figure 2.14: Comparison of the association of phenolic compounds with fibre in low fibre foods (left) and fibre-rich food sources (right) [adapted from Palafox‐Carlos et al (2011)] 26 Figure 2.15: Schematic illustration of the absorption pathway of polyphenols in humans (D’Archivio et al., 2010) 31 Figure 2.16: Illustration of the metabolism of various polyphenol classes such as (A) flavonols, (B) flavanol, flavone and flavanone, (C) anthocyanidin, and (D) phenolic acids by colonic microbiota (adapted from Aura (2008)) 36

CHAPTER 3

xvii

Figure 3.1: Scanning electron micrographs of the architecture of (A) native BC with entrapped bacterial cells (2-3 µm long rod-like shapes) inside the cellulose matrix, and purified cellulose fibres after NaOH treatment

at (B) 0.1 M and (C) 0.5 M 48 Figure 3.2: 13C CP/MAS NMR spectra of bacterial cellulose Samples purified by alkali solutions show no lipid or protein signal derived from bacteria 49 Figure 3.3: The adsorption of diverse phenolic compounds from 1 mM solutions to native and alkali-treated cellulose after various contact times 52 Figure 3.4: Comparison of the binding capacity of diverse polyphenols to different cellulose substrates at 1

mM polyphenols after 24 h contact (A) on a weight basis and (B) on a molar basis 53 Figure 3.5: Effects of initial polyphenol concentration on the binding of phenolic compounds to 0.1 M NaOH-treated cellulose (A) on a weight basis, and (B) on a molar basis Langmuir binding isotherms (C) are presented as a function of the concentration of free solute polyphenols at pH 5.0 (pH 3.4 for Cya-3-glc), after

2 h of incubation at 4ºC The symbols represent experimental data and the lines correspond to Langmuir adsorption isotherms obtained by applying a non-linear regression statistical analysis 57 Figure 3.6: CSLM images of (A) cellulose composite stained with Calcofluor White ST dye compared with auto-fluorescence from (B) Cya-3-glc and (C) ferulic acid bound to 0.1M NaOH-treated cellulose Intense auto-fluorescent spots in (C) are ascribed to trapped fragments of bacteria not removed by alkali treatment.59

CHAPTER 4

Figure 4.1: The effects of the main single factors on the adsorption of three diverse polyphenols onto cellulose 69 Figure 4.2: Contour-surface plots of the combined effects of binding factors on the adsorption of (a, b, c) Cya-3-glc, (d, e, f) ferulic acid, and (g, h, i) (+/-)-catechin onto cellulose 73

CHAPTER 5

Figure 5.1: Representative SEM micrographs show the differences in the micro-architectures amongst various plant cell wall models, including (A) BC, (B) BC-XG, (C) BC-AX, (D) BC-Pectin, and (E–F) ACW White arrows indicate apparent cross-links of xyloglucan between cellulose ribbons and deposition of arabinoxylan onto cellulose micro-fibril’s surface 86 Figure 5.2: Polyphenol adsorption, expressed as [A] µg adsorbed polyphenols per mg dry weight of plant cell wall models, and [B] µg adsorbed polyphenols per mg cellulose of different plant cell wall models, for different exposure times at 1 mM concentration of polyphenols (PCW: plant cell wall) 89 Figure 5.3: Comparison of the binding capacity and Langmuir binding isotherms curves of diverse polyphenols to different plant cell wall models at different concentrations of free solute polyphenols after 2 h contact and at 4ºC, calculated as (A) µg adsorbed polyphenol per mg dry weight plant cell wall models and

xviii

(B) µg adsorbed polyphenol per mg cellulose The symbols represent experimental data and the lines correspond to Langmuir adsorption isotherms which were obtained from apparent Langmuir binding parameters given in Table 5.2 93 Figure 5.4: Illustration of the difference in the apparent binding between auto-fluorescence of (A) Cya-3-glc (green) and (B) ferulic acid (blue) to (C, D) control bright-field images of ACW at the concentration of polyphenols of 15 mM Observations were conducted under a 63x objective lens in oil emulsion 97 Figure 5.5: 13C CP/MAS NMR spectra of (A, B, C) bacterial cellulose and (D, E, F) apple cell wall with different associated polyphenol compounds 99

CHAPTER 6

Figure 6.1: Illustration of (A) fermentation bottles ready for a pig faecal inoculum, and (B) the automatic gas reading system with bottles in position 109 Figure 6.2: Estimation of cell wall polysaccharides in apple cell wall extracts 112 Figure 6.3: The release of adsorbed polyphenols, including (A) ferulic acid, (B) (+/-)-catechin and (C) Cya-

3-glc, from different PCW models during in vitro stomach and small intestinal digestion 115

Figure 6.4: Representative cumulative gas production (DMCV) profiles of apple cell wall substrates with and without added polyphenols 118 Figure 6.5: Total short-chain fatty acids (SFCA) production profiles and branch-chain ratio values of (A, C) apple cell wall substrates and (B, D) bacterial cellulose substrates 120 Figure 6.6: The percentage of polyphenols present in fermentation fluids during 72 h fermentation as a result

of release and metabolism The values were calculated based on the released amounts compared with the initial binding amounts of polyphenols given in Table 6.1 123 Figure 6.7: Schematic representation of the proposed microbial metabolism pathways of Cya-3-glc, (+/-)- catechin and ferulic acid The intermediates and end-products in black were detected in the fermentation fluids in the present study The compounds in grey are expected to be involved in the bioconversion process but were not detected The dashed arrows represent multiple biotransformation steps for production of m- hydroxybenzoic acid 126 Figure 6.8: Time course profiles of ferulic acid and metabolites during metabolism by pig faecal bacteria 128 Figure 6.9: Time course profiles of Cya-3-gl and metabolites during metabolism by pig faecal bacteria 130 Figure 6.10: Time course profiles of (+/-)-catechin and metabolites during metabolism by pig faecal bacteria 132

xix

APPENDICES

Figure A.1: Representative UPLC chromatograms of all (+/-)-catechin substrates and the controls at 0 h 158 Figure A.2: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of (+/-)-catechin, compared to the controls at 6 h 159 Figure A.3: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of (+/-)-catechin, compared to the controls at 12 h 160 Figure A.4: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of (+/-)-catechin, compared to the controls at 72 h 161 Figure A.5: Representative UPLC chromatograms of all ferulic acid substrates and the controls at 0 h 162 Figure A.6: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of ferulic acid, compared to the controls at 6 h 163 Figure A.7: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of ferulic acid, compared to the controls at 12 h 164 Figure A.8: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of ferulic acid, compared to the controls at 72 h 165 Figure A.9: Representative UPLC chromatograms of all Cya-3-glc substrates and the controls at 0 h and 280

nm 166 Figure A.10: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of Cya-3-glc, compared to the controls at 6 h and 280 nm 167 Figure A.11: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of Cya-3-glc, compared to the controls at 12 h and 280 nm 168 Figure A.12: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of Cya-3-glc, compared to the controls at 72 h and 280 nm 169 Figure A.13: Representative UPLC chromatograms of all Cya-3-glc substrates and the controls at 0 h and

499 nm 170 Figure A.14: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of Cya-3-glc, compared to the controls at 6 h and 499 nm 171 Figure A.15: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of Cya-3-glc, compared to the controls at 12 h and 499 nm 172 Figure A.16: Representative UPLC chromatograms show the formation/degradation of possible metabolites from the microbial metabolism of Cya-3-glc, compared to the controls at 72 h and 499 nm 173 Figure A.17: UV spectra of possible metabolites scanned from 250 nm – 350 nm 174

CHAPTER 4

Table 4.1: Box-Behnken design matrix and results obtained for polyphenol adsorptions 65 Table 4.2: Summary of regression analysis and ANOVA of associated RSM factors fitted to second-order polynomial equation 68 Table 4.3: Optimised binding parameters and validation test results 74

CHAPTER 5

Table 5.1: Monosaccharide composition of different plant cell wall models expressed as mol% of total polysaccharides 84 Table 5.2: Apparent Langmuir isotherms parameters for the binding of diverse polyphenols to different plant cell wall models 96

CHAPTER 6

Table 6.1: Substrates subjected to the in vitro digestion and the in vitro colonic fermentation experiments.114

Table 6.2: End-product parameters after 72 h fermentation of different substrates 117 Table 6.3: Possible microbial metabolites identified in fermentation fluids during a 72 h fermentation 125

xxi

List of Abbreviations

UPLC-PDA Ultra Performance Liquid Chromatography- Photodiode Array Detector CP/MAS NMR Cross Polarisation and Magic-angle Spinning Nuclear Magnetic Resonance

FA-MET Possible microbial metabolites of ferulic acid

Cat-MET Possible microbial metabolites of (+/-)-catechin

Cyan-MET Possible microbial metabolites of cyanidin-3-glucoside

1

CHAPTER 1: INTRODUCTION

1.1 Project background

Current dietary advice emphasises food consumption that is rich in fruits and vegetables, because of

a beneficial correlation between fruit and vegetable intake and the decreased risk of chronic diseases Many epidemiological studies have elucidated the biological functions of food phenolic compounds, referred to as “lifespan essential” compounds, in health promotion and preventing certain diseases (e.g cardiovascular disease and cancers) (Saura-Calixto et al., 2007, Martin and Appel, 2010, Velderrain-Rodriguez et al., 2014) Since a large portion of nutritionally-beneficial polyphenols are primarily located within the vacuoles of plant cells, in cell walls as bound forms, and also in tannosome organelles, phenolic compounds must pass through the cell wall barrier, whenever they are released from the intact cells for consumption

The primary PCW of most plants is principally composed of cellulose micro-fibrils, which are interlinked with hemicelluloses, and embedded in a pectic matrix phase (Cosgrove, 2005) As a result, those cell wall components can play important roles in regulating, firstly the binding of cellular phytonutrients to cell wall materials, and later the release of bound polyphenols (i.e polyphenol bioaccessibility) in the digestive tract from the mouth to the colon Recently, the interactions between cellulose/pectin cell wall analogues and anthocyanins and phenolic acids derived from purple carrot purée (Padayachee et al., 2012a, Padayachee et al., 2012b), and the adsorption of extracted apple procyanidins to apple cell wall materials (Le Bourvellec et al., 2005,

Le Bourvellec and Renard, 2005, Renard et al., 2001) have been demonstrated Interestingly, the extent of binding can be high, with up to 80% of anthocyanins and 60% of procyanidins interacting with PCW analogues and apple cell wall materials, respectively (Renard et al., 2001, Padayachee et al., 2012a)

Several authors have suggested that the association between polyphenols and non-starch polysaccharides (dietary fibres or cell wall material) can take place via non-covalent interactions, including hydrogen bonding, hydrophobic or ionic interactions (Le Bourvellec and Renard, 2012, Pinelo et al., 2006, Bordenave et al., 2014) The binding can happen quickly and spontaneously within 30 s after exposure, followed by a slow increase in the extent of interaction after long time contact (Padayachee et al., 2012a, Padayachee et al., 2012b), suggesting that these interactions can

be relevant in real food systems during human chewing of intact tissues (e.g fruits and vegetables) and particularly when plant tissues are processed and stored before consumption Current findings

2

support the hypothesis that dietary fibres (PCW components) can control the bioaccessibility of phenolic compounds, if the binding is strong enough to resist conditions in the gastro-intestinal tract

From a nutritional perspective, food nutrients, including phenolic compounds, need to be bioaccessible for absorption and metabolism in the human body after food ingestion It is well-known from epidemiology studies that the ingestion of dietary fibre and polyphenols (e.g catechin, anthocyanin, resveratrol) is beneficial for human health (Anderson et al., 2009, Jacobs et al., 2009,Yang et al., 1998, Acquaviva et al., 2003, Atten et al., 2005,Manach et al., 2005) However, if polyphenols can interact with dietary fibre during food consumption, the polyphenol-cell wall complexes may be resistant to human upper gastro-intestinal digestion because there is no digestion

of PCWs in the stomach and small intestine This leads to a limitation in the bioaccessibility of phenolic compounds in the upper gastro-intestinal tract, and therefore dietary fibre will act as a phytonutrient carrier that transports these bioactive compounds to the colon In the large intestine, colonic microbiota ferment polysaccharide substrates, resulting in the release of bound polyphenols from dietary fibre for absorption or microbial fermentation (Martin and Appel, 2010, Manach et al., 2004) The consequences of this for human nutrition and health are currently unknown

Recently, a few studies dealing with the release of bound polyphenols from polyphenol–PCW components in various types of plant materials have been conducted (Fogliano et al., 2011, Mandalari et al., 2010, Padayachee et al., 2013) However, there is no detailed published

information concerning polyphenol release during colonic fermentation, from either in vitro or in vivo studies By applying in vitro gastric and small intestinal digestion, Padayachee et al

(2013)found a minimal release of anthocyanins (ca 5%) and limited release of phenolic acids (ca 30%) from PCW analogues from purple carrot juice for several exposure times, or from reconstituted purple carrot purée This finding suggests that most of the food phenolic compounds, which are bound to PCWs, are not bioaccessible in the upper digestive tract, and will be subsequently delivered to the colon for further degradation by the action of colonic bacteria

Although the interactions between polyphenols and PCWs have been initially established, literature regarding the extent and the biological functions of polyphenols released from dietary fibre in the large intestine is insufficient to identify potential benefits from such interactions In addition, the extent and the binding mechanisms of molecularly diverse polyphenols to various cell wall polymers have not been determined Furthermore, the release behaviour of different polyphenol classes under the conditions of the digestive system, and the important roles of colonic microbiota

in releasing bound polyphenols are still unknown The effects on the binding of phenolic

i) The binding capacity and binding affinity of different polyphenol classes to PCW polymers vary due to the differences in the chemical properties of polyphenols

ii) Environmental factors such as pH, temperature and the presence of NaCl potentiallyinfluence the binding capacity of polyphenols

iii) Polyphenols selectively bind to different plant cell wall components

iv) Most polyphenols associated with dietary fibre are retained in the polyphenol-cell wall complexes during gastric and small intestinal digestion, and are released and metabolized by the action of colonic microbiota

1.3 Objectives

The specific objectives of this project are as follow:

 Study 1: To investigate and quantitatively compare the binding of examples of diverse polyphenol classes to the main structural cell wall component, cellulose (as incorporated in Chapter 3)

 Study 2: To characterise the mechanisms involved in polyphenol-cellulose interactions (as incorporated in Chapter 3 and Chapter 4)

 Study 3: To examine the effects of environmental factors (pH, temperature, NaCl addition) on the binding capacity of different polyphenols to cellulose, and to optimize binding conditions (as incorporated in Chapter 4)

 Study 4: To examine the binding capacity of various PCW components (e.g pectin, arabinoxylan, xyloglucan) by employing cellulose/hemicellulose and cellulose/pectin composites (as incorporated in Chapter 5)

4

 Study 5: To investigate the bioaccessibility of polyphenols associated with cellulose during in vitro gastric and small intestinal digestion (as incorporated in Chapter 6)

 Study 6: To examine the roles of colonic bacteria in releasing residual bound polyphenols after

gastro-intestinal digestion in vitro using a pig faecal inoculum (as incorporated in Chapter 6)

5

CHAPTER 2: LITERATURE REVIEW

2.1 Primary plant cell walls: molecular components, structure and functions

The following review focuses on the structure and functions of primary PCWs Detailed information on the biochemistry of PCW components, including cellulose, hemicellulose and pectin, and the structural organization of such cell wall components during the biosynthesis of PCWs is described In addition, the potential application of bacterial cellulose in producing PCW analogues is reviewed and discussed

2.1.1 Plant cell wall structure and functions

In terms of both structure and function, all living things on the earth are composed of fundamental units called cells, and plants are no exception The plant cells are divided into two main separate parts: a cell wall and a protoplast (Figure 2.1).The PCW is an extracellular polymer matrix that covers an inner cell membrane and the protoplast inside From a nutrition viewpoint, the PCW is also considered as the main dietary fibre source in the human diet, because it contains non-starch polysaccharides that are “resistant to digestion and absorption in the human small intestine with complete or partial fermentation in the large intestine” (Harris and Smith, 2006)

Regarding the functions of PCWs during the growth, development and survival of plants, the PCW acts as a frontier zone, that provides strength and shape to the cell, as well as protecting the cell from outside forces (Brett and Waldron, 1996) While the PCW has skeletal support functions, the protoplast, consisting of a nucleus, cytoplasm, chloroplasts and a large vacuole (Figure 2.1), actively photosynthesises and performs other metabolic processes in the cell and the whole plants (Hall et al., 1981) The large vacuole contains an aqueous medium of various organic compounds,including acids, sugars, polyphenols, and other compounds Therefore, these phenolic compounds can bebioaccessible for adsorption in the human body only after the PCWs are disrupted

6

Figure 2.1: General plant cell structure

(Source: https://dmohankumar.files.wordpress.com/2010/07/plant-cell.jpg, retrieved December 13th 2015)

Since the PCW completely encloses the plant cell, any substances that penetrate or are released from cells must pass through the cell wall barrier Interactions between food phenolic compounds and PCWs were observed initially in several studies in either PCW materials (Renard et al., 2001,

Le Bourvellec et al., 2004, Wu et al., 2011, Bindon et al., 2010, Ruiz-Garcia et al., 2014) or mimetic PCW models (Padayachee et al., 2012a, Padayachee et al., 2012b) As a consequence of such interactions, the PCW is considered to be a key factor that is responsible for the bioaccessibility of plant-derived nutrients in the human digestive tract during the consumption of fruits and vegetables In order to determine the mechanism of the associations between the PCW and phenolic compounds, a review of the structure of the PCW and the biochemical properties of cell wall components is presented

From the viewpoint of morphological characteristics, PCWs are normally composed of a layered structure that can include a primary cell wall and a secondary wall (Brett and Waldron, 1996) The primary cell wall is initially deposited during the cell enlargement, whereas the secondary cell wall is developed in some tissues inside the primary cell wall, after the cell has stopped growing The secondary cell walls are normally thicker and more rigid than the primary walls (Lee et al., 2011) The zone between two adjacent cells, called the middle lamella, is typically enriched with pectin, enabling this layer to work as an adhesive gel that links neighbouring cells to form a tissue (Cosgrove, 2005, Brett and Waldron, 1996)

multi-Although all PCWs normally have the same basic structure and functions, they are different in the wall matrix composition and structural organization depending on the cell types and the plant species(Harris and Smith, 2006) This is especially true for fruits, wherein the secondary cell wall,

7

for instance, is typically not found (Cosgrove, 2005, Sarkar et al., 2009) Almost all fruits and vegetables eaten by humans predominantly contain primary cell walls An absence of the secondary cell wall in fruits results in many advantages for the reproduction of the plants (Sarkar et al., 2009)

As a consequence of this feature, the present project will mainly focus on the influences of the primary cell wall on the bioaccessibility of food phenolic compounds

2.1.2 Molecular components of primary plant cell walls

Generally, the basic components of primary PCWs are non-starch polysaccharides, including crystalline cellulose microfibrils embedded in a matrix phase of non-cellulosic polysaccharide mixtures and a minor amount of structural protein (Cosgrove, 2005, Cosgrove, 1997, Harris and Smith, 2006) Figure 2.2 shows a popular model used to describe the assembled structural network

of cell wall components in the primary PCW, which are composed of cellulose, hemicelluloses and pectic polysaccharides The proportions of cell wall components can vary from species to species For example, the primary cell walls of fruits and vegetables contain approximately 35% cellulose, 15% hemicelluloses, 40% pectin, 5% structural protein and 5% phenolics (on a dry weight basis) (Brett and Waldron, 1996), whereas that of maize coleoptiles consist of 55% hemicelluloses, 25% cellulose and only 10% pectin (Cosgrove, 1997) Many questions concerning the synthesis of cell wall components in growing plant cells, and how cross-linkages are generated between them to form a structural network, continue to be challenges for cell wall researchers

Figure 2.2: A simplified model of the primary cell wall (Smith, 2001)

8

2.1.2.1 Cellulose

Cellulose, the mainly structural unit of the PCW, is composed of an unbranched 1,4-linked glucan chain (Figure 2.3) (Sarkar et al., 2009, Brett and Waldron, 1996) Cellulose polymers are extruded into the extracellular space by enzymatic complexes (cellulose synthase) which span the plasma membrane (Cosgrove, 1997, Smith, 2001, Cosgrove, 2005, Somerville et al., 2004) After they are synthesised, the cellulose chains aggregate together by means of hydrogen-bonding, either within chains (intra-molecular hydrogen-bond) or with other cellulose chains (inter-molecular hydrogen-bond) and Van de Waals forces, in order to form a crystalline microfibrillar structure (Somerville et al., 2004) As constructed by this way, cellulose fibres are insoluble, tough and resistant to breakage; therefore, they give rigidity to cells Humans cannot digest cellulose due to a lack of digestive enzymes that are able to break down the beta-glucosidic linkages in the molecular structure Because cellulose resists both acid degradation and enzymatic digestion in the human gastro-intestinal system, it is not damaged by the stomach and small intestine, entering the large intestine unchanged For that reason, cellulose may be able to deliver food nutrients to the large intestine, if there is any interaction between cellulose and plant-derived food nutrients

Figure 2.3: The chemical structure of cellulose [adapted from Sarkar et al (2009)]

Cellulose is normally structured by two distinct regions, crystalline and amorphous domains, with the ratio between crystalline and amorphous phase varying depending on the cellulose source For example, the crystallinity index of cotton wool has been reported to be higher than that of wood pulp (0.81 – 0.95 vs 0.5 – 0.7, Quiroz-Castañeda and Folch-Mallol, 2013) In 1949, Hermans and Weidinger found the degree of crystallinity of native cellulose to be approximately 70%, with the remaining cellulose as amorphous by applying X-ray scattering techniques In terms of the physical properties of cellulose, there is a relationship between the proportion of the crystalline and amorphous phases and the mechanical properties of cellulosic materials, since crystalline regions give strong mechanical strength, whereas amorphous cellulose exhibits important viscoelastic properties (Mazeau and Heux, 2003) In addition, interactions between solid cellulosic materials with adsorptive substances can occur in the non-crystalline regions and/or on the surface of

9

cellulose crystallites (Ciolacu et al., 2011) Therefore, determination whether the roles of two distinct cellulose domains in controlling the interaction between adsorptive substances such as polyphenols and cellulose would be an interesting topic for future research

The polymorphism of cellulose has been studied intensively by many researchers There are six polymorphs of cellulose, including cellulose I, II, III1, III11, IV1, and IV11, of which cellulose I is the most abundant form found in nature (O’Sullivan, 1997) Cellulose II can be formed either by regeneration or mercerization of cellulose I, whereas the other polymorphs of cellulose (III1, III11) and cellulose (IV1, and IV11) can be obtained by treatments of cellulose I and II with liquid ammonia or by heat treatment of cellulose (III1, III11), respectively (O’Sullivan, 1997) Amongst cellulose polymorphs, cellulose I and II are the most important and extensively studied forms Using CP/MAS 13C NMR, Atalla and Vanderhart (1984) were the first to demonstrate that the crystalline phases of native cellulose I are composed of a mixture of two allomorphs Iα and Iβ Recently, different characteristics of the crystalline packing types, molecular conformation, and hydrogen bonding of cellulose Iα and Iβ have been reviewed by Nishiyama (2009)

2.1.2.2 Hemicelluloses

Hemicelluloses include a range of branched polysaccharides that consist of a β-(14)-linked backbone of sugars in an equatorial configuration (Scheller and Ulvskov, 2010) Unlike cellulose which contains only glucoses in the structural chain, hemicelluloses can be constructed by different types of sugar monomers such as glucose, mannose or xylose, and include relatively short side chains (Figure 2.4) (Scheller and Ulvskov, 2010, Sarkar et al., 2009) Xyloglucan and arabinoxylan/mixed linkage glucan are the predominant hemicelluloses in many primary PCWs of fruits/vegetables and grains, respectively Other less common forms of hemicelluloses in edible tissue, including glucuronoarabinoxylan, glucomannan, and galactomannan, are normally found in small quantities in primary PCWs, but in greater amounts in secondary PCWs

10

Figure 2.4: Monomer sugar precursors and chemical structures of major hemicelluloses [adapted

from Sarkar et al (2009)]

The branched or non-regular nature of hemicelluloses means that after they are synthesised in the Golgi apparatus, self-aggregation to form microfibrils does not occur as it does for cellulose However, the structural similarity to the β-(14)-linked backbone of cellulose means that hemicelluloses have the potential to tightly bind to the surface of the existent cellulose microfibrils

by numerous hydrogen bonds, and to tether adjacent microfibrils together (Kączkowski, 2003,

Cosgrove, 2005, Scheller and Ulvskov, 2010) In an in vitro assembly study, Whitney et al (1995)

observed the visual alterations of the bacterial cellulose microfibrillar network by adding xyloglucan to the fermentation media Due to this special feature, hemicelluloses are also called cross-linking glucans In addition, the presence of cross-linkages between such wall components is also suggested to inhibit the microfibrillar shifting, and to give more structural support to the walls (Kączkowski, 2003)

2.1.2.3 Pectin

Besides hemicelluloses, pectin - another cell wall component - is present in the matrix polysaccharide phase of many primary cell walls Pectins are located within the primary cell walls and the middle lamella, and function to control the wall porosity, to glue adjacent cells together, and

to maintain the cell hydration (Kączkowski, 2003) Pectin is a heterogeneous polysaccharide with a high content of the acidic sugar, galacturonic acid There are three identified classes of pectic polysaccharides: homogalacturonan, rhamnogalacturonans, and substituted galacturonans (Caffall and Mohnen, 2009) Among them, homogalacturonan and rhamnogalacturonan are particularly predominant in primary cell walls (Harris and Smith, 2006).Whilst cellulose is insoluble and forms

11

a fibrous structure, pectin forms a gel phase that can be at least partially extracted from the cell walls by hot water, hot dilute acid, alkaline solutions or calcium chelators (Cosgrove, 1997) Because of its gelling properties, pectin is extracted from plant-derived by-product materials and is widely applied in food processing as a thickening ingredient or as a stabiliser

Figure 2.5: illustrates the egg-box model and the arrangement of major components in the cell wall

[adapted from Tho et al (2005) and Cosgrove (1997)]

Pectic polysaccharides are initially secreted from the Golgi apparatus in highly methylesterified forms, after which pectic methylesterases partially remove the methyl groups from these polymers (Caffall and Mohnen, 2009) After some of the methyl groups have been removed, the pectic chains become negatively charged and have a high affinity to bind to calcium and other multivalent cations by ionic-covalent bonds Thus, in the presence of calcium, pectic polysaccharides form a hydrophilic gel phase in which the cellulose-hemicellulose network is embedded Although cross-linkages between pectin and hemicelluloses may also occur, it is still unclear and more work is needed in order to confirm such a relationship (Caffall and Mohnen, 2009) However, the ionic interactions between divalent calcium and two monovalent carboxyl groups in the linear backbones

of the pectic polysaccharides, normally called the ‘egg-box model’ (Tho et al., 2005), is a likely gelling mechanism of pectin networks in the PCWs (Figure 2.5)

2.2 Bacterial cellulose – a plant cell wall model

2.2.1 Bacterial cellulose

It is well known that cellulose is not only synthesised by plants, but also by bacteria and is called bacterial cellulose (BC) Recently, there has been an increasing interest in BC because it has been

12

widely applied in various fields, such as for medical, cosmetic, and food applications, by the paper industry, and for laboratory research As it is composed of extracellular β-1,4-glucan chains, BC is chemically identical to plant-derived cellulose (Ross et al., 1991, Klemm et al., 2001) On the other hand, BC also exhibits a higher purity, mechanical strength, crystallinity and water adsorption capacity, as well as the degree of polymerization when compared to cellulose produced from plants (Klemm et al., 2001, Sheykhnazari et al., 2011)

Since BC is of high chemical purity, it is free of hemicelluloses, pectin, lignin, or any other natural components associated with plant cellulose Consequently, it has been employed as a novel approach for studies that simulate the assembly of PCW architecture (Whitney et al., 1995, Whitney

et al., 1999, Szymańska-Chargot et al., 2011, Mikkelsen and Gidley, 2011, Chanliaud and Gidley, 1999), and for investigations of the mechanical properties of primary PCWs (Whitney et al., 1999,

de Souza et al., 2012, Chanliaud et al., 2002) Figure 2.6 indicates that BC can exhibit a structural resemblance to plant cellulose architecture, with a fine fibrous network of fibres More recently, BC has also been used for examination of the adsorption of plant-derived phytonutrients to PCW

(Padayachee et al., 2012a, Padayachee et al., 2012b) and in vitro fermentation of dietary fibres

(Mikkelsen et al., 2011)

Figure 2.6: Scanning electron microscopy images of micro-architecture of bacterial cellulose (A)

[scale bar = 100nm, (Mikkelsen et al., 2009)] and extracted plant-derived cellulose (B) [scale bar =

200 nm, (Eichhorn et al., 2010)]

2.2.2 Cellulose synthesis using Gluconacetobacter xylinus

Although BC can be produced by the biosynthesis of several bacterial species such as

Gluconacetobacter, Agrobacterium, Aerobacter, Achromobacter, Azotobacter, Rhizobium, Sarcina, and Salmonella(Jonas and Farah, 1998, Römling, 2002, Chawla et al., 2009), not all of these species

13

are able to secrete the extracellular cellulose as fibres (Klemm et al., 2001) As a result, special attention has been given to Gluconacetobacter xylinus (previously named as A xylinus or A xylinum) due to the special features of this microorganism Ga xylinus can synthesise extracellular

cellulose that has a high degree of crystallinity and an ultrafine purified fibrous network (Ross et al., 1991) In addition, this microorganism has been reported as a productive microbial producer of cellulose (Bielecki et al., 2005)

Ga xylinus - a rod shaped, gram negative, and obligate aerobic bacteria - was first described by A

J Brown in 1886, when he found a gelatinous supernatant film in a liquid medium (Jonas and Farah, 1998, Bielecki et al., 2005).In 1954, Hestrin and Schramm cultured this microorganism in Hestrin-Schramm liquid medium (HS medium), consisting of D-glucose, peptone, yeast extract, with citric acid and sodium hydrogen phosphate as a buffer system at pH 6.0, to investigate

cellulose synthesis Ga xylinus can produce BC from a variety of carbon sources (e.g glucose,

fructose, sucrose or mannitol) in the presence of oxygen, at ambient temperatures between 25 and 30⁰C, over a wide pH scale from 3 to 7, and either under static or agitated conditions (Sheykhnazari

et al., 2011) Many studies have investigated the optimization of cultured media (Jonas and Farah,

1998, Mikkelsen et al., 2009, Kurosumi et al., 2009, Sheykhnazari et al., 2011), the screening of high cellulose-producing strains (Nguyen et al., 2008), and the structural alteration of bacteria cellulose microfibrils (Haigler et al., 1980, Whitney et al., 1999, Suzuki et al., 2012, Ruka et al.,

2013) Therefore, Ga xylinus is well suited to produce BC materials for further examination of the

interactions between polyphenols and PCW polymers

Since BC exhibits chemically identical and similar fibre structures to plant-derived cellulose, it is used as a model system for studies on how extracellular cellulose is synthesised, and how sub-elementary fibrils of BC form the micro-architecture network The biosynthesis of BC from glucose precursors to fibrous microfibrils happens between the outer cell membrane and the cytoplasmic membrane of the bacterial cell through a multi-enzymatic process (Ross et al., 1991, Jonas and Farah, 1998, Klemm et al., 2001) Initially, uridine-diphosphoglucose (UDP-glucose), an immediate sugar nucleotide precursor of cellulose, is transformed from available D-glucose in the cultured media by several enzymes Subsequently, the polymerization of synthesized UDP-glucose takes place by the action of cellulose synthase to produce 1,4β-D-glucose chains According to Klemm

et al (2001), approximately 6 - 8 corresponding glucose chains are able to associate together to form a very thin sub-elementary fibril, which is extruded through pores on the bacteria cell membrane These sub-elementary fibrils are assembled together to form microfibrils, which is followed by the aggregation of such microfibrils to create a flat ribbon (Figure 2.7) The final step

2.2.3 The role of Gluconacetobacter xylinus in the synthesis of plant cell wall analogues

The biosynthesis pathway of BC in Ga xylinus is relatively similar to the mechanism of cellulose

production by plants, with the exception that plant-derived cellulose is deposited into an environment containing a complex mixture of other polymers (Mikkelsen and Gidley, 2011) So far, BC offers a novel model system that has been applied in many research areas (e.g cellulose synthesis, molecular assembly, architecture, and mechanical behaviour of the primary cell wall) This is useful for situations where it is not possible to use isolated cellulose from a plant source, since harsh chemical conditions (e.g 6 M alkali) are needed to (incompletely) separate cellulose

from other PCW components There have been a number of studies of the in vitro formation of cellulose-based composites with hemicelluloses and pectins, and in situ modification/isolation of

bacterial cellulose micro-architecture, significantly contributing valuable knowledge to the field of cellulose and PCW research

Recent evidence suggests that Ga xylinus can form a PCW analogue if the bacterium is cultured in

a growth medium containing pectic polysaccharides and/or hemicelluloses (Whitney et al., 1995, Whitney et al., 1999, Mikkelsen and Gidley, 2011, Szymańska-Chargot et al., 2011, Gu and Catchmark, 2012, Chanliaud and Gidley, 1999) Whitney and co-authors (1995) observed an association of xyloglucan with cellulose through the formation of cross-linkages between

15

xyloglucan and cellulose ribbons by scanning electron microscopy They found that supplemented xyloglucan can either bind to the cellulose surface or be woven into the cellulose ribbons Chanliaud and Gidley (1999) conducted an experiment on the synthesis of bacterial cellulose

incorporated with pectin It was suggested that Ga xylinus can create an interpenetrating

cellulose/pectin network, where cellulose microfibrils deposited by bacteria are embedded into the pectin ‘egg-box’ gel phases in the presence of an appropriate amount of calcium Moreover, no direct molecular interaction between pectin and cellulose was observed in this study By employing Raman and Fourier Transform Infrared (FT-IR) spectroscopy, Szymańska-Chargot et al (2011) suggested that adding both xyloglucan and pectin to the cultured medium caused an increase in the degree of crystallinity, and the structure of bacterial cellulose cultured in the presence of both xyloglucan and pectin was found to be similar to those observed in the primary cell wall

Although it is believed that pectin has no direct molecular interaction with bacterial cellulose, pectin might cause considerable effects on the assembly process of cellulose ribbons after they are extruded to the cultured medium (Figure 2.8) (Gu and Catchmark, 2012)

Figure 2.8: illustrates the visual changes in the formation of cellulose-xyloglucan composites as

well as cellulose-pectin composites when adding xyloglucan and pectin, respectively, at different times of cultivation under an agitated condition [adapted from (Gu and Catchmark, 2012)]

Note: (a, b, e) and (f, g, j) are control, xyloglucan, and pectin samples when they were added in the cultured medium at the beginning and after 5 days of cultivation, respectively

Together with research on the influences of hemicelluloses and pectin on the assembly of cellulose microfibrils, investigation of the factors affecting the process of glucose chain aggregation, ribbon assembly, and the crystallization of bacterial celluloses has also been of interest In order to interrupt cellulose ribbon assembly, some water-soluble agents including fluorescent brighteners

16

(e.g Calcofluor White ST) (Haigler et al., 1980, Suzuki et al., 2012) or polymers (e.g carboxymethylcellulose (CMC), Tween 80, hydroxypropyl methyl cellulose (HPMC)) (Whitney et al., 1995, Huang et al., 2010, Huang et al., 2011)can be added to the cultured medium at an appropriate level to trap isolated uncrystallized sub-elementary fibrils and prevent them from aggregating The structural changes of the bacterial cellulose induced by additives have been visually observed by electron microscopy It was proposed that fluorescent brighteners have the ability to create hydrogen bonds immediately with the glucose chains as they are extruded from the bacterial cells Consequently, fluorescent brightener chemicals altered the cellulose assembly and crystallization in the further steps of the cellulose matrix formation (Haigler et al., 1980, Suzuki et al., 2012) CMC can also prevent the normal cellulose ribbon assembly not only by disrupting the hydrogen bond formation between the β-glucan chains of CMC and cellulose, but also by steric hindrance or electrostatic repulsion due to the presence of charged substituent groups (Whitney et al., 1995) Aside from water-soluble substances, Ruka et al (2013) found that the water-insoluble polymer, poly-3-hydroxybutyrate, can also modify the morphology and properties of bacterial cellulosethrough an interaction between neat PHB matrix and bacterial cellulose microfibrils

The high chemical purity together with the capacity of BC to form cross-linkages with other cell wall polysaccharides provides an opportunity to investigate the binding of polyphenols to different PCW analogues This approach could allow evaluation of whether there are any differences in the binding behaviour of polyphenols to different chemical cell wall components

2.3 Phenolic compounds in plant foods

Phenolic compounds or polyphenols are abundant phytonutrients in the plant kingdom and are also major products of plant secondary metabolism Polyphenols are ubiquitous and widely distributed throughout most plant tissues Evidence for their roles in decreasing the risk of current lifestyle-related conditions such as certain types of cancer and cardiovascular diseases are emerging (Manach et al., 2004, Quiñones et al., 2013, Araújo et al., 2011, Wang and Stoner, 2008) Presently, more than 8,000 distinct phenolic compounds have been identified in various plant species (Martin and Appel, 2010) Although polyphenols are present in fruits and vegetables at relatively low concentrations from 2 – 7500 mg/kg fresh weight or mg/L (Manach et al., 2004) depending on phenolic compounds and different sources of plant-based food products, they can contribute to the pigmentation, the astringency, bitter taste, and the antioxidant activities of many plant food products (Vicente et al., 2009) The following section presents the general chemical structure and potential health benefits of dietary polyphenols

17

2.3.1 General structure and classification

In terms of molecular structure, phenolic compounds are a chemical family that can vary from a simple structure with a single aromatic benzene ring (e.g simple phenolic acids) to highly polymerized molecules (e.g tannins) Polyphenols are usually found in a conjugated form such as glycosides, esters or methyl esters, rather than in the free form (Vermerris and Nicholson, 2007) The term glycoside is used for polyphenols having sugar moieties linked to hydroxyl groups; in contrast, the term aglycone refers to polyphenols without attached sugars

Based on chemical structures, polyphenols can be classified into four distinct classes including phenolic acids, flavonoids, stilbenes and lignans (Figure 2.9) (Manach et al., 2004, Spencer et al., 2008) According to Martin and Appel (2010), simple phenolic derivatives and flavonoids are the most common and important phenolic compounds in plant foods, with phenolic acids representing one-third of the dietary intake of polyphenols, while flavonoids account for most of the remaining two-thirds of the total dietary intake of polyphenols Therefore, Phenolic acids and flavonoids (anthocyanins and flavanols) are important representatives of plant-derived water-soluble polyphenols, and are therefore relevant to theinvestigation of the binding of diverse polyphenols to different PCW components

Figure 2.9: Classification and chemical structures of major classes of polyphenols (Spencer et al.,

2008)

2.3.1.1 Flavonoids

Flavonoids, a large group of phenolic compounds, are composed of two aromatic rings associated together through a three carbon bridge (C6-C3-C6), namely a 3C-oxygenated heterocyclic ring (or

Anthocyanins

Anthocyanins, a prominent sub-class amongst the flavonoids, are responsible for the red, purple or blue hues in many fruits, vegetables, cereal grains and flowers, with more than 600 molecular structures having been identified to date (Konczak and Zhang, 2004) Anthocyanins are composed

of a backbone of anthocyanidins (or aglycone) consisting of a C6-C3-C6 basic structure (Figure 2.10) Native anthocyanins are typically glycosylated polyhydroxy or polymethoxy derivatives of the 2-phenylbenzopyrylium cation More than 90% of all anthocyanins identified in nature are based on the basic structures of the six common anthocyanidins shown in Figure 2.10 (Cavalcanti et al., 2011) The structure of anthocyanins can be varied by the presence of different glycosidic substitutions on the benzene rings, or by the acylation of sugar groups with acids (Lee et al., 2008) Whilst the differences in the presence of additional hydroxyl groups or methoxyl groups, together with the placement of such groups on the aromatic benzene ring, result in variations in the stability and the antioxidant capacity of anthocyanin molecules, the color of the pigments depends on pH and the presence of metal ions in the environment (Castañeda-Ovando et al., 2009, Cavalcanti et al.,

2011, Vermerris and Nicholson, 2007)

Figure 2.10: General structure of six common anthocyanidins

(Source http://www.chemistryland.com/CHM130FieldLab/Lab12/Lab12.html, retrieved on March 7th 2016)

Unlike other flavonoids, anthocyanins possess a positive charge in the central ring of the structure, and they are thus cations The charged nature of anthocyanins leads them to be susceptible to pH, with the reversible transformation of their molecular structures resulting in different colors (Figure