Advances in Enzyme Biotechnology Advances in Enzyme Biotechnology

Pratyoosh Shukla Brett I. Pletschke Editors

Pratyoosh Shukla • Brett I. Pletschke Editors

Advances in Enzyme

Biotechnology Editors Pratyoosh Shukla Brett I. Pletschke Department of Microbiology Department of Biochemistry, Maharshi Dayanand University Microbiology and Biotechnology Rohtak, Haryana , India Rhodes University Grahamstown, South Africa

ISBN 978-81-322-1093-1 ISBN 978-81-322-1094-8 (eBook) DOI 10.1007/978-81-322-1094-8 Springer New Delhi Heidelberg New York Dordrecht London

Library of Congress Control Number: 2013945175

© Springer India 2013 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifi cally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfi lms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection with reviews or scholarly analysis or material supplied specifi cally for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer. Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable to prosecution under the respective Copyright Law. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specifi c statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com) Foreword

This book is a collection of few recent discoveries in enzyme biotechnology by leading researchers in Enzyme Technology and further some selected contributions presented at the 51st Annual Conference of the Association of Microbiologists of India (AMI-2010) which was organized at the beautiful campus of Birla Institute of Technology in Mesra, Ranchi, India, during December 14–17, 2010. The book is edited by Dr. Pratyoosh Shukla, one of the executive members of the Organizing Committee and Prof. Brett I. Pletschke from Rhodes University in Grahamstown, South Africa, who was one of the leading invited speakers of the meeting. The meeting was attended mainly by participants from India but was also made international by a number of invited speakers from abroad. The meeting covered various ﬁ elds of microbiology, including agricultural and soil microbiology, algal biotechnology, biodiversity, biofuel and bioenergy, bioinformatics and metagenomics, environmental microbiology, enzyme technology, and food and medical microbiology. An important feature of the meeting was participation of industrial researchers which contributed to fruitful interactions between industrial and academic research indispensable for the development of new progressive biotechnologies. The majority of chapters in the book are dedi- cated to industrially important enzymes modifying plant polysaccharides and lignin. On one hand, the chapters review the current state of the art in the areas of production and application of glycoside hydrolyses, esterases, and lignin-degrading enzymes, while on the other hand, they describe modern trends in the development of enzyme technologies, including the computational enzyme design and enzyme mutations. The fact that most of the chapters originate in India demonstrates rapid emergence of research activity and enormous interest leading to the development of new enzyme technologies in the country. As we know India is a country which is heavily populated, and the sustainability of this country is very strongly dependent on environment friendly biotechnologies. Finally , I would also like to emphasize the general tone of the meeting which was optimistic and enthusiastic about emerging novel applications of enzymes and processes producing usable energy for the future. This book also represents a powerful exposure of important research of the present time to young researchers who ﬁ lled the

v vi Foreword meeting/lecture rooms. The hard work of the organizers of the meeting and the editors of this volume is greatly appreciated.

Slovak Academy of Sciences RNDr. Peter Biely, DrSc Institute of Chemistry, Center for Glycomics Bratislava , Slovakia September 11, 2012 Preface

There has been a rapid expansion of the knowledge base in the fi eld of enzyme biotechnology over the past few years. Much of this expansion has been driven by the bio-discovery of many new enzymes from a wide range of environments, some extreme in nature, followed by subsequent protein (enzyme) engineering. These enzymes have found a wide range of applications, ranging from bioremediation, biomonitoring, biosensor development, bioconversion to biofuels and other biotechnologically important value- added products, etc. The major goal of this book is to provide the reader with an updated view of the latest developments in the area of enzyme biotechnology. This book presents an exceptional combination of fascinating topics and the reader will be pleased to see that the latest technologies available for an improved understanding of enzymes are included in the book. For example, a thermostable enzyme with sugar metabolic activity is improved by targeted mutagenesis (Chap. 1 ). The reader will note that there is a signifi cant focus on the role of hydrolases (Chap. 2) and other depolymerising enzymes in this book, as these enzymes form a major component of the annual revenue generated by industrial enzymes. The various other topics ranging from the synthesis of prebiotic galacto-oligosaccharides (Chap. 3 ), biomass-degrading enzymes, in general, mannanases (Chap. 4 ), glycoside hydrolases and their synergistic interactions (Chap. 5), manganese peroxidases (Chap. 6) to the modern trends in experimental techniques in enzyme technology (Chap. 7 ) are also covered in the present book. Further, the most up-to-date studies related to an overview of the methodologies available for motif fi nding in biological sequences (Chap. 8 ), characteristic molecular features and functional aspects of chitin deacetylases (Chap. 9 ), the role of enzymes in plant–microbe interactions (Chap. 10 ) and the bioprospecting of industrial enzymes in various grain-processing industries (Chap. 11 ) have also been included. Moreover, the readers of the book will be delighted to see that the most up to date technologies available for a better understanding of enzymes are included in this book to enhance the learning skills in key facets of research in enzyme biotechnology.

vii viii Preface

We hope that the reader will ﬁ nd the information presented here valuable and stimulating. We acknowledge and are indebted to all those who have generously contributed to the completion of this book, and welcome com- ments from all those who use this book.

Haryana, India Pratyoosh Shukla Grahamstown, South Africa Brett I. Pletschke Contents

1 Improvement of Thermostable Enzyme with Sugar Metabolic Activity by Targeted Mutagenesis ...... 1 Yutaka Kawarabayasi 2 Glycoside Hydrolases for Extraction and Modifi cation of Polyphenolic Antioxidants ...... 9 Kazi Zubaida Gulshan Ara , Samiullah Khan , Tejas S. Kulkarni , Tania Pozzo , and Eva Nordberg Karlsson 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides ...... 23 Barbara Rodriguez-Colinas , Lucia Fernandez-Arrojo , Miguel de Abreu , Paulina Urrutia , Maria Fernandez-Lobato , Antonio O. Ballesteros , and Francisco J. Plou 4 Microbial Mannanases: Properties and Applications ...... 41 Hemant Soni and Naveen Kango 5 Enzyme Synergy for Enhanced Degradation of Lignocellulosic Waste ...... 57 J. Susan van Dyk and Brett I. Pletschke 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications ...... 67 Samta Saroj , Pragati Agarwal , Swati Dubey , and R. P. Singh 7 Advance Techniques in Enzyme Research ...... 89 Debamitra Chakravorty and Sanjukta Patra 8 Regulatory Motif Identifi cation in Biological Sequences: An Overview of Computational Methodologies ...... 111 Shripal Vijayvargiya and Pratyoosh Shukla 9 Chitin Deacetylase: Characteristic Molecular Features and Functional Aspects ...... 125 Nidhi Pareek , V. Vivekanand , and R. P. Singh 10 Role of Enzymes and Proteins in Plant-Microbe Interaction: A Study of M. oryzae Versus Rice ...... 137 Jahangir Imam , Mukund Variar , and Pratyoosh Shukla

ix x Contents

11 Industrial Enzyme Applications in Bioreﬁ neries for Starchy Materials ...... 147 Vipul Gohel , Gang Duan , and Vimal Maisuria

About the Editors...... 175 Improvement of Thermostable Enzyme with Sugar Metabolic 1 Activity by Targeted Mutagenesis

Yutaka Kawarabayasi

Abstract It was well known that improvement of enzymatic activity and stability is very diffi cult. For most enzymes, introduction of mutation into the amino acid residues located within the reaction center usually disappears their activity. Conversely, it should be useful for application of enzymes if enzymatic activity and stability are artifi cially enhanced. The enzyme isolated from thermophilic archaea generally possesses absolute stability. The nucleotide-sugar molecule is a powerful material for artifi cial construction of polymer structure of sugar. The ST0452 protein, an enzyme with sugar- 1-phosphate nucleotidylyltransferase activity from Sulfolobus tokodaii , was chosen as target for introduction of targeted mutagenesis into the reaction center. All mutant ST0452 enzymes exhibited the same thermostability as shown by the parental ST0452 enzyme. Among 11 mutant ST0452 proteins with substitution of the amino acid residues located at the reaction center by alanine and other amino acids, fi ve mutant ST0452 proteins showed kcat values larger than the original value, revealing that in these mutant ST0452 proteins, reactions progress faster than the original enzyme. Even though these mutant ST0452 proteins showed higher K m values than that of the original enzyme, these improved mutant ST0452 proteins were capable of exhibiting a higher activity than that of the wild-type ST0452 protein under the presence of high concentration of substrate. These results indicate that thermostable enzymes with higher activity were constructed from S. tokodaii ST0452 enzyme by substitution of amino acid residues at the reaction center. These improved enzymes are expected to be useful for application.

Keywords Thermophilic archaea • Thermostable protein • Sugar-nucleotide • Targeted mutagenesis • Improvement

Y. Kawarabayasi, Ph.D. Laboratory for Functional Genomics of Extremophiles, Faculty of Agriculture, National Institute of Advanced Industrial Science Kyushu University, Hakozaki 6-10-1, Higashi-ku, and Technology (AIST) , Tsukuba , Japan Fukuoka 812-8581, Japan e-mail: [email protected]

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 1 DOI 10.1007/978-81-322-1094-8_1, © Springer India 2013 2 Y. Kawarabayasi

(Niemetz et al. 1997 ; Kandler and König 1998 ). Introduction In eukarya, the activated molecule is essential for the synthesis of chitin, a major component of the Carbohydrate molecules are included within fungal cell wall (Cabib et al. 1982 ), and the glyco- many different types of compounds as a compo- sylphosphatidylinositol linker, a molecule anchor- nent: outermost structures on a microorganisms’ ing a variety of cell surface proteins to the plasma cellular surface which separates the cellular inner membrane (Udenfriend and Kodukula 1995 ). The and outer environment, source for the energy GlcNAc moiety is found in the polycarbohydrate metabolism which is important for obtaining structure N- or O-linked to the proteins as a energy (e.g., TCA cycle), polymer structure for posttranslational modifi cation (Guinez et al. 2005 ; storage of energy (e.g., glycogen), and heredity Slawson et al. 2006 ; Taniguchi et al. 2001 ; Spiro molecules as a part of DNA and RNA. Also poly- 2004 ). As glycosylation is the most important mer form of carbohydrate molecule modifi es the modifi cation for activating peptide drugs, UDP- function and stability of protein by binding with GlcNAc is thought to be important for future the protein molecule (Udenfriend and Kodukula development of effective drugs. 1995 ). Many different types of modifi ed sugar GlcNAc-1-P uridyltransferase activity was molecules are necessary to maintain these bio- identifi ed on the thermostable ST0452 protein logical processes. In most microorganisms, many from an acidothermophilic archaeon, Sulfolobus important modifi ed sugars are synthesized from tokodaii strain 7. The mutation was introduced simple sugar molecules that are incorporated into into this ST0452 enzyme for improvement of the cells from the surrounding environment. activity. Among modifi ed sugar molecules, nucleotide- In this chapter, at fi rst the feature of an acido- sugar molecules play one of the most important thermophilic crenarchaeon S. tokodaii strain 7, roles for construction of polymer structure of car- from which the thermostable GlcNAc-1-P uridyl- bohydrate. The nucleotide-sugar, an activated transferase was isolated, will be shown. Then, form of sugar molecule, is the sole substrate for features of the enzymatic activity of this ST0452 construction of polymer structure including a protein and summary on the improvement of the variety of sugar molecules. ST0452 protein by targeted mutagenesis will be Uridine diphosphate N -acetyl-d -glucosamine described. (UDP-GlcNAc) is synthesized by a four-step reaction from fructose-6-phosphate, which is catalyzed by glutamine:fructose-6-phosphate The Feature of Thermophilic amidotransferase (EC:2.6.1.16), phosphoglucos- Archaeon Sulfolobus tokodaii amine mutase (EC:5.4.2.10), glucosamine-1- strain 7 phosphate acetyltransferase (EC:2.3.1.157), and N -acetyl-d -glucosamine-1-phosphate uridyl- Sulfolobus tokodaii strain 7, an acidothermo- transferase (EC:2.7.7.10). UDP-GlcNAc, the philic archaeon, was used for isolation of the fi nal product of this biosynthetic pathway, is an enzyme with the sugar-1-phosphate nucleotidylyl- activated form of GlcNAc constructed by combi- transferase activity. This microorganism was nation of GlcNAc-1-phosphate with UTP. This isolated from Beppu hot springs located at molecule is required for constructing many kinds Kyushu in Japan (Suzuki et al. 2002 ). As this of polymer structures of carbohydrates. In bacte- microorganism was isolated from hot spring, ria, UDP-GlcNAc is required for synthesis of the microorganism is able to grow between lipopolysaccharides, peptidoglycan, enterobacte- 70 °C and 85 °C and between pH 2.5 and 5.0 rial common antigen, and teichoic acid (Frirdich with the optimal growth condition at 80 °C et al. 2004 ; vanHeijenoort 2001 ; Harrington and and 2.5–3.0, respectively. This microorganism Baddiley 1985 ). In archaea, the GlcNAc moiety was isolated from the geothermal environment; is a major component of the cell surface structure thus, this micro organism can grow under aerobic 1 Improvement of Thermostable Enzyme with Sugar Metabolic Activity by Targeted Mutagenesis 3 condition. The phylogenetic analysis showed that are predicted as that present only in one species this micro organism is included in the kingdom (Chen et al. 2005 ). Crenarchaeota of the domain Archaea. The The genomic data of S. tokodaii was used for microorganism grows chemoheterotrophically identifi cation of the useful enzymes. In the fol- under aerobic respiration condition. Autotrophic lowing sections, a brief identifi cation of the growth of this microorganism was not observed enzyme with sugar-1-phosphate nucleotidylyl- in minimal media supplied with elemental sulfur, transferase activity and improvement of the use- although several strains isolated as genus Sulfolobus ful activity are indicated. are known to be capable of growing autotrophically. Phylogenetic analysis by 16S rDNA sequences indicated that the sequence of this microorgan- Sugar Metabolic Enzyme from ism is most closely related to that of Sulfolobus an Acidothermophilic Archaeon, yangmingensis (Suzuki et al. 2002 ). S. tokodaii The entire genomic sequence of S. tokodaii was already determined (Kawarabayasi et al. Although many gaps are remaining in the meta- 2001). The size of the genome of S. tokodaii is bolic pathway constructed from the genomic data 2,694,756 bp long, and the G+C content is of S. tokodaii, the four genes for TDP-rhamnose approximately 32.8 %. Within this genomic biosynthesis pathway from glucose-1-phosphate sequence, over 2,800 open reading frames and TTP were predicted from the genomic data of (ORFs) were predicted as potential protein- S. tokodaii . Also, genes similar to the fi rst enzyme coding regions, and 32.2 % of these are pre- in this biosynthetic pathway, glucose-1-phosphate dicted of their functions (annotatable), 32.6 % thymidylyltransferase, were detected on the of these are related to the conserved but genome. Among these genes located at other unknown ORFs, and 5.1 % of these contain position than the fi rst enzyme within the TDP- some motif sequences. Among 46 tRNA genes rhamnose biosynthesis pathway, the ST0452 gene predicted within the genomic sequence, 24 was chosen for analysis of its activity and function, tRNA genes are shown as the interrupted tRNA because of the presence of the long C-terminal genes which contain the intron within their domain which was not present in the other similar genes. The CCA sequence is required for bind- genes. Thus, the gene encoding the ST0452 protein ing with amino acid, and this CCA sequence is was cloned and expressed in E. coli . As shown in not included in most tRNA genes predicted in Fig. 1.1 , the forward and reverse direction of Glc- this genomic sequence of S. tokodaii . Also the 1-phosphate thymidylyl-transferase activity was tRNA nucleotidylytransferase, which is used for detected on the purifi ed ST0452 protein. The pro- addition of CCA sequence posttranscriptionally, tein exhibited utilization of multiple metal ions, was predicted on the genomic sequence of S. absolute thermostability with retaining 50 % of tokodaii. These features are closely similar to maximum activity after 180 min treatment at 80 °C, that of eukaryote. and relative high activity from pH 5.0 to 8.5 with Already entire genomic sequences of two maximum activity at pH 7.5 (Zhang et al. 2005 ). similar species, Sulfolobus solfataricus and By analysis of substrate specifi city, it was indi- Sulfolobus acidocaldarius , were determined (She cated that multiple sugar-1-phosphate and NTP et al. 2001 ; Chen et al. 2005 ). The genome size plus dNTP substrates were acceptable for the and the number of the predicted protein-coding sugar-1-phosphate nucleotidylyltransferase activ- regions of S. solfataricus and S. acidocaldarius ity of the ST0452 protein as shown in Table 1.1 . are 2,992,245 bp and 2,977 and 2,225,959 bp and Among these, GlcNAc-1-phosphate uridyltrans- 2,292, respectively. Among these potential ferase activity was one of the most important protein-coding regions, approximately 1,600 sugar-1-phosphate nucleotidylyltransferase activ- genes are conserved within three Sulfolobus ities, because the GlcNAc moiety is usually found species. Approximately from 400 to 900 genes at the most fundamental position of polysaccharide. 4 Y. Kawarabayasi

acTDP-glucose Table 1.2 Kinetic properties for the N -acetyl-d - TTP glucosamine-1-phosphate uridyltransferase activity of the ST0452 protein Substrate Km a kcat a ST0452 protein UTP 1.00 1.00 GlcNAc-1-P 4.65 0.72 E. coli enzymeb UTP 9.83 6.80 b d TDP-glucose GlcNAc-1-P 10.40 5.95 TTP TDP-Glucose a The relative values are expressed as a proportion of that detected on UTP and the ST0452 protein TTP b The kinetic parameters for E. coli enzyme is according to the results described by Gehring et al. (1996 )

protein were obtained. Compared with those of the similar enzyme in E. coli, both Km and kcat Fig. 1.1 HPLC elution profi le of the products by glucose- values for this activity of the ST0452 protein are 1- phosphate nucleotidylyltransferase activity of the lower than those of E. coli as shown in Table 1.2 . ST0452 protein. The HPLC elution profi les for the prod- It means that the ST0452 protein is capable of ucts before ( a and c) and after (b and d) incubation for 20 min at 80 °C with the ST0452 protein. The glucose-1- binding with low concentration of substrates, but phosphate was added into the reaction solution as sub- the turnover rate of reaction is slower than that of strate ( a and b), and TDP-glucose and PPi were added as the similar E. coli enzyme. The low turnover rate substrates (c and d ) for proceeding the reaction is not convenient for production of nucleotide- sugar molecules in application. Table 1.1 Substrate specifi city of the sugar-1-phosphate Conversely, thermostability is benefi cial for nucleotidylyltransferase activity of the ST0452 protein industrial application; therefore, it was attempted Relative to increase the sugar-1-phosphate nucleotidylyl- Substrate A Substrate B activity dTTP d -Glucose- 1-phosphate 100 transferase activity, especially GlcNAc-1-phosphate dATP 35 uridyltransferase activity, of the ST0452 protein. dCTP 7 dGTP 1 UTP 130 Improvement of the Archaeal ATP/CTP/GTP NDa Enzymatic Activity by Targeted dTTP N -Acetyl-d -glucosamine -1- 320 phosphate Mutagenesis d -Glucosamine- 1-phosphate NDa d -Galactose- 1-phosphate NDa For increase of the sugar-1-phosphate nucleotidylyl- d -Mannose- 1-phosphate NDa transferase activity of the ST0452 protein, the UTP N -Acetyl-d -glucosamine -1- 540 substitution of amino acid residues without dimin- phosphate ishing the thermostability was planned fundamen- d-Glucosamine- 1-phosphate ND a d-Galactose- 1-phosphate ND a tally according to the expectation that the d-Mannose- 1-phoqsphate ND a substitution of the amino acid residues located a ND: not detected within the reaction center should not affect the thermostability of the protein, because the reac- Therefore, this activity was expected to catalyze tion center is allocating at the relatively inside of the last reaction in the UDP- GlcNAc biosynthesis the protein like a pocket. Thus, it was expected pathway from fructose-6-phosphate. that the substitution of the amino acid residues The kinetic parameters for the GlcNAc-1- within the reaction center should not affect the phosphate uridyltransferase activity of the ST0452 overall structure and thermostability of the protein. 1 Improvement of Thermostable Enzyme with Sugar Metabolic Activity by Targeted Mutagenesis 5

Table 1.3 . It indicated that fi ve mutant ST0452 proteins exhibited higher kcat values than parental wild-type ST0452 protein. However, the K m values for the GlcNAc-1-phosphate uridyltransferase activity of these mutant ST0452 proteins also changed to more larger than that of the wild- Thr80 type ST0452 protein as shown in Table 1.3 . These Tyr97 results revealed that these mutant ST0452 pro- Glu146 teins required higher concentration of substrate Asp208 Gly9 Asp99 for effi cient binding, but reaction proceeds faster than wild-type ST0452 protein. Thus, their Lys147 Arg13 activities under presence of the high concen- Lys23 tration of substrate were analyzed. The results Arg21 indicated that when high concentration of Leu14 GlcNAc-1-phosphate and UTP were supplied into the reaction mixture, fi ve mutant ST0452 proteins exhibited the higher relative activities Fig. 1.2 Proposed 3D structure of the sugar-1-phosphate nucleotidylyltransferase reaction center of the ST0452 than that of the parental wild-type ST0452 pro- protein. The amino acid residues participating in binding tein (Fig. 1.5 ). The result revealed that the substi- with nucleoside triphosphate substrates, sugar-1- tution of the amino acid residues within reaction phosphate substrates, N -acetyl portion of GlcNAc-1- center by alanine or other amino acid is effective phosphate, and metal ions are indicated by red, blue, green, and magenta , respectively. The region from Leu14 and useful for improvement of the enzyme with to Arg21 indicating high conservation with the corre- the GlcNAc-1-phosphate uridyltransferase activ- sponding sequences of E. coli RmlA is indicated by cyan . ity from an acidothermophilic archaeon S. toko- The metal ion is indicated by brown daii. As similar results were obtained from other enzymes isolated from S. tokodaii (data not Therefore, the substitution of the amino acid shown), it can be said that this exclusive feature residues located around the reaction center was is common for proteins in this microorganism. attempted. As shown in Fig. 1.2, the amino acid The amino acid residues effective for improve- residues, shown by color character, surrounding ment of kcat values of the GlcNAc-1-phosphate the reaction center of the ST0452 were changed to uridyltransferase activity of the ST0452 protein alanine or other amino acid. Total 11 mutant by target mutagenesis were shown by enclosure ST0452 proteins were constructed as shown in of red lines in Fig. 1.6 . These effective residues Fig. 1.3 . Analysis of the thermostability of these are located at relatively apart from the reaction mutant ST0452 proteins, SDS-polyacrylamide center. Thus, it is thought from improvement of gel electrophoresis of these proteins after treat- the activity of the ST0452 protein, that the amino ment at 80 °C for 30 min, indicated that thermo- acid residues located relatively surrounding the stability of all mutant ST0452 proteins was not area of the reaction center should play an impor- affected by substitution of the amino acid residues tant role for the turnover rate of the activity. within reaction center (Fig. 1.4 ). As all mutant ST0452 proteins exhibited same thermostability as parental wild-type ST0452 protein as expected, Discussion and Perspective all mutant ST0452 proteins were used for detailed analyses of their sugar-1-phosphate nucleotidylyl- Some number of improvements of enzymatic transferase activity (Zhang et al. 2007 ). activity was already reported (Sun et al. 2011 ; Relative values of kinetic parameters for Qi et al. 2012). However, targets of these experiments GlcNAc-1-phosphate uridyltransferase activity are enzymes from mesophilic microorganism. of the mutant ST0452 proteins are shown in Therefore, the result shown for the ST0452 6 Y. Kawarabayasi

EcRmlA 12 GSGTRLHPATLAVSK 26 83 QPS-PDGL 89 ST0452 protein 9 GSGERLEPITHTRPK 23 73 QKDDIKGT 80 EcGlmU 14 GKGTRMY--S-DLPK 25 76 QAE-QLGT 82 L LEE- KPLEPKSN EcRmlA 107 LVL-GDN 112 160 170 - ST0452 protein 94 LIIYGDL 100 144 IIE KPEIPPSN 154 EcGlmU 99 LMLYGDV 106 152 IVEHKDATDEQR 163 F

EcRmlA 220 RGYAWLDTG 228 ST0452 protein 202 EGY-WMDIG 210 EcGlmU 222 EVEG-VNNR 229

Fig. 1.3 Sequence alignment of ﬁ ve highly conserved within boxes indicate the residues conserved within domains among the ST0452 protein and E. coli glucose- three proteins. The amino acid residues chosen for the 1-phosphate thymidylyltransferase and N -acetyl- d - construction of mutant proteins are indicated by sym- glucosamine-1-phosphate uridyltransferase. EcRmlA bol, and amino acid residues introduced into the mutant and EcGlmU indicate the glucose-1-phosphate thymi- ST0452 proteins other than alanine are shown below dylyltransferase from E. coli (GenBank accession num- symbols. The numerals indicate the coordinates of the ber P37744) and N -acetyl- d -glucosamine-1-phosphate two ends of each domain from the N-terminus of each uridyltransferase from E. coli (NC_000913). The letters protein

Table 1.3 Kinetic properties for the N -acetyl-d - glucosamine-1-phosphate uridyltransferase activity of the wild-type and mutant ST0452 proteins K m for Proteins K m for UTP GlcNAc-1- P kcat ST0452 1.00 1.00 1.00 G9A 5.88 271.25 4.07 R13A 1.00 193.75 0.80 K23A 1.53 833.75 0.05 T80A 1.59 43.75 1.51 Fig. 1.4 SDS-PAGE analysis of the mutant ST0452 pro- T80L 8.24 443.75 0.30 teins produced in E. coli. The wild-type and mutant Y97A 8.24 8.25 2.11 ST0452 proteins expressed in E. coli were subjected to the 12 % of polyacrylamide gel containing 0.1 % of SDS after Y97F 8.82 4.63 4.95 treatment at 80 °C for 20 min. Lane M: lane for molecular D99A 0.94 1107.5 0.012 marker E146A 2.94 386.25 0.73 K147A 1.65 312.50 3.23 protein was thought to be the ﬁ rst result of D208A 3.88 58.75 0.74 improvement of thermostable enzyme. The relative values are showed as a proportion of that The results described in this chapter propose detected on the wild-type protein the opportunity that activity of the thermostable protein is able to be improved by introduction of many target proteins from thermophilic archaeal the targeted mutagenesis at the amino acid resi- species. If this feature will be detected on many dues allocating around the reaction center. If it is target proteins, introduction of this type of muta- general for the thermostable proteins from archaea, tion will become a powerful tool for improving the it is convenient for application in industry to pro- thermostable enzymes isolated from thermophilic vide an enzymatic activity with high turnover rate archaea. This will be helpful for making a consti- by introduction of targeted mutagenesis. Therefore, tutively developing society in this planet for the it is planned to attempt to check this possibility for next generation. 1 Improvement of Thermostable Enzyme with Sugar Metabolic Activity by Targeted Mutagenesis 7

250

200

150

100 Relative activity (%)

ST0452 G9A R13A T80A Y97A Y97F E146A K147A D208A

Fig. 1.5 N-Acetyl- d -glucosamine-1-phosphate uridyl- μMN -acetyl- d -glucosamine-1-phosphate (hatched transferase activity of the mutant ST0452 proteins bars), and 100 μM UTP plus 10 mMN -acetyl- d - under three different conditions. N -acetyl-d -glucosamine- glucosamine-1-phosphate (closed bars). The relative 1-phosphate uridyltransferase activities of each mutant activity is expressed as a percentage of the activity protein indicated were measured in the reaction solu- detected on the wild-type ST0452 protein under the tion with 5 μM UTP plus 50 μM N -acetyl-d -glucosamine- condition containing 100 μM UTP plus 10 1-phosphate (open bars), 100 μM UTP plus 50 mMN -acetyl- d -glucosamine-1-phosphate

Acknowledgements I appreciate ﬁ ve postdoctoral fellows, Tsujimura M., Zhang Z., Akutsu J., Sasaki M., and Md. M. Hossain, working in my laboratory for research in this area. This work was ﬁ nancially supported by special grants for the Protein 3,000 project, a basic knowledge project by New Energy and Industrial Technology Development Organization and a Grant-in-Aid for Research of the Ministry of Education, Culture, Sports, Science and Technology.

Thr80 Tyr97 Glu146 References Asp208 Gly9 Asp99 Cabib E, Roberts R, Bowers B (1982) Synthesis of the Lys147 Arg13 yeast cell wall and its regulation. Annu Rev Biochem 51:763–793 Lys23 Arg21 Chen L, Brügger K, Skovgaard M, Redder P, She Q, Leu14 Torarinsson E, Greve B, Awayez M, Zbat A, Klenk HP, Garrett RA (2005) The Genome of Sulfolobus acidocaldarius , a model organism of the Crenarchaeota. J Bacteriol 187:4992–4999 Fig. 1.6 Proposed 3D structure of the sugar-1-phosphate Frirdich E, Vinogradov E, Whitﬁ eld C (2004) Biosynthesis nucleotidylyltransferase reaction center of the ST0452 of a novel 3-deoxy- d- manno-oct-2-ulosonic acid- protein with marking of the amino acid residues working containing outer core oligosaccharide in the lipopoly- as improving its activity by substitution. The colored saccharide of Klebsiella pneumoniae . J Biol Chem amino acid residues are shown as legend of Fig. 1.2 . The 279:27928–27940 amino acid residues important for improving the sugar-1- Gehring AM, Lees WJ, Mindiola DJ, Walsh CT, Brown phosphate nucleotidylyltransferase activity of the ST0452 ED (1996) Acetyltransfer precedes uridylyltransfer are enclosed by red lines in the formation of UDP-N -acetylglucosamine in 8 Y. Kawarabayasi

separable active sites of the bifunctional GlmU protein Slawson C, Housley MP, Hart GW (2006) O -GlcNAc of Escherichia coli . Biochemistry 35:579–585 cycling: how a single sugar post-translational modifi - Guinez C, Morelle W, Michalski JC, Lefebvre T (2005) cation is changing the way we think about signaling O-GlcNAc glycosylation: a signal for the nuclear networks. J Cell Biochem 97:71–83 transport of cytosolic proteins?.Int J Biochem Cell Spiro RG (2004) Role of N -linked polymannose oligosac- Biol 37: 765–774 charides in targeting glycoproteins for endoplasmic Harrington CR, Baddiley J (1985) Biosynthesis of wall reticulum-associated degradation. Cell Mol Life Sci teichoic acids in Staphylococcus aureus H, Micrococcus 61:1025–1041 varians and Bacillus subtilis W23. Involvement of Sun Y, Yang H, Wang W (2011) Improvement of the ther- lipid intermediates containing the disaccharide mostability and enzymatic activity of cholesterol oxi- N-acetylmannosaminyl N-acetylglucosamine. Eur dase by site-directed mutagenesis. Biotechnol Lett J Biochem 153:639–645 33:2049–2055 Kandler O, König H (1998) Cell wall polymers in Archaea Suzuki T, Iwasaki T, Uzawa T, Hara K, Nemoto N, (Archaebacteria). Cell Mol Life Sci 54:305–308 Kon T, Ueki T, Yamagishi A, Oshima T (2002) Kawarabayasi Y, Hino Y, Horikawa H, Jin-no K, Takahashi Sulfolobus tokodaii sp. nov. (f. Sulfolobus sp. M, Sekine M, Baba S, Ankai A, Kosugi H, Hosoyama A, strain 7), a new member of the genus Sulfolobus iso- Fukui S, Nagai Y, Nishijima K, Otsuka R, Nakazawa H, lated from Beppu Hot Springs, Japan. Extremophiles Takamiya M, Kato Y, Yoshizawa T, Tanaka T, Kudoh 6:39–44 Y, Yamazaki J, Kushida N, Oguchi A, Aoki K, Taniguchi N, Ekuni A, Ko JH, Miyoshi E, Ikeda Y, Ihara Masuda S, Yanagii M, Nishimura M, Yamagishi A, Y, Nishikawa A, Honke K, Takahashi M (2001) Oshima T, Kikuchi H (2001) Complete genome A glycomic approach to the identifi cation and char- sequence of an aerobic thermoacidophilic crenar- acterization of glycoprotein function in cells trans- chaeon, Sulfolobus tokodaii strain7. DNA Res fected with glycosyltransferase genes. Proteomics 8:123–140 1:239–247 Niemetz R, Kärcher U, Kandler O, Tindall BJ, König H Udenfriend S, Kodukula K (1995) How glycosyl- (1997) The cell wall polymer of the extremely phosphatidylinositol-anchored membrane proteins are halophilic archaeon Natronococcus occultus . Eur made. Annu Rev Biochem 64:563–591 J Biochem 249:905–911 vanHeijenoort J (2001) Formation of the glycan chains in Qi X, Guo Q, Wei Y, Xu H, Huang R (2012) Enhancement the synthesis of bacterial peptidoglycan. Glycobiology of pH stability and activity of glycerol dehydratase 11:25R–36R from Klebsiella pneumoniae by rational design. Zhang Z, Tsujimura M, Akutsu J, Sasaki M, Tajima H, Biotechnol Lett 34:339–346 Kawarabayasi Y (2005) Identifi cation of an extremely She Q, Singh RK, Confalonieri F, Zivanovic Y, Allard G, thermostable enzyme with dual sugar-1-phosphate Awayez MJ, Chan-Weiher CCY, Curtis BA, Moors nucleotidyltransferase activities from an acidothermo- AD, Erauso G, Fletcher C, Gordon PMK, Heikamp-de philic archaeon, Sulfolobus tokodaii strain 7. J Biol Jong I, Jeffries AC, Kozera CJ, Peng X, Thi-Ngoc HP, Chem 280:9698–9705 Redder P, Schenk ME, Theriault C, Theriault C, Zhang Z, Akutsu J, Tsujimura M, Kawarabayasi Y Tolstrup N, Charlebois RL, Doolittle WF, Douguest (2007) Increasing in archaeal GlcNAc-1-P uridyl- M, Gaasterland T, Garrett RA, Ragan MA, Sensen transferase activity by targeted mutagenesis while CW, van der Oost J (2001) The complete genome of retaining its extreme thermostability. J Biochem the crenarchaeon Sulfolobus solfataricus P2. Proc Natl 141:553–562 Acad Scie USA 98:7835–7840 Glycoside Hydrolases for Extraction and Modifi cation 2 of Polyphenolic Antioxidants

Kazi Zubaida Gulshan Ara , Samiullah Khan , Tejas S. Kulkarni , Tania Pozzo , and Eva Nordberg Karlsson

Abstract Antioxidants are important molecules that are widely used by humans, both as dietary supplements and as additives to different types of products. In this chapter, we review how fl avonoids, a class of polyphenolic antioxidants that are often found in glycosylated forms in many natural resources, can be extracted and modifi ed using glycoside hydrolases (GHs). Glycosylation is a fundamental enzymatic process in nature, affecting function of many types of molecules (glycans, proteins, lipids as well as other organic molecules such as the fl avonoids). Possibilities to control glycosylation thus mean possibilities to control or modify the function of the molecule. For the fl avonoids, glycosylation affect both the antioxidative power and solubility. In this chapter we overview results on in vitro deglycosylation and glycosylation of fl avonoids by selected GHs. For optimal enzymatic performance, desired features include a correct specifi city for the target, combined with high stability. Poor specifi city towards a specifi c substituent is thus a major drawback for enzymes in particular applications. Efforts to develop the enzymes as conversion tools are reviewed.

Keywords Glucosidase • Cellulase • Amylase • Glycosynthase • GH • Flavonoid • Quercetin

Introduction

The increased concern about scarcity of fossil K. Z. G. Ara • S. Khan • T. S. Kulkarni • T. Pozzo resources has lately put the use of renewable • E. Nordberg Karlsson (*) resources by biotechnological methods in focus, Biotechnology, Department of Chemistry , as these are predicted to have an increased impor- Lund University , P.O. Box 124 , SE-22100 Lund , Sweden tance in production of food, additives and chemi- e-mail: [email protected] cals. Antioxidants can be foreseen to play a role as

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 9 DOI 10.1007/978-81-322-1094-8_2, © Springer India 2013 10 K.Z.G. Ara et al. bio-based ingredients in food (as well as other) diffi cult in many formulations of food, pharma- products, both as preservatives, replacing agents ceutical and nutraceutical products (Ishihara and with negative health aspects, and as nutraceuticals. Nakajima 2003 ). Improvement of the hydrophilic Flavonoids are polyphenolic compounds and a nature, biological properties and stability of fl a- class of secondary metabolites that are widely dis- vonoids can be achieved by enzymatic structural tributed in nature. The benefi cial properties of fl a- modifi cation (Haddad et al. 2005). In nature, vonoids are mainly referred to their ability to enzyme function has however evolved according counteract oxidative stress caused by oxygen spe- to the conditions in the living cells and may not cies and metal ions (Lin and Weng 2006 ; Havsteen be perfect in specifi c biotechnological applica- 2002 ), and they are shown to play a protective role tions. In this chapter, we review the current use of against neoplasia, atherosclerosis and neurodegen- glycoside hydrolases (GHs) in fl avonoid extrac- erative diseases (Lee and Lee 2006 ; Boudet 2007 ). tions and conversions along with efforts to Because of these exclusive properties, fl avonoids develop GHs (especially β-glucosidase and endo- have received great attention and the industrial glucanase) for deglycosylation and glycosylation interest is growing rapidly. Apart from this role, of these polyphenolic compounds. antioxidants can also be added to food and other types of products to prolong their shelf-life. Currently over 6,500 fl avonoids have been identi- Structural Overview of Flavonoids fi ed (Corradini et al. 2011 ), and they are commonly and Different Flavonoid Glycosides found in plants, fruits, vegetables, ferns, stems, roots, tea, wine and also from bark (Nijveldt et al. The core structure of a fl avonoid is 2-phenyl 2001 ). Their role in plants is to protect against benzopyranone, also known as 2- phenyl-1, UV-radiation diseases, infections and insect inva- 4-benzopyrone (Fig. 2.1 ), in which the three-carbon sion (Corradini et al. 2011). The content of fl avo- bridge between phenyl groups is cyclised with noids varies, dependent on the source, but is oxygen (Corradini et al. 2011 ). Flavonoids are normally in the mg-range per kg raw material. For divided into fl avones, isofl avones, fl avonols, fl a- example, the content of the fl avonoid quercetin is vanones, fl avan-3-ols and anthocyanidins based around 300 mg/kg of onion (Griffi ths et al. 2002 ), on their degree of unsaturation and oxidation of 100 mg/kg of broccoli, 50 mg/kg of apples, 40 mg/kg the three-carbon segment (Hughes et al. 2001 ) of blackcurrants and 30 mg/kg of green tea (Table 2.1). They are generally found as glyco- (Hollman and Arts 2000 ). sidic conjugates with sugar residues, and Problems with many fl avonoids are, however, sometimes they can also exist as free aglycones low solubility and poor stability (in both polar (Stobiecki et al. 1999 ). For example, quercetin and nonpolar media) which make their uses exists mostly in the form of glycosides (Fig. 2.1 ).

Fig. 2.1 General structure of the fl avonoid backbone arrows. A quercetin molecule (right ) is also shown with ( left ), shown with backbone numbering. The most com- the substituents present in this type of fl avonoid. R and R′ mon hydroxyl positions for glycosylation (3 and 7) are are hydrogens in the deglycosylated form. In glycosylated indicated with black arrows , and the 5 and 4′ hydroxyls forms isolated from onion, R and/or R′ represents gluco- that are sometimes glycosylated are indicated with grey syl groups 2 Glycoside Hydrolases for Extraction and Modifi cation of Polyphenolic Antioxidants 11

Table 2.1 Chemical structures of subclasses of ﬂ avonoids Flavonol R1 R2 Quercetin OH H Kaempferol H H Myricetin OH OH Isorhamnetin OMe H

Flavone R1 Apigenin H Luteolin OH

Flavanones R1 R2 Eriodictyol OH OH Hesperetin OH OMe Naringenin H OH

Flavan-3-ols R1 (+) Catechin H

Anthocyanidin R1 R2 Cyanidin OH H Delphinidin OH OH Malvidin OMe OMe Petunidin OMe OH

The addition of the glycoside conjugates or gly- vonoids in the cell vacuole (Cuyckens et al. cosylation makes the fl avonoid less reactive and 2003 ). The development of fl avonoid-O- glycosides more polar, leading to higher water solubility. includes one or more of the aglycone hydroxyl Hence, this is the most important modifi cation groups bound to a sugar with formation of an that occurred in plants to protect and store the fl a- O-C acid-labile acetal bond. The glycosylation 12 K.Z.G. Ara et al. does not occur in each hydroxyl groups but in done by using different extraction methods certain favoured positions: 3- and 7-hydroxyls (Fig. 2.2). In processing of renewable resources are common glycosylation sites, but glycosyl- such as agricultural by-products or bark, enzy- ations are also reported at 5-hydroxyls in antho- matic hydrolysis can be coupled with the extrac- cyanidins and 4′-hydroxyls in the fl avonol tion process, and GHs (sometimes also termed quercetin (Cuyckens et al. 2003 ; Iwashina 2000 ; glycosidases) are commonly used for these pur- Robards et al. 1997 ). The most encountered sugar poses. These enzymes are generally easy to han- is glucose, followed by galactose, arabinose, dle, as they do not require cofactors and they can rhamnose and xylose, while glucuronic and be used at an early stage on the readily available galacturonic acids are quite rare. Further, some material found in the forest and agricultural sec- disaccharides are also found in conjugation with tors (Turner et al. 2007 ). GHs are hydrolases fl avonoids, like rutinose (6-O-L -rhamnosyl-d - responsible for the transfer of glycosyl moieties glucose ) and neohesperidose (2-O-L -rhamnosyl- from a donor sugar to an acceptor and have either d - glucose) (Robards et al. 1997 ). an inverting or retaining (Fig. 2.3) reaction mechanism, and in hydrolysis the acceptor is water (Ly and Withers 1999 ). The hydrolysed Glycoside Hydrolases glycosidic bond can be located between two or as Extraction Aids more carbohydrates (e.g. polysaccharides) but also between a carbohydrate and a non-carbohy- By-products from agriculture, food and forest drate moiety (e.g. glycosylated antioxidants). In industries have the potential to become a major these types of applications, enzymes can (depen- source of fl avonoids. Isolation of the polypheno- dent on their specifi city) thus be used both in lic compounds from the plant sources is usually pretreatment of the raw materials – acting on the

Fig. 2.2 Schematic overview of an extraction process to obtain antioxidants with desired glycosylation patterns. The possibilities to use glycoside hydrolases in pretreatment and in conversions to modify the glycosylation are indicated 2 Glycoside Hydrolases for Extraction and Modiﬁ cation of Polyphenolic Antioxidants 13

Fig. 2.3 The double displacement mechanism of retain- the acceptor molecule. The covalent glycosyl-enzyme ing glycoside hydrolases. HO-R1 represents the group intermediate is boxed. For hydrolysis reactions HO-R2 is cleaved from the donor substrate, while HO-R2 represents a water molecule and R2=H polysaccharide fi bres to simplify release of the glycosylated fl avonoids. Extractions from biomass secondary metabolites (the antioxidants) in the often benefi t from high-temperature processing, as following extraction (Fig. 2.2 ) but also to change this aids in loosening recalcitrant polysaccharide the glycosylation pattern (described in more fi bre structures. A step in this direction is also detail in section “ Glycoside Hydrolases in taken in fl avonoid extractions, in which novel Flavonoid Conversions ”) on the polyphenolic technologies striving to increase the environmen- products. Pretreatment with different types of tal performance have been used that replace tradi- polysaccharide- degrading glycoside hydrolases tionally used extraction solvents (e.g. methanol [cellulases, hemicellulases (e.g. xylanases and and where deglycosylation is made by acid) with mannanases) and pectinases] before the extrac- pressurised hot water where deglycosylation is tion has, for example, been reported to promote made in an enzymatic step (Turner et al. 2006 ; release of the desired secondary metabolite fl a- Lindahl et al. 2010 ). The high-temperature extrac- vonoids from matrices of different sources con- tion method puts in a need of a thermostable taining complex polysaccharides (Fu et al. 2008 ; enzyme, which in this case was obtained from a Kapasakalidis et al. 2009 ; Landbo and Meyer thermophilic microorganism (Thermotoga nea- 2001 ; Lin et al. 2009 ; Maier et al. 2008 ; Zheng politana ) but which also can be developed from et al. 2009 ). Sources investigated include fruits enzymes originally active at ambient temperatures and berries, e.g. apples (Zheng et al. 2009 ), by mutagenesis. In the latter case, both rational blackcurrants (Landbo and Meyer 2001 ) and and random methods have been utilised, but due to grapes (Maier et al. 2008 ), but also agricultural relatively straight forward screening possibilities products such as pigeon peas (Fu et al. 2008 ) (often relying on incubations and activity assays at or products from forestry, such as pine (Lin the desired temperature), different random strate- et al. 2009 ). gies are frequently utilised (Fig. 2.4 ). Successful combinatorial designs for enhanced (thermal) stability development have, for instance, been Development of Thermostability: reviewed by Bommarius et al. (2006 ). A Means to Improve GHs as Extraction Aids Glycoside Hydrolases in Flavonoid Thermostable GHs have been well documented for Conversions use in low-value, high-volume applications, such as starch degradation and the conversion of ligno- Use of GHs as specifi c catalysts to modify the cellulosics (Turner et al. 2007 ), but they are still substituents on the target product is currently relatively rarely used in extractions/conversions of also gaining attention. Taking advantage of the 14 K.Z.G. Ara et al.

Fig. 2.4 Strategies for mutagenesis of enzymes by rational and random methodologies

possibilities to utilise retaining enzymes for both compounds is regulated in living organisms synthesis and hydrolysis, GHs can be used to (Yang et al. 2007 ) and will also increase water either remove glycoside substituents (by hydro- solubility of the molecule. Many well-designed lysis using water as acceptor) or to add substitu- chemical glycosylation methods are available, ents (using in this case fl avonoid acceptor but due to limitation of acceptor, it is not pos- molecules) (Fig. 2.3 ). Hydrolases express cata- sible to obtain regioselective glycosylation by lytic activity also in media with low water con- using those methods (Davis 2000 ; Kong 2003 ). tent such as organic solvents (resulting in less The delicate selectivity of biocatalysts can competition with water as acceptor molecule) instead be used for this purpose, and as stated and may under these conditions catalyse new above, GHs provide versatile tools for both gly- reactions (Klibanov 2001 ). GHs however work cosylation and deglycosylation. Below, a few rather poorly in organic media (compared to examples of GHs used (i) for hydrolysis of gly- other hydrolytic enzymes, e.g. lipases) due to the cosidic groups and developed to improve degly- requirement of higher thermodynamic water cosylation of fl avonoids and (ii) for synthesis activity (Ljunger et al. 1994 ). The reasons for this (introducing new glycosidic groups) and devel- are largely unknown, but indicate that water mol- oped to increase glycosylation on fl avonoid ecules have a role in interactions between sub- backbones are given. For hydrolysis, the exam- strate and enzyme. Use of thermostable GHs, ples focus on β-glucosidases, while for the syn- when organic media are used, may again be thesis the examples shown mainly concerns advantageous as these enzymes are often resis- endoglucanases but also mention use of tant to denaturation by organic solvents, espe- α-amylase. cially when run below their temperature optima for activity. Enzymatic hydrolysis of fl avonoid glyco- Deglycosylation of Flavonoids Using sides is dependent on the aglycone moiety, type β-Glucosidases of sugar and linkage, and is, e.g., used to obtain uniform fl avonoid molecules with often higher Glycosylated fl avonoids are the favoured forms antioxidising power than their glycosylated for uptake in the human intestine but are in the counterparts (Turner et al. 2006 ; Lindahl et al. body converted to the aglycone and free carbohy- 2010). On the other hand, glycosylation of fl a- drates in hydrolysis reactions. The hydrolysis vonoids is one of the predominant approaches reactions are mainly catalysed by β-glucosidases by which the biological activity of these natural (Walle 2004 ). The β-glucosidases are also helpful 2 Glycoside Hydrolases for Extraction and Modifi cation of Polyphenolic Antioxidants 15 catalysts in analysis of fl avonoid content in food and act by releasing the nonreducing terminal glucosyl residue leaving a uniform aglycone that is easier to quantify. The β-glucosidases (EC 3.2.1.21) are widely distributed in nature and found in all domains of living organisms and are classifi ed under six GH families (GH1, GH3, GH5, GH9, GH30 and GH116), of which GH1,

GH5 and GH30 display related (β/α)8 barrel structures and are classifi ed as GH clan A (Ketudat Cairns and Esen 2010 ). Five of the families host enzymes with retaining reaction mechanism (only GH9 holds inverting enzymes), but examples of β-glucosidases used for deglucosylation of fl avonoids mainly include enzymes from GH1 and GH3. From GH3 a few examples of enzymes used to deglycosylate fl avonoids are published [including thermostable enzymes from T. neapolitana (termed Tn Bgl3B) and Dictyoglomus turgidum (Turner et al. 2006 ; Kim et al. 2011 )], showing that glycosidic groups at position 7 and 4′ on Fig. 2.5 Spacefi ll representations of the human cytosolic the fl avonoid backbone (Fig. 2.1) were readily β-glucosidase ( top ) and the thermostable β-glucosidase hydrolysable. from Thermotoga neapolitana ( bottom ) facing the active site. The wider active site opening of the fl avonoid From GH1 which includes the largest number 3- glucoside hydrolysing T. neapolitana enzyme is clearly of characterised β-glucosidases, examples visible include enzymes from different domains of life, such as another enzyme from the prokaryotic thermophile T. neapolitana ( Tn Bgl1A) (Turner wider opening of the active site cleft in Tn Bgl1A, et al. 2006 ) but also eukaryotic enzymes of fun- making it possible for the 3-glucoside to fi t gal as well as human origin. Tn Bgl1A is a GH1 (Khan et al. 2011 ). enzyme, which (like the GH3 candidates) effi - ciently hydrolyses glucosylations at the 4′- and Increased Flavonoid Hydrolysis in GH1 7-positions, but in this case hydrolysis of gluco- by Structure-Based Site-Directed sides at the 3-position was also recognisable, Mutagenesis although less effi cient (Lindahl et al. 2010 ). The catalytic domain regions in GH1 are well The two intracellular GH1 β-glucosidases from conserved, but the enzymes in the family have the fungus Penicillium decumbens named GI varying substrate specifi cities, with some enzymes and GII also displayed low activity towards the very specifi c for only one sugar (e.g. true cello- 3- glucoside and were most active towards fl a- biases) or one aglycone (i.e. aryl-β- glucosidases), vonoids glycosylated at the 7-position (Mamma while others have a broad range of specifi cities et al. 2004 ). The human β-glucosidase (hCBG) for the glycones, the aglycones or both and are also preferred deglycosylation at the 7-position, broad substrate specifi city enzymes (e.g. using but in this case glucosides located at the cellulose and β-glucan as well as fl avonoids as 3-position were not hydrolysed (Berrin et al. substrates) (Bhatia et al. 2002 ). These differences 2003 ). Comparison of the 3D structures make enzymes from GH1 interesting models for (Fig. 2.5 ) showed that this difference between studies of the relationship between structure and Tn Bgl1A and hCBG can be attributed to the activity (Chuenchor et al. 2008 ). 16 K.Z.G. Ara et al.

Most genetic engineering studies done on the introducing the mutations, it was revealed that GH1 enzymes in relation to fl avonoid hydrolysis mutant N221S led to a signifi cant increase in con- have thus far focused on identifying residues of version of quercetin-3-glucosides to quercetin, importance for activity and mainly involve while no effect was observed for F219L and a TnBgl1A (Khan et al. 2011 ) and the human limited increase was seen after the G222 muta- cytosolic β-glucosidase hCBG (Berrin et al. tions. The effect of the N221S mutation was 2003 ), while a multitude of GH1 enzymes have suggested to occur via a loss in backbone car- been studied and mutated for other purposes. The bonyl interactions that resulted in an increased studies on Tn Bgl1A and hCBG are site-directed fl exibility of the parallel β-sheets, which could be mutagenesis studies, focusing on residues identi- a reason for the observed increase in catalytic fi ed by analysis of enzymes with known 3D effi ciency towards quercetin-3-glucosides (Khan structure. In our laboratory, the aim was to also et al. 2011 ). improve the hydrolysis of quercetin-3-glucosides Mutation of a neighbouring residue in the using Tn Bgl1A as model enzyme. To start this human enzyme (hCBG, F225S) resulted in work, bioinformatic analysis of Tn Bgl1A was almost complete loss of activity of the enzyme made, comparing this broad specifi city enzyme (Berrin et al. 2003; Tribolo et al. 2007). To with GH1 enzymes active on other bulky phenol- elucidate the role of the corresponding residue containing substrates (e.g. isofl avonoids and the (N220 in Tn Bgl1A), this was also selected for alkaloids strictosidine and raucaffricine) and mutagenesis and was mutated to S as well as to F oligosaccharide- specifi c enzymes. The analysis (which was originally present in the human was made to identify differences between speci- enzyme). In case of Tn Bgl1A, the N220S fi city groups in regions around the +1 and +2 mutation increased the catalytic effi ciency sugar-binding subsites in GH1 enzymes. To towards quercetin-3- glucosides compared to the locate these subsites, 3D structure-determined wild-type enzyme (a result of a combined drop enzymes were included in the analysis and the in KM value together with threefold increase in structure of Tn Bgl1A was determined (Khan the turnover number). Moreover, replacement et al. 2011 ; Kulkarni et al. unpublished ). From of the hydrophilic amino acid residue N220 by the comparison, nonconserved amino acid resi- the aromatic hydrophobic residue F resulted dues located in β-strand 5 (spanning the +1 and in drastic drop in the hydrolysis of both fl avo- +2 subsites) were mutated to residues occurring noid glycosides as well as smaller model sub- in the enzymes identifi ed to use polyphenolic strates like para -nitrophenyl-β-D -glucopyranoside substrates. Different fl avonoid glucosides like (Kulkarni et al. unpublished ). A similar effect quercetin-3-glucoside, quercetin-3,4′-diglucoside was also seen at position 221. The mutation and quercetin-4′-glucoside were also docked in N221F, which introduced small local changes in the active site Tn Bgl1A in order to identify puta- the range of 0.4–0.7 Å, led to loss of catalytic tive interactions of the amino acid residues cho- activity compared to the wild type. This con- sen for mutagenesis. For example, the GH1 fi rmed that also N220 plays a role in hydrolysis sequences showed variability at positions 219, but that the interactions vary between specifi c 221 and 222 ( TnBgl1A numbering), and the GH1 enzymes and that the neighbouring resi- mutations F219L, N221S, G222Q and G222M dues N220 and N221 in TnBgl1A may display a were made on the basis of residues found in similar rearrangement upon a single mutation of enzymes hydrolysing the bulky phenol-containing the respective residue. The corresponding N245 substrates. The site-directed mutagenesis meth- (at the +2 sugar-binding site) in the homologous odology used was a ligation-independent whole rice β-glucosidase Os3BGlu7 was mutated to plasmid amplifi cation methodology (Fig. 2.4 ) M and resulted in 6.5-fold loss of catalytic effi - utilising a methylated template that was selec- ciency towards laminaribiose and 17–30-fold tively degraded by a methylation-specific loss of catalytic effi ciency for cellooligosaccha- restriction enzyme after PCR amplifi cation. After rides with degree of polymerisation >2. On the 2 Glycoside Hydrolases for Extraction and Modifi cation of Polyphenolic Antioxidants 17 other hand, the corresponding mutation M251N Cellulases are GHs that catalyse the hydroly- in Os3BGlu6 led to 15-fold increase in the cata- sis of the 1,4-β-D -glycosidic linkages in cellulose lytic effi ciency for laminaribiose and 9–24-fold (and other related substrates, e.g. lichenan and increase in catalytic effi ciency for cellooligo- cereal β-glucans). The name cellulase can refer to saccharides with degree of polymerisation >2 different types of enzymes acting on cellulose but (Sansenya et al. 2012 ). These observations is most commonly used for endoglucanases show this position to be important for substrate (endo-1,4-β-D -glucanases, EC 3.2.1.4), which interactions in GH1 enzymes. Presence of a can be found in at least 17 different GH families bulkier hydrophobic group in the local area (see http://www.cazy.org/Glycoside-Hydrolases. around the amino acids 220 and 221 in TnBgl1A html ). was not favoured neither for the hydrolysis of For the synthesis of fl avonoid glycosides, a long-chain cellooligosaccharides nor for the cellulase from Aspergillus niger has been used to bulkier fl avonoid glucosides. add a fucose sugar moiety to a catechin backbone (Fig. 2.6 ). Catechin is an antioxidant that is, for instance, found in bark, a renewable resource of Glycosylation of Flavonoids Using signifi cant volume. During the reaction para - Cellulase and Amylase nitrophenyl - β -D -fucopyranoside was acting as a donor, while catechin monohydrate acted as Glycosylation of fl avonoids is one of the predom- acceptor resulting in a 26 % yield of catechin-β-d - inant approaches by which the biological activity fucopyranoside (Gao et al. 2000 ). In another of these natural compounds is regulated in living study, an α-amylase from Trichoderma viride organisms (Yang et al. 2007 ). Many well- was reported to show transglycosylation activity designed chemical glycosylation methods are towards both catechin and epigallocatechin gal- available, but due to limitation of acceptor, it is late using dextrin (α-1,4-linked oligosaccharides not possible to obtain regioselective glycosyl- resulting from starch degradation) as donor ation by using those methods (Davis 2000 ; Kong substrate (Noguchi et al. 2008 ). α-Amylases 2003 ). However, to overcome this problem, the (EC 3.2.1.1) are enzymes classifi ed under the delicate selectivity of a biocatalyst can be used, large glycoside hydrolase family 13, and their and glycosyltransferases and GHs can assist as main activity is hydrolysis of α-1,4-bonds of useful tools for synthesis of defi ned glycosylated starch and glycogen. fl avonoids (Hancock et al. 2006 ). In terms of substrate specifi city, a glycosyltransferase is usually Glycosynthases: Application strict and requires a complex sugar nucleotide as of a Nucleophile-Mutated Cellulase the donor for the catalysis reaction (Wang and in Flavonoid Glycosylation Huang 2009 ). On the other hand, GHs can also The application of GHs in synthesis of carbohy- catalyse transglycosylation reactions. If the water drate or non-carbohydrate substrates has two molecule in the hydrolysis reaction is replaced by major limitations: mainly low transglycosylation another acceptor molecule (e.g. a sugar molecule yield and secondary hydrolysis of the product. In or a fl avonoid), the double displacement mecha- order to overcome this problem, the invention of nism of retaining GHs will result in transfer of glycosynthases was a major development (Ly and the covalently bound glycosyl group from the Withers 1999 ). Glycosynthases is a class of donor substrate to the acceptor molecule. unique GH mutants (mutated in the catalytic Transglycosylation is kinetically controlled, and nucleophile) that can promote glycosidic bond during the reaction it is assumed that there is formation in the presence of an activated glyco- competition between the nucleophilic water and syl donor, and there is no further hydrolysis of the the acceptor substrate at the glycosyl-enzyme newly formed glycosidic linkage (Ly and Withers intermediate (Nakatani 2001 ; Park et al. 2005 ; 1999; Wang and Huang 2009). Drawbacks with Hancock et al. 2006 ). this methodology are however the necessity to 18 K.Z.G. Ara et al.

Fig. 2.6 Glycosylation of (+) catechin monohydrate. Synthesis of (+) catechin-β- D -fucopyranoside in the presence of cellulase at 37 °C while using pNP-β-D-fucopyranoside as donor (Modiﬁ ed from Gao et al. 2000 )

produce activated glycosyl donors, with sometimes Conclusions limited stability. Glycosynthases have been studied for the synthesis of complex glycoconjugates, To increase extraction yields and allow modifi ca- and in a recent report it was shown that a glyco- tion of antioxidants, new biocatalysts have the synthase mutant E197S of the Humicola insolens potential to adapt the compounds to different cellulase Cel7B was able to glycosylate fl avo- applications. Biocatalysis also has the potential noids (Yang et al. 2007 ). Cel7B is a GH7 retaining of being a sustainable technology – something endoglucanase, and the predominant activity of that is given increased attention today with cur- the wild-type enzyme is hydrolysis of the α-1, rent concerns about climate change and scarcity 4-linked glucosidic bond in cellulose, which in the of fossil resources. In this chapter, we have mutant is circumvented by replacing the catalytic reviewed emerging attempts to use and develop nucleophile (E197) with S. In this study a high- GHs as biocatalysts for modifi cation of polyphe- throughput MS-based method was used to screen nolic antioxidants classifi ed under the fl avonoid 80 different acceptors and more than 20 glycosyl group. One aspect is the use of the GHs as extrac- donors for substrate activity. In this screening tion aids, and for this purpose an important prop- process, a subclass of fl avonoids was identifi ed as erty of the enzyme is its stability. Concerning the an acting acceptor for Cel7B- E197S during fl avonoid product, both the antioxidative power transglycosylation reaction (Fig. 2.7 ). According and stability are affected by glycosylation. to kinetic studies, the rate of glycosylation by Currently studies to specifi cally modulate this Cel7B-E197S was comparable to glycosyltrans- (both remove and add glycosidic groups) have ferases which is very promising for the synthesis started. Although work in this fi eld is just of glycosylated fl avonoids (Yang et al. 2007 ). emerging, positive results have been published, 2 Glycoside Hydrolases for Extraction and Modifi cation of Polyphenolic Antioxidants 19

Fig. 2.7 Glycosylation of fl avonoids by Cel7B-E197S gly- the Cel7B-E197S mutant in yields of 72–95 %. The synthe- cosynthase (Adapted from Yang et al. 2007 ; Wang and sis was stereoselective (only β-glycosides) and regioselec- Huang 2009 ). The nucleophile (E197) of the GH7 cellulase tive for the glycosylation using the hydroxyl group at 4′ ( as from Humicola insolens is mutated to S. The lactosyl fl uo- in A2–A4 ), while in absence of the hydroxyl group at this ride (LacF) was the disaccharide donor, and transfer of lac- position (as in A1 ), the 6-position was glycosylated instead. tosyl from LacF to a number of fl avonoids was catalysed by A1 = baicalein, A2 = luteolin, A3 = quercetin, A4 = fi setin where the action of GHs allows modifi cation at Bhatia Y, Mishra S, Bisaria VS (2002) Microbial β uncommon positions or with new sugar moieties. -glucosidases: cloning, properties, and applications. Crit Rev Biotechnol 22:375–407 Further development in this fi eld is expected, Bommarius AS, Broering JM, Chaparro-Riggers JF, allowing increased development and use of Polizzi KM (2006) High-throughput screening for novel ingredients and bioactive compounds from enhanced protein stability. Curr Opin Biotechnol poylphenolics. 17:606–610 Boudet AM (2007) Evolution and current status of research in phenolic compounds. Phytochemistry Acknowledgements The authors wish to thank Formas 68:2722–2735 2009-1527 (SuReTech), the European project AMYLOMICS Chuenchor W, Pengthaisong S, Robinson RC, Yuvaniyama J, and the Antidiabetic Food Centre, a VINNOVA VINN Oonanant W, Bevan DR, Esen A, Chen C-J, Opassiri Excellence Centre at Lund University. R, Svasti J, Cairns JRK (2008) Structural insights into rice bglu1 β-glucosidase oligosaccharide hydrolysis and transglycosylation. J Mol Biol 377:1200–1215 Corradini E, Foglia P, Giansanti P, Gubbiotti R, References Samperi R, Lagana A (2011) Flavonoids: chemical properties and analytical methodologies of identifi - Berrin JG, Czjzek M, Kroon PA, McLauchlan WR, cation and quantitation in foods and plants. Nat Puigserver A, Williamson G, Juge N (2003) Substrate Prod Res 25:469–495 (aglycone) specifi city of human cytosolic β-glucosidase. Cuyckens F, Shahat AA, Van den Heuvel H, Abdel- Biochem J 373:41–48 Shafeek KA, El-Messiry MM, Seif-El Nasr MM, 20 K.Z.G. Ara et al.

Pieters L, Vlietinck AJ, Claeys M (2003) The application Kong F (2003) Regio- and stereoselective synthesis of oli- of liquid chromatography-electrospray ionization gosaccharides with unprotected or lightly protected sug- mass spectrometry and collision-induced dissociation ars as glycosyl acceptors. Curr Org Chem 7:841–865 in the structural characterization of acylated fl avonol Kulkarni TS, Khan S, Mahmood T, Sundin A, Lindahl S, O-glycosides from the seeds of Carrichtera annua . Turner C, Logan DT, Nordberg Karlsson E (unpub- Eur J Mass Spectrom 9:409–420 lished) Structure of Thermotoga neapolitana Davis BG (2000) Recent developments in oligosaccharide β-glucosidase 1A and comparison of active site synthesis. J Chem Soc Perkin Trans 1:2137–2160 mutants in hydrolysis of pNPGlc and quercetin-3 Fu Y-J, Liu W, Zu Y-G, Tong M-H, Li S-M, Yan M-M, glucosides Efferth T, Luo H (2008) Enzyme assisted extraction of Landbo A-K, Meyer AS (2001) Enzyme-assisted extrac- luteolin and apigenin from pigeonpea [ Cajanus cajan tion of antioxidative phenols from black currant juice (L.) Millsp.] leaves. Food Chem 111:508–512 press residues ( Ribes nigrum ). J Agric Food Chem Gao C, Mayon P, MacManus DA, Vulfson EN (2000) Novel 49:3169–3177 enzymatic approach to the synthesis of fl avonoid glyco- Lee KW, Lee HJ (2006) The roles of polyphenols in can- sides and their esters. Biotechnol Bioeng 71:235–243 cer chemoprevention. Biofactors 26:105–121 Griffi ths G, Trueman L, Crowther T, Thomas B, Smith B Lin JK, Weng MS (2006) Flavonoids as nutraceuticals. In: (2002) Onions—a global benefi t to health. Phytother Grotewold E (ed) The science of fl avonoids. Springer, Res 16(7):603–615. doi: 10.1002/ptr.1222 New York, pp 213–238 Haddad AQ, Venkateswaran V, Viswanathan L, Teahan SJ, Lin S-C, Chang C-MJ, Deng T-S (2009) Enzymatic hot Fleshner NE, Klotz LH (2005) Novel antiprolifera- pressurized fl uids extraction of polyphenolics from tive fl avonoids induce cell cycle arrest in human Pinus taiwanensis and Pinus morrisonicola . J Taiwan prostate cancer cell lines. Prostate Cancer Prostatic Inst Chem Eng 40:136–142 Dis 9:68–76 Lindahl S, Ekman A, Khan S, Wennerberg C, Borjesson P, Hancock SM, Vaughan MD, Withers SG (2006) Sjoberg PJR, Nordberg Karlsson E, Turner C (2010) Engineering of glycosidases and glycosyltransferases. Exploring the possibility of using a thermostable Curr Opin Chem Biol 10:509–519 mutant of β-glucosidase for rapid hydrolysis of quer- Havsteen BH (2002) The biochemistry and medical cetin glucosides in hot water. Green Chem 12: significance of the flavonoids. Pharmacol Ther 159–168 96:67–202 Ljunger G, Adlercreutz P, Mattiasson B (1994) Enzymatic Hollman PCH, Arts ICW (2000) Flavonols, fl avones synthesis of octyl-β-glucoside in octanol at controlled and fl avanols – nature, occurrence and dietary bur- water activity. Enzyme Microb Technol 16:751–755 den. J Sci Food Agric 80:1081–1093 Ly HD, Withers SG (1999) Mutagenesis of glycosidases. Hughes RJ, Croley TR, Metcalfe CD, March RE (2001) A Ann Rev Biochem 68:487–522 tandem mass spectrometric study of selected charac- Maier T, Göppert A, Kammerer D, Schieber A, Carle R teristic fl avonoids. Int J Mass Spec 210/211:371–385 (2008) Optimization of a process for enzyme-assisted Ishihara K, Nakajima N (2003) Structural aspects of acyl- pigment extraction from grape (Vitis vinifera L. ) pom- ated plant pigments: stabilization of fl avonoid gluco- ace. Eur Food Res Technol 227:267–275 sides and interpretation of their functions. J Mol Catal Mamma D, Hatzinikolaou DG, Christakopoulos P (2004) B Enzym 23:411–417 Biochemical and catalytic properties of two intracel- Iwashina T (2000) The structure and distribution of the lular β-glucosidases from the fungus Penicillium fl avonoids in plants. J Plant Res 113:287–299 decumbens active on fl avonoid glucosides. J Mol Catal Kapasakalidis PG, Rastall RA, Gordon MH (2009) Effect B Enzym 27:183–190 of a cellulase treatment on extraction of antioxidant Nakatani H (2001) Analysis of glycosidase-catalyzed phenols from black currant ( Ribes nigrum L.) pomace. transglycosylation reaction using probabilistic model. J Agric Food Chem 57:4342–4351 Arch Biochem Biophys 385:387–391 Ketudat Cairns J, Esen A (2010) β-Glucosidases. Cell Mol Nijveldt RJ, van Nood E, van Hoorn DE, Boelens PG, van Life Sci 67:3389–3405 Norren K, van Leeuwen PA (2001) Flavonoids: a Khan S, Pozzo T, Megyeri M, Lindahl S, Sundin A, Turner review of probable mechanisms of action and potential C, Nordberg Karlsson E (2011) Aglycone specifi city applications. Am J Clin Nutr 74:418–425 of Thermotoga neapolitana β-glucosidase 1A modi- Noguchi A, Inohara-Ochiai M, Ishibashi N, Fukami H, fi ed by mutagenesis, leading to increased catalytic Nakayama T, Nakao M (2008) A novel glucosylation effi ciency in quercetin-3-glucoside hydrolysis. BMC enzyme: molecular cloning, expression, and character- Biochem 12:11 ization of Trichoderma viride Jcm22452 α-amylase Kim Y-S, Yeom S-J, Oh D-K (2011) Characterization of a and enzymatic synthesis of some fl avonoid monoglu- GH3 family β-glucosidase from Dictyoglomus turgi- cosides and oligoglucosides. J Agric Food Chem dum and its application to the hydrolysis of isofl avone 56:12016–12024 glycosides in spent coffee grounds. J Agric Food Park T-H, Choi K-W, Park C-S, Lee S-B, Kang H-Y, Chem 59:11812–11818 Shon K-J, Park J-S, Cha J (2005) Substrate specifi c- Klibanov AM (2001) Improving enzymes by using them ity and transglycosylation catalyzed by a thermosta- in organic solvents. Nature 409:241–246 ble β-glucosidase from marine hyperthermophile 2 Glycoside Hydrolases for Extraction and Modifi cation of Polyphenolic Antioxidants 21

Thermotoga neapolitana . Appl Microbiol Biotechnol Turner P, Mamo G, Nordberg Karlsson E (2007) Potential 69:411–422 and utilization of thermophiles and thermostable Robards K, Li X, Antolovich M, Boyd S (1997) enzymes in biorefi ning. Microb Cell Fact 6:9 Characterisation of citrus by chromatographic analy- Turner C, Turner P, Jacobson G, Almgren K, Waldeback M, sis of fl avonoids. J Sci Food Agric 75:87–101 Sjoberg P, Nordberg Karlsson E, Markides KE (2006) Sansenya S, Maneesan J, Ketudat Cairns JR (2012) Subcritical water extraction and β-glucosidase- Exchanging a single amino acid residue generates catalyzed hydrolysis of quercetin glycosides in onion or weakens a +2 cellooligosaccharide binding sub- waste. Green Chem 8:949–959 site in rice beta-glucosidases. Carbohydr Res Walle T (2004) Absorption and metabolism of fl avonoids. 351:130–133 Free Radic Biol Med 36(7):829–837 Stobiecki M, Malosse C, Kerhoas L, Wojlaszek P, Einhorn J Wang L-X, Huang W (2009) Enzymatic transglycosyl- (1999) Detection of isofl avonoids and their glycosides ation for glycoconjugate synthesis. Curr Opin Chem by liquid chromatography/electrospray ionization Biol 13:592–600 mass spectrometry in root extracts of lupin ( Lupinus Yang M, Davies GJ, Davis BG (2007) A glycosynthase albus ). Phytochem Anal 10:198–207 catalyst for the synthesis of fl avonoid glycosides. Tribolo S, Berrin JG, Kroon PA, Czjzek M, Juge N Angew Chem 46:3885–3888 (2007) The crystal structure of human cytosolic Zheng H-Z, Hwang I-W, Chung S-K (2009) Enhancing poly- β-glucosidase unravels the substrate aglycone phenol extraction from unripe apples by carbohydrate- specifi city of a family 1 glycoside hydrolase. J Mol hydrolyzing enzymes. J Zhejiang Univrsity Sci B Biol 370:964–975 10:912–919 On the Enzyme Specifi city for the Synthesis of Prebiotic 3 Galactooligosaccharides

Barbara Rodriguez-Colinas , Lucia Fernandez-Arrojo , Miguel de Abreu , Paulina Urrutia , Maria Fernandez- Lobato , Antonio O. Ballesteros , and Francisco J. Plou

Abstract β -Galactosidases catalyze transgalactosylation reactions in which lactose as well as the glucose and galactose released by hydrolysis serve as galactosyl acceptors yielding a series of galactooligosaccharides (GOS). GOS constitute the major part of oligosaccharides in human milk and are responsible of the formation of a Biﬁ dus microbiota in the intestine of milk-fed babies. The bioactive properties of GOS depend on their chemical composition, structure, and polymerization degree. We have analyzed the product speciﬁ city of various β-galactosidases, namely, those from Kluyveromyces lactis, Bacillus circulans, and Aspergillus oryzae. The major products synthesized by B. circulans β-galactosidase contained only β-(1 → 4) bonds, whereas the enzyme from K. lactis synthesized GOS with major presence of β-(1 → 6) linkages. The A. oryzae β-galactosidase formed preferentially β-(1 → 6) bonds, with minor proportion of β-(1 → 3). B. circulans and K. lactis β-galactosidases produce nearly 45–50 % (w/w) GOS, whereas the A. oryzae enzyme produces less than 30 % (w/w). Another difference between the three enzymes was the polymerization degree of products; in particular, for a GOS mixture enriched in disaccharides, K. lactis and A. oryzae β-galactosidases are the best choices. In contrast, the B. circulans enzyme would be preferable for a GOS product with a high trisaccharides and tetrasaccharides content.

B. Rodriguez-Colinas • L. Fernandez-Arrojo P. Urrutia • A. O. Ballesteros • F. J. Plou (*) Instituto de Catálisis y Petroleoquímica , Instituto de Catálisis y Petroleoquímica , CSIC , CSIC , Cantoblanco, Marie Curie 2 , Madrid Cantoblanco, Marie Curie 2 , Madrid 28049 , Spain 28049 , Spain e-mail: [email protected] School of Biochemical Engineering , M. de Abreu • M. Fernandez-Lobato Pontiﬁ cia Universidad Católica de Valparaíso , Centro de Biología Molecular Severo Ochoa Valparaíso , Chile (CSIC-UAM) , Universidad Autónoma de Madrid , Madrid 28049 , Spain

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 23 DOI 10.1007/978-81-322-1094-8_3, © Springer India 2013 24 B. Rodriguez-Colinas et al.

Keywords Glycosidase • Galactooligosaccharides • Prebiotics • Transglycosylation • Beta-galactosidase • Oligosaccharides • Microbiota

benefi cially affects the host by selectively stim- Introduction ulating the growth and/or the activity of certain types of bacteria in the colon, basically of the β-galactosidases (β-D -galactoside galactohydrolases, genera Bifi dobacterium and Lactobacillus EC 3.2.1.23), also called lactases, catalyze the (Gibson and Ottaway 2000 ). Prebiotics produce hydrolysis of the galactosyl moiety from the positive effects on human health as the metabo- nonreducing end of various oligosaccharides. lism of these bacteria releases short-chain fatty β-galactosidases are retaining glycosidases, result- acids (acetate, propionate and butyrate) and ing in the net retention of the anomeric confi gura- L-lactate (Roberfroid 2007 ). Among them, they tion (Plou et al. 2007 ). The β-galactosidases exert protective effects against colorectal cancer are members of GH1 and GH2 families of the and bowel infectious diseases by inhibiting glycoside hydrolases (Cantarel et al. 2009 ). putrefactive and pathogen bacteria, improve the β-galactosidases have attracted attention from bioavailability of essential minerals, reduce industry as they can be used in different applica- the level of cholesterol in serum or enhance tions. Due to the defi ciency of human lactase, the glucid and lipid metabolism (Tuohy et al. many people in the world are intolerant to lactose 2005). GOS, fructooligosaccharides (FOS), present in dairy products; the problem is currently isomaltooligosaccharides (IMOS), lactulose, solved by means of treatment of lactose- rich prod- soybean oligosaccharides, lactosucrose, gen- ucts with microbial lactases (Adam et al. 2004 ). tiooligosaccharides, and xylooligosaccharides On the other hand, several companies have recently are among the most consumed prebiotics established projects for the transformation of whey (Playne et al. 2003 ). (one of the most important by-products of the It is well reported that the chemical structure agrofood industry, with a high lactose content), of the obtained oligosaccharides (composition, into bioethanol by means of a pretreatment with number of hexose units, and types of linkages β-galactosidases to hydrolyze lactose. Another between them) may affect their fermentation application of β-galactosidases is related to pattern by probiotic bacteria in the gut (Cardelle- transgalactosylation reactions in which lactose Cobas et al. 2011 ; Martinez-Villaluenga et al. (as well as the released glucose and galactose) 2008). Varying the source of β-galactosidase, the serve as galactosyl acceptors yielding a series yield and composition of GOS can be modifi ed of disaccharides, trisaccharides and higher (Iqbal et al. 2010 ; Maischberger et al. 2010 ; oligosaccharides called galactooligosaccharides Rodriguez-Colinas et al. 2011 ; Splechtna et al. (GOS) (Park and Oh 2010 ; Torres et al. 2010 ). 2006). In general, complex mixtures of GOS of GOS constitute the major part of oligosaccharides various chain lengths and glycosidic bonds in human milk (Gosling et al. 2010 ; Rastall et al. are commonly obtained. The most studied 2005 ; Shadid et al. 2007 ). The formation of a β-galactosidases are those from Kluyveromyces benefi cial Bifi dus microfl ora in the intestine of lactis (Chockchaisawasdee et al. 2005 ; Martinez- milk-fed babies seems to be related with the Villaluenga et al. 2008 ; Maugard et al. 2003 ), prebiotic effect of GOS present in the maternal Aspergillus oryzae (Albayrak and Yang 2002c ; milk. Guerrero et al. 2011; Iwasaki et al. 1996 ), In the context of functional foods, a pre- Bacillus circulans (Gosling et al. 2011 ) and biotic is a nondigestible food ingredient that Bifi dobacterium sp. (Hsu et al. 2007 ). 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 25

In our laboratory, we have studied the transga- pH 5.5 and 40 °C. We analyzed in detail the lactosylation activity of several β-galactosidases. synthesis of galactooligosaccharides (GOS) In this chapter, we will summarize the structural catalyzed by Biolactase using 400 g/l lactose features of the GOS synthesized by the different and 1.5 U/ml (β-galactosidase activity towards enzymes. In addition, the kinetics of transgalac- ONPG), in particular the selectivity of the bonds tosylation and the GOS yields will be compara- formed. tively discussed.

Speciﬁ city of B. circulans Bacillus circulans β-Galactosidase β-Galactosidase

Different isoforms of the β-galactosidase from The complete identifi cation of the GOS synthe- Bacillus circulans have been reported in the sized by a particular enzyme is a diffi cult task. In commercial preparation Biolacta (Daiwa Kasei). fact, considering the formation of Gal-β(1 → 2), At least three isoforms with distinctive behavior Gal-β(1 → 3), Gal-β(1 → 4), and Gal-β(1 → 6) in GOS production have been characterized: bonds, the theoretical number of synthesized β-galactosidase-1 showed very low transglyco- GOS accounts for 7 disaccharides, 32 trisaccha- sylation activity (Mozaffar et al. 1984 ), rides, 128 tetrasaccharides, and so on. β-galactosidase-2 contributed most signifi cantly Figure 3.1 shows the chromatogram obtained to GOS synthesis (Mozaffar et al. 1984 , 1986 ), by high-performance anion-exchange chroma- and β-galactosidase-3 was able to produce GOS tography coupled with pulsed amperometric with β(1 → 3) bonds (Fujimoto et al. 1998 ). More detection (HPAEC-PAD) of the reaction mixture recently, Song et al. (2011a ) described in Biolacta with B. circulans β-galactosidase close to the four isoforms differing in their molecular size: time of maximum GOS concentration. Peaks 1 , β-gal-A (189 kDa), β-gal-B (154 kDa), β-gal-C 2, and 4 corresponded to galactose, glucose, and (134 kDa), and β-gal-D (91 kDa). The transferase lactose, respectively. As illustrated in the chro- activity of β-galactosidase from B. circulans has matogram, the two main products present in the been also applied to the synthesis of lactosucrose reaction mixture were peaks 11 and 17 . Using a (Wei et al. 2009 ), N-acetyl-lactosamine (Bridiau commercial standard, peak 11 was identified et al. 2010 ), and other galactosylated derivatives as the trisaccharide 4-galactosyl-lactose [Gal- (Farkas et al. 2003 ). Interestingly, the enzyme is β(1 → 4)-Gal-β(1 → 4)-Glc]. Peak 17 was purifi ed also able to catalyze the galactosylation of differ- by semipreparative chromatography using an ent acceptors in the presence of high percentages amino column, as the sugar concentration in of organic cosolvents, up to 50 % v/v (Usui et al. samples for HPAEC-PAD analysis was too low 1993 ). The B. circulans β-galactosidase has been for an effi cient scaling up . The mass spectrum of immobilized on different supports (Mozaffar et al. peak 17 showed that it was a tetrasaccharide. 1986 ; Torres and Batista-Viera 2012 ). However, The NMR data for peak 17 was consistent with a only partial analysis of the GOS formed in the molecule that presented two galactosyl moieties transglycosylation reaction with lactose has been β-(1 → 4)-linked to the O-4 of the galactose described (Mozaffar et al. 1986 ), probably owing unit of lactose, resulting in the tetrasaccharide to the complexity of the reaction mixture derived Gal- β (1 → 4)-Gal-β(1 → 4)-Gal-β(1 → 4)-Glc. from the presence of several isoforms with differ- Commercially available standards and other ent regiospecifi city. GOS purified in our laboratory allowed us A novel commercial preparation of to identify in the chromatograms the disaccha- β-galactosidase from B. circulans (Biolactase) rides allolactose [Gal-β(1 → 6)-Glc] (peak 3 ), was recently studied in our laboratory. Its volu- 3-galactobiose [Gal-β(1 → 3)-Gal] (peak 5 ), metric activity towards o -nitrophenyl- β-D - 4-galactobiose [Gal-β(1 → 4)-Gal] (peak 6 ) galactopyranoside (ONPG) was 2,740 U/ml at and Gal-β(1 → 3)-Glc (peak 8), as well as the 26 B. Rodriguez-Colinas et al.

8/9

2 4

1 6 11

3 7

5 10 12 14 16 17 18 13 15

0 10203040506070 Retention time (min)

Fig. 3.1 HPAEC-PAD analysis of the reaction of lactose 15 Gal-β(1 → 4)-Gal-β(1 → 3)-Glc, 17 Gal-β(1 → 4)-Gal- with B. circulans β-galactosidase (Biolactase). The peaks β(1 → 4)-Gal- β(1 → 4)-Glc, and 9 , 10 , 12 , 13 , 14 , 16 other correspond to: 1 galactose, 2 glucose, 3 allolactose, GOS (unknown). The chromatogram corresponds to the 4 lactose, 5 3-galactobiose, 6 4-galactobiose, 7 6-galactosyl- reaction mixture after 6.5 h with Biolactase. Conditions: lactose, 8 3-galactosyl-glucose, 11 4-galactosyl-lactose, 400 g/l lactose, 0.1 M sodium acetate buffer (pH 5.5), 40 °C

trisaccharide 6-galactosyl-lactose [Gal-β(1 → 6)- found signifi cant differences in total GOS yield, Gal- β (1 → 4)-Glc] (peak 7 ). We also purifi ed the structural analysis of the synthesized com- peak 15 by semipreparative hydrophilic interac- pounds was not reported. tion chromatography (HPLC-HILIC), whose mass and NMR spectra indicated that it was the trisaccharide Gal-β(1 → 4)-Gal-β(1 → 3)-Glc. Peaks GOS Production with B. circulans 9, 10, 12, 13, 14, and 16 remained unknown. β-Galactosidase It is worth noting that two of the major products synthesized by B. circulans β-galactosidase In the presence of lactose, most of the (peaks 11 and 17) contained only β-(1 → 4) bonds β-galactosidases catalyze both GOS synthesis (Fig. 3.2 ). Yanahira et al. ( 1995 ) were the fi rst in and lactose hydrolysis. The transferase/hydrolase performing structural analysis of the GOS formed ratio, which determines the maximum yield by B. circulans β-galactosidase employing of GOS, depends basically on two parameters: Biolacta from Daiwa Kasei; they reported that (a) the concentration of lactose and (b) the intrinsic the main product was 4-galactosyl-lactose, but enzyme properties, that is, the ability of the the formation of tetrasaccharides was not men- enzyme to bind the nucleophile (to which a galac- tioned. In that paper, several disaccharides and tosyl moiety is transferred) and to exclude H2 O trisaccharides were purifi ed and characterized; from the acceptor binding site. Maximum GOS the authors reported the presence of various GOS production for any particular enzyme depends with β-(1 → 2) bonds (Yanahira et al. 1995 ), which basically on the relative rates of the transgalacto- may correspond to some of the unidentifi ed peaks sylation and hydrolysis reactions. As the lactose in our study. Recently, Song et al. (2011b ) analyzed is consumed, the concentration of GOS increases the GOS production by the different isoforms of until it reaches a maximum. At this point, the rate B. circulans β-galactosidase; although the authors of synthesis of GOS products equals its rate of 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 27

a OH OH 6''

4'' 5'' O

HO O OH 2'' 6' OH 3'' 1'' 6 OH 4 5' O 4' 5 O b OH OH O 6''' HO 2' OH 1' HO 2 5''' O 3' 4''' 3 1 OH OH HO O 2''' 3''' 1''' OH OH 6'' 4'' 5'' O

HO O OH 2'' 6' OH 3'' 1'' 6 OH 4 5' O 4' 5 O O HO 2' HO OH 3' 1' 2 3 1 OH OH

Fig. 3.2 Galactooligosaccharides synthesized by β-galactosidase from B. circulans containing only β(1 → 4) bonds: (a ) Gal-β(1 → 4)-Gal-β(1 → 4)-Glc; (b ) Gal-β(1 → 4)-Gal-β(1 → 4)-Gal-β(1 → 4)-Glc

hydrolysis. Subsequently, kinetic control is lost, (10 g/l, 2.5 % of total carbohydrates) is signifi cantly and the reaction must be stopped quickly before lower than the obtained at 1.5 U/ml. This effect product hydrolysis becomes the major process indicates that the stability of B. circulans and a thermodynamic equilibrium is reached. β-galactosidase is only moderate under typical The existence of this maximum explains why GOS formation conditions; at 1.5 U/ml, the reaction transglycosylation results in higher yields of con- is stopped before reaching the fi nal composition. densation products compared with equilibrium- The upper range of GOS yield reported is controlled processes. close to 40–45 % (Hansson and Adlercreutz We performed the GOS synthesis at 400 g/l 2001; Rabiu et al. 2001 ; Splechtna et al. 2006 ), lactose with a biocatalyst concentration of 15 U/ml. which is lower than that obtained in the case of It has been widely reported that, working under other prebiotics such as fructooligosaccharides kinetic control conditions, enzyme concentration (approx. 65 %) (Alvaro-Benito et al. 2007 ; Ghazi has no effect on the maximum GOS yield as long et al. 2006). In consequence, the GOS production as no enzyme inactivation takes place and it only with B. circulans β-galactosidase is one of the exerts a marked infl uence on the reaction time highest values reported to date. at which the maximum oligosaccharide concentration is achieved (Buchholz et al. 2005 ; Chockchaisawasdee et al. 2005 ). Figure 3.3 shows Kluyveromyces lactis that the maximum GOS production at 15 U/ml β-Galactosidase was achieved in 6.5 h, with a yield of 198 g/l. This value corresponds to 49.4 % (w/w) of total The major commercial source of β-galactosidase sugars, which was higher than the value obtained by far is the mesophile yeast Kluyveromyces at 1.5 U/ml (165 g/l, 41.3 %). Interestingly lactis (Chockchaisawasdee et al. 2005 ; enough, the remaining lactose at the equilibrium Martinez- Villaluenga et al. 2008 ; Maugard 28 B. Rodriguez-Colinas et al.

400 Galactose 350 Lactose Total GOS 300 Glucose

250

200

150 Concentration (g/l) 100

0 0 1020304050607080 Reaction time (h)

Fig. 3.3 Kinetics of GOS formation at 15 U/ml using 400 g/l lactose catalyzed by β-galactosidase from B. circulans (Biolactase). Reaction conditions: 0.1 M sodium acetate buffer (pH 5.5), 40 °C

et al. 2003 ; Pal et al. 2009). Due to its intracel- Specifi city of K. lactis β-Galactosidase lular nature, production of cell-free K. lactis β-galactosidase is limited by the high cost associ- Figure 3.4 shows the HPAEC-PAD chromato- ated to enzyme extraction and downstream program of the reaction mixture with K. lactis cessing as well as the low stability of the enzyme β-galactosidase (Lactozym 3000L) after 7 h. (Park and Oh 2010 ; Pinho and Passos 2011 ). The Peaks 1 , 2 , and 5 were assigned to galactose, production of GOS in batch and continuous bio- glucose, and lactose, respectively. We found that reactors of K. lactis β-galactosidase has been the most abundant GOS synthesized by this described (Chockchaisawasdee et al. 2005 ; enzyme corresponded to peak 3 (the disaccharide Martinez-V illaluenga et al. 2008 ). Several 6- galactobiose [Gal-β(1 → 6)-Gal]), peak 4 approaches for immobilization of this enzyme (allolactose, Gal-β(1 → 6)-Glc), and peak 8 . have been proposed using different carriers Purification of peak 8 was performed by (Maugard et al. 2003 ; Zhou et al. 2003 ; Zhou and semipreparative HPLC-HILIC. Its mass spec- Chen 2001 ). Recently, the crystallization and trum indicated that it was a trisaccharide, and its resolution of its three-dimensional structure 2D-NMR spectra showed a galactosyl moiety has been reported (Pereira-Rodriguez et al. β-(1 → 6) linked to the galactose ring of lactose 2010 ). (Rodriguez-Colinas et al. 2011 ). We studied in detail the transgalactosylation In consequence, the three major products syn- activity of K. lactis β-galactosidase. Different thesized by K. lactis cells contained a β-(1 → 6) enzyme preparations were assayed: (1) ethanol- bond between two galactoses or between one permeabilized cells of a strain of K. lactis and galactose and one glucose (Fig. 3.5 ). This product (2) two soluble β-galactosidases from K. lactis selectivity may have signifi cant implications, as commercially available (Lactozym 3000L and it was reported that β(1 → 6) linkages are cleaved Maxilact LGX 5000). Structural characterization very fast by β-galactosidases from Bifi dobacteria of the synthesized galactooligosaccharides was (Martinez-Villaluenga et al. 2008 ), a key factor in carried out. the prebiotic properties. 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 29

8 5

2 1

9 3 10 6 7 11 12 13 14

0 10203040506070 Retention time (min)

Fig. 3.4 HPAEC-PAD analysis of the reaction of lactose 10 3-galactosyl-glucos, and 9 , 11 , 12 , 13 , 14 other GOS with K. lactis β-galactosidase (Lactozym 3000L): 1 galactose, (unknown). The chromatogram corresponds to the reaction 2 glucose, 3 6-galactobiose, 4 allolactose, 5 lactose, mixture after 7 h. Conditions: 400 g/l lactose, 0.1 M 6 3-galactobiose, 7 4-galactobiose, 8 6-galactosyl-lactose, sodium phosphate buffer (pH 6.8), 40 °C

OH OH 6' OH O 4' 5' 6 O O HO 4 5 2' 3' 1' OH HO 2 OH b 3 1 OH OH OH 6' O 4' 5' O HO 6 2' 3' 1' 4 OH 5 O HO HO 2 OH c 3 1 OH

OH OH 6'' OH O OH 4'' 5'' 6' 6 O O 4 HO 4' 5' O 2'' 5 3'' 1'' O OH HO 2' HO OH 3' 1' 2 OH 3 1 OH

Fig. 3.5 Major galactooligosaccharides synthesized by β-galactosidase from K. lactis : (a ) 6-galactobiose [Gal- β(1 → 6)-Gal], (b ) allolactose [Gal-β(1 → 6)-Glc], (c ) 6-galactosyl-lactose [Gal-β(1 → 6)-Gal-β(1 → 4)-Glc] 30 B. Rodriguez-Colinas et al.

Chockchaisawasdee et al. (2005 ) were the fi rst In this context, ethanol-permeabilized K. lactis in performing a preliminary structural analysis cells have been evaluated for lactose hydrolysis of the GOS formed by soluble K. lactis (Fontes et al. 2001 ; Genari et al. 2003 ; Panesar β-galactosidase (Maxilact L2000), concluding et al. 2007 ; Siso et al. 1992 ) and for the biocon- that the major bonds were β-(1 → 6). Maugard version of lactose and fructose to the disaccha- et al. (2003 ) and Cheng et al. (2006 ) detected the ride lactulose (Lee et al. 2004 ). We investigated formation of disaccharides by K. lactis , but the use of K. lactis permeabilized cells for the no structural identifi cation was done. Martinez- synthesis of GOS. Although permeabilized cells Villaluenga et al. (2008 ) reported the formation of Kluyveromyces marxianus were also recently of products with β-(1 → 6) bonds with K. lactis employed for GOS production, the structure of β-galactosidase, in accordance with our fi ndings, the synthesized products was not reported but the formation of other GOS was not (Manera et al. 2010 ). considered. Different methods of permeabilizing cells that The product specificity of K. lactis do not signifi cantly affect enzymatic activity β-galactosidase contrasts with that of its B. circu- have been described, including drying, treatment lans counterpart. The former exhibits a tendency with solvents or surfactants, lyophilization, ultra- to synthesize β-(1 → 6) bonds (Martinez- sonic treatment, and mechanical disruption Villaluenga et al. 2008 ; Rodriguez-Colinas et al. (Panesar et al. 2007 , 2011 ). In our work, a double 2011 ), whereas the latter prefers the formation treatment (ethanol incubation and lyophilization) of β-(1 → 4) bonds. Another difference between was applied to harvested K. lactis CECT 1,931 both enzymes deals with the formation of disac- cells. Although cells treated with short-chain charides, because B. circulans β-galactosidase alcohols may die in many cases, intracellular yields a moderate amount of allolactose, enzymes may resist such treatment and, in conse- 4-galactobiose, and 3-galactosyl-glucose, whereas quence, are not inactivated. In addition, the the K. lactis enzyme is able to use effi ciently permeability barriers of cell walls are lowered, so free galactose and glucose as acceptors yielding the cells themselves can be utilized as a biocata- 6-galactobiose and allolactose, respectively, with lyst repeatedly. A scanning electron micrograph notable yields (Rodriguez-Colinas et al. 2011 ). (SEM) of the permeabilized cells obtained in our experiments is shown in Fig. 3.6 . Figure 3.7 shows the reaction progress with GOS Production with Permeabilized permeabilized cells. The maximum GOS concen- Cells of K. lactis tration (177 g/l, which represents 44 % (w/w) of the total carbohydrates in the reaction mixture) Biochemical reactions using whole cells have was observed at 6 h. After that, the amount of advantages over soluble enzymes in many indus- GOS diminished by the hydrolytic action of trial bioconversion processes, allowing the reuse β-galactosidase until reaching an equilibrium of the biocatalyst and continuous processing. GOS concentration of 105 g/l. Interestingly, lac- However, the permeability barrier of the cell tose disappeared almost completely at the end of envelope for substrates and products often causes the process. As illustrated in Fig. 3.7, GOS con- very low reaction rates, especially in yeasts. taining β-(1 → 6) bonds contributed substantially In order to increase the volumetric activity, to total GOS throughout the process. At the point permeabilization of yeast cells is an economical, of maximal GOS concentration (6 h), the weigh simple, and safe process that usually facilitates composition of the mixture was: monosaccharides substrate access to the intracellular enzymes (Fontes (32 %), lactose (24 %), 6-galactobiose (6 %), et al. 2001 ; Kondo et al. 2000 ; Siso et al. 1992 ). allolactose (10 %), 6-galactosyl-lactose (16 %), and The permeabilizing agent may decrease the phos- other GOS (12 %). Martinez-Villaluenga et al. pholipids content in the membrane thus allowing ( 2008), using a soluble K. lactis β-galactosidase, the transit of low molecular weight compounds in reported a mixture formed by monosaccharides and out of the cells (Manera et al. 2010 ). (49 %), residual lactose (21 %), the disaccharides 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 31

Fig. 3.6 SEM micrograph of permeabilized K. lactis cells at 1,000x

400 Lactose 350 Total GOS 300 GOS with β(1-6) bonds

250

200

150

Concentration (g/l) 100

0 0 5 10 15 20 25 30 35 40 45 50 55 Reaction time (h)

Fig. 3.7 Galactooligosaccharides production from lactose by permeabilized K. lactis cells. Experimental conditions: 400 g/l lactose in 0.1 M sodium phosphate buffer (pH 6.8), 1.2–1.5 U/ml, 40 °C

6-galactobiose and allolactose (13 %), and in previous studies with the same enzyme 6-galactosyl-lactose (17 %). (Chockchaisawasdee et al. 2005 ; Martinez- In our analyzes with K. lactis β-galactosidase, Villaluenga et al. 2008 ; Maugard et al. 2003 ), GOS yield was slightly higher than the reported which could be the consequence of: (1) the 32 B. Rodriguez-Colinas et al.

ab 400 400 350 350 300 300 250 250 200 200 150 150 100 100 Concentration (g/l) Concentration (g/l) 50 50 0 0 0 5 10 15 20 25 30 35 40 45 50 55 60 0 5 10 15 20 25 30 35 40 45 50 55 60 Reaction time (h) Reaction time (h)

Fig. 3.8 Galactooligosaccharides production from lactose by (a ) Lactozym 3000L and (b ) Maxilact LGX 5000. Experimental conditions: 400 g/l lactose in 0.1 M sodium phosphate buffer (pH 6.8), 1.2–1.5 U/ml, 40 °C

overestimation of lactose in previous works due pattern in which a maximum GOS yield is to its tendency to coelute with allolactose, or followed by the hydrolysis of the synthesized (2) the contribution of minor GOS to the total products. Secondly, lactose is not completely yield is not considered in such works. consumed at the equilibrium, with a remaining concentration of 21 and 65 g/l for Lactozym and Maxilact, respectively, whereas only 3 g/l was GOS Production with Soluble K. lactis determined for permeabilized cells. Maximum β-Galactosidase GOS yield was slightly lower for soluble β-galactosidases (160 g/l for Lactozym and We studied the behavior of two commercially 154 g/l for Maxilact, which represent 42 % and available soluble β-galactosidases (Lactozym 41 % w/w of the total carbohydrates present in 3000L and Maxilact LGX 5000). The experi- the mixture) compared with permeabilized cells ments were carried out at 1.2–1.5 U/ml, which is (177 g/l, 44 % w/w). The above results seem to a lower enzyme concentration than the typically indicate that, due to the low stability of soluble used (3–12 U/ml) in similar experiments with β-galactosidases, the reaction fi nishes before K. lactis β-galactosidase (Chockchaisawasdee reaching the fi nal equilibrium. In contrast, the et al. 2005 ; Lee et al. 2004 ; Martinez-Villaluenga higher stability of permeabilized cells is demon- et al. 2008 ). At higher enzyme concentrations, strated by the characteristic reaction profi le – with reactions are so fast (less than 240 min) that is a maximum – when a competition exists between diffi cult to observe any effect of enzyme stability hydrolysis and transglycosylation (kinetic control) on reaction progress. The rest of the conditions (Rodriguez-Alegria et al. 2010 ). used in our experiments were similar to those Figure 3.9 illustrates the GOS synthesis vs. employed for soluble K. lactis β-galactosidase lactose conversion with the three K. lactis bio- (400 g/l lactose, pH 6.8, 40 °C). catalysts. The maximum GOS concentration with The activity of Lactozym and Maxilact permeabilized cells was obtained when lactose towards ONPG was 645 and 2,145 U/ml, respec- conversion was 76 %. Lactozym and Maxilact tively. The reaction profi le of both enzymes in the reached 95 % and 83 % of lactose conversion, GOS synthesis is shown in Fig. 3.8 . Several without displaying a maximum in GOS concentra- differences were found in the behavior of soluble tion. This could be indicating that the microenvi- β-galactosidases compared with permeabilized ronment of the permeabilized cells also exerts an cells. Firstly, Fig. 3.8 does not show the typical infl uence on the transglycosylation/hydrolysis ratio. 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 33

200 Permeabilized cells 175 Lactozym 3000 L HP G Maxilact LGX 5000 150

125

100

[GOS] (g/l) 75

0 0 102030405060708090100 Lactose conversion (%)

Fig. 3.9 Analysis of GOS formation as a function of conditions: 400 g/l lactose in 0.1 M sodium phosphate lactose conversion catalyzed by β-galactosidase from K. buffer (pH 6.8), 1.2–1.5 U/ml, 40 °C lactis in soluble form or in permeabilized cells. Reaction

Neri et al. 2011 ; Sheu et al. 1998 ) or combined Aspergillus oryzae β-Galactosidase ionic adsorption and cross-linking (Guidini et al. 2011 ). The β-galactosidase from Aspergillus oryzae is a monomeric enzyme whose molecular mass and isoelectric point are 105 kDa (Tanaka et al. 1975 ) Specifi city of A. oryzae and 4.6 (Ansari and Husain 2010 ; Yang et al. β-Galactosidase 1994 ), respectively. The biocatalyst optimum temperature is in the range 45–55 °C (Guidini We have studied the synthesis of GOS catalyzed et al. 2010; Guleç et al. 2010) and shows an opti- by a preparation of β-galactosidase from A. oryzae mum pH of 4.5 with ONPG as substrate and 4.8 commercially available (Enzeco® Fungal towards lactose (Tanaka et al. 1975 ). Lactase) using 400 g/l lactose and 15 U/ml The β-galactosidase from A. oryzae has been (β-galactosidase activity towards 30 mM ONPG at applied to the synthesis of different transgalac- 25 °C and pH 4.5 in citrate-phosphate buffer tosylated products such as GOS (Albayrak and 0.1 M). We analyzed by HPAEC-PAD the product Yang 2002b ; Iwasaki et al. 1996 ; Vera et al. specifi city of A. oryzae β-galactosidase (Urrutia 2012 ), lactulose (Guerrero et al. 2011 ; Mayer et al. 2013 ). Figure 3.10 shows the chromatogram et al. 2004 ), and galactosyl-polyhydroxyalcohols obtained with this enzyme after 4 h of reaction. (Irazoqui et al. 2009 ; Klewicki 2007 ), using free The main galactooligosaccharide synthesized by and immobilized enzyme. The β-galactosidase this enzyme was the trisaccharide 6-galactosyl- from A. oryzae has been immobilized by different lactose (peak 8 ). Several disaccharides contain- strategies including entrapment in alginate (Freitas ing different bonds were also identifi ed: et al. 2011 ), covalent attachment onto various Gal- β (1 → 6)-Gal (peak 3 ), Gal-β(1 → 6)-Glc carriers (Gaur et al. 2006 ; Huerta et al. 2011 ; (peak 4 ), Gal-β(1 → 3)-Gal (peak 7 ), and 34 B. Rodriguez-Colinas et al.

2 5

9 12 7 4 11 6 10 15 3 14 16 17 18 13

0 10203040506070 Retention time (min)

Fig. 3.10 HPAEC-PAD analysis of the reaction of lactose 11 n.d., 12 n.d., 13 n.d., 14 Gal-β(1 → 4)-Gal-β(1 → 4)-Glc, with A. oryzae β-galactosidase. The peaks correspond to 15 n.d., 16 n.d., 17 n.d., and 18 n.d. The chromatogram the following: 1 galactose, 2 glucose, 3 6-galactobiose, 4 corresponds to the reaction mixture after 4 h. Reaction allolactose, 5 lactose, 6 3-galactobiose, 7 n.d., 8 Gal- conditions: 400 g/l lactose, 0.1 M citrate-phosphate buffer β(1 → 6)-Gal- β(1 → 4)-Glc, 9 n.d., 10 Gal-β(1 → 3)-Glc, (pH 4.5), 40 °C. n.d. not detected

Gal- β (1 → 3)-Glc (peak 10 ). The enzyme was also information on its composition and structure able to synthesize tetrasaccharides (at least peaks was given. 17Ð18 , considering their retention times). One of the main contributions of our work is Toba et al. (1985 ) were the fi rst to report the the identifi cation of at least four disaccharides in characterization of some of the tri-, tetra-, and the reaction mixture, two of them containing a pentasaccharides synthesized by A. oryzae β(1 → 6) bond and the other two with a β(1 → 3) β-galactosidase. In particular, they identifi ed bond. The specifi c features of this enzyme indi- three trisaccharides, 6-galactosyl-lactose cate a tendency to form β(1 → 6) bonds followed (peak 8 in Fig. 3.10 ), 3-galactosyl-lactose, and by β(1 → 3) linkages, with a minor contribution Gal- β (1 → 4)-Gal-β(1 → 6)-Glc (the two later of β(1 → 4) bonds. not identifi ed by us in the chromatograms); two tetrasaccharides, Gal-β(1 → 6)-Gal-β(1 → 6)-Gal- β(1 → 4)-Glc and Gal-β(1 → 6)-Gal-β(1 → 3)-Gal- GOS Production with A. oryzae β(1 → 4)-Glc; and one pentasaccharide, β-Galactosidase Gal-β(1 → 6)-Gal- β (1 → 6)-Gal-β(1 → 6)-Gal- β(1 → 4)-Glc. Neri et al. ( 2011 ) characterized Figure 3.11 illustrates the kinetics of GOS pro- several of the GOS obtained with a preparation of duction using 400 g/l lactose and an enzyme con- A. oryzae β-galactosidase covalently immobilized centration of 15 U/ml. GOS reached a maximum onto a hydrazide-Dacron-magnetite composite. concentration of 107 g/l at 8 h of reaction, which In particular, they identifi ed three of the GOS corresponded to 26.8 % of total carbohydrates in previously characterized by Toba et al. (1985 ): the mixture. From that point of maximum yield, the trisaccharides 6-galactosyl-lactose and Gal- the GOS concentration slowly decreased down to β(1 → 4)-Gal-β(1 → 6)-Glc and the tetrasaccharide values of approximately 60 g/l. Gal-β(1 → 6)-Gal-β(1 → 6)-Gal-β(1 → 4)-Glc. It has been proved that the increase of lactose The authors also mentioned the presence of a concentration [up to 30–40 % (w/v)] has a strong disaccharide containing a β(1 → 6) bond, but no positive effect on the maximum GOS obtained 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 35

400 Galactose Glucose 350 Lactose Total GOS 300

250

200

150 Concentration (g/L)

100

0 0 102030405060 Reaction time (h)

Fig. 3.11 Kinetics of GOS formation at 15 U/ml using 400 g/l lactose catalyzed by β-galactosidase from A. oryzae . Reaction conditions: 0.1 M citrate-phosphate buffer (pH 4.5), 40 °C

(Iwasaki et al. 1996 ; Matella et al. 2006 ); in the amounts of di-, tri-, and tetrasaccharides consequence, initial lactose concentration should (expressed in weight percentage referred to the be considered for an adequate comparison with total sugars) were 9.9 %, 15.2 % and 1.8 %, other studies. Albayrak and Yang (2002a ) used an respectively. initial lactose concentration of 500 g/l obtaining Regarding the effect of pH and temperature on a maximum GOS concentration of 27 % (w/w), of GOS synthesis, it has been found that even which more than 70 % was formed by trisaccha- though both parameters affect reaction rate, they rides. In agreement with these results, Neri et al. do not modify maximum GOS concentration ( 2011 ) obtained 20.2 % trisaccharides and 5.9 % (Albayrak and Yang 2002b ; Neri et al. 2011 ). tetrasaccharides using 500 g/l lactose and 40 °C, which accounted for a total yield of 26.1 % (130 g/l). No significant differences were Conclusions observed when employing the free or immobilized enzyme. Gaur et al. ( 2006 ), starting with Signifi cative differences have been found in 200 g/l lactose, reported that A. oryzae the behavior of β-galactosidases from Bacillus β-galactosidase formed only trisaccharides, with circulans , Kluyveromyces lactis, and Aspergillus a maximum concentration of 22.6 % (w/w) for oryzae for the synthesis of GOS. First of all, the soluble enzyme and 25.5 % (w/w) for chitosan- product specifi city was very dependent on the immobilized β-galactosidase. Guleç et al. (2010 ) enzyme origin. Thus, K. lactis β-galactosidase reported the production of mainly trisaccharides, exhibits a tendency to synthesize β-(1 → 6) bonds, with a maximum of GOS concentration of 20.8 % whereas the B. circulans counterpart prefers the (w/w) using 320 g/l lactose and 55 °C. formation of β-(1 → 4) bonds. The β-galactosidase In our study, we noted that the contribution of from A. oryzae displays a marked preference to disaccharides to total GOS was very signifi cant, a form β-(1 → 6) linkages followed by β-(1 → 3). fi nding that had not been previously described for Maximum GOS concentration is also depen- A. oryzae β-galactosidase. In fact, at the point of dent on biocatalyst. The GOS production with maximum GOS concentration (7 h, Fig. 3.11 ), the three enzymes – at the point of maximum 36 B. Rodriguez-Colinas et al.

20 Total GOS (%) 15

0 Bacillus circulans Kluyveromyces lactis Aspergillus oryzae

35.0 Disaccharides Trisaccharides

30.0 Tetrasaccharides

25.0

20.0

15.0 GOS (%)

10.0

5.0

0.0 Bacillus circulans Kluyveromyces lactis Aspergillus oryzae

Fig. 3.12 Comparison of the three β-galactosidases studied in this work: (top ) maximum yield of GOS, (bottom ) distribution of di-, tri-, and tetrasaccharides at the point of maximum GOS production concentration – is represented in Fig. 3.12 (top). choice, as the amount of disaccharides almost B. circulans and K. lactis β-galactosidases equals that of trisaccharides. However, the produce nearly 45–50 % (w/w) GOS referred to B. circulans enzyme would be preferable if a the total amount of carbohydrates in the system. GOS with a high tri- and tetrasaccharides content In contrast, A. oryzae β-galactosidase produces is desirable. less than 30 % (w/w) GOS. Another difference As the properties of GOS may depend signifi - between the three enzymes deals with the cantly on their degree of polymerization and product distribution, as illustrated in Fig. 3.12 chemical structure, the selection of the appropriate (bottom). For a product enriched in disaccharides, enzyme could have a considerable effect on the K. lactis β-galactosidase would be the best bioactivity of the resulting product. However, 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 37 more structure-function studies (in vitro and in Cardelle-Cobas A, Corzo N, Olano A, Pelaez C, Requena vivo) assaying different GOS are required to T, Avila M (2011) Galactooligosaccharides derived from lactose and lactulose: infl uence of structure on determine the target molecules to be synthesized. Lactobacillus, Streptococcus and Bifi dobacterium growth. Int J Food Microbiol 149:81–87 Acknowledgements We thank Ramiro Martínez Cheng CC, Yu MC, Cheng TC, Sheu DC, Duan KJ, Tai (Novozymes A/S, Madrid, Spain) for supplying Lactozym WL (2006) Production of high-content galacto- and for useful suggestions. We thank DSM Food oligosaccharide by enzyme catalysis and fermentation Specialties for supplying Maxilact LGX 5000. Projects with Kluyveromyces marxianus . Biotechnol Lett BIO2010-20508-C04-01 and BIO2010-20508- C04-04 28:793–797 from Spanish Ministry of Science and Innovation sup- Chockchaisawasdee S, Athanasopoulos VI, Niranjan K, ported this research. B.R.C and M.A. were supported by Rastall RA (2005) Synthesis of galactooligosaccharide fellowships from the Spanish Ministries of Science and from lactose using beta-galactosidase from Innovation (FPI program) and Education and Culture Kluyveromyces lactis : studies on batch and continuous (FPU program), respectively. P. U was supported by a UF membrane-fi tted bioreactors. Biotechnol Bioeng PhD Fellowship and international internship from 89:434–443 Conicyt-Chile. Farkas E, Schmidt U, Thiem J, Kowalczyk J, Kunz M, Vogel M (2003) Regioselective synthesis of galactosylated tri- and tetrasaccharides by use of β-galactosidase from Bacillus circulans . Synthesis 5:699–706 References Fontes EAF, Passos FML, Passos FJV (2001) A mecha- nistical mathematical model to predict lactose hydro- Adam AC, Rubio-Texeira M, Polaina J (2004) Lactose: lysis by β-galactosidase in a permeabilized cell the milk sugar from a biotechnological perspective. mass of Kluyveromyces lactis : validity and sensitivity Crit Rev Food Sci Nutr 44:553–557 analysis. Process Biochem 37:267–274 Albayrak N, Yang ST (2002a) Immobilization of Freitas FF, Marquez LDS, Ribeiro GP, Brandao GC, Aspergillus oryzae beta-galactosidase on tosylated Cardoso VL, Ribeiro EJ (2011) A comparison of the cotton cloth. Enzyme Microb Technol 31:371–383 kinetic properties of free and immobilized Aspergillus Albayrak N, Yang ST (2002b) Immobilization of beta- oryzae β-galactosidase. Biochem Eng J 58–59:33–38 galactosidase on fi brous matrix by polyethyleneimine Fujimoto H, Miyasato M, Ito Y, Sasaki T, Ajisaka K for production of galacto-oligosaccharides from (1998) Purifi cation and properties of recombinant lactose. Biotechnol Prog 18:240–251 β-galactosidase from Bacillus circulans. Glycoconj J Albayrak N, Yang ST (2002c) Production of galacto- 15:155–160 oligosaccharides from lactose by Aspergillus oryzae Gaur R, Pant H, Jain R, Khare SK (2006) Galacto- beta-galactosidase immobilized on cotton cloth. oligosaccharide synthesis by immobilized Aspergillus Biotechnol Bioeng 77:8–19 oryzae beta-galactosidase. Food Chem 97:426–430 Alvaro-Benito M, De Abreu M, Fernandez-Arrojo L, Plou Genari AN, Passos FV, Passos FML (2003) Confi guration FJ, Jimenez-Barbero J, Ballesteros A, Polaina J, of a bioreactor for milk lactose hydrolysis. J Dairy Sci Fernandez-Lobato M (2007) Characterization of a 86:2783–2789 b-fructofuranosidase from Schwanniomyces occiden- Ghazi I, Fernandez-Arrojo L, Gomez de Segura AG, talis with transfructosylating activity yielding the pre- Alcalde M, Plou FJ, Ballesteros A (2006) Beet sugar biotic 6-kestose. J Biotechnol 132:75–81 syrup and molasses as low-cost feedstock for the enzy- Ansari SA, Husain Q (2010) Lactose hydrolysis by matic production of fructo-oligosaccharides. J Agric β-galactosidase immobilized on concanavalin Food Chem 54:2964–2968 A-cellulose in batch and continuous mode. J Mol Gibson GR, Ottaway RA (2000) Prebiotics: new develop- Catal B: Enzym 63:68–74 ments in functional foods. Crit Rev Foof Sci. Chandos, Bridiau N, Issaoui N, Maugard T (2010) The effects of Oxford organic solvents on the effi ciency and regioselectivity Gosling A, Stevens GW, Barber AR, Kentish SE, Gras SL of N-acetyl-lactosamine synthesis, using the (2010) Recent advances refi ning galactooligosaccha- β-galactosidase from Bacillus circulans in hydro- ride production from lactose. Food Chem 121:307–318 organic media. Biotechnol Prog 26:1278–1289 Gosling A, Stevens GW, Barber AR, Kentish SE, Gras SL Buchholz K, Kasche V, Bornscheuer UT (2005) (2011) Effect of the substrate concentration and water Biocatalysts and enzyme technology. Wiley-VCH, activity on the yield and rate of the transfer reaction of Weinheim β-galactosidase from Bacillus circulans. J Agric Food Cantarel BL, Coutinho PM, Rancurel C, Bernard T, Chem 59:3366–3372 Lombard V, Henrissat B (2009) The Carbohydrate- Guerrero C, Vera C, Plou F, Illanes A (2011) Infl uence of Active EnZymes database (CAZy): an expert reaction conditions on the selectivity of the synthesis resource for glycogenomics. Nucleic Acids Res of lactulose with microbial beta-galactosidases. J Mol 37:D233–D238 Catal B: Enzym 72:206–212 38 B. Rodriguez-Colinas et al.

Guidini CZ, Fischer J, Resende MMD, Cardoso VL, Manera AP, De Almeida Costa FA, Rodrigues MI, Kalil Ribeiro EJ (2011) β-Galactosidase of Aspergillus ory- SJ, Maugeri Filho F (2010) Galacto-oligosaccharides zae immobilized in an ion exchange resin combining production using permeabilized cells of Kluyveromyces the ionic-binding and crosslinking methods: kinetics marxianus . Int J Food Eng 6, article 4 and stability during the hydrolysis of lactose. J Mol Martinez-Villaluenga C, Cardelle-Cobas A, Corzo N, Catal B: Enzym 71:139–145 Lano A, Villamiel M (2008) Optimization of condi- Guidini CZ, Fischer J, Santana LNS, Cardoso VL, Ribeiro tions for galactooligosaccharide synthesis during EJ (2010) Immobilization of Aspergillus oryzae lactose hydrolysis by beta-galactosidase from β-galactosidase in ion exchange resins by combined Kluyveromyces lactis (Lactozym 3000 L HP G). Food ionic-binding method and cross-linking. Biochem Eng Chem 107:258–264 J 52:137–143 Matella NJ, Dolan KD, Lee YS (2006) Comparison of Guleç HA, Gürdaç S, Albayrak N, Mutlu M (2010) galactooligosaccharide production in free-enzyme Immobilization of Aspergillus oryzae β-galactosidase ultrafiltration and in immobilized-enzyme systems. on low-pressure plasma-modifi ed cellulose acetate J Food Sci 71:C363–C368 membrane using polyethyleneimine for production Maugard T, Gaunt D, Legoy MD, Besson T (2003) of galactooligosaccharide. Biotechnol Bioproc Eng Microwave-assisted synthesis of galacto- 15:1006–1015 oligosaccharides from lactose with immobilized beta- Hansson T, Adlercreutz P (2001) Optimization of galac- galactosidase from Kluyveromyces lactis . Biotechnol tooligosaccharide production from lactose using Lett 25:623–629 beta-glycosidases from hyperthermophiles. Food Mayer J, Conrad J, Klaiber I, Lutz-Wahl S, Beifuss U, Biotechnol 15:79–97 Fischer L (2004) Enzymatic production and complete Hsu CA, Lee SL, Chou CC (2007) Enzymatic production nuclear magnetic resonance assignment of the sugar of galactooligosaccharides by beta-galactosidase from lactulose. J Agric Food Chem 52:6983–6990 Bifi dobacterium longum BCRC 15708. J Agric Food Mozaffar Z, Nakanishi K, Matsuno R (1984) Purifi cation Chem 55:2225–2230 and properties of β-galactosidases from Bacillus Huerta LM, Vera C, Guerrero C, Wilson L, Illanes A circulans . Agric Biol Chem 48:3053–3061 (2011) Synthesis of galacto-oligosaccharides at very Mozaffar Z, Nakanishi K, Matsuno R (1986) Continuous high lactose concentrations with immobilized production of galacto-oligosaccharides from lactose β-galactosidases from Aspergillus oryzae . Process using immobilized beta-galactosidase from Bacillus Biochem 46:245–252 circulans . Appl Microbiol Biotechnol 25:224–228 Iqbal S, Nguyen TH, Nguyen TT, Maischberger T, Neri DFM, Balcao VM, Cardoso SM, Silva AMS, Haltrich D (2010) β-Galactosidase from Lactobacillus Domingues MDRM, Torres DPM, Rodrigues LRM, plantarum WCFS1: biochemical characterization and Carvalho LB, Teixeira JAC (2011) Characterization of formation of prebiotic galacto-oligosaccharides. galactooligosaccharides produced by β-galactosidase Carbohydr Res 345:1408–1416 immobilized onto magnetized Dacron. Int Dairy J Irazoqui G, Giacomini C, Batista-Viera F, Brena BM, 21:172–178 Cardelle-Cobas A, Corzo N, Jimeno ML (2009) Pal A, Pal V, Ramana KV, Bawa AS (2009) Biochemical Characterization of galactosyl derivatives obtained by studies of β-galactosidase from Kluyveromyces lactis . transgalactosylation of lactose and different polyols J Food Sci Technol 46:217–220 using immobilized β-galactosidase from Aspergillus Panesar R, Panesar PS, Singh RS, Kennedy JF, Bera MB oryzae . J Agric Food Chem 57:11302–11307 (2007) Production of lactose-hydrolyzed milk using Iwasaki KI, Nakajima M, Nakao SI (1996) Galacto- ethanol permeabilized yeast cells. Food Chem oligosaccharide production from lactose by an enzymic 101:786–790 batch reaction using β-galactosidase. Process Biochem Panesar R, Panesar PS, Singh RS, Kennedy JF (2011) 31:69–76 Hydrolysis of milk lactose in a packed bed reactor Klewicki R (2007) The stability of gal-polyols and system using immobilized yeast cells. J Chem Technol oligosaccharides during pasteurization at a low pH. Biotechnol 86:42–46 LWT- Food Sci Technol 40:1259–1265 Park AR, Oh DK (2010) Galacto-oligosaccharide pro- Kondo A, Liu Y, Furuta M, Fujita Y, Matsumoto T, Fukuda duction using microbial β-galactosidase: current H (2000) Preparation of high activity whole cell biocata- state and perspectives. Appl Microbiol Biotechnol lyst by permeabilization of recombinant fl occulent yeast 85:1279–1286 with alcohol. Enzyme Microb Technol 27:806–811 Pereira-Rodriguez A, Fernández-Leiro R, Cerdán E, Lee YJ, Kim CS, Oh DK (2004) Lactulose production by Becerra M, Sanz-Aparicio J (2010) Crystallization beta-galactosidase in permeabilized cells of Kluyvero- and preliminary X-ray crystallographic analysis of myces lactis . Appl Microbiol Biotechnol 64:787–793 β-galactosidase from Kluyveromyces lactis . Acta Maischberger T, Leitner E, Nitisinprasert S, Juajun O, Crystallogr Sect F F66:297–300 Yamabhai M, Nguyen TH, Haltrich D (2010) Pinho JMR, Passos FML (2011) Solvent extraction of β-Galactosidase from Lactobacillus pentosus : β-galactosidase from Kluyveromyces lactis yields a purification, characterization and formation of stable and highly active enzyme preparation. J Food galactooligosaccharides. Biotechnol J 5:838–847 Biochem 35:323–336 3 On the Enzyme Specifi city for the Synthesis of Prebiotic Galactooligosaccharides 39

Playne MJ, Bennett LE, Smithers GW (2003) Functional Splechtna B, Nguyen TH, Steinbock M, Kulbe KD, dairy foods and ingredients. Aust J Dairy Technol Lorenz W, Haltrich D (2006) Production of prebiotic 58:242–264 galacto-oligosaccharides from lactose using beta- Plou FJ, de Gómez Segura A, Ballesteros A (2007) galactosidases from Lactobacillus reuteri . J Agric Application of glycosidases and transglycosidases for Food Chem 54:4999–5006 the synthesis of oligosaccharides. In: Polaina J, Tanaka Y, Kagamiishi A, Kiuchi A, Horiuchi T (1975) MacCabe AP (eds) Industrial enzymes: structure, Purifi cation and properties of β-galactosidase from function and application. Springer, New York Aspergillus oryzae . J Biochem 77:241–247 Rabiu BA, Jay AJ, Gibson GR, Rastall RA (2001) Toba T, Yokota A, Adachi S (1985) Oligosaccharide struc- Synthesis and fermentation properties of novel tures formed during the hydrolysis of lactose by galacto-oligosaccharides by beta-galactosidases from Aspergillus oryzae β-galactosidase. Food Chem Bifi dobacterium species. Appl Environ Microbiol 16:147–162 67:2526–2530 Torres DP, Goncalves M, Teixeira JA, Rodrigues LR Rastall RA, Gibson GR, Gill HS, Guarner F, Klaenhammer (2010) Galacto-oligosaccharides: production, proper- TR, Pot B, Reid G, Rowland IR, Sanders ME (2005) ties, applications, and signifi cance as prebiotics. Modulation of the microbial ecology of the human Compr Rev Food Sci Food Saf 9:438–454 colon by probiotics, prebiotics and synbiotics to Torres P, Batista-Viera F (2012) Immobilization of enhance human health: an overview of enabling β-galactosidase from Bacillus circulans onto epoxy- science and potential applications. FEMS Microbiol activated acrylic supports. J Mol Catal B: Enzym Ecol 52:145–152 74:230–235 Roberfroid M (2007) Prebiotics: the concept revisited. J Tuohy KM, Rouzaud GCM, Bruck WM, Gibson GR Nutr 137:830S–837S (2005) Modulation of the human gut microfl ora Rodriguez-Alegria ME, Enciso-Rodriguez A, Ortiz-Soto towards improved health using prebiotics – assessment ME, Cassani J, Olvera C, Munguia AL (2010) of effi cacy. Curr Pharm Des 11:75–90 Fructooligosaccharide production by a truncated Urrutia P, Rodriguez-Colinas B, Fernandez-Arrojo L, Leuconostoc citreum inulosucrase mutant. Biocatal Ballesteros AO, Wilson L, Illanes A, Plou FJ (2013) Biotransform 28:51–59 Detailed analysis of galactooligosaccharides synthesis Rodriguez-Colinas B, De Abreu MA, Fernandez-Arrojo with β-galactosidase from Aspergillus oryzae . J Agric L, De Beer R, Poveda A, Jimenez-Barbero J, Haltrich Food Chem 61:1081–1087 D, Ballesteros AO, Fernandez-Lobato M, Plou FJ Usui T, Kubota S, Ohi H (1993) A convenient synthesis of (2011) Production of galacto-oligosaccharides by the β-D -galactosyl disaccharide derivatives using the β-d - β-galactosidase from Kluyveromyces lactis: compara- galactosidase from Bacillus circulans. Carbohydr Res tive analysis of permeabilized cells versus soluble 244:315–323 enzyme. J Agric Food Chem 59:10477–10484 Vera C, Guerrero C, Conejeros R, Illanes A (2012) Shadid R, Haarman M, Knol J, Theis W, Beermann C, Synthesis of galacto-oligosaccharides by Rjosk-Dendorfer D, Schendel DJ, Koletzko BV, β-galactosidase from Aspergillus oryzae using par- Krauss-Etschmann S (2007) Effects of galactooligo- tially dissolved and supersaturated solution of lactose. saccharide and long-chain fructooligosaccharide Enzyme Microb Technol 50:188–194 supplementation during pregnancy on maternal and Wei L, Xiaoli X, Shufen T, Bing H, Lin T, Yi S, Hong Y, neonatal microbiota and immunity – a randomized, Xiaoxiong Z (2009) Effective enzymatic synthesis of double-blind, placebo-controlled study. Am J Clin lactosucrose and its analogues by β-galactosidase Nutr 86:1426–1437 from Bacillus circulans . J Agric Food Chem Sheu DC, Li SY, Duan KJ, Chen CW (1998) Production 57:3927–3933 of galactooligosaccharides by beta-galactosidase Yanahira S, Kobayashi T, Suguri T, Nakakoshi M, Miura immobilized on glutaraldehyde-treated chitosan S, Ishikawa H, Nakajima I (1995) Formation of oligo- beads. Biotechnol Tech 12:273–276 saccharides from lactose by Bacillus circulans Siso MIG, Cerdán E, Picos MAF, Ramil E, Belmonte ER, β-galactosidase. Biosci Biotechnol Biochem Torres AR (1992) Permeabilization of Kluyveromyces 59:1021–1026 lactis cells for milk whey saccharification: a com- Yang ST, Marchio JL, Yen JW (1994) A dynamic light parison of different treatments. Biotechnol Tech scattering study of β-galactosidase: environmental 6:289–292 effects on protein conformation and enzyme activity. Song J, Abe K, Imanaka H, Imamura K, Minoda M, Biotechnol Prog 10:525–531 Yamaguchi S, Nakanishi K (2011a) Causes of the Zhou QZ, Chen XD, Li X (2003) Kinetics of lactose production of multiple forms of β-galactosidase by hydrolysis by β-galactosidase of Kluyveromyces lactis Bacillus circulans. Biosci Biotechnol Biochem immobilized on cotton fabric. Biotechnol Bioeng 75:268–278 81:127–133 Song J, Imanaka H, Imamura K, Minoda M, Katase T, Zhou QZK, Chen XD (2001) Effects of temperature and Hoshi Y, Yamaguchi S, Nakanishi K (2011b) Cloning pH on the catalytic activity of the immobilized and expression of a β-galactosidase gene of Bacillus β-galactosidase from Kluyveromyces lactis . Biochem circulans . Biosci Biotechnol Biochem 75:1194–1197 Eng J 9:33–40 Microbial Mannanases: Properties and Applications 4

Hemant Soni and Naveen Kango

Abstract

Mannans are a major constituent of the hemicellulose fraction of lignocelluloses. Mannans perform distinct functions as structural components in cell walls of softwoods and storage functions in seeds. Enzymatic hydrolysis of mannan involves the backbone hydrolyzing endo-β-mannanases and β-mannosidases. Mannans are heteropolymeric and their hydrolysis also requires the action of β-glucosidases and side-chain cleaving α-galactosidases and acetyl mannan esterases. Microorganisms are therefore explored for the production of such repertoire of enzymes so that effective mannan hydrolysis can be achieved. The present chapter discusses the occurrence and structural properties of mannans in plant materials and its hydrolysis using enzymes sourced from various fungi and other microorganisms. The production and properties of mannanolytic enzymes, their cloning and expression in heterologous hosts, and their application have also been discussed .

Keywords Hemicellulose • Mannans • β -Mannanase • β -Mannosidase • Locust bean gum

Introduction the lignocellulosic biomass. Lignocellulose is abundant and represents one of the major natural Hemicelluloses are structural polysaccharides of renewable resources and a dominating waste the plant cell wall. Hemicellulose is associated material from agriculture. This renewable resource with cellulose and lignin and forms about 30 % of can be used in several industries, including the pharmaceutical, biofuel, and pulp and paper industries, and many more (Kango et al. 2003 ; Kango H. Soni • N. Kango (*) 2007). The generation of feedstock is possible Department of Applied Microbiology by hydrolysis of lignocellulosic biomass using and Biotechnology , Dr. Hari Singh Gour Vishwavidyalaya , Sagar 470003 , MP , India various microorganisms and their enzymes. e-mail: [email protected] The hydrolysis of lignocellulose has become the

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 41 DOI 10.1007/978-81-322-1094-8_4, © Springer India 2013 42 H. Soni and N. Kango

“hot spot” and a crucial part of lignocellulose bio- a technology. According to Chaikumpollert et al. (2004 ), hemicelluloses form about one third of all b the components available in plants and are the second major heteropolymer present in nature. Distribution of hemicellulose in gymnosperms c and angiosperms varies. Hemicelluloses consist of different heterogeneous polymers of sugars such d as xylose, arabinose, mannose, glucose, galactose, and sugar acids. These hemicelluloses are named Fig. 4.1 Structure of different types of mannans found according to their main sugar component (80– in nature: (a ) Linear mannan (b ) Glucomannan (c ) 90 %) which is present in its backbone, e.g., man- Galactomannan (d ) Galactoglucomannan. Mannose nose is present in mannan hemicelluloses. Glucose Galactose Enzyme-based hydrolysis of hemicelluloses, especially mannan and xylan, signifi cantly affects the glucomannan, the main chain consists of ran- prospects of biobleaching and saccharifi cation of domly β-1,4-linked D-mannose and D-glucose lignocellulosic biomass (Viikari et al. 1993 ). residues, while in galactomannan, the galactose sugars are present as single side chains substituted on the main-chain sugar, mannose (Fig. 4.1 ). Mannan: Occurrence and Structure In galactoglucomannan, galactose sugars are present as single side chains in α-1,6-linkage Most of the main-chain sugars in hemicellulose with the main chain, which consists of both man- are linked together by β-1,4-glycosidic bonds. nose and glucose. Mannans are one of the most important constitu- The ratio of sugars present in mannan varies ents of hemicelluloses in the wall of higher with respect to the different sources from which plants. Mannan is composed of repeating units it is obtained, which indicates the polydiversity of mannose (a second carbon epimer of glucose) of the polymer. In linear mannans, mannose is linked by β-1,4-glycosidic linkages. Besides predominantly present, while in galactomannan D-mannose, other sugars like glucose, galactose, the ratio of galactose to mannose is 1:3. True and acetyl groups can be present in various man- galactomannan is considered to contain more nans. Mannans are further classifi ed on the basis than 5 % of galactose residues in side chains. In of other sugars present in the structure, e.g., galactoglucomannan, the ratio of galactose/glu- glucose-containing mannan is called glucoman- cose/mannose is observed to be 1:1:3. Mannans nan. Similarly, when galactose is present as a are found in nature as part of the hemicellulose side chain linked to the main chain, the polymer fraction of hardwood and softwood. It is predom- is called galactomannan, and when both glucose inantly found in the endosperm of copra, locust and galactose are linked to the mannose sugars bean, guar beans, seeds of other leguminous of the main-chain backbone, it is called galacto- plants, coffee beans, roots of the konjac tree, and glucomannan. Mannan exists in both linear and ivory nuts. Locust bean gum (LBG) is a galacto- branched forms with a β-1,4-linked backbone. mannan, while mannan from konjac trees is a Mannans are an important part of the hemicel- glucomannan. Linear mannans are the major lulose family, which are further classifi ed as structural units in woods, in seeds of ivory nut, linear mannan, glucomannan, galactomannan, and in green coffee beans. Petkowicz et al. (2007 ) and galactoglucomannan. Linear mannans are separated mannans from ivory nuts into two com- homopolysaccharides which have a main chain ponents, mannan I and mannan II. Mannan I, composed of 1,4-linked β- d-mannopyranosyl extracted with alkali, displayed a crystalline (mannose-mannose) residues. The percentage of structure, while mannan II was not amenable to galactose in linear mannans is 5 % or less. In direct extraction and displayed a less crystalline 4 Microbial Mannanases: Properties and Applications 43

Table 4.1 Mannan content of some plants Source of mannan Plant part Type of mannan Ratio of sugars Ceratonia siliqua Endosperm of seed Galactomannan ~1:4 (Gal:Man) (carob or locust bean) Phytelephas macrocarpa Endosperm of seed Linear mannan Homopolymer (mannose) (ivory nut) Schizolobium amazonicum (a) Seed coat side Linear mannan Homopolymer (mannose) or exterior section (b) An intermediate section Rich in galactomannan ~1:3 (Gal:Man) of seed endosperm Schizolobium parahybum Endosperm of seed Galactomannan ~1:3 (Gal:Man) Carum carvi Endosperm of seed Linear mannan Homopolymer (mannose) Cyamopsis tetragonolobus Endosperm of seed Galactomannan ~1:2 (Gal:Man) (guar seed) Amorphophallus konjac Roots Glucomannan ~3:4 (Glu:Man) Coffea arabica (coffee bean) Endosperm Galactomannan ~1:2 and ~1:7 (Gal:Man)a Aloe barbadensis (acemannan) Leaves Linear mannanb Homopolymer (mannose) Cesalpinia spinosa (tara tree) Endosperm of seed Galactomannan ~1:3 (Gal:Man) a Developmentally regulated b Associated with acetyl group structure. Molecular size also varied as mannan I are crucial for seed germination, remain active. was smaller compared to mannan II. Linear man- Liepman et al. ( 2007 ) have showed some evi- nans are present in the seed coat or exterior sec- dence that mannan also functions as a signaling tion of leguminous plants, while galactomannan molecule in plant growth and development. occurs in the intermediate section of the seed Three-dimensional structure studies of guaran or endosperm. Various gum extracts from plants are guar fi bers (Cyamopsis tetragonolobus ) were conspicuous sources for galactomannan, for done by Chandrasekaran et al. (1998 ) using x-ray example, locust bean gum, tara gum, fenugreek diffraction, and they revealed that the hydrogen gum, and guar gum. In these sources, the main of the galactosyl side chain interacted with the chain of galactomannan contains 1,4-linked β-d- mannan backbone and provided structural stabil- mannopyranosyl residues with side chains of ity. The structure showed a fl at twofold helix with single 1,6-linked α-d -galactopyranosyl groups a pitch of 10.38 Å. Glucomannans are the princi- attached along the main chain (Fig. 4.1 ). The dis- pal components of softwood hemicelluloses and tribution of galactose in galactomannan varies consist of β-1,4-linked D-mannose and D-glucose between mannans obtained from different residues with a 3:1 ratio. Hongshu et al. ( 2002 ) sources. It is observed that all types of galacto- obtained glucomannan from ramie (Boehmeria mannans have more than 5 % of galactose resi- nivea ) which contained 95–99 % of D-glucose dues as side chains. Galactomannan obtained and D-mannose residues with a ratio of 1.3–1.7:1. from the endosperm of locust bean or Ceratonia The presence of D-galactose residues in gluco- siliqua (carob) has a ratio of 1:4 (galactose/man- mannan is very rare, but Puls and Schuseil (1993 ), nose). The galactomannan of the intermediate working with softwood, observed D-galactose section of the seed endosperm of Schizolobium residues attached to the main-chain mannose amazonicum and Cesalpinia spinosa has a sugar residues with α-1,6-linked terminal units and ratio of 1:3 (galactose/mannose) (Table 4.1 ). The observed a ratio of mannose/glucose/galactose as function of galactomannan in seeds, in addition 3:1:0.1. The glucomannan from Amorphophallus to the retention of water by solvation, is to pre- (konjac) showed an association with starch- vent complete drying of seeds in high atmo- like α-glucan, comprised of 1,4-linked β-d- spheric temperatures so that the enzymes, which mannopyranose and D-glucopyranose in 70 % 44 H. Soni and N. Kango and 30 %, respectively (Aspinall 1959 ). Kenne dues with a difference in the side-chain pattern et al. (1975 ) studied distribution of the O -acetyl at O-2 and O-4 instead of O-6, which is observed groups in glucomannan from pine and observed in various galactomannan structures. Singh and that acetyl groups are irregularly distributed in Malviya (2006 ) observed D-glucopyranosyl units pine glucomannan. In galactoglucomannans, in glucomannan from seeds of a medicinal plant, galactose residues are attached to both D-glucosyl Bryonia laciniosa , which displayed α-1-6- linkages and D-mannosyl units with a α-1,6-linkage, and in the main chain with a 1:1.01 ratio of glucose and mannosyl units also have partial substitution by mannose. O -acetyl groups. Several reports showed that The degree of polymerization (DP) of any about 60–65 % of mannose residues in galacto- macromolecule is a manifestation of the approx- glucomannan from native Norway spruce wood imate number of monomer units present in poly- and pulp were acetylated at either the C-2 or C-3 mer. The DP of any polymer infl uences its position (Aspinall et al. 1962 ; Popa and Spiridon various properties such as colligative properties, 1998 ; Timell 1967 ; Willfor et al. 2003 ). Lundqvist boiling point, freezing point, solubility, viscos- et al. ( 2002 , 2003 ) extracted galactoglucomannan ity, toughness, and somatic pressure. The DP from spruce (Picea abies ) by heat fractionation also helps in calculating the average molecular at different temperatures and characterized it. weight of the polymer. Petkowicz et al. (2007 ) Galactoglucomannan from spruce contained isolated mannan from ivory nuts and observed about one third of D-mannosyl units substituted two types, viz., mannan I and II in which man- by O-acetyl groups with an equal distribution nan I has a lower molecular weight and a DP of between C-2 and C-3 and a molar ratio of 0.1:1:4 ~15, while mannan II had a DP of about ~80 (galactose/glucose/mannose). Various types of with higher molecular weight. Softwood glucomannans with their sources and sugar ratios are mannans with a 3:1 ratio of mannose:glucose listed in Table 4.1 . exhibit higher DPs of more than 200. Softwood Solubility among mannans towards water varies galactoglucomannans with a 1:1:3 ratio for due to the presence of D-galactose side chains. The galactose/glucose/mannose exhibit a DP solubility of galactomannan and galactoglucoman- between 100 and 150. Enzymes play a crucial nan is higher in comparison to linear and glu- role in the modifi cation of polymers and its comannan homopolymers. The D-galactose side structural analysis. Analysis or sequencing of chains prevent alignment of macromolecules and mannan requires several mannanases from lead to formation of strong hydrogen bonds (Timell legume seeds and microorganisms to act on the 1965 ). In addition to aforesaid structures, mannans various mannans. Selection of enzymes is also display a range of curious structures and important because their differential activity confi gurations. For instance, Ishurd et al. (2004 ) towards substrates reveals the structural differ- observed and isolated galactomannan from Retama ence. A mannanase from Trichoderma reesei raetam (Fabaceae). Its backbone consisted mostly was able to hydrolyze fi ber-bound galactogluco- of 1-3-linked β-d -mannopyranosyl residues with mannan from pine kraft pulp, while an enzyme attachment of galactopyranosyl residues observed from Bacillus subtilis was not effective for its at C-6. Nunes et al. (2005 ) observed arabinosyl and hydrolysis (Ratto et al. 1993 ). Tenkanen et al. glucosyl residues in galactomannan from green ( 1997 ) studied the action of a mannanase from and roasted coffee infusions. The acetyl groups T. reesei on galactoglucomannan in pine kraft were present in the main chain of mannan at the pulp and analyzed the hydrolysate by O-2 position of mannose residues, while arabinose 1H NMR spectroscopy and high-performance residues were at O-6 of mannose residues as side anion-exchange chromatography (HPAEC-PAD). chains. Omarsdottir et al. (2006 ) isolated galacto- The relative amount of sugar residues in the mannan from a number of lichen species like foli- hydrolysate of pine kraft pulp, after extensive ose lichen ( Peltigera canina). The backbone of hydrolysis by a mannanase from T. reesei, was these mannans displayed odd structures, being analyzed. The molar percentage of mannose, glu- composed of α-1,6-linked mannopyranosyl resi- cose, and galactose was 73.4, 20.4, and 5.8, 4 Microbial Mannanases: Properties and Applications 45

Table 4.2 Some microbial sources of mannanases Microorganisms Major groups Genus Species References Fungi Malbranchea M. cinnamomea Maijala et al. (2012 ) Myceliophthora M. fergusii Maijala et al. (2012 ) Aspergillus A. niger Benech et al. (2007 ) A. fumigatus Puchart et al. (2004 ) A. aculeatus Setati et al. (2001 ) A. niger Ademark et al. (1998 ) Trichoderma T. harzianum Ferreira and Filho (2004 ) T. reesei Stalbrand et al. (1993 ) Penicillium P. oxalicum Kurakake et al. (2006 ) P. citrinum Yoshida et al. (1993 ) Sclerotium S. coffeicola Groβwidnhager et al. (1999 ) S. rolfsii Gubitz et al. (1996 ) Yeast Saccharomyces S. cerevisiae Setati et al. (2001 ) Candida C. albicans Reyna et al. (1999 ) Bacteria Bacillus B. subtilis WY34 Jiang et al. (2006 ) Enterococcus E. casseliﬂ avus Oda et al. (1993 ) Actinomycetes Streptomyces Streptomyces sp. S27 Shi et al. ( 2011 ) S. galbus NR Kansoh and Nagieb (2004 ) S. lividans Arcand et al. (1993 ) Thermomonospora T. fusca Hilge et al. (1998 ) Cellulomonas Cellulomonas sp. Takegawa et al. ( 1989 )

respectively, as determined by 1 H NMR spectros- called as mannanases. Complete biodegradation of copy, and 71.8, 20.3, and 6.9 by the HPAEC-PAD mannans necessitates the use of various enzymes. method. LBG hydrolysis products of the recombi- Enzymes that actively participate in mannan nant man5S27 enzyme were analyzed using hydrolysis include β-mannanase (1,4-β-D-mannan HPAEC. The approximate percentages of man- mannohydrolase, EC 3.2.1.78), β-mannosidase nose, mannobiose, mannotriose, mannotetraose, (1,4-β-d -mannopyranoside hydrolase, EC 3.2.1.25), mannopentaose, and other sugar oligosaccharides β-glucosidase (1,4-β-D- glucoside glucohydrolase, were 3.23, 0.74, 22.14, 2.21, 6.89, and 64.79 (Shi EC 3.2.1.21), α-galactosidase (1,4-α-d -galactoside et al. 2011 ). Analysis of the sequence and percent- galactohydrolase, EC 3.2.1.22), and acetyl esterase age of sugars in mannan requires a particular (EC 3.1.1.6). A number of microorganisms, includ- enzyme or enzymes with their homo- and hetero- ing fungi, yeast, bacteria, and actinomycetes, pro- synergistic actions leading to the hydrolysis of duce β-mannanases and other accessory enzymes. hemicelluloses. The analysis of hydrolysate needs Among these, fungi have been investigated by vari- a suitable method or a combination of methods ous workers for mannanase production (Dhawan like TLC, HPLC, and NMR spectroscopy. and Kaur 2007 ; Moreira and Filho 2008 ; Van Zyl et al. 2010 ). Some of the prominent mannanase producers are listed in Table 4.2 . Mannan-Degrading Enzymes β-Mannanase distribution is also observed in and Their Sources plants. Seeds are the most preferred sources for isolation of mannanases. However, other plant Mannans can be present as linear and branched, organs like fruits also displayed the presence of homo- as well as heteropolymers. In general, mannanases (Bourgault et al. 2001 ). Schroder enzymes involved in the hydrolysis of mannan are et al. ( 2006) obtained endo-β-mannanase from 46 H. Soni and N. Kango ripe tomato fruit. For mannanase production, Mannan composition also affects the action of microorganisms which are selected from various enzymes, and to achieve complete degradation of sources (soil, compost, water, agriculture waste) heteromannan like locust bean gum, fungi and are grown on basal media containing mannan bacteria have to produce three enzymes, namely, (LBG) as sole carbon source (Ratto and Poutanen β-mannanase, β-mannosidase, and α-galactosidase 1988 ; Puchart et al. 2004 ; Maijala et al. 2012 ). (Hilge et al. 1998 ). These hemicellulases also show synergistic action. When β-mannanase and β-mannosidase (main-chain cleaving enzymes) Mode of Action of Mannanases or α-galactosidase and acetyl mannan esterase (side-chain cleaving enzymes) cooperate, it is Polysaccharides like mannans can exist in linear, called homosynergistic action. Heterosynergy homo, hetero, or branched form. β-Mannanases refers to the interaction of main- and side-chain ﬁ nd application in the extraction of vegetable oil, cleaving enzymes working together (Fig. 4.3 ). in the manufacture of instant coffee as a viscosity Homosynergy between β-mannosidase and two reducer agent for coffee extract, nutraceuticals β-mannanases obtained from the enzyme extract such as the production of MOS (mannose oli- of Sclerotium rolfsii (Gubitz et al. 1996 ; Moreira gosaccharides), pharmaceuticals, food and and Filho 2008 ) and heterosynergy between feed, production of second-generation biofuels, β-mannanase, β-mannosidase, and α-galactosidase paper and pulp, and various other industries have been observed in enzyme extractions of (Sachslehner et al. 2000 ; Van Zyl et al. 2010 ). At Thermotoga neapolitana 68 on galactomannan least one main-chain hydrolyzing enzyme, like (Duffaud et al. 1997 ). β-mannanase, and one side-chain hydrolyzing enzyme, like α-galactosidase, are required for the a breakdown of branched mannan (LBG). Endo b-Mannanase β-Mannanase cleaves internal β-1,4-linked residues of mannose/glucose in the mannan back- Exo b-Mannosidase bone. This enzyme mainly produces oligomannan/ oligoglucomannan and is very effective on linear mannan and glucomannan (homopoly- Mannose Mannobiose mer), although the hydrolysis action of this b-Glucosidase a-Galactosidase enzyme is affected in galactomannan due to the b presence of side chains. β-Mannosidase helps to remove mannose from the nonreducing end of mannan and cleaves β-1,4-linked mannose resi- Fig. 4.2 (a ) Action of β-mannanase and β-mannosidase dues. Similarly, β-glucosidase removes glucose on linear mannan. ( b ) Action of debranching enzyme α-galactosidase and β-glucosidase on galactoglucoman- residue from the nonreducing end of the oligo- nan. Mannose Glucose Galactose glucomannan and cleaves 1,4-β-d -glucopyranose. Besides these main-chain enzymes, two side- b b b chain cleaving enzymes are very important for -Glucosidase -Mannanase -Mannosidase complete biodegradation of mannan, viz., a α-galactosidase cleaves the α-1,6 glycosidic b a-Galactosidase bonds between galactose and the main-chain sug- -Mannanase b-Mannosidase ars (mannose/glucose) and leads to hydrolysis of b D-galactopyranosyl side chains of galactomannan and galactoglucomannan. Acetyl mannan Fig. 4.3 ( a) Homosynergetic actions of β-mannanase, β β esterase removes acetyl groups from galactoglu- -mannosidase, and -glucosidase on glucomannan. ( b ) Heterosynergetic action of β-mannanase and comannan. The delineation of various man- β-galactosidase on galactoglucomannan. Mannose nanases action is shown in Fig. 4.2 . Glucose Galactose 4 Microbial Mannanases: Properties and Applications 47

Microbial Production of Mannanases process gives a good outcome in terms of enzyme activity at optimum temperature, pH, and other The best and richest sources of enzymes are factors. The effect of the agitation speed, aeration microorganisms (Kirk et al. 2002 ). For man- rate, and temperature on the production of nanase production, mainly fungi and some bacte- β-mannanase by Bacillus licheniformis NK 27 in ria are used at a commercial level and their a batch fermenter was studied by Feng et al. enzyme systems are reported to be inducible. (2003 ). They concluded that temperature was the Hemicelluloses like xylan are not able to cross most signifi cant factor in β-mannanase produc- cell walls; therefore, small oligosaccharides tion. Feng et al. ( 2003) obtained a maximum formed as a result of xylan degradation act as an activity of 212 U/ml in 36 h at an aeration rate of inducer and also play an important role as a regu- 0.75 vvm, agitation of 600 rpm, and a constant lation factor for xylanase biosynthesis (Singh temperature of 30 °C. Mannanase production by et al. 2003 ). Both submerged and solid-state fer- microorganisms is infl uenced by the media com- mentation (SSF) have been examined for man- position, mostly carbon and nitrogen (Kataoka nanase production. The cost of enzymes remains and Tokiwa 1998 ; Dhawan and Kaur 2007 ). a bottleneck in realizing their application on a Groβwindhager et al. ( 1999 ) used glucose and large scale. Use of inexpensive substrates, such cellulose for S. rolfsii , while Ademark et al. as by-products of agro-industries and forestry (1998 ) and Gomes et al. (2007 ) used locust bean waste, can effectively subsidize the recurring cost gum (LBG) for A. niger and a thermophilic fun- of enzyme production. Ratto and Poutanen gus, Thermoascus aurantiacus. Ferreira and (1988 ) have used wheat bran with locust bean Filho (2004 ) have used wheat bran as the carbon galactomannan for mannanase production by source for the production of β-mannanase from bacteria and fungi. Mannanase activities were mesophilic fungus Trichoderma harzianum strain found to be 256, 34, and 24 nkat ml −1 with T4. Besides the carbon source, various organic or Bacillus subtilis , Aspergillus awamori , and inorganic nitrogen sources play an important role T. reesei , respectively. Abdeshahian et al. ( 2010 ) in mannanase production. Organic nitrogen used palm kernel cake (PKC) as a substrate for sources like peptone, yeast autolysate, corn steep β-mannanase production by Aspergillus niger liquor (CSL), and beef extract are preferred FTCC5003 through solid-state fermentation. (Puchart et al. 2004 ; Zhang et al . 2006; Cui et al . Production was evaluated by response surface 1999 ; Kataoka and Tokiwa 1998 ), while inor- methodology on the basis of a central composite ganic nitrogen sources like ammonium sulfate, face-centered (CCF) design with three indepen- diammonium hydrogen phosphate, ammonium dent variables, namely, incubation, temperature, dihydrogen phosphate, and sodium nitrate have initial moisture content of substrate, and air- been found to play an effective role (Zakaria flow rate. The highest level of β-mannanase et al . 1998; Perret et al. 2004). Gomes et al. ( 2007 ) (2,117.89 U/g) was obtained when the incubation achieved the highest β-mannanase and temperature, initial moisture level, and aeration β-mannosidase activity by the thermophilic fun- rate were 32.5 °C, 60 %, and 0.5 l/min, respec- gus (Thermoascus aurantiacus) with soya meal tively, during SSF. There are many species of as nitrogen source, supplemented with LBG as fungi reported to produce signifi cantly high man- carbon source. Recently, Mohamad et al. ( 2011 ) nanase activity. For instance, Lin and Chen performed a comparison study of different car- ( 2004) observed 27.4 U/ml mannanase activity in bon and nitrogen sources for their effect on a submerged culture of Aspergillus niger NCH mannan-degrading enzyme production by 189. Similarly, Hossain et al. (1996 ) obtained Aspergillus niger . They revealed in their result about 90 U/ml mannanase activity in submerged that guar gum (GG) and bacteriological peptone conditions using Bacillus sp. KK01. Production supported the highest β-mannanase activity. of enzymes is affected by temperature, pH, They achieved β-mannanase activities equiva- agitation, and aeration. The overall production lent to 1,495, 1,148, 10.7, 8.8, and 4.6 nkat ml −1 48 H. Soni and N. Kango with guar gum (GG), LBG, α-cellulose, glucose, production started only when the glucose con- and carboxymethyl cellulose as carbon sources, centration in the medium dropped low. High respectively. Activity levels equivalent to 1,744, mannanase activity (240 Uml−1 ) by S. rolfsii 1,168, 817, 241, 113, and 99 nkat ml −1 were CB5191.62 was achieved in a glucose fed-batch achieved with bacteriological peptone, yeast system in which glucose concentration in the extract, ammonium sulfate, ammonium nitrate, media was maintained low (Groβwindhager et al. and ammonium chloride as nitrogen sources, 1999 ). Table 4.3 displays an overview of produc- respectively. The above results showed that man- tion and properties of mannanases from various nanase production by A. niger can be enhanced microorganisms. with GG and LBG. Inorganic nitrogen sources reduced β-mannanase production greatly, while organic nitrogen sources enhanced β-mannanase Heterologous Production production. In contrast, Kalogeris et al. (2003 ) have obtained better production of cellulases by Higher yield, ease of operational conditions, sim- Thermoascus aurantiacus using inorganic nitro- ple recovery, and downstream processing have gen sources. Various industries like paper and prompted several workers towards cloning and pulp and detergent industries need enzymes that heterologous production of mannanases. The function well at a high pH. Alkaline β-mannanase recombinant DNA technique provides enormous was obtained for the fi rst time from alkaliphilic opportunity to make genetically modifi ed micro- Bacillus sp. AM001 by Akino et al. (1987 ). This bial strains. More than 50 % of mannanase- mannanase showed a pH optimum between 7.0 producing microorganisms, which are being used and 9.0. Mudau and Setati (2006 ) have studied at industrial level, are genetically engineered endo-mannanase-producing molds from hypersa- (Dhawan and Kaur 2007 ). line environments and observed the effect of salt S. cerevisiae is not known for production of (NaCl) on growth and enzyme production. All mannanase by itself, but the heterologous pro- four isolates, Scopulariopsis brevicaulis LMK002, duction of endo-β-1,4-mannanase has been done S. candida LMK004, S. candida LMK008, and using S. cerevisiae as a genetically modifi ed host Verticillium dahliae LMK006, showed growth on by Setati et al . (2001 ). Similarly, Qiao et al. NaCl concentrations of up to 10 %. Endo- (2008 ) have used Pichia pastoris as a host for mannanase production by Scopulariopsis isolates expression of MAN gene of Bacillus subtilis . It was found to increase with NaCl concentration. has been observed that, if the same gene encod- Groβwindhager et al. (1999 ) have shown effi cient ing mannanase is expressed in different hosts, the β-mannanase production by Sclerotium rolfsii and resultant recombinant enzymes show somewhat S. coffeicola under derepressed condition by using different properties. For instance, MAN1 gene of cellulose- and glucose-based media. They have Aspergillus aculeatus MRC 11624 was cloned concluded that cellulose is the best inducer for and expressed in S. cerevisiae , A. niger , and both S. rolfsii and S. coffeicola strains for man- Y. lipolytica. Besides higher enzyme activity, the nanase production with maximum activities of 677 resultant recombinant enzymes showed different and 461 Uml −1, respectively. In a glucose-based temperature and pH optima as compared to the medium, activities were 96.6 and 67.7 Uml−1 . native enzymes. Isolation and cloning of genes Glucose is an easily metabolizable substrate, and encoding mannanases and their expression in a in the presence of this substrate, glycosyl hydro- suitable host play an important role in the molec- lase systems get repressed (Ronne 1995 ; Ruijter ular and structural studies of enzyme proteins and Visser 1997 ). However, both the strains S. rolf- and protein engineering thereof. Eight essentially sii and S. coffeicola were observed to produce conserved active site residues of β-mannanases, mannanase activity when a typical repressing sub- viz., Arg-83, His-119, Asn-157, Glu-158, His- strate, glucose, was used as sole carbon source in 224, Tyr-226, Gly-254, and Trp-283, are reported batch cultivation. Mannan-degrading enzyme in Bacillus N16-5 mannanase (Ma et al . 2004 ). 4 Microbial Mannanases: Properties and Applications 49 ) ) ) ) ) ) 2001 ) 1999 1999 1999 1988 1988 ) ) 1990 ) ) 2004 2004 1990 2006 Reference Ratto and Poutanen ( Großwindhager et al. ( Großwindhager et al. ( Großwindhager et al. ( Ratto and Poutanen ( Araujo and Ward ( Araujo and Ward Kurakake and Komaki ( and Komaki Kurakake M2-30 M3-55 M4-89 Substrate Mr (kDa) Incubation Rev/min PI Optimum pH T °C 1,105 U/ml – 50 Konjac powder 39.6 4 days 200 – Jiang et al. ( Production 28 U/ml 40 U/ml 5 – 30 45 Defatted copra Glucose/LBG – 60/63 75 h 7 days 0.60/0.66 IU/ml 120 200 – – 28 4.9/4.7 Puchart et al. LBG/Avicel ( Lin and Chen ( – 6–8 days 100 6.5 Torrie et al. ( E 58 Man I/Man II wheat bran wheat bran CBS 667.85

WB CBS147082 CBS 151.31 WY 34 Production and properties of mannanases from various microorganisms microorganisms Production and properties of mannanases from various NCH-189 -Mannosidase 0.001 U/ml 5 45 LBG 72 48 200 d coffee waste, waste, coffee -Mannanase -Mannosidase 50 U/g 1.4 U/g – – 30 30 CW + WB CW + WB - 88 h 88 h β β Mannanase Mannanase Mannanase Mannanase Mannanase Mannanase 209 U/ml 8.2 U/ml Mannanase 348 5 U/ml 95.8 U/ml 5 30 30 461 5 U/ml β Cellulose 5 67.7 U/ml Glucose 30 Mannanase 30 5 Cellulose 5 6.7 U/ml Mannanase Glucose 30 30 Cellulose 13 5 Glucose days 13 days 45 150 34 nkat/ml 13 days LBG 150 13 days 24 nkat/ml 150 5.3 13 days 150 28 13 days 5.3 M1-52 150 WB + LBG 28 150 48 WB + LBG 200 Bacillus subtilis

Microorganism Microorganism Table Table 4.3

A. niger fumigatus Aspergillus rolfsii Sclerotium rolfsii Sclerotium coffeicola Sclerotium harzianum Trichoderma awamori Aspergillus reesei Trichoderma CW Thielavia terrestris Aspergillus awamori Aspergillus 50 H. Soni and N. Kango

A 1,345 bp gene encoding mannanase (ManN) production of mannanases, vectors employed, from Aspergillus sulphureus was expressed in and the properties of recombinant enzyme are Pichia pastoris (Chen et al. 2007 ). Alkaline presented in Table 4.4 . Heterologous expression β-mannanase (ManA) was cloned by Ma et al. allows a simpler and cheaper means of produc- (2004 ) from Bacillus sp. N165, and its overpro- tion using desired hosts, while induction of duction and optimization have been studied by β-mannanases by native organisms needs mostly Lin et al. (2007 ). They achieved a maximum expensive and complex medium components yield of 310 U/ml after optimization. Recently, (Kote et al . 2009; Viniegra et al . 2003 ; Van Zyl Pan et al . ( 2011) demonstrated heterologous et al . 2010 ). expression of alkaline β-mannanase by a yeast expression system. Pan et al. (2011 ) have used Kluyveromyces cicerisporus Y179U and Pichia Applications of Mannanases pastoris GS115 for expression of MAN 330 (truncated β-mannanase) and MAN 493. Mannan-degrading enzymes fi nd various uses in MAN330 and MAN 493 genes were amplifi ed different industries. More recently, mannanases and alkaline mannanase was successfully have been used for the production of manno-oli- expressed using Y179U/pUKD-S-MAN 330 and gosaccharides (MOS), feed upgradation, bio- GS115/pPIC-9 k MAN 493 (vectors), and high bleaching, and detergents. The details of these yields of 1,378 and 1,114 U/ml in shake fl asks applications are detailed below. were obtained, respectively. Both enzymes had a maximum activity at pH 9.5 and 70 °C. β-Mannanase from Bacillus subtilis has been Production of Manno- purifi ed and characterized (Jiang et al. 2006 ). Oligosaccharide Recently, a thermostable β-mannanase from Bacillus subtilis BCC41051 was expressed in Mannanases degrade mannans and produce E. coli and Bacillus megaterium (Summpunn manno-oligosaccharides (MOS) and mannose. et al . 2011 ). The MOS contribute to human health and are The open reading frame of the gene coding considered to confer prebiotic benefi ts. Fan et al. β-mannanase was amplifi ed by PCR using Man- (2009a , b) have showed that glucomannan enhanced CHF (5′-GTACGCCATATGTTTAAGAAACATAC fecal probiotics. Hydrolyzed glucomannan can GATCTCTTTGC-3′) and Man-CHR (5′-GTACG be used as a prebiotic to augment growth of fecal CCTCGAGTTCAACGATTGGCGTT probiotics. MOS is demonstrated to confer a sim- AAAGAATC-3′) primers, and the recombinant ilar result as oligofructose prebiotics. Kobayashi vector pEManAhis was transfected into E. coli et al. (1987 ) have noticed that oligosaccharides, BL21. The gene was expressed and induced by which are used as prebiotics to enhance growth IPTG (isopropyl-β- d-1thiogalactopyranoside), of human intestinal microfl ora, including manno- and the highest activity of 415.18 U/ml was oligosaccharide. MOS also used as functional obtained. For expression in B . megaterium , E. coli , food ingredients. Bacillus shuttle, and expression vector, Pxb was used. The gene coding for β-mannanase was amplifi ed with the primer Man-F1 (5′-GTACGCG Coffee and Coconut Oil Extraction GATCCGACAAATGTTTAAGAAACATA CGATC-3′) and Man-R1 (5′-CTGATTCATT In coffee extract, mannan is present as the main CAACGATTGG-3′) and transformed into polysaccharide and this mannan increases the B. megaterium with the help of a pXManA plas- viscosity of the coffee extract, which unfavorably mid. The expression of the cloned gene was affects instant coffee preparation. β-Mannanase induced by xylose to obtain 359 U/ml enzyme is used for reducing the viscosity of coffee activity. Various examples of heterologous extract, because it hydrolyzes mannan into simple Table 4.4 Heterologous production and properties of expressed mannanase Optima for Stability for Vmax Activity Microorganism Promoter/ enzymes pH enzyme μmolmin−1 (fl ask) −1 −1 (gene source) Host for cloning Host for expression terminator Vector/plasmid Temp °c pH Temp °c KM mg/ml mg nkatml Reference B. subtilis – E. coli BL21 (DE3) – pET24b(+) – – – – ~6,930 Summpunn BCC41051 B. megaterium – pXb – – – – ~5,995 et al. (2011 ) Bacillus sp. N16-5 – Kluyveromyces – pUKD-S- MAN330 9.5 70 5–11 60–70 ~23,012 Pan et al. (MAN330) cicerisporus (2011 ) Bacillus sp. N16-5 _ Pichia pastoris – pPIC-9 K-MAN493 9.5 70 5–11 60–70 ~18,603 (MAN493) Streptomyces sp. E. coli JM109 E. coli BL21 (DE3) – pET-30a(+) 7 65 5–9 50 0.16 3,739 ~447.2 Shi et al. S27 (man5) ( 2011 ) Paenibacillus sp. E. coli DH5α E. coli BL21 (DE3) – pGEX-6P-1 4.5 60 4–8.5 35–65 3.80 91.70 – Xiaoyu BME-14 (man26B) et al. (2010 ) A. aculeatus E. coli JM109 A. niger gpdp/glaAT pGT-man1 3.8 ~80 – – – 16,596 Van Zyl MRC11624 (man 1) et al. (2009 ) Aspergillus aculeatus E. coli DH5α Yarrowia lipolytica hp4dp/- pYL-man1- HmA – – 5.5 28 – – 13,073 Roth et al. MRC11624 man1 ( 2009 ) A. sulphureus E. coli Top10 Pichia pastoris – pPICZαA 2.4 50 2.2–8.0 <40 0.93 ~344U/mg ~ 1,603 Chen et al. (MANN) ( 2007 )

A. aculeatus S. cerevisiae S. cerevisiae ADH2P/ADH2 T pMES1 3 60 – – – 521 Setati et al. MRC11624 PGK1P/PGK1 pMES2 379 (2001 ) 52 H. Soni and N. Kango sugars. The hydrolyzed mannan is also benefi cial (Gubitz et al. 1997 ; Puchart et al. 2004 ). to consumer health as it decreases fat utilization. Mannanases also help in removal of lignin from Endo-mannanase is also applicable in coconut pulps by biobleaching without affecting the extraction. Mannan is present as the main com- quality of fi bers. Here are so many thermophilic ponent in the cell wall of coconuts, so the use microorganisms like Thermomyces lanugino- of mannanase helps to achieve a higher yield sus , Malbranchea cinnamomea , Myceliophthora of oil and improves the refi ning properties of fergusii , and Bacillus subtilis which are able to oil. Commercially, endo-β-mannanase from produce thermostable mannanases (Maijala Aspergillus niger is marketed as “GAMANASE” et al. 2012 ; Summpunn et al. 2011 ). The ther- for coconut oil extraction (Novo Nordisk, mostable β-mannanases and β-mannosidases Denmark). offer a signifi cant advantage for biobleaching at elevated temperatures. A thermostable extracellular β-mannanase from A. niger was studied by Feed Upgradation Naganagouda et al. (2009 ). They have shown that this β-mannanase was active over a wide pH Mannan is commonly present in feed ingredi- and temperature range. Due to these properties ents such as soybean meal, copra meal, and of this enzyme, it is potentially useful in bio- palm kernel meal. Among these meals, soy- bleaching and food processing. Tenkanen et al. bean meal is mostly used as a protein source in (1997 ) have shown that mannanases can be used poultry feed. This meal is the product remain- as an alternative in place of hydrogen peroxide ing after complete oil extraction from soybean in biobleaching. seeds. Due to high levels of mannan, this meal negatively affects growth performance of animals because of its indigestive nature. This Detergent and Textile Industries mannan component can be hydrolyzed by addition of endo-β-mannanase in the meal. These Amylases and cellulases are well-known com- additions of β-mannanase improve digestibility mon enzymes which are used for many decades by hydrolysis of mannan and thereby enhance in detergents. Gums are used as stabilizing and the performance of poultry. Mannanase pro- thickening agents worldwide in many foods and duced from Trichoderma longibrachiatum and in various household products. These gums con- B. lentus is marketed as “Hemicell” by tain mannans like galactomannan, guar gum, and ChemGen, a US-based company, and is poten- tara gum. Mannanases can be added in detergents tially used as an animal feed supplement. to remove this gummy matter from clothes. Detergents or cleaners, which are used to clean contact lenses, and hard surface cleaners also Biobleaching contain mannanases. It is observed that detergents containing mannanase can improve the The use of xylanase in biobleaching of pulp is whiteness of cellulosic material. But utilization well known (Viikari et al. 1993 ). Mannanase is of mannanase has been limited because most of also used in enzymatic bleaching of softwood the mannanases have their optima around neutral pulp. The extraction of lignin from softwood is or somewhat acidic pH (4.0–6.0). Some fungal important. For this purpose, alkaline treatment β-mannanases can tolerate an alkaline pH up to is performed for hydrolyzing hemicelluloses eight. β-Mannanase from alkaliphilic Bacillus sp. and removal of lignin. This alkaline treatment of N16-5, cloned and expressed in Kluyveromyces wood pulps creates environment pollution, cicerisporus and Pichia pastoris, had a pH especially water pollution. Application of man- optimum of 9.5 and was stable over a pH range nanases has signifi cantly reduced the use of of 5.0–11.0 (Pan et al. 2011 ). A mannanase alkaline chemicals in treatment of wood pulps preparation, “MANNAWAY,” used in washing 4 Microbial Mannanases: Properties and Applications 53 detergents is marketed by Novozymes, a Akino T, Nakamura N, Horikoshi K (1987) Production of β β US-based company. Galactomannans such as -mannosidase and -mannanase by an alkaliphilic Bacillus sp. Appl Microbiol Biotechnol 26:323–327 guar gum and LBG are widely employed as print Araujo A, Ward OP (1990) Hemicellulases of Bacillus paste in textile printing. Mannanase, used in species: preliminary comparative studies on produc- enzyme consortia, helps to degrade mannan after tion and properties of mannanases and galactanases. the printing of cloth. J Appl Bacteriol 68:253–261 Arcand N, Kluepfel D, Paradis FW, Morosoli R, Shareck F (1993) Beta-mannanase of Streptomyces lividans 66: cloning and DNA sequence of the Other Applications manA gene and characterization of the enzyme. Biochem J 290:857–863 Aspinall GO (1959) Structural chemistry of the hemicel- Mannanase is also used in oil drilling or gas oil luloses. Adv Carbohydr Chem 14:429–468 well stimulation. There are many fungal and bac- Aspinall GO, Begbie R, McKay JE (1962) The hemi- terial β-mannanases available which have high- celluloses of European larch (Larix decidua). Part temperature optima. Heterologous production is II. J Chem Soc 214:219 β Benech RO, Li X, Patton D, Powlowski J, Storms R, providing access to thermostable -mannanases. Bourbonnais R, Paice M, Tsang A (2007) Recombinant To open crevices and cracks for oil and gas fl ow, expression, characterization, and pulp prebleaching mannans like guar gum and tara gum are used property of a Phanerochaete chrysosporium endo-β- as polymer solutions with sand particles. Polymer 1,4-mannanase. Enzyme Microb Technol 41:740–747 Bourgault R, Bewley JD, Alberici A, Decker D (2001) solutions are hydrolyzed with a suitable thermo- Endo-β-mannanase activity in tomato and other ripen- stable mannanase for better oil and gas recovery. ing fruits. Hortscience 36:72–75 Organisms from hydrothermal vents are isolated and Chaikumpollert O, Methacanon P, Suchiva K (2004) used as a source of endo- and exo-mannanase, Structural elucidation of hemicelluloses from Vetiver grass. Carbohydr Polym 57:191–196 and this preparation is marketed by companies, Chandrasekaran R, Radha A, Okuyama K (1998) Morpho- for example, pyrolase 160 and pyrolase 200 logy of galactomannans: an X-ray structure analysis of products are marketed by Diversa Company with guaran. Carbohydr Res 306:243–255 a 37–93 °C recommended temperature. In the Chen X, Cao Y, Ding Y, LuW LD (2007) Cloning, functional expression and characterization of Aspergillus production of second-generation fuel (bioetha- sulphureus β-mannanase in Pichia pastoris. J Biotechnol nol, biodiesel), mannanases are crucial with 8:452–461 other enzymes. Synergy between mannan- Cui F, Shi J, Lu Z (1999) Production of neutral beta-man- degrading enzymes and cellulases was demon- nanase by Bacillus subtilis and its properties. Wei Sheng Wu Xue Bao 39(1):60–63 strated with a fi vefold increase in glucose yield Dhawan S, Kaur J (2007) Microbial mannanases: an over- from lignocellulose polysaccharides (Jorgensen view of production and applications. Crit Rev et al . 2010 ). Partially hydrolyzed guar gum Biotechnol 27(4):197–216 (PHGG) is used as relief agent for irritable bowel Duffaud GD, McCutchen CM, Leduc P, Parker KL, Kelly RM (1997) Purifi cation and characterization of syndrome (IBS). extremely thermostable β-mannanase, β-mannosidase, and α-galactosidase from the hyperthermophilic Eubacterium Thermotoga neapolitana 5068. Appl Environ Microbiol 63:169–177 References Fan Z, Wagschal K, Chen W, Montross MD, Lee CC, Yuan L (2009a) Multimeric hemicellulases facilitate biomass Abdeshahian P, Samat N, Hamid AA, Yusoff WM conversion. Appl Environ Microbiol 75:1754–1757 (2010) Utilization of palm kernel cake for produc- Fan Z, Werkman JR, Yuan L (2009b) Engineering of a tion of beta-mannanase by Aspergillus niger FTCC multifunctional hemicellulose. Biotechnol Lett 5003 in solid substrate fermentation using an aerated 31:751–757 column bioreactor. J Ind Microbiol Biotechnol Feng Y, He Z, Ong SL, Hu J, Zhang Z, Ng WJ (2003) 37(1):103–109 Optimization of agitation, aeration and temperature Ademark P, Varga A, Medve J, Harjunpää V, Drakenberg conditions for maximum β-mannanase production. T, Tjerneld F, Stålbrand H (1998) Softwood hemicel- Enzyme Microb Technol 32:282–289 lulose-degrading enzymes from Aspergillus niger : Ferreira HM, Filho EXF (2004) Purifi cation and charac- purifi cation and properties of a β-mannanase. terization of a β-mannanase from Trichoderma harzia- J Biotechnol 63:199–210 num strain T4. Carbohydr Polym 57:23–29 54 H. Soni and N. Kango

Gomes J, Terler K, Kratzer R, Kainz E, Steiner W (2007) Kataoka N, Tokiwa Y (1998) Isolation and characteriza- Production of thermostable β-mannosidase by a strain tion of an active mannanase-producing anaerobic bac- of Thermoascus aurantiacus : isolation, partial purifi - terium, C. tertium KT-5A from lotus soil. J Appl cation and characterization of the enzyme. Enzyme Microbiol 84:357–367 Microb Technol 40:969–975 Kenne L, Rossel K-G, Svensson S (1975) Studies on the Großwindhager C, Sachslehner A, Nidetzky B, Haltrich D distribution of the O-acetyl groups in pine glucoman- (1999) Endo- β-1,4 mannanase is effi ciently produced nan. Carbohydr Res 44:69–76 by Sclerotium (Athelia) rolfsii under derepressed con- Kirk O, Borchet TV, Fuglsang CC (2002) Industrial ditions. J Biotechnol 67:189–203 enzyme applications. Curr Opin Biotechnol Gubitz GM, Hayn M, Sommerauer M, Steiner W (1996) 13:345–351 Mannan-degrading enzymes from Sclerotium rolfsii : Kobayashi Y, Echigen R, Mada M, Mutai M (1987) Effect characterization and synergism of two endo of hydrolyzate of konjac mannan and soybean oligo- β-mannanases and β-mannosidase. Bioresour saccharides on intestinal fl ora in man and rats. In: Technol 58:127–135 Mitsuoka T (ed) Intestinal fl ora and food factors. Gubitz GM, Haltrich D, Latal B, Steiner W (1997) Mode Gakkai Shuppan Centre, Tokyo, pp 79–97 of depolymerisation of hemicellulose by various man- Kote NV, Patil AGG, Mulimani VH (2009) Optimization nanases and xylanases in relation to their ability to of the production of thermostable endo-β-1,4 man- bleach softwood pulp. Appl Microbiol Biotechnol nanases from a newly isolated Aspergillus niger gr and 47:658–662 Aspergillus fl avus gr. Appl Biochem Biotechnol Hilge M, Gloor SM, Rypniewski W, Sauer O, Heightman 152(2):213–223 TD, Zimmerman W et al (1998) High-resolution Kurakake M, Komaki T (2001) Production of β-Mannanase native and complex structures of thermostable and β-Mannosidase from Aspergillus awamori K and β-mannanase from Thermomonospora fusca -sub- their properties. Curr Microbiol 42:377–380 strate specifi city in glycosyl hydrolase family 5. Kurakake M, Sumida T, Masuda D, Oonishi S, Komaki T Structure 6:1433–1444 (2006) Production of galacto-manno-oligosaccharides Hongshu Z, Jinggan Y, Yan Z (2002) The glucomannan from guar gum by β-mannanase from Penicillium oxa- from ramie. Carbohydr Polym 47:83–86 licum SO. J Agric Food Chem 54:7885–7889 Hossain MZ, Abe JI, Hizukuri S (1996) Multiple forms of Liepman AH, Nairn CJ, Willats WGT, Sørensen I, Roberts β-mannanase from Bacillus sp. KK01. Enzyme AW, Keegstra K (2007) Functional genomic analysis Microb Technol 18:95–98 supports conservation of function among cellulose Ishurd O, Kermagi A, Zgheel F, Flefl a M, Elmabruk M, synthase-like A gene family members and suggest Yalin W, Kennedy JF, Yuanjiang P (2004) Structural diverse roles of mannans in plants. Plant Physiol aspects of water soluble galactomannan from the 143:1881–1893 seeds of Retama raetam . Carbohydr Polym 58:41–44 Lin TC, Chen C (2004) Enhanced mannanase production Jiang Z, Wei Y, Li D, Chai P, Kusakabe I (2006) High- by submerged culture of Aspergillus niger NCH-189 level production, purifi cation and characterization of a using defatted copra based media. Process Biochem thermostable β-mannanase from the newly isolated 39(9):1103–1109 Bacillus subtilis WY34. Carbohydr Polym 66:88–96 Lin SS, Dou WF, Xu HY, Li HZ, Xu ZH, Ma YH (2007) Jorgensen H, Sanadi AR, Felby C, Lange NEK, Fischer Optimization of medium composition for the produc- M, Ernst S (2010) Production of ethanol and feed by tion of alkaline b-mannanase by alkaliphilic Bacillus high dry matter hydrolysis and fermentation of palm sp.N16–5 using response surface methodology. Appl kernel press cake. Appl Biochem Biotechnol Microbiol Biotechnol 75:1015–1022 161:318–332 Lundqvist J, Teleman A, Junel L, Zacchi G, Dahlman O, Kalogeris E, Christakopoulos P, Katapodis P, Alexiou A, Tjerneld F, Stalbrand H (2002) Isolation and charac- Vlachou S, Kekos D, Macris BJ (2003) Production terization of galactoglucomannan from spruce (Picea and characterization of cellulolytic enzymes from the abies ). Carbohydr Polym 48:29–39 thermophilic fungus Thermoascus aurantiacus under Lundqvist J, Jacobs A, Palm M, Zacchi G, Dahlman O, solid state cultivation of agricultural wastes. Process Stålbrand H (2003) Characterization of galactogluco- Biochem 38:1099–1104 mannan extracted from spruce ( Picea abies) by heat- Kango N (2007) Lignocellulose biotechnology: harness- fractionation at different conditions. Carbohydr Polym ing renewable plant materials for energy and chemi- 51:203–211 cals. Everyman’s Sci XLII(4):194–197 Ma YH, Xue YF, Dou YT, Xu ZH, Tao WY, Zhou PJ Kango N, Agrawal SC, Jain PC (2003) Production of (2004) Characterization and gene cloning of a novel xylanase by Emericella nidulans NK-62 on low- b-mannanase from alkaliphilic Bacillus sp. N16–5. value lignocellulosic substrates. World J Microbiol Extremophiles 8:447–454 Biotechnol 19(7):691–694 Maijala P, Kango N, Szijarto N, Viikari L (2012) Kansoh AL, Nagieb ZA (2004) Xylanase and mannanase Characterization of hemicellulases from thermophilic enzymes from Streptomyces galbus NR and their use fungi. Anton van Leeuwonhoek 101(4):905–917 in biobleaching of softwood kraft pulp. Anton van Mohamad SN, Nagasundara R, Mohamad RR, Ariff Leeuwonhoek 85:103–114 AB (2011) Improved mannan-degrading enzymes 4 Microbial Mannanases: Properties and Applications 55

production by Aspergillus niger through medium Ratto M, Siika-aho M, Buchert J, Valkeajarvi A, Viikari optimization. New Biotechnol 28. doi:10.1016/j. L (1993) Enzymatic hydrolysis of isolated and fi bre- nbt.2010.10.008 bound galactoglucomannans from pine-wood and Moreira LRS, Filho EXF (2008) An overview of mannan pine raft pulp. Appl Microbiol Biotechnol 40: structure and mannan-degrading enzyme systems. 449–454 Appl Microbiol Biotechnol 79:165–178 Reyna ABV, Noyola PP, Calvo-Méndez C, López-Romero Mudau MM, Setati ME (2006) Screening and identifi cation E, Flores-Carreón A (1999) Purifi cation and biochem- of endomannanase-producing microfungi from hyper- ical characterization of two soluble α-mannosidases saline environments. Curr Microbiol 52:477–481 from Candida albicans. Glycobiology 9:533–537 Naganagouda K, Salimath PV, Mulimani VH (2009) Ronne H (1995) Glucose repression in fungi. Trends Purifi cation and characterization of endo-β -1,4 man- Genet 11:12–17 nanase from Aspergillus niger gr for application in Roth R, Moodley V, van Zyl P (2009) Heterologous food processing industry. J Microbiol Biotechnol expression and optimized production of an Aspergillus 19(10):1184–1190 aculeatus endo-1,4-β-mannanase in Yarrowia lipolyt- Nunes FM, Domingues MR, Comibra MA (2005) ica . Mol Biotechnol 43:112–120 Arabinosyl and glucosyl residues as structural features Ruijter GJG, Visser J (1997) Carbon repression in of acetylated galactomannans from green and roasted Aspergilli . FEMS Microbiol Lett 151:103–114 coffee infusions. Carbohydr Res 340:1689–1698 Sachslehner A, Foidl G, Foidl N, Gubitz G, Haltrich D Oda Y, Komaki T, Tonomura K (1993) Production of (2000) Hydrolysis of isolated coffee mannan and coffee β-mannanase and β-mannosidase by Enterococcus extract by mannanases of Sclerotium rolfsii. J Biotechnol casselifl avus FL2121 isolated from decayed konjac. 80:127–134 Food Microbiol 10:352–358 Schroder R, Wegrzyn TF, Sharma NN, Atkinson RG Omarsdottir S, Petersen BO, Barsett H, Paulsen BS, Duus (2006) LeMAN4 endo-β-mannanase from ripe tomato JØ, Olafsdottir ES (2006) Structural characterization fruit can act as a mannan transglycosylase or hydro- of a highly branched galactomannan from the lichen lase. Planta 224:1091–1102 Peltigera canina by methylation analysis and NMR- Setati ME, Ademark P, VanZyl WH, Hahn-Hagerdal B, spectroscopy. Carbohydr Polym 63:54–60 Stalbrand H (2001) Expression of the Aspergillus acu- Pan X, Zhou J, Tian A, Le K, Yuan H, Xue Y, Ma Y, Lu H leatus endo-β -1,4-mannanase encoding gene (man1 ) in (2011) High level expression of a truncated β-mannanase Saccharomyces cerevisiae and characterization of the from alkaliphilic Bacillus sp. N16-5 in Kluyveromyces recombinant enzyme. Protein Expr Purif 21:105–114 cicerisporus . Biotechnol Lett 33:565–570 Shi P, Yuan T, Zhao J, Huang H, Luo H, Meng K, Wang Y, Perret S, Belaich A, Fierobe HP, Belaich JP, Tardif C Yao B (2011) Genetic and biochemical characteriza- (2004) Towards designer cellulosomes in Clostridia : tion of a protease-resistant mesophilic β-mannanase mannanase enrichment of the cellulosomes produced from Streptomyces sp. S27. J Ind Microbiol Biotechnol by Clostridium cellulolyticum. Biotechnol Appl 38:451–458 Biochem 40:255–259 Singh V, Malviya T (2006) A non-ionic glucomannan Petkowicz CLO, Schaefer S, Reicher F (2007) The from the seeds of an indigenous medicinal plant: mannan from Schizolobium parahybae endosperm Bryonia laciniosa . Carbohydr Polym 64:481–483 is not a reserve polysaccharide. Carbohydr Polym Singh S, Madlala AM, Prior BA (2003) Thermomyces 69:659–664 lanuginosus : properties of strains and their hemicel- Popa VI, Spiridon J (1998) Hemicelluloses: structure and lulases. FEMS Microbiol Rev 27:3–16 properties. In: Dumitriu S (ed) Polysaccharides: struc- Stalbrand H, Siika-aho M, Tenkanen M, Viikari L (1993) tural diversity and functional versatility. Marcel Purifi cation and characterization of two β-mannanases Dekker, New York, pp 297–311 from Trichoderma reesei . J Biotechnol 29:229–242 Puchart V, Vrsanska M, Svoboda P, Pohl J, Biely P Summpunn P, Chaijan S, Isarangkul D, Wiyakrutta S, (2004) Purifi cation and characterization of two forms Meevootisom V (2011) Characterization, gene cloning of endo- β -1,4-mannanase from a thermotolerant and heterologous expression of β-mannanase from a fungus, Aspergillus fumigatus IMI 385708 (formerly thermophilic bacillus subtilis . J Microbiol 49:86–93 T. lanuginosus IMI 158749). Biochim Biophys Acta Takegawa K, Miki S, Jikibara T, Iwahara S (1989) 1674:239–250 Purifi cation and characterization of exo-α- d- Puls J, Schuseil J (1993) Chemistry of hemicellulose: mannosidase from a Cellulomonas sp. Biochim relationship between hemicellulose structure and Biophys Acta 991:431–437 enzyme required for hydrolysis. In: Coughlan MP, Tenkanen M, Makkonen M, Perttula M, Viikari L, Hazlewood GP (eds) Hemicellulose and hemicellu- Teleman A (1997) Action of Trichoderma reesei lases. Portland Press, London, pp 1–27 mannanase on galactoglucomannan in pine kraft Qiao Y, Chen X, Ding H, Yue M (2008) Heterologous pulp. J Biotechnol 57:191–204 expression and characterization of man gene from Timell TE (1965) Wood hemicelluloses: part II. Carbohydr Bacillus subtilis in Pichia pastoris . Front Biol China Chem 20:409–483 3:26–31 Timell TE (1967) Recent progress in the chemistry of Ratto M, Poutanen K (1988) Production of mannan- wood hemicelluloses. Wood Sci Technol 1:45–70 degrading enzymes. Biotechnol Lett 10(9):661–664 56 H. Soni and N. Kango

Torrie JP, Senior DJ, Saddler JN (1990) Production of Willfor S, Sjoholm R, Laine C, Roslund M, Hemming β-mannanases by Trichoderma harzianum E58. Appl J, Holmbom B (2003) Characterisation of water- Microbiol Biotechnol 34:303–307 soluble galactoglucomannan from Norway spruce Van Zyl PJ, Moodley V, Rose SH, Roth RL, Van Zyl WH wood and thermomechanical pulp. Carbohydr (2009) Production of the Aspergillus aculeatus endo- Polym 52:175–187 1,4-β-mannanase in A. niger. J Ind Microbiol Xiaoyu F, Huang X, Liu P, Lin L, Wu G, Li C, Feng C, Biotechnol 36:611–667 Hong Y (2010) Cloning and characterization of a Van Zyl WH, Rosea SH, Trollopeb K, Gorgensb JF (2010) novel mannanase from Paenibacillus sp. BME-14. Fungal β-mannanases: mannan hydrolysis, heterolo- J Microbiol Biotechnol 20(3):518–524 gous production and biotechnological applications. Yoshida T, Inoue T, Ichishima E (1993) 1,2-i-d -Manno- Process Biochem 45:1203–1213 sidase from Penicillium citrinum : molecular and Viikari L, Tenkanen M, Buchert J, Ratto M, Bailey M, enzymic properties of two isoenzymes. Biochem Siika-Aho M, Linko M (1993) Hemicellulases for J 290:349–354 industrial applications. In: Saddler JN (ed) Zakaria MM, Ashiuchi M, Yamamoto S, Yagi T (1998) Bioconversion of forest and agricultural plant resi- Optimization for beta-mannanase production of a psy- dues. CAB International, Wallingford, pp 131–182 chrophilic bacterium, Flavobacterium sp. Biosci Viniegra-Gonzalez G, Favela-Torres E, Aguilar CN, Biotechnol Biochem 62:655–660 Romero-Gomez SJ, Diaz-Godinez G, Augur C (2003) Zhang Q, Yan X, Zhang L, Tang W (2006) Cloning, Advantages of fungal enzyme production in solid sequence analysis, and heterologous expression of a state over liquid fermentation systems. Biochem Eng β-mannanase gene from Bacillus subtilis Z-2. Mol J 13:157–167 Biol 40:368–374 Enzyme Synergy for Enhanced Degradation of Lignocellulosic 5 Waste

J. Susan van Dyk and Brett I. Pletschke

Abstract Lignocellulosic substrates are very complex with a number of different components associated within a three-dimensional network. The complexity of the substrate, as well as the close association between components, makes the substrate recalcitrant to degradation. Such substrates require a large number of hydrolytic enzymes working in synergy to achieve complete degradation. The use of synergy studies allows us to determine whether enzymes display cooperation in hydrolysis of lignocellulosic substrates. This allows us to evaluate which enzymes should possibly be included in designer enzyme cocktails for lignocellulose hydrolysis. It also provides us with insight into areas of substrate recalcitrance, including obstacles such as physical association between components which prevent access of enzymes to their substrate.

Keywords Cellulose • Enzymes • Hemicellulase • Lignocellulose • Synergy

Introduction Lignin is a polyphenolic compound that provides rigidity to plants and can make up to Lignocellulose contains approximately 75 % 20 % of the lignocellulose composition in sec- polysaccharides in the form of cellulose and ondary plant cell walls. It has a very complex hemicellulose, and its degradation represents a structure and only a few microorganisms in sustainable option for obtaining large quantities nature (white- rot fungi) produce the enzymes of sugars for fermentation into bioethanol required to degrade this component. These (Gomez et al. 2008 ). enzymes are listed in Table 5.1 . Lignin represents the main obstacle to lignocellulose hydrolysis as it prevents hydrolytic enzymes from accessing J. S. van Dyk • B. I. Pletschke (*) the cellulose and hemicellulose polysaccharides Department of Biochemistry, Microbiology (Varnai et al. 2010 ). Biological degradation of lig- and Biotechnology , Rhodes University , PO Box 94 , 6140 Grahamstown , South Africa nin in biofuel technologies is not commonly used, e-mail: [email protected] and generally, a physico-chemical pretreatment

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 57 DOI 10.1007/978-81-322-1094-8_5, © Springer India 2013 58 J.S. van Dyk and B.I. Pletschke

Table 5.1 Enzymes required to degrade lignocellulose linkages. Due to the variety of sugars and linkages, components a large number of enzymes are required for the Component Enzymes required for degradation degradation of these substrates as enzymes may Lignin Laccase, manganese peroxidase, have specifi city for particular sugars and/or lignin peroxidase particular linkages. Hemicellulose composition Cellulose Cellobiohydrolase, endoglucanase, varies between primary and secondary cell walls β-glucosidase and between sources of lignocellulose such as Hemicellulose: Endo-xylanase, acetyl xylan xylan esterase, β-xylosidase, grasses, hardwoods and softwoods. Therefore, α-L- arabinofuranosidase, different hemicellulase enzymes may be required α-glucuronidase, ferulic acid for each type of substrate. Table 5.1 lists some of esterase, p-coumaric acid esterase the enzymes required to degrade the most com- Hemicellulose: Endo-mannanase, β-mannosidase, mannan α-galactosidase, β-glucosidase, mon hemicelluloses. The hydrolysis of hemicel- acetyl mannan esterase lulose can release a number of sugars such as Hemicellulose: Endo-arabinase, xylose, arabinose, mannose and galactose which arabinan α-L-arabinofuranosidase could provide an additional source of substrate Pectin Pectin methylesterase, pectin for ethanol production, although fermentation and pectate lyases, endo- and has to be carried out by other yeasts such as exopolygalacturonases, rhamnogalacturonan lyases Pichia stipitis. This could potentially enhance and hydrolases the overall ethanol yield from lignocellulose substrates compared to using only the cellulose fraction. method is used to remove the lignin or to break An additional component that further increases the linkages between the lignin and polysaccha- the complexity of lignocellulose hydrolysis is the ride components (see review by Hendriks and presence of pectin. Pectin may be present at high Zeeman 2009 ). levels in the primary cell walls of plants, but Cellulose consists of chains of glucose con- becomes less than 10 % of secondary cell walls nected with β-1,4-linkages that are packed into (Albersheim et al. 2011 ). However, in agricul- microfi brils into the plant cells and is the main tural wastes such as citrus peel or apple pomace, polysaccharide component in most plants. A high concentrations of pectin may be found minimum of three enzymes are required to which necessitates additional enzymes for their degrade cellulose to glucose, namely, cello- degradation. Three different types of pectin are biohydrolases (exoglucanases), endoglucanases found in plants, namely, homogalacturonan, and β-glucosidases. As glucose is easily converted rhamnogalacturonan I and rhamnogalacturonan into ethanol by Saccharomyces cerevisiae , cellu- II. Table 5.1 lists some of the main enzymes used lose hydrolysis is the main target in bioconver- to degrade pectins. sion. The hydrogen bonding between adjacent After removal of lignin, the main obstacle to cellulose chains, the crystalline structure and the the degradation of the polysaccharides in the sub- insolubility of the cellulose present great diffi - strate lies in the nature of the three-dimensional culty for the enzymes to hydrolyse this substrate. structure and association between the different Hemicellulose is defi ned as the component of components. Short sections of hemicellulose the plant that hydrogen bonds to cellulose within hydrogen bond with adjacent cellulose fi brils, the three-dimensional structure of the plant masking the cellulose as well as forming cross- (Albersheim et al. 2011 ). It is a broad term that linkages between cellulose fi brils within the plant refers to various polysaccharides such as arabi- structure, resulting in a complex, interwoven noxylan, xyloglucan, galactoglucomannan, glu- structure (Albersheim et al. 2011 ). If pectin is comannan and arabinan, each consisting of further present, it will be found within this matrix different combinations of sugars with a variety of between the cellulose and hemicellulose. The linkages, including α-1,2, α-1,3, α-1,6 and β-1,4 unravelling of this network requires that each 5 Enzyme Synergy for Enhanced Degradation of Lignocellulosic Waste 59 enzyme obtains access to its particular substrate is not dependent on the other. It is not an indication and bonds within the plant cell wall. This is that the one enzyme should be discarded from the achieved through cooperation between enzymes, eventual enzyme cocktail as it may still contrib- referred to as synergistic interactions. ute to the overall degradation of the substrate. By studying synergistic interactions between This could be evaluated in a different way by enzymes, we are able to identify the physical bar- measuring the impact of the enzymes on increased riers to access of enzymes or the types of bonds yields of product. that contribute the most to recalcitrance. Studying The third category of synergy is where the the ratios of enzymes that provide the highest degree of synergy is calculated to be above 1. degree of synergy allows the optimisation of This is a clear indication that the enzymes are enzyme loadings which leads to reduced cost for cooperating in the degradation of the substrate the enzyme input. This forms the basis for the which may take place in different ways. One design of enzyme cocktails for the hydrolysis of enzyme may be hampered by substituents on the different complex substrates. substrate which is removed by the other enzyme. Within the three-dimensional structure of lignocellulose, the one enzyme may be removing a Enzyme Synergy Studies portion of the substrate which masks the access of enzymes to other parts of the substrate. Enzyme synergy studies investigate the interac- The degree of synergy is an indication of the tion between two or more enzymes on a substrate enzyme cooperation required for the degradation to determine whether the enzymes display coop- of the substrate. A high degree of synergy indi- erative behaviour in the degradation of the cates the extent to which the enzymes are mutually substrate. The degree of synergy/synergism is dependent on each other for access to their sub- calculated as the observed activity of the strate or bond. enzyme mixture, divided by the theoretical sum The degree of synergy is dependent on the of the individual activities of the same enzymes concentration and ratio of the enzymes, as well as on the same substrate. The calculated value for the specifi c characteristics of the enzymes. degrees of synergy can be divided into three cat- Synergy is generally observed only at low con- egories which provide different interpretations of centrations of enzyme which retards the degra- the interactions taking place. dation in order to allow observation of the A degree of synergy of below 1 is an indication interaction between the enzymes (Andersen et al. that no synergy took place and that the combined 2008 ). At high concentrations of enzyme, synergy use of the enzymes resulted in a lower activity may not be observed. than the theoretical sum of their individual activi- The ratios in which enzymes are used have ties. This may indicate that the enzymes were an impact on the degree of synergy observed. competing for the same substrate sites and there- Synergistic interactions become more important fore did not enhance the degradation of the sub- when an enzyme is only present at a low ratio. strate. Thus, enzymes with similar or overlapping Ratios are important to optimise enzyme load- functions could display this degree of synergy, ings and also provide an indication of the level of indicating that one of the enzymes is probably not recalcitrance associated with specifi c bonds or required in a cocktail of multiple enzymes for specifi c components of the substrate. The contri- degradation of the specifi c substrate. bution of the enzymes is not necessarily related A degree of synergy of 1 indicates that no syn- to the percentage composition of the substrate, ergy took place when the enzymes were used in and only direct biochemical assays can determine combination and that the one enzyme did not optimal ratios (Banerjee et al. 2010 ). A substrate contribute to the other enzyme’s activity on the may contain only a small percentage of xylan, substrate. The enzymes most likely act indepen- yet the synergistic relationship may indicate dently on the substrate and the action of the one a requirement for endo-xylanase supplementation 60 J.S. van Dyk and B.I. Pletschke at much greater levels. An optimal ratio between polysaccharides. In a similar manner, synergy two or more enzymes may change if a further was found between endo-xylanases and acetyl enzyme is added to the combination. xylan esterases. Raweesri et al. (2008 ) found that Substrate characteristics may also have an the release of arabinose and acetyl groups from impact on the degree of synergy, for example, the substituted xylan before using a depolymerising crystallinity of the substrate or the degree of endo-xylanase resulted in greater levels of syn- polymerisation (Andersen et al. 2008). The spe- ergy, indicating that the depolymerisation was cifi c characteristics of enzymes have an impact restricted by the presence of substituents on the on the degree of synergy observed. When study- backbone. ing synergistic interactions, it is therefore not Another enzyme characteristic that is impor- suffi cient to have a general classifi cation of the tant is the presence of carbohydrate-binding hydrolytic enzymes involved. It is important that domains (CBMs), although their impact on syn- the family of the enzyme and its substrate speci- ergistic relationships has received limited atten- fi city is known. If this information is not avail- tion. These domains provide an interaction able, it may be diffi cult to draw clear conclusions between the enzyme and the substrate that is from the study. For example, most xylanases are separate from the active site and allows the found in two families, namely, family 10 and 11, enzyme to bind to the substrate. A large number although some xylanases also fall within fami- of CBM families exist and enzymes may have no lies 5, 8 and 43 (Kolenova et al. 2006 ). Endo- CBM, one CBM or multiple CBMs. While they xylanases from the glycosyl hydrolase family 10 may increase the activity of an enzyme on a sub- (GH10) have a smaller active site and a low substrate, they could have a specifi c impact on syner- strate specifi city and are able to cleave linkages gistic interactions between enzymes. If two in the xylan backbone close to any substituents enzymes have different substrate specifi cities, present. On the other hand, endo-xylanases from but similar CBMs, they could display competi- glycosyl hydrolase family 11 (GH11) have a tive behaviour as they will both be binding to the higher substrate specifi city with a larger active same sites (Andersen et al. 2008 ). The rate of site. Therefore, it is probable that substituents adsorption or desorption of enzymes could also will pose a greater hindrance to GH11 endo- result in competitive behaviour between enzymes xylanases as cleavage of the backbone can only (Andersen et al. 2008 ). take place at a number of residues away from the In addition, time can have an impact on the substituent. Therefore, it is expected to fi nd a degree of synergy observed. The general pattern high degree of synergy between the depoly- that has been found is that the highest degree of merising (GH11) and the debranching enzymes synergy is observed at the early stages of hydro- (such as α-L-arabinofuranosidases and acetyl lysis with the degree of synergy becoming lower xylan esterases). during later stages (Andersen et al. 2008 ). It is Studies in this regard have focused on arabi- argued that the reason for this pattern is that the nose and acetyl substituents on arabinoxylan. greatest level of cooperation between enzymes is From these studies, it is apparent that it is not just required at the early stages of deconstruction of the characteristics of the endo-xylanases but also the substrate. As the enzymes cooperate to the α-L-arabinofuranosidases and acetyl xylan unravel the substrate, more binding sites become esterases that affect synergistic behaviour. Shi available over time, requiring a lower degree of et al. ( 2010 ) used a GH51 arabinofuranosidase cooperation between the enzymes and a lower with an endo-xylanase in synergy studies and degree of synergy. found that the arabinofuranosidase was unable to Synergistic cooperation may be different cleave arabinose from xylan polymers unless the depending on whether enzymes are added in a endo-xylanase had fi rst cleaved the xylan back- sequential or simultaneous manner. Where one bone into shorter oligosaccharides. The arabino- enzyme relies on the removal of a substituent furanosidase appeared to be unable to bind large before it is able to cleave the substrate, the synergy 5 Enzyme Synergy for Enhanced Degradation of Lignocellulosic Waste 61 may be higher using the sequential method. The high quantities and have great potential as patterns observed in this regard are highly depen- lignocellulosic substrates for biofuel production. dent on the specifi c characteristics of the enzymes The composition of the substrates, from our own involved. Shi et al. (2010 ) found similar degrees analysis and from literature, is provided in of synergy when an endo-xylanase or arabinofu- Table 5.3 as there may be a relationship between ranosidase was used simultaneously or sequen- composition and synergistic interactions. tially, but only when the endo-xylanase was By utilising two or three pure enzymes in used fi rst. combination in synergy studies, we were able to observe the interactions between these enzymes on complex substrates. From the results sum- Synergy Studies in Our Research marised in Table 5.2 , it is apparent that most of Group the enzyme combinations displayed some degree of synergy, indicating that cooperation between Synergy studies were conducted in our research the enzymes took place. The degrees of synergy group using purifi ed enzymes from Clostridium varied with different enzyme combinations, indi- cellulovorans , namely, EngE, ManA, XynA and cating that different levels of cooperation took ArfA in combinations of two or three enzymes on place between the enzymes examined and on the different substrates. EngE is a GH5 cellulosomal different substrates. The relationship between enzyme from C. cellulovorans (Tamaru and Doi synergy and the composition of the substrates 1999 ). Purifi ed EngE was able to hydrolyse CMC (Table 5.3 ) does not display a clear correlation – and lichenan, but had virtually no activity on acid the level at which an enzyme is required is not swollen cellulose, Avicel, laminarin and xylan necessarily related to the percentage of that sugar (Tamaru and Doi 1999 ). ManA is a GH5 cellulo- within the overall composition, as the associa- somal enzyme from C. cellulovorans (Tamaru tions of components within the substrate are not et al. 2000 ). ManA was examined for substrate taken into account by the analysis. specifi city and only had activity on β-1, Synergistic interactions were observed 4-mannosidic linkages and had no activity on between EngE and XynA on all substrates where CMC (Tamaru et al. 2000 ). XynA is an endo-1,4- this combination was tested. In most cases, the β-xylanase with a family 11 (GH11) catalytic degree of synergy between EngE and XynA on domain, a dockerin domain and an acetyl xylan all the substrates tested was quite high relative to esterase module (Kosugi et al. 2002b ). The acetyl the other enzyme combinations. When the com- xylan esterase module was able to remove acetate position of the substrates is examined in Table 5.3 , from acetylated xylan and other acetylated sub- it is apparent that all the substrates contained strates. The xylanase and acetyl xylan esterase cellulose and some degree of xylan, although the module could act synergistically to degrade xylan composition was not characterised in the acetylated xylan (Kosugi et al. 2002b ). ArfA corn stalk, grass and pineapple pulp, but it is is a non-cellulosomal α-L-arabinofuranosidase known that arabinoxylan is the main hemicellu- classifi ed in glycosyl hydrolase family 51 lose in grasses. These results correspond to other (GH51), with specifi city towards p -nitrophenol- reports in literature (Bura et al. 2009 ; Garcia- arabinofuranoside and additional activity on sugar Aparicio et al. 2007 ; Selig et al. 2008) that xylanase beet arabinan (arabinase activity) (Kosugi et al. supplementation improved cellulose hydrolysis 2002a ). regardless of the xylan content. Addition of Synergy studies were conducted on a variety higher concentrations of xylanase allowed lower of complex substrates such as corn stalk, grass, cellulase enzyme loadings while achieving the pineapple pulp, sugarcane bagasse (with various same level of cellulose hydrolysis. Kumar and pretreatments) and lime pretreated sugar beet Wyman (2009 ) demonstrated that supplementing pulp (see Table 5.2 ). Many of these substrates xylanases with cellulase not only improved cellu- are agricultural waste products produced in lose hydrolysis but displayed a linear relationship 62 J.S. van Dyk and B.I. Pletschke

Table 5.2 A summary of synergy studies conducted in our research group Degrees of synergy Substrate Pretreatment used Enzymes used Ratio (DS) Reference Corn stalk Untreated EngE:XynA 75:25 6.5 Olver et al. (2011 ) EngE:ManA 50:50 3.5 Olver et al. (2011 ) ManA:XynA 75:25 2.6 Olver et al. (2011 ) Grass Untreated EngE:XynA 75:25 3.3 Olver et al. (2011 ) EngE:ManA 75:25 2.0 Olver et al. (2011 ) ManA:XynA 75:25 1.2 Olver et al. (2011 ) Pineapple Untreated EngE:XynA 50:50 2.4 Olver et al. (2011 ) pulp EngE:ManA 50:50 2.8 Olver et al. (2011 ) ManA:XynA 75:25 2.5 Olver et al. (2011 ) Sugar Lime pretreatment ArfA:ManA 75:25 1.36 Dredge et al. (2011 ) beet pulp ArfA:XynA No synergy Dredge et al. (2011 ) Sugarcane Untreated EngE:ManA 50:50 2.28 Beukes et al. (2008 ) bagasse EngE:XynA 25:75 4.65 Beukes et al. (2008 ) XynA:ManA 75:25 3.95 Beukes et al. (2008 ) XynA:EngE:ManA 25:25:50 2.73 Beukes et al. (2008 ) Untreated ArfA:XynA 12.5:87.5 2.4 Beukes and Pletschke (2010 ) ArfA:ManA:XynA Various ±1.6 Beukes and Pletschke (2010 ) Lime (CaOH) ManA:XynA 87.5: 12.5 1.4 Beukes and Pletschke (2010 ) ArfA:XynA 50:50 1.3 Beukes and Pletschke (2010 ) ArfA:ManA Various 1.8–2.0 Beukes and Pletschke (2010 ) ArfA:ManA:XynA Various 1.6–2.2 Beukes and Pletschke (2010 )

NH4 OH XynA:ManA 75:25 2.85 Beukes and Pletschke (2011 ) pretreated ArfA:XynA ArfA at 3 Beukes and Pletschke (2011 ) high levels ArfA:ManA:XynA Various 3 Beukes and Pletschke (2011 ) NaOH ArfA:XynA 37.5:62.5 1.6 Beukes and Pletschke (2011 ) ManA:ArfA 75:25 1.8 Beukes and Pletschke (2011 ) ManA:XynA No synergy Beukes and Pletschke (2011 ) ArfA:ManA:XynA Various 1.4–1.6 Beukes and Pletschke (2011 )

between glucose and xylose release. Murashima could predict that the synergy should be sequen- et al. (2003 ) observed that synergistic interac- tial, rather than simultaneous. Selig et al. ( 2008 ), tions between cellulases and xylanases were only however, came to the conclusion that the cellu- present in simultaneous, not sequential, reac- lose also masked the xylan in a similar manner as tions. The mechanism behind this type of syn- the xylan masked the cellulose. The simultaneous ergy is generally considered to be the fact that the interactions between the two types of enzymes xylan masks the cellulose and prevents access of allow for the synchronised hydrolysis of both cel- cellulases to their substrate. As the xylanases lulose and xylan. hydrolyse the xylan components in the substrate, The role of XynA on untreated substrates the cellulose is exposed, allowing the cellulases becomes particularly important as XynA con- to access the cellulose. Based on this model, one tains an acetyl xylan esterase module. Acetate 5 Enzyme Synergy for Enhanced Degradation of Lignocellulosic Waste 63

Table 5.3 Composition of complex substrates studied in our research group Substrate Pretreatment Composition Reference Corn stover Drying and grinding 18 % lignin, 41.7 % cellulose, Merino and Cherry (2007 ) 20.5 % hemicellulose Grass Drying and grinding 10–30 % lignin, 25–40 % cellulose, Howard et al. (2003 ) 25–50 % hemicellulose (average values) Pineapple pulp Drying and grinding 19 % cellulose, 22 % hemicellulose, Ban-Kofﬁ and Han (1990 ) Soluble sugars removed 5 % lignin and 53 % cell soluble matters Sugar beet pulp Lime pretreated Rhamnose 2.4 %, fucose 0.2 %, Micard et al. (1996 ) arabinose 20.9 %, xylose 1.7 %, mannose 1.1 %, galactose 5.1 %, glucose 21.1 %, galacturonic acid 21.1 % Bagasse Untreated 42.4 % lignin, 1.1 % arabinose, Beukes and Pletschke (2010 ) 28.8 % glucose, 14.2 % xylose, 7.5 % acetate Lime pretreated 32.8 % lignin, 1.0 % arabinose, 34 % glucose, 15 % xylose, 5.6 % acetate

NH4 OH 8.8 % lignin, 1.24 % arabinose, Beukes and Pletschke (2011 ) 54.64 % glucose, 15.46 % xylose NaOH 15.65 % lignin, 0 % arabinose, 42.88 % glucose, 17.5 % xylose

groups on acetylated xylan have been demonstrated lower than the interactions between EngE and to hinder the hydrolysis of xylan, contributing to XynA, except in pineapple pulp which probably substrate recalcitrance (Selig et al. 2009 ). Once contains high levels of mannan. This is in agree- alkaline pretreatments have been carried out, ace- ment with Tenkanen et al. (1999 ) who indicated tate groups are generally removed, as can be that the effect of mannanase on cellulose hydro- observed from Table 5.3 for the NH4 OH and lysis is much smaller than the effect of xylanase. NaOH treatments of bagasse, although only a Varnai et al. ( 2011 ) also found a linear relationship limited reduction in acetate was observed with between the hydrolysis of mannan and cellulose, lime pretreatment. indicating that the mannan and xylan formed a Combinations of EngE and ManA also dis- “network of polysaccharides around the cellu- played synergistic interactions in all cases where losic ﬁ bres”. They also found that the addition of these enzymes were used together (Olver et al. a mannanase had an impact on hydrolysis of the 2011 ; Beukes et al. 2008 ). This indicates that substrate even when the mannan content was mannan and cellulose interact within the three- only 0.23 % of the dry weight. This may indi- dimensional structure of the lignocellulose sub- cate a requirement for the addition of an endo- strate and probably hydrogen bonds with the mannanase to an optimised enzyme cocktail for cellulose in a similar manner to xylan. This certain mannan-containing substrates such as results in the mannan masking the cellulose and softwoods. preventing access of the cellulases to their sub- Further synergistic relationships that were strate. Combining these enzymes in assays on observed in studies in our laboratory occurred lignocellulose substrates produces a synergistic between different depolymerising hemicellulases, interaction, even when negligible amounts of such as XynA, ManA and ArfA. ArfA acts as mannan appeared to be present within the substrate. both a depolymerising hemicellulase (on arabinan) The degrees of synergy found were generally and a debranching enzyme (on arabinoxylan). 64 J.S. van Dyk and B.I. Pletschke

Limited studies have been carried out on interactions between depolymerising hemicellulases. Conclusions and Gaps in the Field Our studies demonstrate that XynA and ManA combinations all displayed synergistic interac- Enzyme synergy studies are an important tool to tions, indicating that some association exists demonstrate the dependence of enzymes on between mannan and xylan within lignocellulose mutual cooperation with other enzymes to degrade substrates. In one case, on NaOH pretreated the same substrate, such as arabinoxylan, as well bagasse, no synergy was observed between XynA as substrates that are closely associated within a and ManA, although high levels of synergy were larger network, such as lignocellulose. Our stud- present for this enzyme combination on untreated ies have demonstrated the importance of hemicel- bagasse which were lower in the lime pretreated lulases, in combination with cellulases or other bagasse. The absence of synergy on NaOH pre- hemicellulases, cooperating in order to degrade treated bagasse indicates that the pretreatment complex substrates. The role of hemicellulases, disrupted the association between the xylan and such as mannanases, in combination with cellu- mannan within the substrate. It is interesting that lases, is still an important area that has received the three different alkaline pretreatments dis- limited attention. Similarly, the cooperation played different results with respect to the xylan required between different depolymerising hemi- and mannan association. Varnai et al. (2011 ) also cellulases has received very limited attention and found a synergistic relationship between man- can provide an understanding of the mutual asso- nanase and xylanase and concluded that hydroly- ciation between different hemicellulose substrates sis of mannan improved the hydrolysis of xylan. within the framework of lignocellulose. This is an They concluded that this indicated that the xylan important area requiring further investigation. and mannan were also closely associated within the substrate and “partially cover each other”. Acknowledgments JS van Dyk was supported by the The exact nature of the association is not clear. Claude Leon Foundation. This material is based upon Synergy between ArfA and ManA on lime work supported by the National Research Foundation (NRF) of South Africa and Rhodes University Joint pretreated sugar beet pulp was found by Dredge Research Committee (JRC). Any opinion, fi ndings and et al. (2011 ) although the degree of synergy was conclusions or recommendations expressed in this mate- low (1.36). In the case of sugar beet pulp, the rial are those of the author(s) and therefore the NRF does main hemicellulose in this substrate is arabinan not accept any liability in regard thereto. (see Table 5.3 ), and thus, the arabinose residues are not present as substituents on xylan, but as polymers of arabinose. As the ArfA has prominent References activity on arabinase, it was able to depolymerise Albersheim P, Darvill A, Roberts K, Sederoff R, the arabinan. Based on the synergy found between Staehelin A (2011) Plant cell walls: from chemistry ArfA and ManA, it is therefore hypothesised that to biology. Garland Science, Taylor & Francis Group, the arabinan and mannan are also closely associ- New York ated within the sugar beet pulp substrate. Andersen N, Johansen KS, Michelsen M, Stenby EH, Krogh KB, Olsson L (2008) Hydrolysis of cellulose Further hemicellulase interactions found in using mono-component enzymes shows synergy dur- our studies included combinations of ArfA, ManA ing hydrolysis of phosphoric acid swollen cellulose and XynA, as well as ArfA and XynA. ArfA, (PASC), but competition on Avicel. Enzyme Microb acting as an α-L-arabinofuranosidase, cleaves Technol 42:362–370 Banerjee G, Scott-Craig JS, Walton JD (2010) Improving arabinose substituents from arabinoxylan. There enzymes for biomass conversion: a basic research are several reports in literature confi rming the perspective. Bioenerg Res 3:82–92 synergistic interaction between these two Ban-Koffi L, Han YW (1990) Alcohol production from enzymes (Shi et al. 2010). The removal of arabi- pineapple waste. World J Microbiol Biotechnol 6(3): 281–284 nose substituents by the ArfA assists the action of Beukes N, Chan H, Doi RH, Pletschke BI (2008) Syner- the XynA on the xylan backbone. gistic associations between Clostridium cellulovorans 5 Enzyme Synergy for Enhanced Degradation of Lignocellulosic Waste 65

enzymes XynA, ManA and EngE against sugarcane Merino ST, Cherry J (2007) Progress and challenges in bagasse. Enzyme Microb Technol 42:492–498 enzyme development for biomass utilization. Adv Beukes N, Pletschke BI (2010) Effect of lime pre- Biochem Eng/Biotechnol 108:95–120 treatment on the synergistic hydrolysis of sugarcane Micard V, Renard CMGC, Thibault JF (1996) Enzymatic bagasse by hemicellulases. Bioresour Technol saccharifi cation of sugar-beet pulp. Enzyme Microb 101:4472–4478 Technol 19:162–170 Beukes N, Pletschke BI (2011) Effect of alkaline pre- Murashima K, Kosugi A, Doi RH (2003) Synergistic treatment on enzyme synergy for effi cient hemicellu- effects of cellulosomal xylanase and cellulases from lose hydrolysis in sugarcane bagasse. Bioresour Clostridium cellulovorans on plant cell wall degrada- Technol 102:5207–5213 tion. J Bacteriol 185(5):1518–1524 Bura R, Chandra R, Saddler J (2009) Infl uence of xylan Olver B, Van Dyk JS, Beukes N, Pletschke BI (2011) on the enzymatic hydrolysis of steam-pretreated Synergy between EngE, XynA and ManA from corn stover and hybrid poplar. Biotechnol Prog Clostridium cellulovorans on corn stalk, grass and 25(2):315–322 pineapple pulp substrates. 3 Biotech 1:187–192 Dredge R, Radloff SE, Van Dyk JS, Pletschke BI (2011) Raweesri P, Riangrungrojana P, Pinphanichakam P (2008) Lime pretreatment of sugar beet pulp and evaluation α-L-arabinofuranosidase from Streptomyces sp. PC22: of synergy between ArfA, ManA and XynA from Purifi cation, characterization and its synergistic action Clostridium cellulovorans on the pretreated substrate. with xylanolytic enzymes in the degradation of xylan and 3 Biotech 1:151–159 agricultural residues. Bioresour Technol 99:8981–8986 Garcia-Aparicio MP, Ballesteros M, Manzanares P, Selig MJ, Knoshaug EP, Adney WS, Himmel ME, Decker Ballesteros I, Gonzalez A, Negro MJ (2007) Xylanase SR (2008) Synergistic enhancement of cellobiohydro- contribution to the effi ciency of cellulose enzymatic lase performance on pretreated corn stover by addition hydrolysis of barley straw. Appl Biochem Biotechnol of xylanase and esterase activities. Bioresour Technol 136–140:353–365 99:4997–5005 Gomez LD, Steele-King CG, McQueen-Mason SJ (2008) Selig MJ, Adney WS, Himmel ME, Decker SR (2009) Sustainable liquid biofuels from biomass: the writing’s The impact of cell wall acetylation on corn stover on the wall. New Phytol 178:473–485 hydrolysis by cellulolytic and xylanolytic enzymes. Hendriks AW, Zeeman G (2009) Pretreatments to enhance Cellulose 16:711–722 the digestibility of lignocellulosic biomass. Bioresour Shi P, Li N, Yang P, Wang Y, Luo H, Bai Y, Yao B (2010) Technol 100:10–18 Gene cloning, expression, and characterization of a Howard RL, Abotsi E, Jansen van Rensburg EL, Howard Family 51 α-L-arabinofuranosidase from Streptomyces S (2003) Lignocellulose biotechnology: issues of bio- sp. S9. Appl Biochem Biotechnol 162:707–718 conversion and enzyme production. Afr J Biotechnol Tamaru Y, Doi RH (1999) Three surface layer homology 2(12):602–619 domains at the N terminus of the Clostridium cellulo- Kolenova K, Vrsanska M, Biely P (2006) Mode of action of vorans major cellulosomal subunit EngE. J Bacteriol endo-β-1,4-xylanases of families 10 and 11 on acidic 181(10):3270–3276 xylooligosaccharides. J Biotechnol 121:338–345 Tamaru Y, Karita S, Ibrahim A, Chan H, Doi RH (2000) Kosugi A, Murashima K, Doi RH (2002a) Characterization A large gene cluster for the Clostridium cellulovorans of two noncellulosomal subunits, ArfA and BgaA, cellulosome. J Bacteriol 182(20):5906–5910 from Clostridium cellulovorans that cooperate with the Tenkanen M, Tamminen T, Hortling B (1999) Investigation cellulosome in plant cell wall degradation. J Bacteriol of lignin-carbohydrate complexes in kraft pulps by 184(24):6859–6865 selective enzymatic treatments. Appl Microbiol Kosugi A, Murashima K, Doi RH (2002b) Xylanase Biotechnol 51:241–248 and acetyl xylan esterase activities of XynA, a key Varnai A, Siika-aho M, Viikari L (2010) Restriction of the subunit of the Clostridium cellulovorans cellulosome enzymatic hydrolysis of steam-pretreated spruce by for xylan degradation. Appl Environ Microbiol lignin and hemicellulose. Enzyme Microb Technol 68(12):6399–6402 46:185–193 Kumar R, Wyman CE (2009) Effects of cellulase and Várnai A, Huikko L, Pere J, Siika-aho M, Viikari L (2011) xylanase enzymes on the deconstruction of solids Synergistic action of xylanase and mannanase from pretreatment of poplar by leading technologies. improves the total hydrolysis of softwood. Bioresour Biotechnol Prog 25:302–314 Technol 102:9096–9104 Manganese Peroxidases: Molecular Diversity, Heterologous 6 Expression, and Applications

Samta Saroj, Pragati Agarwal, Swati Dubey, and R.P. Singh

Abstract Manganese peroxidases (MnPs) are a fascinating group of biocatalysts with various ecological and biotechnological implications. They are involved in the biodegradation of lignocellulose and lignin and participate in the bioconversion of other diverse recalcitrant compounds, like polycyclic aromatic hydrocarbons, chlorophenols, industrial effluents (mostly from the paper and pulp), and textile and petrochemical industries, and bioremediation of contaminated soils. This chapter presents an overview of the structural basis of the catalytic properties of MnPs and the enumeration of the molecular and protein homology characteristics of this enzyme. Multiple developments mainly pertaining to enzyme engineering for improved substrate specificity and stability of MnPs have also been highlighted. Inevitably, the progress in enzyme engineering research and the expression of MnPs have explored the vast genetic diversity of these enzymes with great interest being placed on exploiting these enzymes for a variety of industrial and scientific applications.

Keywords Manganese peroxidases • Properties • Heterologous production • Molecular cloning • Enzyme engineering research • Crystal structure • Industrial applications

Introduction gen peroxide (H2O2) as an electron-accepting cosubstrate, for catalyzing the peroxide-depen- Manganese peroxidase (MnP) [EC 1.11.1.13, dent oxidation of MnII to MnIII. These enzymes MnII:hydrogen-peroxide oxidoreductase] is an have mainly been isolated from white-rot fungal extracellular heme enzyme that utilizes hydro- species like Phanerochaete chrysosporium, Trametes versicolor, Heterobasidion annosum, and Irpex lacteus with the ability to degrade lig- S. Saroj • P. Agarwal • S. Dubey • R.P. Singh (*) nin (Eriksson et al. 1990; Cai and Tien 1993). A Department of Biotechnology, Indian Institute of Technology Roorkee, Roorkee 247667, India great majority of these enzymes contain a protopor- e-mail: [email protected] phyrin IX (heme) prosthetic group. Lignin is a het-

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 67 DOI 10.1007/978-81-322-1094-8_6, © Springer India 2013 68 S. Saroj et al. erogeneous, optically inactive polymer, consisting (Hatakka 1994, 2001; Heinzkill et al. 1998; of phenylpropanoid subunits. These phenylpro- Steffen et al. 2000). C. subvermispora have been panoid subunits cannot be cleaved by hydrolytic found to produce up to 11 different isoforms of enzymes unlike most other natural polymers, e.g., MnP (Lobos et al. 1994; Urzua et al. 1995). Various cellulose, starch, and proteins. Interestingly, some white-rot fungi, which are well characterized for white-rot basidiomycetes that produce peroxidases their ligninolytic ability, belong to phyloge- have ability to degrade lignin (Sarkanen and netically older families such as Meruliaceae Ludwig 1971). Peroxidases were first discovered (P. radiata, P. sordida, P. chrysosporium, Merulius in Phanerochaete chrysosporium (Kuwahara sp. M15), Coriolaceae (B. adusta, C. subvermispora, et al. 1984; Glenn and Gold 1985; Paszcynski C. pruinosum, P. tephropora, A. biennis), and et al. 1985, 1986). During the last decade, work Polyporaceae (T. versicolor, T. gibbosa, T. trogii, has been done on the heterologous expression of T. hirsuta) as well as litter decomposers of peroxidases using X-ray crystallographic studies euagaric families such as Strophariaceae and and active-site engineering to enhance substrate Tricholomataceae have been found to have nota- specificity and the thermal stability of MnP ble expression of MnP (Table 6.1). In addition, genes (Mino et al. 1988; Benner and Gerloff some marine-derived fungal strains and strains 1990; Petersen et al. 1993; Li et al. 2001; dwelling on decaying sea grass (Raghukumar Sundaramoorthy et al. 2010). Attempts have also et al. 1999), cooling-tower wood (Schmidt et al. been made to determine the regulation of MnP 1997), and brown coal (Willmann and Fakoussa gene at transcriptional and translational levels 1997a, b) have MnP production ability. Sequences (Brown et al. 1991; Gettemy et al. 1998; Johansson of different MnPs were retrieved from GenBank et al. 2002). The nonspecific and non-stereoselec- (available at: http://www.ncbi.nlm.nih.gov). For tive nature of MnP allows it to degrade a wide phylogenetic analysis of all MnPs, an alignment range of pollutants, such as polycyclic aromatic was created with ClustalW2 multiple sequence hydrocarbons (PAHs), chlorinated phenols, alignment tool (available at http://www.ebi.ac. polychlorinated biphenyls, dioxins, pesticides, uk/Tools/msa/clustalw2/). The evolutionary his- explosives, and dyes (Levin et al. 2004). tory was determined using the neighbor-joining The main objective of this chapter has been method (Saitou and Nei 1987). The percentage of to highlight the diversity of MnPs among basid- replicate trees in which the associated taxa iomycetes, their heterologous production, phy- clustered together in the bootstrap test (1,000 logenetic analysis, structural characteristics, and replicates) is shown next to the branches molecular features. This chapter also describes (Felsenstein 1985). The evolutionary distances attempts made to engineer this enzyme for were computed using the Poisson correction improved substrate specificity and stability and method (Zuckerkand and Pauling 1965) and are to quantify the utility of this enzyme. in the units of the number of amino acid substitutions per site. The analysis involved 53 amino acid sequences. Evolutionary analyzes were con- Occurrence and Phylogenetic ducted in MEGA5 (Tamura et al. 2011). It can be Analysis of MnPs inferred from the phylogenetic tree that MnPs are divergent among different taxonomic groups. Production of extracellular MnP has been mainly Among the 54 amino acid sequences analyzed, observed from certain basidiomycetes, and thus 44 basidiomycetes were grouped together at the far, no bacterium, yeast/mold, or mycorrhiza- same phenetic unit, whereas apart from main forming basidiomycete have been reported to basidiomycete grouping, Coprinopsis cinerea, produce this enzyme (Cairney and Burke 1998; Arthromyces ramosus, Coprinellus disseminatus, Hatakka 2001). Many ecophysiological groups and Ganoderma sp. arose as distinct genetic of basidiomycetes have been found to secrete groups (Fig. 6.1). The MnP variant from Inonotus isoforms of MnP into their microenvironments hispidus was the single gene variation ungrouped 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 69

Table 6.1 Amino acid sequences retrieved from GenBank and used for alignment and phylogenetic analyzes S. No. Species Phylum GenBank acc. no. 1. Ganoderma applanatum Basidiomycota BAA88392.1 2. Ganoderma australe Basidiomycota ABB77244.1 3. Ganoderma formosanum Basidiomycota ABB77243.1 4. Ganoderma lucidum Basidiomycota ACA48488.1 5. Trametes versicolor Basidiomycota CAA83148.1 6. Phanerochaete chrysosporium Basidiomycota AAA33743.1 7. Pleurotus sp. “Florida” Basidiomycota CAB51617.1 8. Phylloporia ribis Basidiomycota ADK60897.1 9. Pleurotus pulmonarius Basidiomycota AAY42945.1 10. Coprinopsis cinerea Basidiomycota CAA49216.1 11. Lenzites gibbosa Basidiomycota AEX01147.1 12. Bjerkandera adusta Basidiomycota AAY89586.1 13. Ceriporiopsis rivulosa Basidiomycota ABB83813.1 14. Trametes gibbosa Basidiomycota ADW83732.1 15. Agrocybe praecox Basidiomycota ADW41627.1 16. Spongipellis sp. FERM P-18171 Basidiomycota BAE79812.1 18. Phlebia radiata Basidiomycota CAC84573.1 19. Hericium erinaceus Basidiomycota ADK26471.3 20. Agaricus bisporus Basidiomycota CAG27835.1 21. Phlebia sp. MG60 Basidiomycota BAG12560.1 22. Arthromyces ramosus Basidiomycota BAA09861.1 23. Laccaria bicolor S238N-H82 Basidiomycota XP_001888065.1 24. Coprinellus disseminatus Basidiomycota AAZ14938.1 25. Lentinula edodes Basidiomycota BAG72080.1 26. Phlebia sp. b19 Basidiomycota ABR66918.1 27. Phanerochaete flavidoalba Basidiomycota AAM46826.1 28. Phanerochaete sordida Basidiomycota BAC06187.1 29. Dichomitus squalens Basidiomycota AAF31330.1 30. Phellinidium ferrugineofuscum Basidiomycota ACX51165.1 31. Lactarius rufus Basidiomycota ACX51162.1 32. Lactarius fulvissimus Basidiomycota ACX51161.1 33. Hypholoma fasciculare Basidiomycota ACX51160.1 34. Hygrophorus agathosmus Basidiomycota ACX51158.1 35. Gomphus clavatus Basidiomycota ACX51157.1 36. Cortinarius traganus Basidiomycota ACX51156.1 37. Cortinarius malachius Basidiomycota ACX51154.1 38. Cortinarius infractus Basidiomycota ACX51153.1 39. Cortinarius hinnuleus Basidiomycota ACX51152.1 40. Cortinarius armillatus Basidiomycota ACX51151.1 41. Inonotus hispidus Basidiomycota ADK60893.1 42. Fomitiporia mediterranea Basidiomycota ADK60890.1 43. Coriolopsis gallica Basidiomycota AAZ16493.1 44. Phlebia albomellea Basidiomycota ABT17238.1 45. Hymenochaete corrugata Basidiomycota ADK60895.1 46. Polyporus brumalis Basidiomycota AEJ38000.1 47. Cytidia salicina Basidiomycota ABT17236.1 48. Phlebia chrysocreas Basidiomycota ABT17228.1 49. Phlebiopsis gigantea Basidiomycota ABT17220.1 (continued) 70 S. Saroj et al.

Table 6.1 (continued) S. No. Species Phylum GenBank acc. no. 50. Pulcherricium caeruleum Basidiomycota ABT17217.1 51. Cryptoporus volvatus Basidiomycota ABT17214.1 52. Hapalopilus rutilans Basidiomycota ABT17210.1 53. Trametes cinnabarina Basidiomycota ADK60908.1 54. Pleurotus ostreatus Basidiomycota AAA84397.1 from the main cluster as aforementioned. strate for catalyzing the peroxide-dependent Lactarius rufus along with Lactarius fulvissimus oxidation of MnII to MnIII. The MnIII formed is formed the third additional genetic group. highly active, which in turn is stabilized by fun- The maximum parsimony (MP) method was gal chelators like oxalate and malonate. These used to analyze the evolutionary history among chelators act as physiological regulators to the different basidiomycete MnPs. The bootstrap enzyme as they enhance the enzyme activity due consensus tree obtained from 1,000 replicates is to their ability to facilitate the dissociation of taken to represent the evolutionary history of the MnIII from the enzyme. The role of oxalate as an taxa analyzed (Felsenstein 1985). Branches cor- extracellular buffering agent has also been responding to partitions reproduced in less than reported. It facilitates the ability of the fungus to 50 % of bootstrap replicates are collapsed. The control the pH of its environment (Timofeevski percentage of replicate trees in which the associ- and Aust 1997; Zapanta and Tien 1997). Calcium ated taxa clustered together in the bootstrap test sequestration by these chelators acts to increase (1,000 replicates) is shown next to the branches the pore size of the plant cell wall and assist in (Felsenstein 1985). The MP tree was obtained the penetration of enzyme molecules. Oxidation using the close-neighbor-interchange algorithm of oxalic acid by MnIII produces a formate radical − (Nei and Kumar 2000) with search level 1 in (HCO2 ) that reacts with dioxygen to form super- − which the initial trees were obtained with the oxide (O2 ) and subsequently H2O2 (Khindaria random addition of sequences (ten replicates). et al. 1994; Urzua et al. 1998). Chelated MnIII in Parsimony analysis revealed four distinct main turn acts as low molecular weight, diffusible evolutionary lineages among all basidiomycetes redox mediator that attacks phenolic lignin (Fig. 6.2). The MnP from Pleurotus ostreatus and structures and monomeric phenols, e.g., azo Pleurotus sp. Florida evidenced two additional dyes, resulting in the formation of unstable free lineages as they were grouped separately from radicals that tend to disintegrate spontaneously. the main cluster. The MnPs from Pleurotus pul- Characteristic features of the catalytic cycle of monarius, Hygrophorus agathosmus, Laccaria MnP resemble to those of other heme-containing bicolor S238N-H82, Agrocybe praecox, Phelli- peroxidases, such as horseradish peroxidase nidium ferrugineofuscum, and Hymenochaete (HRP), lignin peroxidase, versatile peroxidase, corrugata are evolutionarily closely related, and and chloroperoxidases (Cai and Tien 1993; they arose as the sister group to the other remain- Magliozzo and Marcinkeviciene 1997; Longoria ing groups (Fig. 6.2). et al. 2008). However, MnP is unique in its ability to utilize MnII as a reducing substrate to oxidize it to MnIII (Kishi et al. 1994; Sundaramoorthy et al. Mechanism of Catalysis 1997; Youngs et al. 2000; Deguchi et al. 2002). Spectroscopic studies have revealed that the Manganese peroxidase (MnP) [EC 1.11.1.13, heme iron of the native enzyme is in the ferric, MnII:hydrogen-peroxide oxidoreductase, MnP] is high-spin, pentacoordinate state and is ligated to an extracellular heme enzyme that utilizes hydro- the proximal histidine (Mino et al. 1988; Wariishi gen peroxide (H2O2) as electron-accepting cosub- et al. 1988; Gelpke et al. 2000). The reactions 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 71

33 Cytidia salicina 19 Phanerochaete sordida Ceriporiopsis subvermispora

37 Phanerochaete flavidoalba 91 Phlebiopsis gigantea Pulcherricium caeruleum 84 Hapalopilus rutilans 25 Phanerochaete chrysosporium Cryptoporus volvatus

88 67 Phlebia sp. b19 43 Dichomitus squalens 59 Phlebia chrysocreas Gomphus clavatus 57 Lentinula edodes Fomitiporia mediterranea 33 Hymenochaete corrugata 57 Phellinidium ferrugineofuscum Agaricus bisporus 11 Pleurotus ostreatus 58 Pleurotus sp. Florida 98 Agrocybe praecox Laccaria bicolor S238N-H82 Hypholoma fasciculare Cortinarius infractus 94 Cortinarius armillatus 51 Cortinarius malachius 95 Cortinarius hinnuleus 39 78 Cortinarius traganus 34 Hygrophorus agathosmus Pleurotus pulmonarius 49 Hericium erinaceum 22 Phlebia albomellea Phylloporia ribis 38 37 Ceriporiopsis rivulosa Spongipellis sp. FERMP-18171 14 Bjerkandera adusta Phlebia radiata Phlebia sp. MG60 33 Polyporus brumalis Trametes cinnabarina 73 29 Coriolopsis gallica Trametes versicolor Lenzites gibbosa 66 Trametes gibbosa Inonotus hispidus Lactarius fulvissimus 99 Lactarius rufus 99 Arthromyces ramosus Coprinopsis cinerea Coprinellus disseminatus 81 Ganoderma formosanum Ganoderma lucidum 93 Ganoderma applanatum 94 Ganoderma australe

0.45 0.40 0.35 0.30 0.25 0.20 0.15 0.10 0.05 0.00

Fig. 6.1 Neighbor-joining tree of manganese peroxidases amino acid sequences. Numbers on the branches are bootstrap values (jackknife values in parentheses) obtained for 1,000 pseudoreplications 72 S. Saroj et al.

38 Trametes versicolor 9 Coriolopsis gallica Trametes gibbosa 15 45 Lenzites gibbosa 54 Polyporus brumalis 38 Trametes cinnabarina 28 Phlebia radiata 30 33 Phlebia sp. MG60 Bjerkandera adusta 13 Spongipellis sp. FERM P-18171 31 Ceriporiopsis rivulosa Phylloporia ribis 6 6 Hericium erinaceum 31 Phlebia albomellea

3 Pleurotus pulmonarius 13 Hygrophorus agathosmus 96 Laccaria bicolor S238N-H82 Agrocybe praecox 28 Hypholoma fasciculare 85 Cortinarius infractus 5 53 Cortinarius hinnuleus 81 Cortinarius traganus 42 Cortinarius malachius 53 Cortinarius armillatus Inonotus hispidus 99 Lactarius rufus 10 Lactarius fulvissimus 7 99 Arthromyces ramosus 33 45 Coprinopsis cinerea Coprinellus disseminatus 69 63 Ganoderma lucidum Ganoderma formosanum 97 Ganoderma australe 33 68 Ganoderma applanatum Agaricus bisporus 37 Phellinidium ferrugineofuscum 28 Hymenochaete corrugata Fomitiporia mediterranea 55 Lentinula edodes 21 Gomphus clavatus 42 Dichomitus squalens 42 Phlebia chrysocreas 46 77 Phlebia sp. b19 Cryptoporus volvatus 88 Phanerochaete flavidoalba 66 Phlebiopsis gigantea 18 Ceriporiopsis subvermispora 44 19 Phanerochaete sordida Cytidia salicina 31 Pulcherricium caeruleum 41 Phanerochaete chrysosporium 53 Hapalopilus rutilans Pleurotus sp. Florida Pleurotus ostreatus

Fig. 6.2 Maximum parsimony analysis of taxa from MnP amino acid sequences. Numbers on the branches are bootstrap values (jackknife values in parentheses) obtained for 1,000 pseudoreplications 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 73

Fig. 6.3 MnP-catalyzed oxidation of phenolic arylglycerol β-aryl ether lignin model compound (Modified according to Tuor et al. 1992) involved in the MnP catalytic cycle are (Wariishi reactive radicals in the presence of a second et al. 1992; Gelpke et al. 1999) mediator during the oxidation of non-phenolic substrates (Reddy et al. 2003). The presence of MnPH+®OMnPIH+ O 22 2 thiols, such as glutathione, mediates the oxidation of substituted benzyl alcohols and diarylpro- II III MnPI +®Mn MnPIIM+ n pane structures to their respective aldehydes by MnIII (Reddy et al. 2003). In these reactions, thi- II III MnPIIM+®nMnP ++Mn HO III 2 ols are oxidized to thiyl radicals by Mn , which subsequently removes hydrogen from the substrate to form a benzylic radical. The latter under- Catalysis of Phenolic Substrates goes successive nonenzymatic reactions like

addition of O2 at C1 position of benzylic radical During the oxidation of phenolic compounds, followed by loss of • OOH, and homolytic C–O phenoxy radical intermediates are formed which fission at C2 of benzylic radical expels a phenoxy undergo rearrangements, bond cleavages, and radical that results in the formation of final prod- nonenzymatic degradation to yield various break- ucts (Fig. 6.4) (Wariishi et al. 1989b, c). down products (Fig. 6.3) (Tuor et al. 1992). MnIII Manganese peroxidase generated MnIII has generated by MnP is known to catalyze the oxi- also been coupled with peroxidation of lipids to dation of phenolic substrates, including simple catalyze Cα–Cβ cleavage and to β-aryl ether cleav- phenols, amines, dyes, and also phenolic lignin age of non-phenolic diarylpropane and β-O-4 lig- substructure and dimers (Wariishi et al. 1989a; nin structures, respectively (Fig. 6.4) (Bao et al. Urzua et al. 1995). 1994; Daina et al. 2002; Reddy et al. 2003; Kapich et al. 2005). The steps involved in the mechanism are the following: firstly, hydrogen abstraction Catalysis of Non-Phenolic Substrates from the benzylic carbon via lipid peroxy radicals, and secondly, peroxy radicals are formed

In contrast to LiP-catalyzed reactions, which by addition of O2 and subsequent oxidative cleav- involve electron abstraction from the aromatic age and nonenzymatic degradation. Absence of III ring, forming a radical cation, Mn forms exogenous H2O2 directs the enzyme to oxidize 74 S. Saroj et al.

Fig. 6.4 MnP-catalyzed oxidation of non-phenolic β-O-4 lignin model compound (Modified according to Wong 2009) nicotinamide adenine dinucleotide phosphate et al. 1994). Sakamoto et al. (2009) studied the (NADPH) (reduced form), glutathione, dithioth- transcriptional and translational levels of dif- reitol, and dihydroxymaleic acid to generate ferent isoforms of MnP, i.e., lemnp1 and

H2O2. One can infer (by observing the oxidase lemnp2, in Lentinula edodes using sawdust activity of MnP) that the H2O2 produced may medium and reported lemnp2 as the major become available for the enzyme to start the extracellular enzyme. peroxidase cycle, thereby, assisting in the degra- Heterologous expression of MnP from dation of lignin by this fungal species (Paszczynski Phanerochaete chrysosporium has been reported et al. 1986). in Aspergillus niger, Pichia pastoris, P. chrysosporium adel, Aspergillus oryzae, and Aspergillus nidulans (Pribnow et al. 1989; Mayfield et al. Molecular Cloning and Expression 1994; Stewart et al. 1996; Janse et al. 1998; of MnP Larrondo et al. 2001; Gu et al. 2003). MnPs from Trametes versicolor, P. eryngii, P. ostreatus, In most fungi, MnP appears to be produced as a Dichomitus squalens, and Ceriporiopsis subver- family of isoenzymes, which may be encoded by mispora have also been cloned and heterolo- structurally related genes (Larrondo et al. 2001; gously expressed in T. versicolor 9522-1, Coprinus Sakamoto et al. 2009). Commercial applications cinereus, P. chrysosporium, and Aspergillus of MnP require significantly higher levels of nidulans, respectively (Ogawa et al. 1998; extracellular enzyme production by fungal Camarero et al. 2000; Li et al. 2001; Larrondo strains; however, the yield of MnP in its native et al. 2001; Kim et al. 2005; Yeo et al. 2007). To hosts is too low (1.5–5 mgl−1) (Stewart et al. date there have been no reports of successful 1996; Li et al. 2001). Therefore, improvement in expression of fungal MnPs in bacterial expres- the yield and reduction in the production cost are sion hosts. the major goals to be focused for commercial Although the MnP production levels have often exploitation of MnP. been improved significantly by expression in heter- The first extracellular fungal MnP that was ologous hosts, the reported levels are still rather characterized and homologously expressed was low (100–400 mgl−1) for use in industrial applica- obtained from Phanerochaete chrysosporium tions (Conesa et al. 2000; Punt et al. 2002; Espinosa (Glenn and Gold 1985). It was found that high et al. 2012). The capability of the P. pastoris strain concentrations of carbon and nitrogen in the to perform various posttranslational modifications, medium significantly affected the transcription such as heme insertion, glycosylation, folding, and of the mnp l gene in P. chrysosporium (Mayfield protein secretion, had been reported for successful 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 75 production of active MnP to a maximum yield of the mechanism of transcriptional regulation by 120 UL−1 (Gu et al. 2003). As described by Yeo manganese in basidiomycetes. et al. (2007), increase in MnP activity of transformant TF6 was up to 45 % as compared to the recip- ient strain. A thermostable recombinant MnP from Characteristic Features of MnP Gene D. squalens has also been heterologously expressed in P. chrysosporium and purified. The recombinant MnP genes among different basidiomycetes protein appeared similar in kinetic and spectral (starting from the ATG codon) encompass a characteristics to the wild-type MnP from D. squa- genomic region of 1.4–1.9 kbps. Lobos et al. lens (Li et al. 2001), whereas the homologous (1998) had shown that MnP genes contain seven expression of P. chrysosporium recombinant MnP short intervening sequences with sizes ranging resulted in 30 % of the level of MnP activity between 52 and 60 bp. The last intron restrained expressed under by wild-type strain (Mayfield by mnp1 and mnp2 genes of P. chrysosporium et al. 1994). Data indicated that the addition of segregates a codon for proline to give different exogenous MnII, CdII, and ZnII conferred additional isozymes (Fig. 6.5) (Godfrey et al. 1990; thermal stability to MnP from D. squalens and Mayfield et al. 1994). Sequences at the intron P. chrysosporium (Li et al. 2001). Aspergillus splicing junctions adhere to the GT–AG rule. In species have proven to be excellent hosts for the turn, three putative internal lariat formation sites expression of heterologous proteins such as those match the consensus sequence CTRAY (Padgett of A. oryzae. The same has been shown to be effec- et al. 1989). After structural comparison of mnp tive in the expression of P. chrysosporium mnp1 genes from the Cs-mnp1 gene of C. subvermis- with notable yields (Stewart et al. 1996). pora and fivemnp genes from different basidio- Isozyme multiplicity of MnPs has been mycetes, Lobos et al. (1994) revealed an almost observed in various strains of C. subvermispora perfect alignment between Cs-mnp1 and mnp2 of and P. chrysosporium. As an expression host, P. chrysosporium (Fig. 6.5), and both genes have Aspergillus nidulans proved to be convenient sys- an additional intron splitting exon 3 of mnp1 and tem for MnPs. It has also been demonstrated that mnp3 of P. chrysosporium at the codon for the MnII is the key component that regulates the tran- distal histidine, H46 (Mayfield et al. 1994). The scription of different recombinant MnP isoforms pattern of differential distribution of introns in carbon-limited cultures (Banci et al. 1992; observed in the P. chrysosporium mnp genes Alic et al. 1997; Larrondo et al. 2001). Nutrient might be because the mnp2 represents the ances- limitation such as Mn concentration, culture agi- tral gene structure and the mnp1and mnp3 genes tation, heat shock, H2O2 concentration, and other arose with the loss of one intron each (Alic et al. chemical stresses have also been reported to 1997). In contrast, only one of the 15 introns significantly regulate transcription of different (intron 12) in the P. ostreatus mnp gene aligns isozymes of MnP (mnp1, mnp2, and mnp3) in exactly with a C. subvermispora mnp1 intron P. chrysosporium (Janse et al. 1998). In contrast (intron 6), and none of the 15 introns of the to manganese regulation, Tello et al. (2000) have P. ostreatus mnp gene align precisely with introns proposed that the putative MREs (metal response of the P. chrysosporium mnp genes. elements) found in the upstream region of MnP The regulatory sequence of Cs-mnp2B contains genes in P. chrysosporium might have a role in the a TATA box and an inverted CAAT (ATTG) ele- regulation of transcription of genes coding for ment located 92 and 191 bp upstream of the ATG MnP in filamentous fungi. Metallothionein genes codon (Tello et al. 2000), respectively. Examination are known to be regulated by MREs through vari- of the promoter regions of the mnp1 and mnp2 ous metals in animal cells, although these sites do genes (Godfrey et al. 1990; Gold and Alic 1993; not respond to manganese. Therefore, further Mayfield et al. 1994) revealed the presence of work is required to decipher both the role of putative MREs within 800 bp of the translation MREs in the upstream region of these genes and initiation codon. These sequences are identical 76 S. Saroj et al.

Fig. 6.5 Intron/exon structure of MnP genes mnp1, T. versicolor. The exons are indicated by open boxes, mnp2, and mnp3 from P. chrysosporium, Cs-mnp1 from whereas the solid black boxes correspond to the introns C. subvermispora, mnp from P. ostreatus, and mpg1 from (Modified after Lobos et al. 1998) to cis-acting MRE sequences responsible for Orth and coworkers (1994) have analyzed the heavy-metal induction of animal cell metallothio- organization of the MnP gene family of P. chrys- nein genes (Gettemy et al. 1998). Interestingly, osporium BKM1767 and concluded that the closer examination of the mnp1 promoter region λMP-1 and λMP-2 genes hybridized to 3.6 and also revealed the presence of putative HSEs within 3.8 Mb of DNA fragments located on separate 400 bp upstream of the mnp1 translation initiation chromosomes and in contrast to five LiP genes codon (Lobos et al. 1998). Mn2+ regulation of that are localized to a dimorphic chromosome of MnPs has been previously highlighted by different about 3.7 and 3.5 Mb (Gaskell et al. 1991). research groups (Bonnarme and Jeffries 1990). Godfrey et al. (1990) stated that 1,500 bp of sequence immediately upstream of the MnP trans- Crystallographic Analysis of MnP lation start site is sufficient to regulate theural reporter in a manner analogous to the regulation of MnP from different basidiomycetes has been crys- the endogenous MnP genes with respect to Mn, tallized and subsequently analyzed (Poulos et al. nutrient nitrogen levels, and metabolic phase of 1993; Sundaramoorthy et al. 1994a, b, 1995; Duenas growth. et al. 1999). MnP is a glycoprotein with the molec- The translocation of 48-bp fragment of pro- ular weight of 46 kDa and contains one heme group moter region of MnP isozyme 1 to a site 120 bp (Sundaramoorthy et al. 1994a, b). The structural downstream of its original location has been features of manganese peroxidase (pdb 1mnp) shown to regulate Mn2+-dependent expression of from P. chrysosporium are displayed/shown in downstream genes; this suggests the possibility Fig. 6.6, with a resolution of 2.06 Å (retrieved from of the presence of at least one Mn2+-responsive http://www.rcsb.org/pdb). The substrate-bound cis element in the fragment. However, deletion of MnP (Mn–MnP) consists of 357 amino acids, a 48-bp fragment, located at 521 bp upstream of three sugar residues, a heme prosthetic group, the translation start codon in the mnp1 promoter, two structural calcium ions, substrate MnII ion, and or replacement of this fragment with an unrelated 478 solvent molecules, including two glycerol sequence resulted in egfp expression under nitro- molecules (Sundaramoorthy et al. 2010). The gen limitation, both in the absence and presence sequence of the heme distal helix Glu35–Ala48 is of exogenous Mn2+ (Ma et al. 2004). highly conserved in the key residues (Selvaggini 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 77

to the presence of a Mn binding site (involving Glu36, Glu40, and Asp179) which enables the oxidation of this cation by the internal heme propionate (Sundaramoorthy et al. 1997). Multiple sequence alignment of amino acid sequences (retrieved from www.ncbi.nlm.nih.gov/ protein/) conducted on several characteristic peroxidases (P. chrysosporium MnP (378 amino acids) and LiP (371 amino acids), T. versicolor MnP (365 amino acids) and LiP (372 amino acids), P. ostreatus MnP (361 amino acids), P. brasiliensis Pb01 cytochrome c peroxidase (374 amino acids), and P. eryngii versatile peroxidase (370 amino acids), Brassica rapa TP7 (296 amino acids), and Fig. 6.6 The overall structure of P. chrysosporium MnP (pdb 1mnp) as analyzed by UCSF chimera 1.4.1. The Arthromyces ramosus peroxidase (364 amino cyan spheres are structural CaII ions conserved in extracel- acids)) indicated significant conserved structural lular heme peroxidases. The location of the substrate, residues (Fig. 6.7). All the enzymes have the con- +2 Mn , near the heme, is indicated. The purple color shows served proximal histidine, the distal histidine and active-site structure of MnP. This architecture is highly conserved in heme peroxidase the distal arginine. The distal arginine residue has been proposed to participate in the formation of the peroxide-binding pocket together with distal et al. 1995). The active site consists of a proximal histidine, while Asn131 has been revealed to be His ligand H-bonded to an Asp residue, and a distal the only carbohydrate binding site in MnP1 side peroxide-binding pocket consisting of a cata- (Sundaramoorthy et al. 1994a, b, 2005). The iron lytic His and Arg is the same among all peroxidases coordination and the residues involved in the (Fig. 6.7). MnP differs with respect to having five active site are the most conserved region among all rather than four disulfide bonds. The additional the aforementioned fungal peroxidases, as well as disulfide bond is located near the C-terminus of the cytochrome c peroxidase. The two glutamic acid polypeptide chain. The ligands constituting the and the aspartic acid residues present in the man- Mn2+ binding site include Asp179, Glu35, Glu39, ganese binding site are also conserved. It could be a heme propionate, and two water molecules. The inferred from Fig. 6.7 that heme binding sites are overall structure is similar to that of two other conserved among all the characteristic peroxidases fungal peroxidases, i.e., lignin peroxidase from with some variations found in the case of the tur- Phanerochaete chrysosporium and Arthromyces nip (Brassica rapa) and cytochrome c peroxidases. ramosus peroxidase. Like the other fungal peroxi- The distal Ca2+-binding residues are conserved in dases, MnP also has two structural calcium ions all fungal peroxidases (coordination being com- and N-acetylglucosamine residues N-linked to pleted by two water molecules), whereas some dif- Asn131 (Sundaramoorthy et al. 1994a, b). ferences exist in the residues binding Ca2+ at the Proximal pocket of MnP active site contains proximal side. His173 that coordinates to heme atom and the side chain of Asp242 interacts with His173 through H bond. The Phe190 residue is positioned Comparative Analysis of MnP, LiP, just below the heme ring, near the His173 residue. and VP There is a second opening between the heme propionate residues; this region is polar and char- Comparative studies were performed between acterized by residues Arg177, Asp179, Glu35, P. chrysosporium MnP (pdb 1mnp), T. cervina and Glu39 (Selvaggini et al. 1995). Its ability to LiP (pdb 3q3u), and P. eryngii VP (pdb 3fmu) on the oxidize Mn2+ with high substrate affinity is related basis of tertiary structure alignment using UCSF 78 S. Saroj et al.

Fig. 6.7 Multiple alignments of amino acid sequences of conserved residues in the substrate binding site. Blank boxes fungal peroxidases. MnP, LiP, VP, CcP TP, and other peroxi- encompass the common residues in heme and substrate dases from P. chrysosporium (Pc) and T. versicolor (Tv), binding sites. The following marks are used in the consensus P. ostreatus (Po), P. brasiliensis Pb01 (Pb), P. eryngii (Pe), line: Mn2+ binding site (▼); distal and proximal histidine Brassica rapa turnip (turnip), and A. ramosus (ArP) are (■). Accession numbers of the amino acid sequences are as compared. The alignment was generated and conserved given below: TvLiP, AAA34049.1; TvMnP, CAA83148.1; residues were identified by using Clustal X 2.0.11. The PcLiP, AAA03748.1; PeVP, AAD54310.1; PoMnP, sequences highlighted in red denote heme binding sites; cal- AAA84397.1; ArP, P28313.3; PcMnP, AAA33743.1; cium binding sites are highlighted in blue. Green shows Turnip, P00434.3; PbCcP, EEH41729.1 chimera 1.4.1 software (Fig. 6.8). Structural manganese peroxidase does not directly bind alignment depicted approximately 50 % similar- aromatic substrates; ity of the residues among these peroxidases. (c) the location of residues potentially able to bind Several functionally relevant structurally impor- Mn2+, spatially positioned on the side of the tant features were also observed: 3-CH3 heme edge. The close sequence similar- (a) the very close structural similarity between ity of all the peroxidases strongly suggests a MnP, LiP, and VP active sites, suggesting a very similar three-dimensional fold. Lignin similar mode of hydrogen-peroxide activa- peroxidase was characterized by two calcium tion. The heme prosthetic group is found binding sites involving the side-chain residues, embedded between the N-terminal and the i.e., Asp194, Ser177, Thr196 (proximal site) C-terminal domains, along with the surround- and Asp65, Asp86 (distal site). The corre- ing conserved residues H46 (MnP)/H47 sponding residues of aligned manganese per- (LiP)/H47(VP), H173 (MnP)/H175 (LiP)/ oxidase were Asp198, Thr199, Thr219 and H169(VP), and R42 (MnP)/R43 (LiP)/ Asp64, Asp85 and of versatile peroxidase were R43(VP); Ser195, Asp194, Lys215 and Asp65, Asp86; (b) the substitution of polar residues for the (d) the high degree of conservation of the aspartic hydrophobic amino acids exposed at the edge acid residue at N + 1 site position after the dis- of the channel involved in substrate recogni- tal histidine suggests that the calcium distal tion in lignin peroxidase, suggesting that site is involved in maintaining the integrity of 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 79

H47

R43 H46 H47

R43

HEME

R42

H173

H169

H175

Fig. 6.8 Tertiary structural comparison of MnP, LiP, and VP from left to right. (b) Structural alignments of MnP, LiP, using UCSF chimera 1.4.1 software. (a) Whole structures of and VP are shown in pink, cyan, and green, respectively. P. chrysosporium MnP (pdb 1mnp), T. cervina LiP (pdb (c) Structural alignment of the region surrounding heme and 3q3u), and P. eryngii VP (pdb 3fmu) are shown respectively the MnP, LiP, and VP residue numbers are shown

the active site is in accordance with Henrisatt Superimposition of the MnP, LiP, and VP struc- et al. (1990) and Welinder (1992); tures clearly depicts the presence of five disulfide (e) the sequence corresponding to the solvent- bonds in MnP in contrast to four disulfide bonds exposed a helix Ala12–Gln25, connected with present in the LiP and VP. The initial four disul- other portions of the protein by disulfide fide bonds in MnP, viz., Cys3–Cys15, Cys33–Cys117, bridges involving Cys3–Cysl5 and Cysl4– Cys14–Cys289, and Cys253–Cys319, are the Cys289, is closely similar to the Ala12–Gln25 same as observed in Lip, VP, and A. ramosus segment of LiP and VP with some closely peroxidase and also had perfect alignment as por- related mutations. The coil region located at trayed by superimposed structures in Fig. 6.8b. residues Ala50–Gly65 in the MnP sequence The additional disulfide bond found in MnP was not completely conserved in the lignin (Cys341–Cys348 located near the C-terminus of peroxidase and versatile peroxidase. The the polypeptide chain) aids in the formation of presence of glycine and proline in this region the MnII binding site and is responsible for push- strongly suggests a coil arrangement (Benner ing the C-terminus segment away from the main 1989; Benner and Gerloff 1990; Branden and body of the protein (Sundaramoorthy et al. 1994a, Tooze 1991). The region Pro163–Val175 b). LiP, VP, and MnP all have a Phe position 190, (heme proximal helix) is highly similar to the but the orientation differs. In LiP the Phe ring is aligned segments of lignin peroxidase and nearly parallel to the proximal His imidazole versatile peroxidase sequences except for ring, whereas in MnP the Phe ring is almost per- some minor changes in this region. pendicular to the plane of the proximal His. 80 S. Saroj et al.

Table 6.2 Helices distribution Helices among MnP, LiP, and VP S.No. MnP LiP VP 1. Ala16-Ile28 Asn11-Leu28 Asn11-Leu28 2. Gly34-Ile49 Glu35-Val50 Gly35-Ile50 3. Gly65-Phe70 Ser53-Ala59 Gly63-Ala79 4. Ile83-His97 Gly69-Thr79 Ile81-His95 5. Ser101-Ser115 Gly86-Gly102 Ser98-Asn113 6. Ser147-Gly159 Ser104-Asn119 Ser144-Ala155 7. Thr162-Leu170 Ser150-Asp160 Ser158-Ile171 8. Thr199-Val205 Ser164-Ile177 Ser195-Thr201 9. Leu239-His247 Asp184-Asp188 Gln229-Asp237 10. Thr251-Gly257 Thr201-Leu208 Thr240-Met247 11. Glu261-Ala277 Gln234-Asp242 Asn250-Leu267 12. – Thr245-Val253 Asp271-Leu275 13. – Asn255-Ala271 Ser279-Ile282 14. – Ile276-Leu280 Ser299-Val303 15. – Ser284-Ile287 – 16. – Ser306-Val310 –

Owing to the differences in orientation of the nities to study unique structural/functional outer propionate in MnP, the distal Arg42 cannot relationships in MnPs and allows the detailed form a hydrogen bond with the propionate, characterization of the MnII binding site. whereas in LiP, the distal arginine directly inter- Kishi et al. (1994) developed a series of acts with the propionate. The second (inner) pro- mutants (E35Q, E39Q, and E35Q-D179N) from pionate interacts with Mn2+ ion and water the gene encoding manganese peroxidase isozyme molecules, and the peptide NH group of a propio- 1 (mnp1) from Phanerochaete chrysosporium, nate together with a main chain peptide nitrogen using site-directed mutagenesis. The mutations is also found in LiP. demonstrated that changing any of the acidic The residues forming the helices found to be amino acid MnII ligands, Asp179, Glu35, or structurally aligned in MnP, LiP, and VP models Glu39, significantly affects the oxidation of MnII, are summarized in Table 6.2. most probably by decreasing the affinity of the enzyme for MnII. Asp179, Glu35, and Glu39 residues at the catalytic site are essentially required MnP and Active-Site Engineering for MnII oxidation, since the double mutation, i.e., E35Q–D179N, had almost completely resulted Major challenges remain in understanding the into the loss of MnII oxidation (Sundaramoorthy role of functional domains and their structural/ et al. 1994a, b). The coordination of MnII at this functional relationships. The desired features site is octahedral, which is typical of MnII coor- required for the commercial exploitation of dination complexes (Demmer et al. 1980). It these enzymes are stability, notable yields, and has been postulated that Glu39 is a Mn ligand enhanced or superannuated activities. Site- and its precise geometry within the Mn binding directed mutagenesis (SDM) or active-site engi- site of MnP is essential for the efficient binding, neering is an invaluable tool to alter bases at oxidation, and release of Mn by this enzyme and precise positions in the gene. Engineered that mutation of this ligand decreases both the enzymes are then subjected to analysis for the Mn binding and the rate of Mn oxidation above-mentioned favorable characteristics using (Martınez 2002; Li et al. 2001). various molecular tools. Site-directed mutagen- Miyazaki and Takahashi (2001) found that the esis is a major approach that provides opportu- site-directed mutagenesis of oxidizable Met273 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 81

located near the H2O2-binding pocket to a non- to the proximal His of MnP, in other peroxidases, oxidizable Leu had resulted into improved stabil- and this conserved Asp, in turn, is hydrogen ity of IMnP, as it retained more than 60 % of its bonded to a Trp. In MnP and other fungal peroxi- initial activity in presence of 1 mM H2O2 and dases, the Trp is replaced by a Phe (F190). Both more than 30 % at a concentration of 3 mM H2O2 residues are thought to have a direct influence on as compared to wild type that was completely the catalytic center of the enzyme. Mutagenesis inactivated by 1 mM H2O2. Stability in the pres- of D242 and F190 has shown that these residues ence of hydrogen peroxide may be attributed to affect the reactivity of the heme active site. The the above-mentioned mutation that makes it changes in the axial ligand H-bonding network resistant to oxidation by the conformational largely influence the reactivity of compound II 4+ stabilization around the H2O2-binding pocket. (Fe ) and have little influence on the reactivity of Manganese peroxidase (MnP) is susceptible to compound I (porphyrin cation radical). thermal inactivation due to the loss of calcium. Zhang et al. (2009) described the role of Arg42 Engineering of a disulfide bond near the distal and Asn131 in the oxidation of 2,6-DMP after calcium binding site of MnP by double mutation performing site-directed mutagenesis with in A48C and A63C showed the improvement in vitro synthesis. As previously described, R177A thermal stability as well as pH stability in com- and R177K mutants of P. chrysosporium had spe- parison to native MnP. The disulfide bond adja- cifically altered binding of Mn, whereas the rate cent to the distal calcium ligands Asp47 and of electron transfer from Mn2+ to the oxidized Asp64 stabilizes the recombinantly expressed heme was apparently not affected (Gelpke et al. MnP against the loss of calcium (Reading and 1999). Whitwam et al. (1997) had suggested that Aust 2000). Arg177 may anchor to the carboxylate of Glu35 Timofeevski et al. (1999) have described that in the Mn2+-occupied closed configuration of the a single mutation (S168W) in rMnP added protein. Shortening the side chain of this residue veratryl alcohol oxidase activity to the enzyme by one methylene in the E35D mutant probably without significantly affecting Mn2+ oxidase does not affect the salt bridge to Arg177, but it activity. This surface tryptophan residue, present may restrain the carboxylate of this ligand from in various LiP isoenzymes but absent in MnP, making a strong bond with the MnII atom. This may be the site of VA binding and oxidation by results in a disruption of the ligation for MnII and LiP. Other research conducted observed how hence in the electron-transfer rate as observed by the hydrophobicity of the heme pocket could Gelpke et al. (1999). affect the reactivity of compound I (formed during MnP catalysis). Leu169 and Ser172 were mutated and converted Phe and Ala, respectively. MnPs and Industrial/Commercial Steady-state kinetics characterization indicated Applications that the Leu169Phe mutation had little effect on activity, whereas the Ser172Ala mutation decreased Industrial applications for manganese peroxi- kcat to 45 s−1 as compared to wild type (449 s−1) dases that have been proposed include bleaching and also the specificity constant (kcat/Km) of of unbleached kraft pulp in pulp and paper indus- Ser172Ala mutant decreased from 1.1 × 107 M−1s−1 tries, treatment of textile industry effluents, in 5 −1 −1 2+ to 5.3 × 10 M s for Mn , but not H2O2. It has generating natural aromatic flavors in foods, deg- been shown/demonstrated that compound II is radation of environmental pollutants like polyaro- the most sensitive to changes in the heme envi- matic hydrocarbons (PAHs), azo dyes, TNT, and ronment when compared to compound I (Balay DTT. Manganese peroxidase is versatile and et al. 2000). energy saving, and its ability to bioremediate dis- The role of the axial ligand hydrogen-bonding plays the capacity of this enzyme to be a signifi- network on heme reactivity was analyzed by cant tool for a number of eco-friendly commercial Whitwam et al. (1999); D242 is hydrogen bonded applications. 82 S. Saroj et al.

Ecotoxic organic chemicals generated from 1996; Sasaki et al. 2001). In vitro depolymeriza- textile, pulp, and paper industry effluents are major tion studies using LiP and MnP showed that the contributors for the environmental pollution and enzymes were able to degrade to a variety of are the major health hazards for a number of verte- aromatic substrates (Conesa et al. 2002). brates and to human population. Detoxification of MnP also demonstrated the ability to deco these compounds poses an immense technical lorize a range of azo and anthraquinone dyes, as challenge (Evans et al. 2004; Brar et al. 2006). well as textile industry effluents in aqueous cul- Conventional methods for treatment of various tures and in packed bed bioreactors (Robinson industrial effluents include physical (adsorption, et al. 2001; Mielgo et al. 2001; Kasinath et al. membrane filtration, ion exchange, irradiation, etc.) 2003; Shin 2004; Yang et al. 2004; Snajdar and and chemical (oxidation, coagulation, electrochemi- Baldrian 2007; Asgher et al. 2008; Sedighi et al. cal, etc.) processes; these methods have earlier been 2009). Susla et al. (2008) have evaluated the reviewed extensively (Hao et al. 2000; Forgacs contribution of MnP from D. squalens and lac- et al. 2004; Joshi et al. 2004; Kuhad et al. 2004). case in degradation of azo, anthraquinone, phtha- The major drawbacks of physicochemical approa- locyanine, and oligocyclic aromatic dyes. MnP ches are that these are prohibitively expensive, less has been observed to be capable of decolorizing efficient, not versatile, and have interference by the mixture of azo dyes at a concentration range other wastewater constituents. Biological methods of 10–200 mgl−1 each (Singh and Pakshirajan consisting of biosorption, biodegradation, and 2010), and also MnP from P. chrysosporium sp. enzymatic processes are eco-friendly, simpler, HSD has been reported to rapidly decolorize a and cost-effective and are receiving greater atten- higher concentration (up to 600 mgl−1) of azo tion for treatment of industrial effluents (Kuhad dyes (Hailei et al. 2009). et al. 2004; Kaushik and Malik 2009). MnP has been shown to have the mineraliza- Manganese peroxidases are the part of the extra- tion ability for many environmental contami- cellular oxidative system which evolved in white- nants. Besides having ability to degrade azo, rot fungi for lignin degradation (Kirk and Cullen heterocyclic, reactive, and polymeric dyes 1991; Hatakka 1994; Vares and Hatakka 1996). (Champagne and Ramsay 2005), it can degrade Lignin is a heterogeneous, optically inactive 1.1.1-trichloro-2.2-bis-(4-chlorophenyl) ethane polymer consisting of phenylpropanoid inter- (DDT), 2.4.6-trinitrotoluene (TNT), and polycy- units, which are linked by several covalent bonds clic aromatic hydrocarbons (PAHs) too (Maciel (e.g., aryl-ether, aryl–aryl, carbon–carbon bonds) et al. 2010). (Hofrichter 2002). The structure of the lignin Manganese peroxidase (MnP) from two meta- polymer implies that lignolytic enzymes possess bolically distinct fungi Phanerochaete chryso- the ability to oxidize substrates of high redox sporium BKM-F-1767 (ATCC 24725) and potential in a nonspecific manner. Paper and pulp Bjerkandera sp. BOS55 (ATCC90940) has the industries employ a combination of chlorine- ability to degrade (98 %) anthracene to generate based chemicals and alkaline extraction multi- anthraquinone in organic solvent mixtures after stage procedures for bleaching of the kraft pulp. 6 h of operation under optimal conditions (Eibes However, chemical bleaching procedures end et al. 2005). Utilization of MnP (from the basid- up with chlorinated organic substances as by- iomycete Bjerkandera adusta) for acrylamide products which contain toxic, mutagenic, and polymerization has also been reported (Iwahara carcinogenic polychlorinated dioxins, dibenzofu- et al. 2000). MnPs from Phanerochaete chryso- rans, and phenols. Discharge of these organic sporium have also been employed in styrene deg- compounds into the effluent generates serious radation, an important industrial polymer used as environmental concern. Partially purified man- a raw material for wrapping and transporting ganese peroxidase in the presence of oxalate goods. Its disposal poses serious environmental preparations is known to be effective in decol- concerns (Soto et al. 1991; Lee et al. 2006). orizing kraft effluents and oxidizing a broad MnP a redox enzyme has the potential of range of xenobiotic compounds (Harazono et al. directly transferring the electrons to the electrodes. 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 83

This enables the use of this enzyme for various Balay KA, Dougherty M, Tien M (2000) Reactivity of applications in the development of biosensors, manganese peroxidase: site-directed mutagenesis of residues in proximity to the porphyrin ring. Arch designing effective biofuel cells, and for selective Biochem Biophys 382:89–94 bioorganic synthesis (Maciel et al. 2010). Banci L, Bertini I, Pease EA, Tien M, Turano P (1992) 1H NMR investigation of manganese peroxidase from Phanerochaete chrysosporium. A comparison with other peroxidases. Biochemistry 31(41):10009–10017 Future Perspectives Bao W, Fukushima Y, Jensen KA, Moen MA, Hammel KE (1994) Oxidative degradation of non-phenolic lig- This review highlights the various developments nin during lipid peroxidation by fungal manganese related to the molecular features, cloning, heter- peroxidase. FEBS Lett 354:297–300 Benner SA (1989) Patterns of divergence in homologous ologous production, crystal structure refinements proteins as indicators of tertiary and quaternary struc- of MnPs, and its possible industrial and biotech- ture. Adv Enzyme Regul 28:219–236 nological applications. MnPs are promising Benner SA, Gerloff D (1990) Patterns of divergence in enzymes and an eco-friendly alternative to the homologous proteins as indicators of secondary and tertiary structure: a prediction of the structure of the conventional physicochemical processes as pres- catalytic domain of protein kinases. Adv Enzyme ently employed for various such as the pulp and Regul 31:121–181 paper, textile, pharmaceutical, and for food Bonnarme P, Jeffries TW (1990) MnII regulation of lignin industries. However, a major challenge in the peroxidases and manganese dependent peroxidases from lignin-degrading white rot fungi. Appl Environ commercialization of the MnPs is due to its lower Microbiol 56:210–217 thermal stability. Despite recent progress, our Branden C, Tooze J (1991) Introduction to protein struc- understanding of the process is still limited due to ture. Garland Publishing, New York/London its substrate complexity and because of multi- Brar SK, Verma M, Surampalli RY, Misra K, Tyagi RD, Meunier N, Blais JF (2006) Bioremediation of hazard- plicity of the peroxidases. Manganese peroxi- ous wastes: a review. Pract Period Hazard, Toxic dases currently generated are so far not promising Radioact Waste Manag 10:59–72 enough for their commercial scale exploitation. Brown JA, Alic M, Gold MH (1991) Manganese peroxidase Although efforts have been made for improving gene transcription in Phanerochaete chrysosporium: activation by manganese. J Bacteriol 173:4101–4106 the thermal stability of MnPs, further efforts are Cai D, Tien M (1993) Lignin-degrading peroxidases of required for the development of designer enzymes Phanerochaete chrysosporium. J Biotechnol 30:79–90 with desired levels of thermal stability for its Cairney JWG, Burke RM (1998) Do ecto- and ericoid industrial applications. Therefore, the tailor- mycorrhizal fungi produce peroxidase activity. Mycorrhiza 8:61–65 made enzymes can be designed using a combina- Camarero S, Ruiz-Duen FJ, Sarkar S, Martinez MJ, tion of molecular approaches like site-directed Martinez AT (2000) The cloning of a new peroxidase mutagenesis, saturation mutagenesis, and found in lignocellulose cultures of Pleurotus eryngii directed evolution enabling the industrial exploi- and sequence comparison with other fungal peroxidases. FEMS Microbiol Lett 191:37–43 tation of the unique catalytic abilities of these Champagne PP, Ramsay JA (2005) Contribution of man- biocatalysts. The various developments in ganese peroxidase and laccase to dye decoloration by enzyme engineering research open a wide spec- Trametes versicolor. Appl Microbiol Biotechnol trum of possible applications in the near future. 69:276–285 Conesa A, van Den Hondel CA, Punt PJ (2000) Studies on the production of fungal peroxidases in Aspergillus niger. Appl Environ Microbiol 66:3016–3023 Conesa A, Punt PJ, vanden Hondel CAMJJ (2002) Fungal References peroxidases: molecular aspects and applications. J Biotechnol 93:143–158 Alic M, Akileswaran L, Gold MH (1997) Characterization Daina S, Orlandi M, Bestetti G, Wiik C, Elegir G (2002) of the gene encoding manganese peroxidase isozyme Degradation of β-5 lignin model dimers by 3 from Phanerochaete chrysosporium. Biochim Ceriporiopsis subvermispora. Enzyme Microb Biophys Acta 1338:1–7 Technol 30:499–505 Asgher M, Bhatti HN, Ashraf M, Legge RL (2008) Recent Deguchi T, Matsubara M, Nishida T (2002) NADH oxi- developments in biodegradation of industrial pollut- dation by manganese peroxidase with or without ants by white rot fungi and their enzyme system. alpha-hydroxy acid. Biosci Biotechnol Biochem Biodegradation 19:771–783 66(4):717–721 84 S. Saroj et al.

Demmer H, Hinz I, Keller-Rudex H, Koeber K, Hailei W, Ping L, Min P, Zhijun Z, Guangli Y, Guosheng Kottelwesch H, Schneider D (1980) In: Schleitzer- L, Jianming Y (2009) Rapid decolourization of azo Rust E (ed) Coordination compounds of manganese, dyes by a new isolated higher manganese peroxidase vol 56, 8th edn. Springer, New York, pp 1–185 producer: Phanerochaete sp. Hsd Biochem Eng J Duenas FJR, Martnez MJ, Martnez A (1999) Molecular 46:327–333 characterization of a novel peroxidase from the ligni- Hao OJ, Kim H, Chiang PC (2000) Decolourization nolytic fungus Pleurotus eryngii. Mol Microbiol of wastewater. Crit Rev Environ Sci Technol 31:223–235 30:449–505 Eibes G, Lú-Chau T, Feijoo G, Moreira MT, Lema JM Harazono K, Kondo R, Sakai K (1996) Bleaching of hard- (2005) Complete degradation of anthracene by wood kraft pulp with manganese peroxidase from Manganese Peroxidase in organic solvent mixtures. Phanerochaete sordida YK-624 without addition of

Enzyme Microb Technol 37:365–372 MnSO(inf4). Appl Environ Microbiol 62(3):913–917 Eriksson KEL, Blanchette RA, Ander P (1990) Microbial Hatakka A (1994) Lignin-modifying enzymes from and enzymatic degradation of wood and wood compo- selected white-rot fungi: production and role in lignin nents. Springer, Berlin/Heidelberg/New York degradation. FEMS Microbiol Rev 13:125–135 Espinosa DVC, Absalón AE, Sanchez N, Loera O, Hatakka A (2001) Biodegradation of lignin. In: Hofrichter Vazquez RR, Fernández FJ (2012) Heterologous M, Steinbüchel A (eds) Biopolymers. Lignin, humic expression of manganese peroxidase in Aspergillus substances and coal. Wiley-VCH, Weinheim, pp niger and its effect on phenanthrene removal from 129–180 soil. J Mol Microbiol Biotechnol 21:120–129 Heinzkill M, Bech L, Halkier T, Schneider P, Anke T Evans FF, Rosado S, Sebastian GV, Casella R, Machado (1998) Characterization of laccases and peroxidases PLOA, Holmstrom C, Kjelleberg S, Van Elsas JD, from wood-rotting fungi (family Coprinaceae). Appl Seldin L (2004) Impact of oil contamination and bios- Environ Microbiol 64:1601–1606 timulation on the diversity of indigenous bacterial Henrisatt B, Saloheimo M, Lavaitte S, Knowles JKC communities in soil microcosms. FEMS Microbiol (1990) Structural homology among the peroxidase Ecol 49:295–305 enzyme family revealed by hydrophobic cluster analysis. Felsenstein J (1985) Confidence limits on phylogenies: an Protein Struct Funct Genet 8:251–257 approach using the bootstrap. Evolution 39:783–791 Hofrichter M (2002) Review: lignin conversion by Forgacs E, Cserhati T, Oros G (2004) Removal of syn- manganese peroxidase (MnP). Enzyme Microb thetic dyes from wastewaters: a review. Environ Int Technol 30:454–466 30:953–971 Iwahara K, Hirata M, Honda Y, Watanabe T, Kuwahara M Gaskell J, Dieperink E, Cullen D (1991) Genomic organi- (2000) Free-radical polymerization of acrylamide by zation of lignin peroxides genes of Phanerochaete manganese peroxidase produced by the white-rot chrysosporium. Nucleic Acids Res 19:599–603 basidiomycete Bjerkandera adusta. Biotechnol Lett Gelpke SMD, Loccoz PM, Gold MH (1999) Arginine 177 22:1355–1361 is involved in MnII binding by manganese peroxidase. Janse BJH, Gaskell J, Akhtar M, Cullen D (1998) Biochemistry 38:11482–11489 Expression of Phanerochaete chrysosporium genes Gelpke MD, Youngs HL, Gold MH (2000) Role of encoding lignin peroxidases, manganese peroxidases, arginine 177 in the MnII binding site of manganese and glyoxal oxidase in wood. Appl Environ Microbiol peroxidase. Studies with R177D, R177E, R177N, and 64:3536–3538 R177Q mutants. Eur J Biochem 267(24):7038–7045 Johansson T, Nyman PO, Cullen D (2002) Differential Gettemy JM, Ma B, Alic M, Gold MH (1998) Reverse regulation of mnp2, a new manganese peroxidase- transcription-PCR analysis of the regulation of the encoding gene from the ligninolytic fungus Trametes manganese peroxidase gene family. Appl Environ versicolor PRL 572. Appl Environ Microbiol 68: Microbiol 64:569–574 2077–2080 Glenn JK, Gold MH (1985) Purification and characteriza- Joshi M, Bansal R, Purwar R (2004) Colour removal from tion of an extracellular MnII-dependent peroxidase from textile effluents. Indian J Fibre Text Res 29:239–259 the lignin-degrading basidiomycete, Phanerochaete Kapich AN, Steffen KT, Hofrichter M, Hatakka A (2005) chrysosporium. Arch Biochem Biophys 242:329–341 Involvement of lipid peroxidation in the degradation Godfrey BJ, Mayfield MB, Brown JA, Gold MH (1990) of a non-phenolic lignin model compound by manga- Characterization of a gene encoding a manganese per- nese peroxidase of the litter-decomposing fungus oxidase from Phanerochaete chrysosporium. Gene Stropharia coronilla. Biochem Biophys Res Commun 93:119–124 330:371–377 Gold MH, Alic M (1993) Molecular biology of the lignin Kasinath A, Novotny C, Svobodova K, Patel KC, Saseck degrading basidiomycetes Phanerochaete chrysospo- V (2003) Decolorization of synthetic dyes by Irpex rium. Microbiol Rev 57:605–622 lacteus in liquid cultures and packed-bed bioreactor. Gu L, Lajoie C, Kelly C (2003) Expression of a Enzyme Microb Technol 32:167–173 Phanerochaete chrysosporium manganese peroxidase Kaushik P, Malik A (2009) Fungal dye decolourization: gene in the yeast Pichia pastoris. Biotechnol Prog recent advances and future potential. Environ Int 19:1403–1409 35:127–141 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 85

Khindaria A, Grover TA, Aust SD (1994) Oxalate- of the basidiomycota: a review. Elect J Biotechnol dependent reductive activity of manganese peroxidase 13(6). (ISSN: 0717-3458) doi:org/10.2225/, full-text2 from Phanerochaete chrysosporium. Arch Biochem Magliozzo RS, Marcinkeviciene JA (1997) The role of Biophys 314:301–306 Mn(II)-peroxidase activity of mycobacterial catalase- Kim Y, Yeo S, Kum J, Song HG, Choi HT (2005) Cloning peroxidase in activation of the antibiotic isoniazid. J of a manganese peroxidase cDNA gene repressed by Biol Chem 272(14):8867–8870 manganese in Trametes versicolor. J Microbiol Martınez AT (2002) Molecular biology and structure- 43:569–571 function of lignin-degrading heme peroxidases. Kirk TK, Cullen D (1991) Enzymology and molecular Enzyme Microb Technol 30:425–444 genetics of wood degradation by white-rot fungi. Mayfield MB, Godfrey BJ, Gold MH (1994) Characterization Wiley, New York of the mnp2 gene encoding manganese peroxidase iso- Kishi K, Wariishi H, Marquez L, Dunford BH, Gold MH zyme 2 from the basidiomycete Phanerochaete chryso- (1994) Mechanism of manganese peroxidase II reduc- sporium. Gene 142:231–235 tion. Effect of organic acid and chelators and pH. Mielgo I, Moreira MT, Feijoo G, Lema JM (2001) A Biochemistry 34:8694–8701 packed-bed fungal bioreactor for the continuous deco- Kuhad RC, Sood N, Tripathi KK, Singh A, Ward OP (2004) lourisation of azo-dyes (Orange II). J Biotechnol Developments in microbial Methods for the treatment 89:99–106 of dye effluents. Adv Appl Microbiol 56:185–213 Mino Y, Wariishi H, Blackburn NJ, Loehrldx TM, Gold Kuwahara M, Glenn JK, Morgan MA, Gold MH (1984) MH (1988) Spectral characterization of manganese Separation and characterization of two extracellular peroxidase, an extracellular heme enzyme from the

H2O2-dependent oxidases from ligninolytic cultures lignin-degrading basidiomycete, Phanerochaete of Phanerochaete chrysosporium. FEBS Lett chrysosporium. J Biol Chem 263:7029–7036 169:247–250 Miyazaki C, Takahashi H (2001) Engineering of the

Larrondo LF, Lobos S, Stewart P, Cullen D, Vicuña R (2001) H2O2-binding pocket region of a recombinant manga- Isoenzyme multiplicity and characterization of recombi- nese peroxidase to be resistant to H2O2. FEBS Lett nant manganese peroxidases from Ceriporiopsis subver- 509(1):111–114 mispora and Phanerochaete chrysosporium. Appl Nei M, Kumar S (2000) Molecular evolution and phyloge- Environ Microbiol 67:2070–2075 netics. Oxford University Press, New York Lee JW, Lee SM, Hong EJ, Jeung EB, Kang HY, Kim Ogawa K, Yamazaki T, Hasebe T, Kajiwara S, Watanabe MK, Choi IG (2006) Estrogenic reduction of styrene A, Asada Y, Shishido K (1998) Molecular breeding of monomer degraded by Phanerochaete chrysosporium the basidiomycete Coprinus cinereus strains with high KFRI 20742. J Microbiol 44:177–184 lignindecolorization and β-degradation activities using Levin L, Papinutti L, Forchiassin F (2004) Evaluation of novel heterologous protein expression vectors. Appl Argentinean white rot fungi for their ability to produce Microbiol Biotechnol 49:285–289 lignin-modifying enzymes and decolorize industrial Orth AB, Rzhetskaya M, Cullen D, Tien M (1994) dyes. Bioresour Technol 94:169–176 Characterization of a cDNA encoding a manganese per- Li D, Youngs HLA, Gold MH (2001) Heterologous oxidase from Phanerochaete chrysosporium: genomic expression of a thermostable manganese peroxidases organization of lignin and manganese peroxidase- from Dichomitus squalens in Phanerochaete chryso- encoding genes. Gene 148:161–165 sporium. Arch Biochem Biophys 385:348–356 Padgett R, Konarska M, Grabowski P, Hardy S, Sharp P Lobos S, Larram J, Salas L, Cullen D, Vicuna R (1994) (1989) Lariat RNAs as intermediates and products in Isoenzymes of manganese dependent peroxidase and the splicing of messenger RNA products. Science laccase produced by the lignin degrading basidiomy- 225:898–903 cete Ceriporiopsis subvermispora. Microbiology Paszcynski A, Huynh V-B, Crawford R (1985) Enzymatic 14:2691–2698 activities of an extracellular, manganese-dependent Lobos S, Larrondo L, Salas L, Karahanian E, Vicun R peroxidase from Phanerochaete chrysosporium. (1998) Cloning and molecular analysis of a cDNA and FEMS Microbiol Lett 29:37–41 theCs-mnp1 gene encoding a manganese peroxidase Paszcynski A, Huynh V-B, Crawford R (1986) isoenzyme from the lignin-degrading basidiomycete Comparison of ligninase-1 and peroxidase-M2 from Ceriporiopsis subvermispora. Gene 206:185–193 the white-rot fungus Phanerochaete chrysosporium. Longoria A, Tinoco R, Duhalt VR (2008) Chloroperoxidase- Arch Biochem Biophys 244:750–765 mediated transformation of highly halogenated mono- Petersen JF, Tams JW, Vind J, Svensson A, Dalbøge H, aromatic compounds. Chemosphere 72:485–490 Welinder KG, Larsen S (1993) Crystallization and Ma B, Mayfield MB, Godfrey BJ, Gold MH (2004) Novel X-ray diffraction analysis of recombinant Coprinus promoter sequence required for manganese regulation cinereus peroxidase. J Mol Biol 232:989–991 of manganese peroxidase isozyme 1 gene expression Poulos TL, Edwards SL, Wariishi H, Gold MH (1993) in Phanerochaete chrysosporium. Eukaryot Cell Crystallographic refinement of lignin peroxidase at 2 3:579–588 Å. J Biol Chem 268:4429–4440 Maciel MJ, Silva AC, Ribeiro HCT (2010) Industrial and Pribnow D, Mayfield MB, Nipper VJ, Brown JA, Gold biotechnological applications of ligninolytic enzymes MH (1989) Characterization of a cDNA encoding a 86 S. Saroj et al.

manganese peroxidase, from the lignin-degrading from “modified” polystyrene. Environ Health Perspect basidiomycete Phanerochaete chrysosporium. J Biol 92:167–173 Chem 264:5036–5040 Steffen K, Hofrichter M, Hatakka A (2000) Mineralization Punt PJ, Biezen NV, Conesa A et al (2002) Filamentous of 14C-labelled lignin and production of ligninolytic fungi as cell factories for heterologous protein produc- enzymes by selected litter decomposing basidiomyce- tion. Trends Biotechnol 20:200–206 tes. Appl Microbiol Biotechnol 54:819–825 Raghukumar C, D’Souza TM, Thorn RG, Reddy CA Stewart P, Whitwam RE, Kersten PJ, Cullen D, Tien M (1999) Lignin-modifying enzymes of Flavodon flavus, (1996) Efficient expression of a Phanerochaete chrys- a basidiomycete isolated from a coastal marine envi- osporium manganese peroxidase gene in Aspergillus ronment. Appl Environ Microbiol 65:2103–2111 oryzae. Appl Microbiol Biotechnol 62:860–864 Reading NS, Aust SD (2000) Engineering a disulfide bond Sundaramoorthy M, Kishi K, Gold MH, Poulos TL in recombinant manganese peroxidase results in (1994a) Preliminary crystallographic analysis of man- increased thermostability. Biotechnol Prog 16:326–333 ganese peroxidase from Phanerochaete chrysospo- Reddy GVB, Sridhar M, Gold MH (2003) Cleavage of rium. J Mol Biol 238:845–848 nonphenolic β-1 diarylpropane lignin model dimers by Sundaramoorthy M, Kishi K, Gold MH, Poulos TL manganese peroxidase from Phanerochaete chryso- (1994b) The crystal structure of manganese peroxi- sporium. Eur J Biochem 270:284–292 dase from Phanerochaete chrysosporium at 2.06-A Robinson T, Chandran B, Nigam P (2001) Studies on the resolution. J Biol Chem 269:32759–32767 production of enzymes by white-rot fungi for the Sundaramoorthy M, Terner J, Poulos TL (1995) The decolourisation of textile dyes. Enzyme Microb crystal structure of chloroperoxidase: a heme peroxi- Technol 29:575–579 dase-cytochrome P450 functional hybrid. Structure Saitou N, Nei M (1987) The neighbor-joining method: 3:1367–1377 a new method for reconstructing phylogenetic trees. Sundaramoorthy M, Kishi K, Gold MH, Poulos TL (1997) Mol Biol Evol 4:406–425 Crystal structures of substrate binding site mutants of Sakamoto Y, Nakade K, Nagai M, Uchimiya H, Sato T manganese peroxidase. J Biol Chem 272:17574–17580 (2009) Cloning of Lentinula edodes lemnp2, a manga- Sundaramoorthy M, Youngs HL, Gold MH, Poulos TL nese peroxidase that is secreted abundantly in sawdust (2005) High-resolution crystal structure of manganese medium. Mycoscience 50:116–122 peroxidase: substrate and inhibitor complexes. Sarkanen KV, Ludwig CH (1971) Lignins: occurrence, for- Biochemistry 44:6463–6470 mation and structure. Wiley–Interscience, New York Sundaramoorthy M, Gold MH, Poulos TL (2010) Sasaki T, Kajino T, Li B, Sugiyama H, Takahashi H Ultrahigh (0.93 Å) resolution structure of manganese (2001) New pulp biobleaching system involving man- peroxidase from Phanerochaete chrysosporium: ganese peroxidase immobilized in a silica support implications for the catalytic mechanism. J Inorg with controlled pore sizes. Appl Environ Microbiol Biochem 104:683–690 67(5):2208–2212 Susla M, Novotny C, Erbanova P, Svobodova K (2008) Schmidt O, Schmitt U, Moreth U, Potsch T (1997) Wood Implication of Dichomitus squalens manganese- decay by the white-rotting basidiomycete Physisporinus dependent peroxidase in dye decolorization and coop- vitreus from a cooling tower. Holzforschung 51:193–200 eration of the enzyme with laccase. Folia Microbiol Sedighi M, Karimi A, Vahabzadeh F (2009) Involvement 53:479–485 of ligninolytic enzymes of Phanerochaete chrysospo- Tamura K, Peterson D, Peterson N, Stecher G, Nei M, rium in treating the textile effluent containing Astrazon Kumar S (2011) MEGA5: molecular evolutionary Red FBL in a packed-bed bioreactor. J Hazard Mater genetics analysis using maximum likelihood, evolu- 169(1–3):88–93 tionary distance and maximum parsimony methods. Selvaggini C, Salmona M, Gioia LD (1995) Manganese Mol Biol Evol 28(10):2731–2739 peroxidase from Phanerochaete chrysosporium. Tello M, Corsini G, Larrondo L, Salas L, Lobos S, Vicun A homology-based molecular model. Eur J Biochem R (2000) Characterization of three new manganese 228:955–961 peroxidase genes from the ligninolytic basidiomycete Shin KS (2004) The role of enzymes produced by white- Ceriporiopsis subvermispora. Biochim Biophys Acta rot fungus Irpex lacteus in the decolorization of the 1490:137–144 textile industry effluent. J Microbiol 42:37–41 Timofeevski SL, Aust SD (1997) Effects of Mn2+ and oxa- Singh S, Pakshirajan K (2010) Enzyme activities and late on the catalytic activity of manganese peroxidase. decolourization of single and mixed azo dyes by the Biochem Biophys Res Commun 239:645–649 white-rot fungus Phanerochaete chrysosporium. Int Timofeevski SL, Nie G, Reading NS, Aust SD (1999) biodeterior Biodegrad 64:1–5 Addition of veratryl alcohol oxidase activity to man- Snajdar J, Baldrain P (2007) Temperature effect the pro- ganese peroxidase by site-directed mutagenesis. duction, activity and stability of ligninolytic enzymes Biochem Biophys Res Commun 256:500–504 in Pleurotus ostreatus and Trametes versicolor. Folia Tuor U, Wariishi H, Schoemaker HE, Gold MH (1992) Microbiol 52(5):498–502 Oxidation of phenolic aryglycerol β- aryl ether lignin Soto AM, Justicia H, Wray JW, Sonnens Chein C (1991) model compounds by manganese peroxidase from p-nonyl-phenol: an estrogenic xenobiotic released Phanerochaete chrysosporium: oxidative cleavage 6 Manganese Peroxidases: Molecular Diversity, Heterologous Expression, and Applications 87

of an α-carbonyl model compound. Biochemistry Whitwam RE, Brown KR, Musick M, Natan MJ, Tien M 31:4986–4995 (1997) Mutagenesis of the Mn2+ binding site of man- Urzua U, Larrondo LF, Lobos S, Larraın J, Vicuna R ganese peroxidase affects oxidation of Mn2+ by both (1995) Oxidation reactions catalyzed by manganese compound I and compound II. Biochemistry 36: peroxidase isoenzymes from Ceriporiopsis subver- 9766–9773 mispora. FEBS Lett 371:132–136 Whitwam RE, Koduri RS, Natan M, Tien M (1999) Role Urzua U, Kersten PJ, Vicuna R (1998) Manganese of axial ligands in the reactivity of Mn peroxidase peroxidase-dependent oxidation of glyoxylic and from Phanerochaete chrysosporium. Biochemistry oxalic acids synthesized by Ceriporiopsis subvermis- 38:9608–9616 pora produces extracellular hydrogen peroxide. Appl Willmann G, Fakoussa RM (1997a) Biological bleach- Environ Microbiol 64:68–73 ing of water-soluble coal macromolecules by a Vares T, Hatakka A (1996) Lignin-degrading activity and basidiomycete strain. Appl Microbiol Biotechnol ligninolytic enzymes of different white-rot fungi: effects 47:95–101 of manganese and malonate. Can J Bot 75:61–71 Willmann G, Fakoussa RM (1997b) Extracellular oxida- Wariishi H, Akaleswaran L, Gold MH (1988) Manganese tive enzymes of coal attacking fungi. Fuel Process peroxidase from the basidiomycete Phanerochaete Technol 52:27–41 chrysosporium: spectral characterization of oxidized Wong DWS (2009) Structure and action mechanism of states and the catalytic cycle. Biochemistry 27: ligninolytic enzymes. Appl Biochem Biotechnol 5365–5370 157:174–209 Wariishi H, Dunford HB, MacDonald ID, Gold MH Yang HS, Yoon JS, Kim MN (2004) Effects of storage of (1989a) Manganese peroxidase from the lignin- a mature compost on its potential for biodegradation degrading basidiomycete Phanerochaete chrysospo- of plastics. Polym Degrad Stab 84:411–417 rium. Transient state kinetics and reaction mechanism. Yeo PSN, Song HG, Choi HT (2007) Generation of a J Biol Chem 264:3335–3340 transformant showing higher manganese peroxidase Wariishi H, Valli K, Renganathan V, Gold MH (1989b) (Mnp) activity by overexpression of Mnp Gene in Oxidative cleavage of a phenolic diarylpropane Trametes versicolor. J Microbiol 45:213–218 lignin model dimer by manganese peroxidase from Youngs HL, Gelpke SMD, Li D, Sundaramoorthy M, Gold Phanerochaete chrysosporium. Biochemistry MH (2000) The role of Glu39 in MnII binding and oxi- 28:6017–6023 dation by manganese peroxidase from Phanerochaete Wariishi H, Valli K, Renganathan V, Gold MH (1989c) chrysoporium. Biochemistry 40:2243–2250 Thiol-mediated oxidation of nonphenolic lignin Zapanta LS, Tien M (1997) The roles of veratryl alcohol model compounds by manganese peroxidase of and oxalate in fungal lignin degradation. J Biotechnol Phanerochaete chrysosporium. J Biol Chem 264: 53:93–102 14185–14191 Zhang X, Wang Y, Wang L, Chen G, Liu W, Gao P (2009) Wariishi H, Valli K, Gold MH (1992) Manganese(II) oxi- Site-directed mutagenesis of manganese peroxidase dation by manganese peroxidase from the basidiomy- from Phanerochaete chrysosporium in an in vitro cete Phanerochaete chrysosporium. J Biol Chem expression system. J Biotechnol 139:176–178 267:23688–23695 Zuckerkand E, Pauling L (1965) Evolutionary divergence Welinder KG (1992) Superfamily of plant, fungal and bac- and convergence in proteins. In: Evolving genes and terial peroxidases. Curr Opin Struct Biol 2:388–393 proteins. Academic, New York Advance Techniques in Enzyme Research 7

Debamitra Chakravorty and Sanjukta Patra

Abstract Enzyme research encompasses the purifi cation, characterization, identifi cation, and applications of enzymes. Enzyme technology has entered a modern era in which bioinformatics, genetic engineering, and high-throughput screening by miniaturization of instruments and robotic handling have altogether made existing processes economical and time saving. These advances resulted in enzymes to be industrial workhorses. This chapter brings forward advances in existing techniques along with novel and innovative methodologies and instrumentations applied in high-throughput enzyme research. Highlights are the applications of advanced recombinant DNA technology like the Gateway® cloning and CODEHOP PCR for the fast and effi cient cloning of enzyme genes; computational biology for in silico enzyme characterization like monitoring behavior of enzymes in real time through molecular dynamics simulations; modern analytical techniques like microfl uidics, affi nity ultrafi ltration, automated electrophoresis, and ATP-affi nity chromatography; and the use of nanoparticles, nanofi bers, and mesoporous silica in enzyme stabilization.

Keywords Enzyme • Research • Cloning • Screening • Puriﬁ cation • Gateway cloning • CODEHOP

Introduction enzymes is expanding exponentially, and a number of biotechnology companies are continuously Enzymes are ubiquitous natural biocatalysts prospecting for new and adapting existing enzymes. with well-recognized potential applications (Kim Enzyme technology is an interdisciplinary et al. 2006a ). The global market for industrial science and is recognized by the Organization for Economic Cooperation and Development (OECD) as an important component of sustainable industrial * D. Chakravorty • S. Patra ( ) development (OECD 1998 ) as they ﬁ nd practical Department of Biotechnology , Indian Institute of Technology , Guwahati , India applications in large number of ﬁ elds; recent devel- e-mail: [email protected] opments are in the areas of organosynthesis,

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 89 DOI 10.1007/978-81-322-1094-8_7, © Springer India 2013 90 D. Chakravorty and S. Patra pharmaceuticals, biosensors, bioremediation, and purifi cation and characterization, whereas the biofuel cells (Kim et al. 2006b ). Enzyme research reverse approach starts with the gene of interest evolved in the twentieth century; since then the followed by its amplifi cation, recombinant scope of enzymology has expanded, from an expression, and purifi cation. early focus on the chemical and catalytic mechanisms of individual enzymes to more recent efforts to understand enzyme behavior in the con- Forward Enzyme Screening Approach text of functional biological systems in the milieu of many interacting enzymes and proteins In the forward enzyme screening approach, (Zalatan and Herschlag 2009 ). This was possible environmental samples are assayed for enzyme of due to a plethora of new enzymes to study and the interest, based on their physicochemical properties curiosity of the researchers to analyze how such as temperature, pH, solvent stability, and enzymes could achieve their enormous rate their substrate specifi city or substrate enantiose- enhancements and exquisite specifi cities. lectivity or stereospecifi city. For this purpose, Enzyme technology thus entered a new era and recent advances in enzymological techniques has become much more economical which is have seen the development of high-throughput shaping the discovery, development, purifi cation, analytical instruments like parallel capillary and application of biocatalysts to a much greater electrophoresis and thermistor arrays to study the extent (Bailen et al. 2002 ). Modern researchers physicochemical properties of particular enzyme focus on complex subjects of enzyme research in of interest. Thermistors are small thermometers the context of the complex biological system that monitor temperature changes in wells of where they are naturally found. Recent trend microtiter plates and provide an advantage over shows development in strategies for enzyme the traditional infrared (IR) thermography, which production, such as modernization of enzyme monitors temperature from IR radiation (Wahler fermentation processes and their recombinant and Reymond 2001 ). On the simple principle that expression in hosts like Escherichia coli. It has enzyme reactions can be assayed by recording also been realized that many enzymes are expen- changes in their physicochemical properties, sive and must be recovered and reused; thus, Copeland and Miller attached the fl uorescent techniques like enzyme stabilization through pH indicator to a solid support to explore the immobilization and enzyme engineering are screening of peptide-based catalysts prepared by being considered (Bailen et al. 2002 ). In this combinatorial chemistry techniques. The indica- chapter we shall discuss the modern technologies tor was either directly attached to the bead or developed for enzyme research. incorporated into a polymeric gel. Klein and Reymond in 2001 performed enantioselective assay for alcohol dehydrogenases using fl uoro- Enzyme Discovery and Screening genic substrates to measure the rate of oxidation of each enantiomer to ketone in the presence of High-throughput screening methods for enzymes bovine serum albumin, which catalyzes have been recently developed to perform a key β-elimination to form the strongly fluorescent role in the search for new enzymes and biocata- product umbelliferone (Klein and Reymond lysts (Wahler and Reymond 2001 ). For such pro- 2001 ). In similar lines Olsen et al. in 2000 cesses the modern trend being set is towards reported an elegant imaging experiment that robotic handling and miniaturization of instru- relies on the fl uorogenic phospholipid substrates ments as it improves reproducibility and authen- for phospholipase A2 (PLA2) to visualize the ticity of the resulting data. Basically there can be activity and localization of this enzyme in zebraf- two main approaches to enzyme screening: one ish larvae, based on the principle of fl uorescence can be the forward approach and the other is the resonance energy transfer (FRET) (Olsen et al. reverse approach. The forward approach starts 2000 ). A similar FRET-based assay has been with the native enzyme of interest followed by its reported to detect enzyme activities in cells using 7 Advance Techniques in Enzyme Research 91

fl uorescence-activated cell sorting (FACS) (Olsen poses of substrate–enzyme complexes (Xu et al. et al. 2000 ). Another interesting example is in the 2009 ). Thus, in the future this approach com- use of fl uorogenic and chromogenic substrates for bined with homology modeling, wherein structure enzyme assays. For example, Wahler et al. employed of an enzyme can be modeled from its primary fl uorogenic substrates to identify caspase activity amino acid sequence based on a template struc- in bacteria (Wahler and Reymond 2001 ). Further ture, can be applied to screen substrates for novel progress in enzyme research in this area was made and newly characterized enzymes whose X-ray by Aw et al. who described continuous assessment structure is still undetermined. of enzymatic activity using nanodroplet microar- Some of the other latest innovations in detection rays (Aw et al. 2010 ). By uniformly coating fl uoro- and quantifi cation of enzymes include the use of genic substrates on slides, they successfully microfl uidics, which involves the manipulation of generated surfaces capable of detecting enzymatic liquid samples in miniaturized systems, often on a activity offering an unprecedented ability for per- chip (Smith 2005). Moreover in 2010, a research forming solution-phase enzymatic assays in nano- team at the Georgia Institute of Technology has liter volumes on microarrays, in contrast to developed a technique called gelatin zymography to microliter volumes typically required in micro- detect and quantify the nature of mature cathepsin K plate-based assays, thereby reducing the amounts in femtomoles (Li et al. 2010 ). Thus, we can use the of reagents required anywhere from a hundred- to a aforementioned techniques with other enzymes too thousandfold. This new approach thus provides a with modifi cations done on the enzyme of interest. potentially more cost-effective and label-free enzyme screening technique. A single slide is able to accommodate several thousand assays, facilitat- Reverse Enzyme Screening Approach ing the assessment of both dose- and time-dependent inhibition parameters in a single run. The recent progress in sequencing of genomes of High-throughput screening for enzymes is various organisms and the availability of this often aimed not only at activity but also at sub- multitude of data conveniently structured in strate specifi city for the application of enzymes freely accessible databases like the National Center in fi ne chemical synthesis. Recent progress in for Biotechnology Information ( www.ncbi.nlm. analysis of substrate specifi city and enantiose- nih.gov/) and the UniProt ( www.uniprot.org/ ) in lectivity of enzymes was by Hwang and Kim the World Wide Web has made it possible to ( 2004 ), who developed a new high-throughput selectively screen for genes expressing enzyme screening method using staining solution of of interest followed by their primer base poly-

CuSO4 /MeOH for screening ω-transaminase for merase chain reaction (PCR) amplifi cation and multiple substrate specifi cities and their enanti- expression in recombinant host. The PCR-based oselectivities based on UV-visible spectrometry amplifi cation of gene can follow three strategies. (Hwang and Kim 2004 ). Staining solution of First is amplifi cation of genes whose sequence is

CuSO4 /MeOH forms a blue complex with the already known by conventional primer design α-amino acid produced during ω-transaminase and by traditional PCR and, second, amplifi cation reaction. This complex can be easily quantifi ed of homologous enzyme genes from uncharacter- using UV–vis spectrophotometer at 595 nm ized genomes though the design of degenerate (Hwang and Kim 2004 ). In similar lines, the idea primers followed by PCR amplifi cation by a pro- of computer-aided substrate screening of enzymes cess known as gene walking. For example, Hallin is new and innovative combination of computa- and Lindgren (1999 ) designed two sets of PCR tional skills and enzymology which was recently primers using consensus regions in gene developed by Xu et al. based on the enzyme sequences encoding the two forms of nitrite structure, and they successfully screened sub- reductase (Nir), a key enzyme in the denitrifi ca- strates of Candida antarctica lipase B (CALB). tion pathway, and successfully amplifi ed nir In this method restricted molecular docking was genes in nine species within four genera of nitri- employed to predict the energetically favorable fying organisms (Hallin and Lindgren 1999 ). 92 D. Chakravorty and S. Patra

The third strategy is amplifi cation of genes negating the requirement for expensive thermal from distantly related organisms. In this regard, cyclers. It is performed by Bst DNA polymerase a recent methodology has been developed to from Bacillus stearothermophilus , with high isolate genes coding for functionally equivalent strand displacement activity and a set of two enzyme from distantly related species by specially designed inner and two outer primers. Rose et al. (2003 ) through a new primer design The inner primers are called the forward inner strategy for PCR amplification, based on primer (FIP) and the backward inner primer consensus- degenerate hybrid oligonucleotide (BIP), respectively, and each contains two distinct primers (CODEHOPs). They have also developed sequences corresponding to the sense and antisense an interactive computational program to design sequences of the target DNA. Detection of ampli- CODEHOP PCR primers from conserved blocks fi cation products of the targeted DNA can be of amino acids within multiple aligned protein easily done by photometry for turbidity caused by sequences. Each CODEHOP consists of a pool increasing quantity of magnesium pyrophosphate of related primers containing all possible 3–4 in solution or by the use of SYBR Green which nucleotide sequences encoding highly conserved specifi cally intercalates only to double-stranded amino acids within a 3′ degenerate core, and a DNA and emits green light (λ max = 520 nm) longer 5′ non-degenerate clamp region contains (Mori et al. 2001 ). the most probable nucleotide predicted for each Another approach of gene amplifi cation, fl anking codon. The primer design software and worth mentioning here, for enzymes, whose the CODEHOP PCR strategy have been utilized sequence is not yet available, is that the sequenc- for the identifi cation and characterization of new ing of the N- and C-terminal of proteins, which gene orthologs and paralogs in different plant, have not yet been sequenced, can lead to the animal, and bacterial species (Rose et al. 2003 ). design of primers and amplifi cation of the gene As seen from the above examples, PCR is the of the enzyme by degenerate primers coupled most valuable tool to amplify genes of enzyme with 3′ or 5′ RACE (rapid amplifi cation of cDNA of interest. Recent developments in conventional ends) PCR. RACE is a technique which can be PCR techniques are self-sustained sequence rep- used to obtain the full-length sequence of an lication (3SR) (Guatelli et al. 1990 ), nucleic acid RNA transcript found within a cell. RACE can sequence-based amplifi cation (NSBA) (Compton provide the sequence of an RNA transcript from 1991 ), Strand Displacement Amplifi cation (SDA) a small known sequence within the transcript all (Walker et al. 1992 ), and loop-mediated isothermal the way to the 5′ end (5′ RACE-PCR) or 3′ end amplifi cation PCR (LAMP) (Notomi et al. 2000 ). (3′ RACE-PCR) of the RNA (Sambrook et al. PCR uses heat denaturation of double-stranded 2001 ). The PCR product can then be used for DNA to promote next round of DNA synthesis; cloning and expression of the enzyme gene. NSBA and 3SR eliminate this step of heat dena- Sakamoto et al. (2011 ) used this technique and turation by using a set of transcription and reverse amplifi ed endo-β-1,3-glucanase (GLU1), from transcription reactions for gene amplifi cation. the fruiting body of the fungi, Lentinula edodes . SDA eliminates the heat denaturation by employ- The glu1 gene was isolated by rapid amplifi ca- ing a set of restriction enzyme digestions with tion of cDNA ends (RACE)-PCR using primers modifi ed nucleotides as substrate. Notomi et al. designed from the N-terminal amino acid ( 2000 ) have developed loop-mediated isothermal sequence of GLU1 protein (Sakamoto et al. 2011 ). amplifi cation PCR (LAMP) which has emerged The next step in the reverse enzyme approach, as the most cost-effective solution for the tech- after amplifi cation of the gene of interest for the nique known as real-time PCR. This method can enzyme, is cloning the amplifi ed gene in a suitable amplify a few copies of DNA to 109 in less vector followed by its transformation into compe- than an hour under isothermal conditions and tent hosts. Traditionally the steps in cloning of an with greater specifi city (Notomi et al. 2000 ). This enzyme gene involve DNA isolation followed by method relies on single temperature incubation restriction digestion of the DNA into desired frag- of 60–65 °C for displacement DNA synthesis ments and then ligating them into the selected 7 Advance Techniques in Enzyme Research 93 vector system. Though these aforementioned processes turn out to be successful, the entire process are fairly time consuming. Thus, suffi cient groundwork and experimentation have been carried out to modernize the entire protocol. The recent advances in the cloning techniques have led to the development of more specifi c, highly repro- ducible, and guaranteed cloning strategies known as the zero background TA cloning, TOPO cloning, and the Gateway® cloning . TA cloning is the cloning technique that negates the use of restriction enzymes. The technique is dependent on the ability of adenine and thymine to hybridize in the presence of a DNA ligase. With the property of Taq polymerase Fig. 7.1 Schematic illustration of the TOPO cloning terminal transferase activity, deoxyadenosine procedure is added to the 3′ end of double-stranded PCR- amplifi ed DNA duplex leaving a 3′ overhang. Such inserts can be directly cloned to linearized The phosphotyrosyl bond between the DNA and vectors with 5′-deoxythymidine overhang with enzyme can subsequently be attacked by the the aid of a DNA ligase (Chen and Janes 5′ hydroxyl of the original cleaved strand, revers- 2002 ). A modifi ed version of TA cloning is ing the reaction and releasing topoisomerase known as the zero background TA (ZeBaTA) (Shuman 1994 ). Figure 7.1 illustrates the TOPO cloning. The improved system takes advantage cloning procedure. of the restriction enzyme XcmI whose restriction Gateway Cloning Technology is a powerful site is CCAATACT/TGTATGG to generate a T new methodology that can greatly facilitate overhang after digestion and the negative selec- enzyme expression, cloning of PCR products, tion marker gene ccdB to eliminate the self-liga- and analysis of gene function by replacing restriction background after transformation which is tion endonucleases and ligase with site-specifi c quite common in the case of simple TA cloning recombination. This technology had been (Chen and Janes 2002). TOPO cloning is also a invented and commercialized by Invitrogen modifi ed version of the TA cloning strategy and since the late 1990s. The overall reaction of has been invented by Invitrogen. In this method Gateway ® cloning has been illustrated in direct insertion DNA takes place in a matter of Fig. 7.2 . It mainly involves two site-specifi c few minutes with the aid of Taq or Pfu polymerase recombination reactions known as the LR reac- and vaccinia virus DNA topoisomerase I which tion and the BP reaction. In the LR reaction a specifi cally recognizes DNA sequence 5′-(C/T) recombination reaction between an Entry Clone CCTT-3′ into linearized plasmid vectors with and a Destination Vector occurs with the action of 3′ T overhangs, without the requirement for LR Clonase. An Entry Clone, containing a gene any ligation reaction. Taq polymerase has a fl anked by recombination sites attL1 and attL2, nontemplate-dependent terminal transferase recombines with a Destination Vector (pDESTTM) activity that adds a single deoxyadenosine (A) to to yield an Expression Clone and a by-product the 3′ ends of PCR products. The topoisomerase plasmid with attB1, attB2 and attP1, attP2 sites, I remains covalently bound to the vector and respectively. The by-product plasmid contains binds to duplex DNA at specifi c sites and cleaves the ccdB gene and hence gives rise to no colo- the phosphodiester backbone after 5′-CCCTT in nies when using standard strains of E. coli . In BP one strand (Shuman 1991 ) of a covalent bond reaction the gene of interest in the Expression between the 3′ phosphate of the cleaved strand and Clone (between attB sites) is transferred into a tyrosyl residue (Tyr-274) of topoisomerase I. a Donor Vector (containing attP sites) to produce 94 D. Chakravorty and S. Patra

abEntry clone Destination Vector Expression clone Donor Vector

L1 L2 R1 R2 B1 B2 P1 P2 GENE ccdB GENE ccdB pENTR gene + pDEST pEXP gene + pDONR

Knr r Ap Knr

LR clonase By Product BP clonase Destination Vector P1 P2 R1 R2 Expression clone ccdB L1 L2 GENE ccdB B1 B2 + GENE + pENTR gene

pEXP gene r Kn r Knr Ap

Apr

E.Coli transformation; Knr colonies E.Coli transformation; Apr colonies

Fig. 7.2 The overall process of Gateway ® cloning. attL1 reacts only with attR1 , and attL2 reacts only with ( a ) The LR reaction. (b ) The BP reaction. The sites labeled attR2 . attB1 reacts only with attP1 and attB2 with attP2 L , R , B , and P are, respectively, the attL , attR , attB , and attP (This figure has been printed with permission from recombination sites for bacteriophage lambda in E. coli . www.lifetech.com/gateway ) a new Entry Clone (attL sites) catalyzed by the only the expression of the enzyme but also the BP Clonase enzyme. way in which the product can be subsequently Another series of dual expression vectors purifi ed. Traditional strategies for recombinant were constructed by Madzak et al. containing an enzyme expression involve transfecting cells hp4d promoter which allows for expression of with a DNA vector that contains the template and more than one expression cassettes in a single then culturing the cells so that they transcribe and vector and expresses different proteins effi ciently translate the desired protein. Typically, the cells and constantly without multiple infl uences by are then lysed to extract the expressed protein for nutritional and environmental factors in medium, subsequent purifi cation. Both prokaryotic and such as carbon/nitrogen sources and pH values eukaryotic in vivo protein expression systems (Madzak et al. 2000 ). This cassette is integrated which consist of a suitable vector with an appro- into the genome of a genetically modified priate promoter and other necessary regulatory Yarrowia lipolytica strain by homologous recom- sequences along with the gene of interest are bination. This effi cient hp4d promoter expression widely used. The selection of the system depends cassette system has been applied to produce various on the type of protein, the requirements for func- enzymes, such as β-galactosidase, prorennin, tional activity, and the desired yield. Escherichia cytokinin oxidase, lipase, and laccases success- coli have been used extensively till date as fully (Chuang et al. 2010 ). expression hosts due to its rapid growth rate, Transformation into competent host cell or capacity for continuous fermentation, and rela- expression host is the major vital step for a tively low cost. Problems associated with such successful cloning experiment, and enzyme host system are codon bias, correct protein fold- expression level control is still a big challenge in ing, and disulfi de bridge formation. Moreover enzyme evolution (Cambon et al. 2010 ). If this formation of inclusion bodies in such prokaryotic fails we will not be able to recover or express the host system is another great obstacle. There are enzyme of interest. The choice of host affects not surplus of information available over the Internet 7 Advance Techniques in Enzyme Research 95 on the modern expression host systems developed endonuclease digestion in the target bacterium in the last decade; some of the most relevant (Suzuki and Yasui 2011 ). Another example of an developments in this area are discussed here. excellent expression system developed by the In this regard the most promising modern company TAKARA ( http://www.takara-bio.com ) expression hosts that were developed are the is the Brevibacillus expression system. Rosetta- gami™ B host strains developed by Brevibacillus is a gram-positive microbe, charac- Novagen, as they successfully express the terized by its ability to secrete/produce large eukaryotic proteins that contain codons rarely amount of proteins. This system is recommended used in E. coli (Ren 2007 ). These strains supply for the production of functional proteins that are tRNAs for AGG, AGA, AUA, CUA, CCC, and naturally produced intracellularly but become GGA codons on a compatible chloramphenicol- insolubilized and cannot undergo in vitro refolding resistant plasmid named pRARE. Elorza et al. when produced by Escherichia coli . The target expressed human alpha galactosidase A, which protein can be easily purifi ed using a histidine- contains many non-optimized codons disulfi de tagged protein purifi cation column. The His-Tag bonds in its native structure in Rosetta-gami™ B can be removed by exposing the purifi ed protein host strains (Xiao et al. 2007 ). Mai et al. have to enterokinase treatment. In similar lines, developed the shuttle vector pIKM1 consisting of Biomedal CASCADETM was developed as a selection marker which is the thermostable bacterial protein expression system that provides kanamycin cassette from Streptococcus faecalis tightly regulated, high-level expression. The system plasmid pKD102for for transformation of makes use of linked regulatory circuits to amplify Thermoanaerobacterium strain JW/SL-YS485 gene expression levels when induced, maintaining (Mai et al. 2000 ). Kushnir et al. developed a new low basal expression levels under noninducing inducible protein expression system which was conditions. Its features are tight control of gene able to overexpress functional proteins, based on expression, activity at low temperature, low sen- the protozoan host Leishmania tarentolae . This sitivity to media formulation, and expression of strain can co-express T7 RNA polymerase and heteromultimeric proteins. Recently a trend has tetracycline repressor and thus can be stably moved towards eukaryotic expression system, transformed with the heterologous target gene and in this regard Drosophila expression system under control of the T7 promoter/TET operator (DES®) by Invitrogen and Leishmania tarentolae assembly, which can initiate transcription upon expression system known as LEXSY by Jena addition of tetracycline to the culture medium Bioscience were developed. (Kushnir et al. 2005 ). Enzymes from extremo- Other than the aforesaid techniques, a new philes are hard to express in normal preexisting system known as cell-free enzyme expression or host systems. Recently Cambon et al. developed a in vitro expression has recently emerged as a new Yarrowia lipolytica JMY1212 expression strong tool in enzyme research as it is simple, system which is an effi cient tool for rapid and inexpensive, and can translate proteins or enzymes reliable kinetic analysis of improved enzymes in hours from PCR fragments without the need (Cambon et al. 2010 ). This strain allows for tar- for Escherichia coli cloning, and thus it is easily geted integration of the expression cassette into adapted for high-throughput, automation, and/or the genome and homogenous transformants miniaturization procedures (He 2008 ). Cell-free (Cambon et al. 2010 ). Most recently a method protein expression makes use of cellular extracts designated Plasmid Artifi cial Modifi cation (PAM) or purifi ed molecular components to direct pro- was proposed by Suzuki and Yasui (2011 ). This tein synthesis from added DNA template in PAM method promises higher transformation test tubes (He 2008). Finally, cell-free systems effi ciency as the presence of PAM plasmid encod- are capable of synthesizing large protein popu- ing the modifi cation enzymes ensures that recom- lations in a single reaction, offering an exploit- binant plasmids harboring the gene of interest are able system for developing proteomic tools modifi ed such that it is protected from restriction (He 2008 ). The cell-free environment is also 96 D. Chakravorty and S. Patra helpful in the study of protein translation and Pyrococcus horikoshii by using certain conserved synthetic biology applications (He 2008 ). genes as genomic landmarks (Ishikawa et al. 2005 ) . PURExpress™ In Vitro Protein Synthesis Kit This method is much more fast and economical from New England Biolab and TNT® Quick than the traditional method used to screen for Coupled Transcription/Translation Systems are restriction endonucleases wherein individual strains examples of such technology. are cultured and their extracts assayed for the Other most promising recently developed and ability to produce specifi c fragments from small high-fi delity reverse enzyme screening methods DNA molecules. The method is also benefi cial in are the processes known as metagenomics, diver- the case when we are dealing with extremophiles sity generation by error-prone polymerase chain who are fastidious and economically hefty to reaction (PCR), and gene shuffl ing of an existing laboratory culture and maintenance. enzyme’s gene or gene family (Wahler and Reymond 2001 ). Metagenomics is the new rising science used to explore the unculturable microbes Enzyme Purifi cation directly from the environmental samples. It is the outcome of the strength of genomics, bioinfor- An apparent quote from Efraim Racker, an eminent matics, and system biology and is a multistage Austrian biochemist, is “Don’t waste clean thinking process that implies direct cloning of environ- on a dirty enzyme . ” This quote stresses on mental DNA into large clone libraries to facilitate enzyme purifi cation as the fi rst criteria before the analysis of the genes and the sequences within analyzing its properties. The science, art, and the libraries (Handelsman 2004 ). In the recent practice of protein and enzyme purifi cation are past, employing metagenomics, Tirawongsaroj known for more than a century, yet, in many and co-workers isolated novel thermostable lipo- respects, the fi eld is only now evolving (Ward and lytic enzymes from a Thailand hot spring metage- Swiatek 2009 ). Each enzyme possesses unique nomic library (Tirawongsaroj et al. 2008 ). Chu properties, which can be exploited for its puri- et al. also identifi ed two novel esterase genes from fication (Puig et al. 2001 ), and modern tech- a marine metagenomic library through functional niques aim for enzyme purifi cation with a screening of lipolytic clones (Chu et al. 2008 ), minimum of labor and cost to remove all con- and Wang et al. with the aid of bioinformatics taminants while retaining as much as possible of and the huge freely accessible databases avail- the enzyme of interest. Chromatography is called able on the web designed metagenomic gene- the protein or enzyme purifi cation workhorse for specifi c primers and isolated 23 truncated lipase both native and recombinant proteins (Smith gene fragments from 60 metagenomic samples 2005). The outline of the general scheme of (Wang et al. 2010 ). They designed primers based enzyme purifi cation is illustrated in Fig. 7.4 below. on well-known conserved motifs of lipase genes. Today numerous companies like Qiagen, From the aforesaid we can say that bioinfor- Applied Biosystems, BioChrom, and Amersham matics or in silico biology related to genes of Biosciences offer improvements to these tech- enzymes can play a prominent role in screening niques by introducing new tweaks to improve and functional analysis of novel enzymes. In 2005 effi cacy of the procedure. Here we discuss the Ishikawa and co-workers discovered a novel recent strategies for purifying native and recom- restriction endonuclease by genome comparison binant enzymes. The generalized steps for and application of a wheat germ-based cell-free enzyme purifi cation are given in Fig. 7.3 . translation assay from the hyperthermophilic Among recently developed column chromato- archaeon Pyrococcus abyssi (Ishikawa et al. 2005 ). graphic matrices, BioChrom has developed a Their methodology involved identifi cation by polymeric material, Hydrocell. The high level of comparison of candidate genes from the already cross-linking in the polymer beads is intended sequence and related genomes of the hyper- to give better and faster protein separation thermophilic archaea Pyrococcus abyssi and compared to conventional silica-based HPLC 7 Advance Techniques in Enzyme Research 97

Enzyme assay Enzyme assay SDS PAGE Native PAGE 1 2 Activity staining Source Crude extract Partially purified

Centrifugation Cell disruption Ammonium sulfate precipitation, Removal of phenols Ultrasonication, 3 Dialysis lipids, Nucleic acids. French Press

Partially purified clear extract

Column chromatography lon exchange Size exclusion 4 Enzyme assay Enzyme assay lsoelectric focussing SDS PAGE SDS PAGE Affinity Native PAGE Native PAGE Hydrophobic interaction Activity staining Activity staining Hydroxyapatite

6 5 Purified and Concentration Buffer Exchange active enzyme & Lyophilization Dialysis Ultrafiltration Gel filtration

Fig. 7.3 Generalized steps for enzyme puriﬁ cation

columns. GE Healthcare has also recently beads have become a popular choice for protein launched a new chromatography matrix for the and enzyme purifi cation due to their speed, ease purifi cation of glutathione S-transferase (GST) of use, and affordability. They possess an iron and His-tagged proteins which combines small- core surrounded by an inert polymer material bead technology with affi nity purifi cation. The (Fig. 7.4). When subjected to magnetic fi eld, the relatively small bead size results in distinctly beads behave like magnets simplifying purifi- separated protein elution peaks. cation procedures negating the use of centrifu- Another relatively new method of enzyme gation. Such beads also come along with specifi c purifi cation is affi nity ultrafi ltration (Galaev molecular tags suitable for affi nity purifi cation. 1999 ). In affi nity ultrafi ltration, affi nity separa- Some companies such as Dynal Biotech, Poly- tion is combined with membrane fi ltration tech- sciences, and BioScience Beads have capitalized nique. The enzyme to be purifi ed is complexed on magnetic beads as one of their leading prod- with a macroligand composed of a soluble poly- ucts (Smith 2005 ). mer along with target protein-specifi c affi nity For purifi cation of small quantity of proteins ligands. The complex is trapped by an ultrafi ltra- and enzymes, there is an increasing trend in the tion membrane, whereas unwanted proteins pass use of mass spectroscopy in protein and enzyme through the membrane (Galaev 1999 ). Several analysis. For this purpose, Vivascience has devel- enzymes like urokinase have been purifi ed oped Vivapure MALDI-Prep Micro spin columns, through this method (Male et al. 1990 ). Two-Step which contain the C-18 (the number of carbons in Affi nity Purifi cation System by Qiagen combines the alkyl chains attached to silica particles in con- sequential affi nity purifi cation steps and yields ventional column chromatography membrane), ultrapure (>98 %) protein. This system can be for desalting and concentrating protein or enzyme useful with enzymes in the presence of high con- digests for mass spectroscopy. Companies like centrations of chelating compounds or from Millipore are providing solutions for purifi cation eukaryotic cell lysates. In similar lines, magnetic of bulk amount of enzymes. They have developed 98 D. Chakravorty and S. Patra

Fig. 7.4 Magnetic Dynabeads from Dynal Biotech (Courtesy of Dynal Biotech) (This fi gure has been reprinted with permission from Smith 2005 ; Striving for purity: advances in protein purifi cation. Nature Methods 2, 71–77 (2005)) the Amicon concentrators which can be used to material (matrix). This matrix is usually packed concentrate and desalt from 4 to 15 ml of solution into a column. Crude cell lysates are loaded onto (Smith 2005 ). A recent breakthrough in mass the column under conditions that ensure specifi c spectrometry was made by scientists at ETH binding of the protein to the immobilized ligand. Zurich, along with AB SCIEX, a global leader Other enzymes that do not bind the immobilized in life science analytical technologies; they ligand are washed through. Elution of the bound have developed MS/MSALL with SWATHTM enzyme of interest can be achieved by changing Acquisition, a groundbreaking that successfully the experimental conditions to favor desorption. quantifies nearly peptides and proteins in a There is a variety of new commercially available sample from a single analysis. These techniques fusion systems where enzyme genes are tagged. employ data-independent methods to systemati- This is designed to ease enzyme purifi cation fol- cally generate complete, high-specifi city fragment lowed by cleavage of enzyme of interest. A variety ion maps that can be queried for the presence and of tags are available for such purpose and are listed quantity of any protein of interest using a targeted in Table 7.1 . The tagged enzyme is purifi ed through data analysis strategy (Framingham, Mass, April washing in an affi nity chromatography medium and 17, 2012 ). This technique will thus aid in enzyme the tagged enzyme is eluted. Modern columns sequencing and identifi cation in a much faster developed in this regard are His GraviTrap for histi- rate in the near future. dine-tagged proteins. HiTrap columns are suitable Recombinant protein expression and purifi ca- for use with a syringe or peristaltic pump for histi- tion calls for production of large amounts of an dine-, GST-, MBP-, and Strep-tagged proteins affi nity-tagged protein so that a single purifi cation (HisTrap, GSTrap™, MBPTrap™, and StrepTrap™ step using affi nity chromatography is suffi cient columns, respectively). to achieve the desired level of purity. In affi nity After affi nity purifi cation of the recombinant chromatography, the protein or enzyme of interest enzyme, cleavage of the protein from the tag is purifi ed by its ability to bind a specifi c ligand system calls for proteases like factor Xa, thrombin, that is immobilized on a chromatographic bead or enterokinase. The limitation is that they may 7 Advance Techniques in Enzyme Research 99

Table 7.1 Fusion vector systems with tags for recombinant protein purifi cation Company Vector system Tags (N/C terminus) Purifi cation Novagen, Inc. T7 pET T7-tag (monoclonal antibody) 1. Ni affi nity column purifi cation (Merck) followed by Exopeptidase cleavage of N-terminal His tag

His Tag (Histidine6 tag 2. Antibody affi nity systems for metal chelation chromatography) S-tag (RNAse S-protein) HSV-tag (monoclonal antibody) pelB/ompT (potential periplasmic localization) Pharmacia Biotech pGEX vectors Glutathione S-transferase Glutathione Sepharose 4B affi nity (GE Healthcare) (GST) chromatography Eastman Kodak Flag system Flag marker octapeptide Affi nity chromatography and an amino-terminal Flag peptide can be removed by the protease, enterokinase Promega PinPoint™ Xa Protein Biotinylated Affi nity-purifi ed using the Purifi cation System SoftLink™ Soft Release Avidin Resin from promega Stratagene VariFlex™ Solubility enhancement Enhances solubility of the fusion tags (SETs) peptide by providing a net negative charge preventing protein aggregation. Proteins can be purifi ed through affi nity chromatography through Streptavidin Resin and tags can be removed by Thrombin digestion Thermo Scientifi c Myc-tag Derived from the c-myc Immunoprecipitation purifi cation gene product Invitrogen Champion™ pET Small ubiquitin like Produces soluble protein in SUMO modifi er (SUMO) fusion Escherichia coli , for purifi cation the SUMO moiety can be cleaved by the highly specifi c and active SUMO (ULP-1) protease at the carboxyl terminal, producing a native protein Takara pCold TF DNA Vector Cold shock expression vector The chaperone ensures co- that expresses Trigger Factor translational protein folding. (TF) chaperone as a soluble tag Affi nity purifi cation followed by along with His-Tag sequence Thrombin or Factor Xa cleavage Takara Single Protein MazF tag Utilizes E. coli protein MazF which Production system is an endoribonuclease and cleaves mRNA at ACA sequence. Thus transcript of interest should lack ACA sequence so that it cleaves ones derived from the host proteins or others at ACA sequences BioRad Profi nity eXact Prodomain of the subtilisin Utilizes a modifi ed form of the fusion-tag system protease subtilisin protease, which is immobilized onto a chromatographic support and used to generate pure, tag-free target protein in a single step (continued) 100 D. Chakravorty and S. Patra

Table 7.1 (continued)

Company Vector system Tags (N/C terminus) Purifi cation European Molecular TAP tag Calmodulin binding peptide Tagged protein binds to beads coated Biology Laboratory (CBP) from the N-terminal, with IgG, the TAP tag is then broken followed by tobacco apart by an enzyme, and fi nally a etch virus protease different part of the TAP tag binds (TEV protease) cleavage reversibly to beads of a different type site and Protein A, which binds tightly to IgG Schaffer et al. 2010 SnAvi-tag Fluorescent tag Enhanced green fl uorescent protein (EGFP) to enable in vivo localization studies, the TEV-protease recognition motif to allow elution under non-denaturing conditions

cleave the protein of interest also. Thus, new In the capture phase, enzymes are isolated, proteases were discovered that have made their concentrated, and stabilized. During the interme- way into commercial expression vector systems. diate purifi cation phase, bulk impurities such PreScission protease by GE Healthcare Life as other proteins and nucleic acids, endotoxins, Sciences (Amersham Pharmacia) is one such and viruses are removed. In the polishing phase, product based on the rhinovirus protease. New trace amounts or closely related substances England Biolabs markets a system named are removed (GE Healthcare 2009 ). Another IMPACT where the cleavage is affected by the worth-mentioning innovation is in the use of activity of a self-splicing protein called an intein adenosine triphosphate (ATP)-affi nity chroma- (Chong et al. 1998 ). For tag-free expression, tography. This has been widely used to purify Wacker Biotech ESETEC® technology provides various ATP- binding enzymes such as kinases an innovative and highly effi cient Escherichia and β- and γ-glutamate decarboxylase using coli expression system, which enables secretion affi nity purifi cation by ATP-matrix attachment of native or recombinant enzyme products into (Jeansonne et al. 2006 ; Bendz et al. 2007 ; Dhillon the fermentation broth. This simplifi es primary et al. 2009 ). recovery and purifi cation processes. The system Recent trend in enzyme detection after purifi ca- is based on a two-step export mechanism: fi rst, tion has seen the development of Caliper Life the target product is transported across the Sciences’ LabChip90, an automated electrophoresis cytoplasmatic membrane into the periplasmatic system and an alternative to the conventional slab space. During this step the signal peptides are gel electrophoresis system such as SDS-PAGE. It is cleaved off, releasing the native product. The sec- less expensive than traditional gel electrophoresis ond step is mediated by a unique feature of the and increases sample throughput by two- to three- proprietary WACKER Secretion Strain. The cor- fold. This system integrates and automates all of the rectly folded product is secreted from the processing steps right from sampling to separation periplasmatic space across the outer membrane to detection and to quantifi cation via the software into the culture broth for easy recovery. Although that comes with the instrument (Smith 2005 ). the tagged system allows for single-step enzyme purifi cation through affi nity chromatography, it suffers from the disadvantage that affi nity Enzyme Engineering tags may sometimes interfere with the post- purifi cation use of the protein. The solution to Enzyme engineering leads to tailor-made physi- such a problem is known as Capture, Intermediate cal and catalytic properties of enzymes. This Purifi cation, and Polishing (CIPP) technique. approach can be used either to study the function 7 Advance Techniques in Enzyme Research 101 of an enzyme or to produce an enzyme of desired genes of over 1 Kb can now be effi ciently and physicochemical properties. Two popular app- economically synthesized in a matter of weeks. roaches are directed evolution and rational In creating the synthetic gene, degeneracy of the design. An effective directed evolution strategy genetic code can be used to generate nucleotide combines different mutagenesis methods with sequences that have useful properties such as effi cient high-throughput screening or selection large numbers of endonuclease restriction sites, assays. Rational design is pillared on the three- optimized primer sites for the polymerase chain dimensional enzyme structure and identifi cation reaction (PCR) and sequencing, and desired of amino acids involved in enzyme functionality levels of GC content and codon bias (Withers- or catalysis and constructing mutants by single- Martinez et al. 1999 ). site or multisite, site-directed mutagenesis (Cambon et al. 2010). Song et al. ( 2008 ) used a modifi ed method of site-directed mutagenesis Enzyme Structure and Function using RecA-mediated homologous recombina- Determination tion and temperature-sensitive replications, site- directed mutagenesis of the catalytic triad amino NMR spectroscopy and X-ray crystallography acid cysteines in tandem ketosynthase domains are the routine approaches that provide an atomic- to test the function they play in OZM biosynthe- level, 3D structural model of a protein or an sis (Song et al. 2008 ). A modifi ed and better enzyme. Crystallization has been automated method for multisite-directed mutagenesis was which allows the use of small protein sample developed by Tian et al. (2010 ) which is based volumes, and development of crystallization on polymerase chain reaction (PCR), DpnI robots has contributed to increasing the effi ciency digestion, and overlap extension. The method of screening crystallization conditions (Liu and does not require 5′-phosphorylated primers and Hsu 2005 ). The applications of these high-resolu- ligation and, thus, signifi cantly simplifi es the tion approaches, however, are limited by enzyme routine work and reduces the experimental cost size, conformational fl exibility, and aggregation for multisite-directed mutagenesis (Tian et al. propensity. Mass spectrometry (MS) has become 2010 ). Today, rational engineering is also bene- the recent method of choice for studying protein fi tted through in silico approaches involving structure, dynamics, interactions, and function. multiple web-based tools and software packages Hamuro et al. (2003 ) published the use of rapid which aid in saving time, effort, and money. For analysis of protein or enzyme structure by example, protein mutant data is being compiled hydrogen/deuterium exchange mass spectrometry in the protein mutant database (PMD). This data- (Hamuro et al. 2003 ). Recent developments in base provides information on the results of func- mass spectrometry as the method of choice for tional and/or structural infl uences that can be the high-throughput identifi cation of proteins and brought about by amino acid mutations. It covers their modifi cations led to the concept of protein natural as well as artifi cial mutants, including cross-linking to determine structural information random and site-directed ones. The advantage of about protein like enzymes in combination with PMD is that it is based on literature, not on pro- mass spectrometry (Leitner et al. 2010 ). The teins (Kawabata et al. 1999). A gist of the vari- overall process in protein structure determination ous other web-based tools has been listed in has been illustrated in Fig. 7.5 . Table 7.2 . The knowledge of three-dimensional structure Apart from the aforesaid, a methodology in and space by these techniques is still limited. enzyme evolution known as synthetic gene tech- Thus, computational methods such as comparative nology has recently evolved which results in de and de novo approaches and molecular dynamic novo synthesis of entire protein-coding sequences simulations are intensively used as alternative tools of the enzyme from preannealed oligonucleotides to predict the three-dimensional structures and (Gustafsson et al. 2004). Through this strategy, dynamic behavior of enzymes (Liu and Hsu 2005 ). 102 D. Chakravorty and S. Patra

Table 7.2 Web tools and software packages for predicting protein and enzyme stability on point mutations Sl No Web tool/software Applications Reference 1 HotSpot Wizard HotSpot Wizard is a web server for automatic Pavelka et al. (2009 ) identiﬁ cation of ‘hot spots’ for engineering of substrate speciﬁ city, activity or enantioselectivity of enzymes and for annotation of protein structures 2 iPTREE-STAB It is a web based tool which discriminates the Huang et al. (2007 ) stability of proteins and predicts their stability changes (ΔΔG) upon single amino acid substitutions from amino acid sequence 3 Eris server It is a protein stability prediction server. Eris server Yin et al. ( 2007 ) calculates the change of the protein stability induced by mutations (ΔΔG) 4 I-Mutant Support Vector Machines based Predictor of Protein Capriotti et al. (2005 ) stability changes upon Single Point Mutation from the Protein Sequence and Structure. The tool correctly predicts whether the protein mutation stabilises or destabilises the protein in 80 % of the cases when the three-dimensional structure is known and 77 % of the cases when only the protein sequence is available 5 CUPSAT A web tool to analyse and predict protein stability Parthiban et al. (2006 ) changes upon point mutations 6 MUpro It is a set of machine learning programs to predict Cheng et al. (2006 ) how single-site amino acid mutation affects protein stability 7 Swiss-PdbViewer This software allows browsing a rotamer library http://spdbv.vital-it.ch/ in order to change amino acids sidechains. This can be very useful to quickly evaluate the putative effect of a mutaion before actually doing the lab work

Computer graphics display of 3-D structure and relational Enzyme selection Analyze structure database of sequence, 3-D structure and theoretical calculation

Gene cloning/or Determine 3-D structure Molecular modeling chemical synthesis Site directed mutagenesis

Produce large amount Crystallize Enzyme NMR 2-D analysis Gene construction of Enzyme

Purification Biological characterization Mutant gene expression

Enzyme-Ligand Modified Enzyme complex

Fig. 7.5 Flowchart showing the overall process of protein or enzyme structure determination (This ﬁ gure has been adapted from Liu and Hsu (2005 ) )

Currently high-throughput genome sequencing formatics tools is necessary to process and priori- programs coupled have provided researchers tize the plethora of data. For this purpose freely with a perplexing array of sequence and biological accessible enzyme databases like BRENDA have data to contend with. Thus, application of bioin- been recently developed that have been classifi ed 7 Advance Techniques in Enzyme Research 103 by the International Union of Biochemistry and alignment and template structure (Venselaar et al. Molecular Biology ( IUBMB ). Every classifi ed 2010 ). Today various software packages exist enzyme is characterized with respect to its catalyzed which involve rigorous mathematical models to biochemical reaction . Kinetic properties of the build the protein model structures. Some of them corresponding reactants , i.e., substrates and are, namely, Modeller (Fiser and Sali 2003 ), products , are also described in detail (Scheer Swiss Model (Arnold et al. 2009 ), and I-Tasser et al. 2011 ). Other existing enzyme databases are (Roy et al. 2010 ). Enzyme nomenclature database ( http://enzyme. Enzyme–substrate or inhibitor binding is of expasy.org/ ) and IntEnz ( http://www.ebi.ac.uk/ critical importance for biomedical research; in intenz/ ). In addition to enzyme databases, soft- vitro analysis of the same is time consuming and ware has been developed to analyze and present requires the assistance from patient unbiased functional and statistical data for enzyme kinetics; researchers. To speed up work and reduce the various software tools have been developed. workload due to trial and error methodologies, EZ-Fit™ by Perrella Scientifi c is one such soft- computational programs have been developed ware. This software has features for nonlinear known as molecular docking to analyze enzyme– regression of enzyme inhibition data. It makes ligand binding. Molecular docking is the compu- data entry simple followed by selection of an tational modeling of the quaternary structure of equation from the menu of commonly used inhi- complexes formed by two or more interacting bition models. EZ-Fit does the rest automatically, biological macromolecules enabling us to predict curve-fi ts the data to one or more models, displays how small molecules, such as substrates or drug the results, and draws a graph of the data and candidates, bind to a receptor of known 3D curve (Perrella 1999 ). VisualEnzymics by protein or enzyme structures. Popular program SoftZymics is another such tool which presents a packages are AutoDock (Goodsell and Olson custom visual interface. Through this software 1990 ), Gold (Jones et al. 1995 ), and FlexX we can see curve fits in real time providing (Rarey et al. 1996 ) and are available to carry out the user with interactive control. It possesses the docking studies. models for one substrate rate saturation data, one Apart from enzyme structure prediction, we substrate one inhibitor data, pH rate profi les, can enjoy the benefi ts of computational program- exponential data, dose response data, two substrate ming and simulate the behavior of enzymes in data, one substrate one activator data, binding real time up to femto (10 −15) second scale. This data, and tight binding data. Users can access the process is known as molecular dynamics simulation. web link http://www.softzymics.com/index.htm Enzyme–substrate binding, enzyme–inhibitor for further details. interaction, enzyme folding, and thermodynam- Nowadays, a plethora of enzyme sequence ics can be investigated through MD simulation. data exist in the NCBI ( http://www.ncbi.nlm.nih. Software packages regularly utilized for MD simu- gov) and UniProt ( www.uniprot.org ) databases lation are NAMD (Nelson et al. 1996 ), AMBER but lack crystallized structure information in the (Ponder and Case 2003), and GROMACS Protein Data Bank ( www.rcsb.org ). Thus, their (Van der Spoel et al. 2005 ). This computational functional elucidation becomes diffi cult. In modern method calculates the time- dependent behavior era to speed up work, protein structure models are of a molecular system, and it is now routinely designed computationally based on the principle used to examine the structure, dynamics, and of homology with known target protein structures, thermodynamics of biological molecules and and the process is known as homology modeling their complexes. Numerous enzymes like protein or comparative modeling. Homology modeling tyrosine phosphatase 1B (Peters et al. 2000 ) and relies on the production of an alignment that metaloenzyme thiocyanate hydrolase (Peplowski maps residues in the query sequence to residues and Nowak 2008 ) have been investigated through in the template sequence. The quality of the this approach. Buch et al. (2011 ) successfully model depends on the quality of the sequence reconstructed the complete binding process of the 104 D. Chakravorty and S. Patra enzyme–inhibitor complex trypsin–benzamidine enable the choice of a wide range of substrates by performing 495 molecular dynamics simulations that is insoluble in most organic solvents (Kim (Buch et al. 2011 ). et al. 2006b ). For example, Erbeldinger and colleagues used 1-butyl-3-methylimidazolium hexafl uorophosphate as the solvent for Enzyme Stabilization thermolysin-catalyzed synthesis of Z-aspartame (Erbeldinger et al. 2000). Along similar lines, It is a fact that enzymes function in industrial lipase-catalyzed transamination of carboxylic solvents with enhanced technological utility acids with ammonia has been established in (Klibanov 2001 ). Enzyme-catalyzed reactions 1-butyl-3-methylimidazolium tetrafl uoroborate in organic solvents and supercritical fl uids have (Kim et al. 2006b ). found numerous potential applications, some of Other than room temperature ionic liquids, which have been successfully commercialized nanostructures like nanoparticles, nanofi bers, (Klibanov 2001 ). Research over the past few mesoporous silica, and nanoparticles prepared decades has shown that in such solvents, via sol–gel encapsulation have been instrumen- enzymes can catalyze reactions which are seem- tal in enzyme stabilization. More recent devel- ingly impossible in water (Klibanov 2001 ). opment in this area is the “single-enzyme A priori, it can be of concern that when an nanoparticles (SENs)” which is a new enzyme enzyme is exposed to an organic solvent, it can composite of nanometer scale by Kim and Grate denature, but enzymes were instead discovered (Kim et al. 2006a ). Nanostructures are better to be much stable and to retain “molecular carriers for immobilization of enzymes since memory” in such solvents. The reason behind is their reduced size can generally improve the that reactions such as deamidation of thermola- effi ciency of immobilized enzymes due to larger bile residues and hydrolysis of peptide bonds do surface area per unit mass, pore sizes tailored to not ensue in solvents in the absence of water. protein molecule dimensions, functionalized The stability of enzymes in organic solvents is surfaces, multiple sites for interaction or attach- also due to their conformational rigidity in the ment, and reduced mass-transfer limitations dehydrated state (Klibanov 2001 ). Furthermore, (Kim et al. 2006a ). A recent report using mag- proteolytic degradation does not occur in the netic nanoparticles for the enzyme immobiliza- absence of water as they are insoluble in organic tion was demonstrated with covalently attached solvents (Klibanov 2001). Irrespective of the lipase on the magnetic γ-Fe 2 O3 nanoparticles aforesaid, large-scale usage of organic solvents and showed stability for a month (Dyal et al. provided with certain limitations, such as 2003). One disadvantage in using nanoparticles increased polarity of organic solvents, typically is that their dispersion in reaction solutions and results in reduced catalytic activity (Lee and the subsequent recovery for reuse are often Dordick 2002 ). Recent advances in nanotechnol- found to be a daunting task. Thus, a replacement ogy have promised solution in this booming of nano-particles can be the electrospun nanofi - area of enzyme stabilization, suggesting the bers which provide a large surface area for the usage of inorganic nonaqueous solvents or the attachment or entrapment of enzymes and the room temperature ionic liquids (RTILs) for enzyme reaction. Another recent achievement in reaction medium engineering or by means of nanoenzymology is the nanoparticles called nanostructures with large surface area for the Single-Enzyme Nano-particles (SENs) (Kim immobilization of enzyme molecules (Kim et al. 2006a). Each enzyme molecule is sur- et al. 2006a ). Ionic liquids have proved to be a rounded with a porous composite organic/ better reaction media for enzymes since they inorganic network of less than a few nanome- have essentially zero vapor pressure, are highly ters thick. Converting free enzymes to SENs can stable to the range of operating conditions in result in signifi cantly more stable catalytic bioreactors, are stable to air and water, and activity, while the nanoscale structure of the 7 Advance Techniques in Enzyme Research 105

SEN does not impose a serious mass-transfer is a separation of the solution into two phases limitation on substrates (Kim et al. 2006a ). (Minuth et al. 1996 ). Interestingly another breakthrough in nanoen- Affi nity chromatography techniques are the zymology was reported in November 2006 when most applied tools available for downstream it was discovered that silica-spun FMS pores processing. However, porous affi nity supports can which are hexagons of approximately 30 nm in only work in a later stage of purifi cation in clear diameter and mimic the crowding of cells can be solutions and not in early stages when fouling used for refolding of enzymes and upon refolding compounds are present in the system. Nonporous the enzyme gets reactivated and becomes capable support particles can replace such a system if and of catalyzing thousands of reactions a second only if the size of nonporous support particles is (PNNL 2006 ). in the range of 0.1–1 μm. Magnetic separation seems to be the only feasible method for the recovery of such small particles from the bio- Improved Bioseparation Techniques: logical debris. For separating enzymes, an appro- Downstream Processing of Enzymes priate affi nity ligand is usually immobilized on a magnetic carrier, such as silanized magnetite, or Downstream processing of enzymes holds an on polymeric magnetic (chitin) or magnetizable important position in industries as the bulk of the particles. product cost is determined by the steps leading to The electrophoretic separation of proteins purifi cation of enzymes. Thus, biotechnology without gels has been an innovative goal in sep- companies are prospecting for new and improved aration research. Electrophoretic separations processing methods. New improved methods are infl uenced by factors such as size (or molec- over the conventional chromatography and mem- ular weight), shape, secondary structure, and brane fi ltration techniques are the cloud point charge of the macromolecule or cell. Ohmic extraction (CPE) and fi eld-assisted separation heating hinders the scale-up of electrophoresis. technologies like MAGSEP (magnetic separator) The heat generated is equal to the product of and ELECSEP (electrophoretic separator) for the the current and voltage, and this heat can cause quantitative separation of enzymes are gaining free convection and mixing within the system. more importance. Too much heat can denature the labile biomole- Cloud point extraction technique has been cules or cells. To overcome such problems, the used to separate proteinase 3 (Pr3) from neutro- multistage electrophoretic method was devel- phil azurophilic granules, tyrosinase from mush- oped which is a combination of free electropho- room pileus by Garcia-Carmona and co-workers, resis and multistage extraction. The apparatus cholesterol oxidase extracted from Nocardia used is known as advanced separation apparatus rhodochrous by Kula and co-workers, and hexo- (ADSEP). This ADSEP was modifi ed to act as an kinase and lactate dehydrogenase from aqueous ELECSEP by replacing the chamber bottoms with solutions using Triton X-114 as the surfactant. metal cover plates. In these systems electrodes The technique in brief can be described as are kept over these cover plates with gaskets surface- active agents (surfactants, detergents) in between them. Each cavity has a height of can aggregate in aqueous solution to form few millimeters ensuring the fl uid within it to colloidal-sized clusters referred to as micelles remain isothermal during the application of an (normal micelles). The minimum concentration electric fi eld, which transfers the molecules of surfactant required for this phenomenon to from the bottom to the top cavity. As each occur is called the critical micelle concentration molecule is transferred to the new cavity, it is (CMC). Upon heating, aqueous solutions of either drawn into the upper cavity by the elec- many nonionic surfactants become turbid at a tric fi eld or left in the lower cavity, depending temperature known as the cloud point (or the on its electrophoretic mobility (Karumanchi lower consulate temperature), above which there et al. 2002 ). 106 D. Chakravorty and S. Patra

Conclusion References

Modern enzyme research is engrossed in Arnold K, Bordoli L, Kopp J, Schwede T (2009) The developing better ways of isolation, purifi cation, SWISS-MODEL workspace: a web-based environment for protein structure homology modelling. characterization, and stabilization of enzymes in Bioinformatics 22(2):195–201 an application-oriented approach. Numerous Aw KL, Yao SQ, Uttamchandani M (2010) Nanodroplet companies like Qiagen, Applied Biosystems, microarrays for high-throughput enzyme screening. BioChrom, and Amersham Biosciences are Methods Mol Biol 669:79–93 van Bailen JB, Li Z (2002) Enzyme technology: an over- involved in such research to improve effi cacy of view. Curr Opin Biotechnol 13(4):338–344 the whole procedure. Improvement of existing Bendz H, Ruhland SC, Pandya M et al (2007) Human heat methods in enzyme research is being carried out shock protein 70 enhances tumor antigen presentation through the interplay of multidisciplinary sub- through complex formation and intracellular antigen delivery without innate immune signaling. J Biol jects and multitasking through a single approach Chem 282(43):31688 or equipment. To mention a few trends, miniatur- Buch I, Giorgino T, De Fabritiis G (2011) Complete ization of equipments, robotic handling of high- reconstruction of an enzyme-inhibitor binding process end equipments, and development of label-free by molecular dynamics simulations. Proc Natl Acad Sci USA 108(25):10184–10189 techniques by the use of biosensors to detect as Cambon E, Piamtongkam R, Bordes F et al (2010) A low as femto (10−15 ) molar quantity of enzymes new Yarrowia lipolytica expression system: an effi - are the modern approaches. Available biological cient tool for rapid and reliable kinetic analysis of databases and knowledge bases of enzymes have improved enzymes. Enzym Microb Technol 47:91–96 added new tweak to existing methodologies by Capriotti E, Fariselli P, Casadio R (2005) I-Mutant2.0: pre- saving time and effort and thus increasing indus- dicting stability changes upon mutation from the pro- trial potential for the applicability of enzymes. tein sequence or structure. Nucleic Acids Res Modern PCR like self-sustained sequence replica- 33:W306–W310 Chen B-Y, Janes HW (2002) PCR cloning protocols. tion (3SR), loop-mediated isothermal amplifi cation Humana Press, New Jersey PCR, RACE, and cloning techniques like Cheng J, Randall A, Baldi P (2006) Prediction of protein Gateway and TOPO cloning strategies is bringing stability changes for single-site mutations using sup- enzyme research to the forefront. Breakthroughs port vector machines. Proteins Struct Funct Bioinform 62(4):1125–1132 in evolution of enzyme research that have been Chong S, Montello GE, Zhang A et al (1998) Utilizing brought about by approaches like metagenomics the C-terminal cleavage activity of a protein splicing are leading to the isolation of enzymes even from element to purify recombinant proteins in a single uncultured sources and cell-free in vitro enzyme chromatographic step. Nucleic Acids Res 26:5109–5115 expression along with their affi nity tag-free Chu X, He H, Guo C, Sun B (2008) Identifi cation of two expression and in silico analysis of their kinetics novel esterases from a marine metagenomic library with substrates and inhibitors are possible derived from South China Sea. Appl Microbiol through the aid of rigorous mathematical algo- Biotechnol 80(4):615–625 Chuang L-T, Chen D-C, Nicaud J-M, Madzak C, Chen rithm packages known as molecular dynamics Y-H, Huang Y-S (2010) Co-expression of heterolo- simulation and molecular docking. Another gous desaturase genes in Yarrowia lipolytica . worth-mentioning approach is the synthetic gene New Biotechnol 27(4):277–282 technology which can be used to design and syn- Compton J (1991) Nucleic acid sequence-based amplifi - cation. Nature 350(6313):91–92 thesize enzyme genes with desired properties. Dhillon AS, Yip YY et al (2009) The C-terminus of Raf-1 Improvements have also occurred in the down- acts as a 14-3-3-dependent activation switch. Cell stream processing and stabilization of enzymes. Signal 21(11):1645 Conclusively we can say that enzyme research Dyal A, Loos K, Noto M, Chang SW, Spagnoli C, Shafi KV, Ulman A, Cowman M, Gross RA (2003) Activity has really evolved and has entered a new era, and of Candida rugosa lipase immobilized on gamma- we expect more innovations in this fi eld in the Fe2O3 magnetic nanoparticles. J Am Chem Soc near future. 125(7):1684–1685 7 Advance Techniques in Enzyme Research 107

Erbeldinger M, Mesiano AJ, Russell AJ (2000) Enzymatic Jeansonne DP, Bordes TJ et al (2006) A rapid ATP catalysis of formation of Z-aspartame in ionic liquid – affi nity- based purifi cation for the human non-receptor an alternative to enzymatic catalysis in organic tyrosine kinase c-Src. Protein Expr Purif 46:240 solvents. Biotechnol Prog 16(6):1129–1131 Jones G, Willett P, Glen RC (1995) Molecular recognition Fiser A, Sali A (2003) Modeller: generation and refi ne- of receptor sites using a genetic algorithm with a ment of homology-based protein structure models. description of desolvation. J Mol Biol 245:43–53 Methods Enzymol 374:461–491. doi: 10.1016/S0076- Kawabata T, Ota M, Nishikawa K (1999) The protein 6879(03)74020-8 mutant database. Nucleic Acids Res 27(1):355–357 Framingham M (April 17, 2012). Business Wire Kim J, Grater JW, Wang P (2006a) Nanostructures for Data-independent SWATH™ acquisition on the enzyme stabilization. Chem Eng Sci 61(3):1017–1026 TripleTOF™ 5600 System enables comprehensive Kim J, Jia H, Wang P (2006b) Challenges in biocatalysis proteome-wide quantitation. Mol Cell Proteomic. for enzyme-based biofuel cells. Biotechnol Adv http://www.businesswire.com/news/home/ 24:296–308 20120417005204/en/Breakthrough-Proteomics- Klein G, Reymond JL (2001) An enzyme assay using Technique- SWATH%E2%84%A2-Acquisition- pM. Angew Chem Int Ed Engl 40:1771–1773 Worldwide- Scientifi c Klibanov AM (2001) Improving enzymes by using them Galaev IY (1999) New methods of protein purifi cation. in organic solvents. Nature 409:241–246 Affi nity ultrafi ltration. Biochemistry (Mosc) 64(8): Karumanchi R, Doddamane SN, Sampangi C, Todd PW 849–856 (2002) Field-assisted extraction of cells, particles and Gateway® Cloning. Life Technologies. USA. www. macromolecules. Trends Biotechnol 20(2):72–78 lifetech.com/gateway Kushnir S, Gase K, Breitling R, Alexandrov K (2005) GE Healthcare (2009) Strategies for protein purifi cation Development of an inducible protein expression system handbook. www.gelifesciences.com/protein-purifi cation based on the protozoan host Leishmania tarentolae . Goodsell DS, Olson AJ (1990) Automated docking of Protein Expr Purif 42(1):37–46 substrates to proteins by simulated annealing. Proteins Lee M-Y, Dordick J-S (2002) Enzyme activation for Struct Funct Genet 8:195–202 nonaqueous media. Curr Opin Biotechnol 13(4): Guatelli JC, Whitfi eld KM, Kwoh DY, Barringer KJ, 376–384 Richman DD, Gingeras TR (1990) Isothermal in vitro Leitner A, Walzthoeni T, Kahraman A (2010) Probing amplifi cation of nucleic acids by a multienzyme native protein structures by chemical cross linking, reaction modeled after retroviral replication. Proc Natl mass spectrometry, and bioinformatics. Mol Cell Acad Sci 87:1874–1878 Proteomic 9:1634–1649 Gustafsson C, Govindarajan S, Minshull J (2004) Codon Li WA, Barry ZT, Cohen JD, Wilder CL, Deeds RJ, bias and heterologous protein expression. Trends Keegan PM, Platt MO (2010) Detection of femtomole Biotechnol 22:346–353 quantities of mature cathepsin K with zymography. Hallin S, Lindgren P-E (1999) PCR detection of genes Anal Biochem 401(1):91–98 encoding nitrite reductase in denitrifying bacteria. Liu H-L, Hsu J-P (2005) Recent developments in struc- Appl Environ Microbiol 65(4):1652–1657 tural proteomics for protein structure determination. Hamuro Y, Coales SJ, Southern MR et al (2003) Rapid Proteomics 5:2056–2068 analysis of protein structure and dynamics by hydrogen/ Madzak C, Tréton B, Blanchin-Roland S (2000) Strong deuterium exchange mass spectrometry. J Biomol hybrid promoters and integrative expression/secretion Tech 14(3):171–182 vectors for quasi-constitutive expression of heterolo- Handelsman J (2004) Metagenomics: application of gous proteins in the yeast Yarrowia lipolytica. J Mol genomics to uncultured microorganisms. Microbiol Microbiol Biotechnol 2(2):207–216 Mol Biol Rev 68(4):669–685 Mai V, Wiegel J (2000) Advances in development of a He M (2008) Cell-free protein synthesis: applications in pro- genetic system for Thermoanaerobacterium spp.: teomics and biotechnology. N Biotechnol 25:126–132 expression of genes encoding hydrolytic enzymes, Huang L-T, Gromiha MM, Ho S-Y (2007) iPTREE- development of a second shuttle vector, and integration STAB: interpretable decision tree based method for of genes into the chromosome. Appl Environ Microbiol predicting protein stability changes upon mutations. 66(11):4817–4821 Bioinformatics 23(10):1292–1293 Male KB, Nguyen AL, Luong JHT (1990) Isolation of Hwang B-Y, Kim B-G (2004) High-throughput screening urokinase by affi nity ultrafi ltration. Biotechnol Bioeng method for the identifi cation of active and enantiose- 35(1):87–93 lective ω-transaminases. Enzym Microb Technol Minuth T, Thommes J, Kula MR (1996) A closed concept 34(5):429–436 for purifi cation of the membrane-bound cholesterol Ishikawa K, Watanabe M, Kuroita T et al (2005) Discovery oxidase from Nocardia rhodochrous by surfactant- of a novel restriction endonuclease by genome based cloud-point extraction, organic-solvent extrac- comparison and application of a wheat-germ-based tion and anion-exchange chromatography. Biotechnol cell-free translation assay: PabI (5′-GTA/C) from Appl Biochem 23:107 the hyperthermophilic archaeon Pyrococcus abyssi . Mori Y, Nagamine K, Tomita N, Notomi T (2001) Nucleic Acids Res 33(13):e112 Detection of loop-mediated isothermal amplifi cation 108 D. Chakravorty and S. Patra

reaction by turbidity derived from magnesium pyro- Sambrook J, Fritsch EF, Maniatis T (2001) Molecular phosphate formation. Biochem Biophys Res Commun cloning: a laboratory manual. Cold Spring Harbor 289:150–154 Laboratory, Cold Spring Harbor Nelson M, Humphrey W, Gursoy A et al (1996) NAMD Schaffer U, Schlosser A, Müller KM, Schäfer A, Katava – a parallel, object-oriented molecular dynamics pro- N, Baumeister R, Schulze E (2010) SnAvi–a new tan- gram. Int J Supercomput Appl High Perform Comput dem tag for high-affi nity protein-complex purifi cation. 10:251–268 Nucleic Acids Res 38(6):e91. doi:10.1093/nar/ Notomi T, Okayama H, Masubuch H et al (2000) gkp1178 Loop-mediated isothermal amplifi cation of DNA. Scheer M, Grote A, Chang A, Schomburg I, Munaretto C, Nucleic Acids Res 28(12):e63 Rother M, Söhngen C, Stelzer M, Thiele J, Schomburg OECD (1998) Biotechnology for clean industrial products D (2011) BRENDA, the enzyme information system and processes. OECD, Paris in 2011. Nucleic Acids Res 39:D670–D676. Olsen M, Iverson B, Georgiou G (2000) High-throughput doi: 10.1093/nar/gkq1089 screening of enzyme libraries. Curr Opin Biotechnol Shuman S (1991) Recombination mediated by vaccinia 11:331–337 DNA topoisomerase I in Escherichia coli is sequence Parrella FW (1999) EZ-Fit™ by Perrella Scientifi c, specifi c. Proc Natl Acad Sci 88:10104–10108 enzyme inhibition kinetics software. http://www.jlc. Shuman S (1994) Novel approach to molecular cloning net/~fperrell/index.htm and polynucleotide synthesis using vaccinia DNA Parthiban V, Gromiha MM, Schomburg D (2006) topoisomerase. J Biol Chem 269:32678–32684 CUPSAT: prediction of protein stability upon point Smith C (2005) Striving for purity: advances in protein mutations. Nucleic Acids Res 34:W239–W242 purifi cation. Nat Methods 2:71–77. doi: 10.1038/ Pavelka A, Chovancova E, Damborsky J (2009) HotSpot nmeth0105-71 Wizard: a web server for identification of hot spots Song D, Coughlin J et al (2008) Alternative method for in protein engineering. Nucleic Acids Res 37(2): site-directed mutagenesis of complex polyketide W376–W383 synthase in Streptomyces albus JA3453. Acta Biochim Peplowski L, Nowak W (2008) Molecular dynamics Biophys Sin 40:319–326 simulations of the metaloenzyme thiocyanate hydrolase Suzuki T, Yasui K (2011) Plasmid artifi cial modifi cation: with non-corrinoid Co(III) in active site. CBSB08 a novel method for effi cient DNA transfer into bacte- Proceed NIC Workshop 40:353–356 ria. Methods Mol Biol 765:309–326 Peters GH, Frimurer TM, Andersen JN, Olsen OH (2000) Tian J, Liu Q, Dong S et al (2010) A new method for Molecular dynamics simulations of protein-tyrosine multi-site-directed mutagenesis. Anal Biochem phosphatase 1B. II. substrate-enzyme interactions and 406(1):83–85 dynamics. Biophys J 78(5):2191–2200 Tirawongsaroj P, Sriprang R, Harnpicharnchai P et al PNNL Media Contacts November 28, 2006. Night of the (2008) Novel thermophilic and thermostable lipolytic living enzyme 509:375–3776 enzymes from a Thailand hot spring metagenomic Ponder JW, Case DA (2003) Force fi elds for protein simu- library. J Biotechnol 133(1):42–49 lations. Adv Protein Chem 66:27–85 Van Der Spoel D, Lindahl E, Hess B, Groenhof G, Puig O, Caspary F, Rigaut G et al (2001) The tandem Mark AE, Berendsen HJ (2005) GROMACS: fast, affi nity purifi cation (TAP) method: a general procedure fl exible, and free. J Comput Chem 26(16):1701–1718. of protein complex purifi cation. Methods 24:218–229 doi: 10.1002/jcc.20291 Rarey M, Kramer B, Langauer T et al (1996) A fast fl exi- Venselaar H, Joosten RB, Vroling B et al (2010) ble docking method using an incremental construction Homology modelling and spectroscopy, a never- algorithm. J Mol Biol 261:470–489 ending love story. Eur Biophys J 39:551–563 Ren X (2007) Rosetta-gami ™ B host strains from Wahler D, Reymond J-L (2001) High-throughput screening Novagen. Biocompare. http://www.biocompare.com/ for biocatalysts. Curr Opin Biotechnol 12:535–544 Articles/ProductReview/981/Rosetta-gami--B-Host- Walker GT, Fraiser MS, Schram JL, Little MC, Nadeau JG, Strains-From-Novagen.html Malinowski DP (1992) Strand displacement amplifi ca- Rose TM, Henikoff JG, Henikoff S (2003) CODEHOP tion–an isothermal, in vitro DNA amplifi cation tech- (COnsensus-DEgenerate Hybrid Oligonucleotide nique. Nucleic Acids Res 20(7):1691–1696 Primer) PCR primer design. Nucleic Acids Res Ward WW, Swiatek G (2009) Protein purifi cation. Curr 31(13):3763–3766 Anal Chem 5(2):1–21 Roy A, Kucukural A, Zhang Y (2010) I-TASSER: a uni- Wang Q, Wu H, Wang A, Du P, Pei X, Li H, Yin X, Huang fi ed platform for automated protein structure and L, Xiong X (2010) Prospecting metagenomic enzyme function prediction. Nat Protoc 5:725–738 subfamily genes for DNA family shuffl ing by a novel Sakamoto Y, Nakade K, Konno N (2011) Endo-β-1,3- PCR-based approach. J Biol Chem 285(53):41509– glucanase GLU1, from the fruiting body of Lentinula 41516. doi: 10.1074/jbc.M110.139659 edodes, belongs to a new glycoside hydrolase family. Withers-Martinez C, Carpenter EP, Hackett F, Ely B, Appl Environ Microbiol 77(23):8350–8354 Sajid M, Grainger M, Blackman MJ (1999) PCR- 7 Advance Techniques in Enzyme Research 109

based gene synthesis as an efﬁ cient approach for Xu T, Zhang L, Wang X, Wei D, Li T (2009) Structure- expression of the A+T-rich malaria genome. Protein based substrate screening for an enzyme. BMC Eng 12:1113–1120 Bioinform 10:257. doi: 10.1186/1471-2105-10-257 Xiao H, Tanaka T, Ogawa M, Yada RY (2007) Expression Yin S, Ding F, Dokholyan NV (2007) Eris: an automated and enzymatic characterization of the soluble recom- estimator of protein stability. Nat Methods 4(6):466–467 binant plasmepsin I from Plasmodium falciparum . Zalatan JG, Herschlag D (2009) The far reaches of enzy- Protein Eng Des Sel 20(12):625–633 mology. Nat Chem Biol 5:516–520 Regulatory Motif Identification in Biological Sequences: An 8 Overview of Computational Methodologies

Shripal Vijayvargiya and Pratyoosh Shukla

Abstract The transcription factor binding sites (TFBS), also called as motifs, are short, recurring patterns in DNA sequences that are presumed to have a biological function. Identification of the motifs from the promoter region of the genes is an important and challenging problem, specifically in the eukaryotic genomes. In this chapter, an overview of motif identification methods has been presented. The computational methods for motif identification are classified as enumerative methods, probabilistic methods, phylogeny-based methods, and machine learning methods. The chapter also presents the standard evaluation scheme for accuracy of prediction.

Keywords TFBS • Motif identification

Introduction trolled mechanism. The main part of regulation is performed by the specific proteins called tran- Understanding the regulatory networks of higher scription factors (TFs) binding to specific tran- organisms is one of the main challenges of func- scription factor binding sites (TFBS) in regulatory tional genomics. Gene regulation is a finely con- regions associated with genes. A TFBS is also known as motif. A motif is a pattern of nucleotide bases or amino acids, which captures a biologi- S. Vijayvargiya (*) cally meaningful feature common to a group of Department of Computer Science and Engineering, nucleic acid or protein sequences. Regulatory Birla Institute of Technology, Mesra Extension Center Jaipur, 27, Malviya Industrial Area, motifs capture the patterns of DNA bases respon- Jaipur 302017, Rajasthan, India sible for controlling when and where a gene is Department of Biotechnology, Birla Institute expressed. Typically, regulatory motifs describe of Technology (Deemed University), TFBSs embedded in the DNA sequences upstream Extension Centre, Jaipur, India of a gene’s transcription start site (TSS). More e-mail: [email protected] rarely, regulatory signals may occur downstream P. Shukla of the TSS and even within coding sequences. Enzyme Technology and Protein Bioinformatics Identification of the regulatory regions and Laboratory, Department of Microbiology, M.D. University, Rohtak, India binding sites is a prerequisite for understanding e-mail: [email protected] gene regulation (Lockhart and Winzeler 2000).

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 111 DOI 10.1007/978-81-322-1094-8_8, © Springer India 2013 112 S. Vijayvargiya and P. Shukla

Initially the experimental techniques like DNase kinds of different tissues, consisting of cells that footprinting assay and the electrophoretic mobility are also dynamic and changing over time, shift assay (EMSA) have been used to discover and although the basic DNA is the same across the analyze DNA binding sites. However, the develop- body and across time. The dynamics of organ- ment of DNA microarrays and fast sequencing isms are handled by the gene regulatory mecha- techniques has led to new methods for in vivo iden- nisms. So understanding the process that tification of binding sites, such as ChIP-chip and regulates gene expression and identification of ChIP-seq (Elnitski et al. 2006). Experimental iden- those regulating element is a major challenge tification and verification of such elements are of biology. The main idea in gene expression is challenging and costly; therefore, much effort has that every gene contains the information to pro- been put into the development of computational duce a protein, which performs most of the bio- approaches. A good computational method can logical functions of an organism. The main part potentially provide high-quality prediction of the of regulation is performed by specific proteins binding sites and reduce the time required for called transcription factors (TFs) that regulate experimental verification. the production of RNA and proteins from genes. Computational discovery of the regulatory This regulation is achieved by the TFs binding elements is possible because they occur several to DNA near genes, thus influencing the recruit- times in the same genome and they may be evo- ment of RNA polymerase. RNA polymerase is a lutionarily conserved (Sandve and Drablos protein that performs the translation of genes 2006). This means that searching for overrepre- into RNA, the first step in translating genes to sented motifs across regulatory regions may dis- proteins. The regions where TFs bind are often cover novel regulatory elements. However, this called regulatory regions. The region just before simple looking problem turns out to be a tough the gene, called upstream region, is the most problem, made difficult by a low signal-to-noise basic regulatory region, but TFs can also bind in ratio. This is because of the poor conservation regulatory region that are situated after the gene and short length of the transcription factor bind- (downstream), within the gene (introns), or fur- ing sites in comparison to the length of promoter ther upstream. sequences. Many computational techniques and Gene regulation is a finely controlled mecha- tools have been developed for the motif identifi- nism, and the TFs do not attach randomly to the cation, but many of the existing tools for regula- DNA. As both the TFs and the DNA are mole- tory motif discovery have some limitations like cules containing a structured organization of posi- the limited applicability of current nucleotide tive and negative charges, binding of a TF to DNA background models, rapid failure with increasing will depend on whether these charges can be sequence length, and a tendency to report false aligned in a complementary way that forms strong positives rather than true transcription factor physical bonds between the molecules. Because binding sites (Tompa et al. 2005; Hu et al. 2005). of this, each TF will have its own sequence- This chapter presents a survey of the experimental specific requirement for binding to DNA. as well as the computational methods developed Determining where in the DNA each TF can for the motif identification. It mainly focuses on bind is important for several reasons. The regula- the computational methods. tion of genes by TFs is a basic component of a very complex system of interactions between genes. Knowing the exact locations where TFs Transcription Factor Binding Sites can bind is an important step toward determining (Motifs) how genes are regulated by a given TF, and it may also explain how slight sequence variations The DNA and genes give only static and general between individuals in the regulatory regions view of the genome. The body of an advanced may influence, for instance, the risk for a specific organism like human is composed of several disease (Fig. 8.1). 8 Regulatory Motif Identification in Biological Sequences… 113

Fig. 8.1 Gene regulation. (a) Promoter region showing binding of TFs with regulatory factors to initiate tran- distribution of motifs in upstream sequence and transcrip- scription (Figure adapted from Wray et al. (2003)) tion unit showing exons, introns, and UTR. (b) Shows

Representations and Scoring color and height proportional to the base pair Matrices frequency and information content for each position by formulas (Schneider and Stephens 1990). Motifs are generally represented as consensus IUPAC strings, position frequency matrices Consensus Sequence (PFMs), position weight matrices (PWMs), or A consensus sequence refers to the most common position-specific scoring matrices (PSSMs) in nucleotide at a particular position after multiple databases. Commonly, motifs in noncoding DNA sequences are aligned. A consensus sequence is a sequences are conserved but still tend to be degen- way of representing the results of a multiple erate, which can influence the interaction between sequence alignment, where related sequences are TFs and motifs. Therefore, after the motif data are compared to each other and similar functional collected and aligned from experimental or com- sequence motifs are found. A consensus sequence putational results, relevant consensus IUPAC shows that which residues are most abundant in strings can be constructed by selecting a degen- the alignment at each position. The consensus eracy base pair symbol for each position in the sequences are represented using the following alignment (Wasserman and Sandelin 2004). The notation: motif data can also be modeled as PFM by align- ACTNAYR ing identified sites and counting the frequency of []{} each base pair at each position of the alignment In this notation, A means that an A is always (Vavouri and Elgar 2005). Moreover, by using found in that position; [CT] stands for either C sequence logos, PWM can be displayed with or T; N stands for any base; and {A} means any 114 S. Vijayvargiya and P. Shukla base except A. Y represents any pyrimidine (C, T), and R indicates any purine (A, G). Here, 0.2 the notation [CT] does not give any indication of 0.1 the relative frequency of C or T occurring at that 0.0 position. -6 -5 -4 -3 -2 -1 +1+2+3+4

Scoring Matrices In probabilistic models the collection of binding Fig. 8.2 A sequence logo showing the most conserved sites is represented using a profile matrix. A pro- bases around the initiation codon from all human mRNAs file is a matrix of numbers containing scores for each residue or nucleotide at each position of a fixed-length motif. There are two types of weight represented as consensus IUPAC strings, posi- matrices: tion frequency matrices (PFMs), position 1. A position frequency matrix (PFM) records weight matrices (PWMs), or position-specific the position-dependent frequency of each res- scoring matrices (PSSMs) in databases. Known idue or nucleotide. PFMs can be computation- regulatory motif profiles are cataloged in data- ally discovered by tools such as MEME using bases such as TRANSFAC (Matys et al. 2003), hidden Markov models. JASPAR (Sandelin et al. 2004; Vlieghe et al. 2. A position weight matrix (PWM) contains 2006), SCPD (Zhu and Zhang 1999), TRRD log-odds weights for computing a match (Kolchanov et al. 2000), TRED (Zhao et al. score. A cutoff is needed to specify whether 2005), and ABS (Blanco et al. 2006). an input sequence matches the motif. PWMs are calculated from PFMs. Experimental Methods for Motif Sequence Logo Identification A sequence logo is a graphical representation of the sequence conservation of nucleotides (in a Since identification of regulatory regions and strand of DNA/RNA) or amino acids (in protein binding sites is a prerequisite for understanding sequences). This is a graphical representation of gene regulation (Lockhart and Winzeler 2000; the consensus sequence in which the size of a Stormo 2000), various experimental and compu- symbol is related to the frequency that a given tational techniques have been employed for this nucleotide (or amino acid) occurs at a certain purpose. Earlier, the experimental techniques of position. In sequence logos the more conserved choice to discover and analyze DNA binding the residue, the larger the symbol for that residue sites have been the DNase footprinting assay and is drawn, the less frequent, the smaller the sym- the electrophoretic mobility shift assay (EMSA). bol (Schneider and Stephens 1990). Sequence However, the development of DNA microarrays logos can be used to represent conserved DNA and fast sequencing techniques has led to new, binding sites, where transcription factors bind massively parallel methods for in vivo identifica- (Fig. 8.2). tion of binding sites, such as ChIP-chip and ChIP-seq (Elnitski et al. 2006). Following is the brief description of experimental strategies. Regulatory Motif Databases Chromatin immunoprecipitation (ChIP) provides a powerful in vivo strategy to determine There are several private and public databases its target locations for a known protein. Using devoted to compilation of experimentally formaldehyde, the proteins are cross-linked to reported, and sometimes computationally pre- the DNA, which is then fragmented into 100–500 bp dicted, motifs for different transcription factors long pieces. A protein-specific antibody, coupled in different organisms. Motifs are generally to a retrievable tag, is used to pull down 8 Regulatory Motif Identification in Biological Sequences… 115

(precipitate) the DNA-protein complex from the 1. Those that use promoter sequences from pool of DNA fragments. Finally, the associated coregulated genes from a single genome DNA is recovered, sequenced, and analyzed – 2. Those that use orthologous promoter sequences either through amplification or through the use of a single gene from multiple species (i.e., of DNA microarrays. phylogenetic footprinting) In a DNA microarray, probe sequences of 3. Those that use promoter sequences of coregu- known DNA molecules are placed on an array lated genes as well as phylogenetic footprinting of inert substrate, thus forming a collection of Motif identification algorithms can also be microscopic spots. By measuring the hybrid- classified based on the representation of motifs ization levels of target sequences, one can and the combinatorial approach used in the determine their enrichment under different design of the algorithms like: conditions or locations. Using the DNA purified 1. Consensus sequence-based or regular by ChIP, the precise location of the binding expression-based counting methods that mostly regions on the sequence can be identified. rely on exhaustive enumeration like Weeder This technique known as ChIP-on-chip pro- (Pavesi et al. 2004), YMF (Sinha and Tompa vides an efficient and scalable way for the 2003), and MITRA (Eskin and Pevzner 2002) identification of binding sites of DNA-binding 2. Matrix-based methods that use probabilistic proteins. Through recently developed genome- sequence models where the model parameters wide analyses, one can determine the binding are estimated using maximum-likelihood sites of a protein throughout the genome (Ren principle, for example, MEME (Bailey et al. et al. 2000; Horak and Snyder 2002). Methods 2006) and Gibbs sampler (Lawrence et al. generating higher resolution and coverage 1993) (Boyer et al. 2005; Odom et al. 2006) have also 3. Feature based (Chin and Leung 2008; Sharon been proposed. The method ChIP-chip has et al. 2008) become popular due to its ability to identify the Motif recognition is NP-complete and there- motifs in an unbiased manner. However, the fore cannot be solved in polynomial time unless dependence on a highly TF-specific antibody is P = NP (Evans et al. 2003). Nonetheless, numer- usually a major hurdle in performing ChIP- ous methods and tools have been developed for chip experiments. the motif identification problem, including MEME (Bailey and Elkan 1995a), AlignACE (Roth et al. 1998), REDUCE (Bussemaker et al. Computational Methods for Motif 2001), Winnower (Pevzner and Sze 2000), Identification PROJECTION (Buhler and Tompa 2002), MITRA (Eskin and Pevzner 2002), MDScan (Liu Experimental identification and verification of et al. 2002), YMF (Sinha and Tompa 2003), motifs are challenging and costly, so much pattern-driven approaches (Sze et al. 2004), effort has been put into the development of com- Weeder (Pavesi et al. 2004), DME (Smith et al. putational approaches. Over the years, many 2005), PSM1 (Rajasekaran et al. 2005), VAS algorithms and computational techniques have (Chin and Leung 2006), MEME (Bailey et al. been proposed for the motif identification. The 2006), RISOTTO (Pisanti et al. 2006), PMSprune computational motif identification schemes can (Davila et al. 2007), the voting algorithm (Chin be classified on the basis of various criteria. One and Leung 2005), MCLWMR (Boucher et al. of them is based on the type of DNA sequence 2007), and Trawler (Ettwiller et al. 2007). Despite information employed by the algorithm to dis- these available tools, the effective and efficient cover the motifs. The algorithms available for identification of motifs within datasets of interest motif finding can be classified into three major remains a challenging problem, particularly classes (Das and Dai 2007): when studying datasets derived from mammals, 116 S. Vijayvargiya and P. Shukla such as those from mice and humans. Tompa size by collecting their occurrences, at a given et al. (2005) have evaluated 13 different motif distance, from a set of coregulated DNA sequences discovery tools and showed that many of the (Carvalho et al. 2006; Marsan and Sagot 2000). tools are inefficient when used on datasets RISOTTO, being a method based on the detection derived from organisms higher than yeast. of overrepresentation of motifs in coregulated DNA sequences, faces problems in detecting weak motifs. Carvalho et al. (2011) extended Enumerative Methods RISOTTO by post-processing its output with a greedy procedure that uses prior information. The word-based enumerative methods guarantee They combined position-specific priors from dif- global optimality, and they are appropriate for ferent sources into a scoring criterion that guides short motifs and are therefore useful for motif the greedy search procedure. The rationale finding in eukaryotic genomes where motifs are behind their approach was that the combinatorial generally shorter than prokaryotes. algorithm could exploit the full space of possible Sagot (1998) introduced a word-based motifs pointing out good candidates. Afterwards approach for motif finding that is based on the a greedy search is performed over these initial representation of a set of sequences with a suffix guesses, and good motifs are up weighted by the tree. Representation of upstream sequences as prior. The reduction of the search space attained suffix trees gives a large number of possible com- in the greedy search by using the output of a combinations; therefore, efficiency of the algorithm is binatorial algorithm improves efficiency of their a challenging issue. The motif-finding algorithms algorithm, called GRISOTTO. Weeder (Pavesi et al. 2001) and MITRA The word-based methods can also be very fast (Mismatch Tree Algorithm) (Eskin and Pevzner when implemented with optimized data struc- 2002) are also based on the suffix tree and its tures such as suffix trees (Sagot 1998) and are a variant. The algorithms WINNOWER (Pevzner good choice for finding totally constrained and Sze 2000) and cWINNOWER (Liang 2003) motifs, i.e., all instances are identical. However, use word-based approach combined with graph- for typical transcription factor motifs that often theoretic methods for motif finding. have several weakly constrained positions, word- Zaslavsky and Singh (2006) used a combinato- based methods can be problematic, and the result rial optimization framework for motif finding that often needs to be post-processed with some clus- couples graph pruning techniques with a novel tering system (Vilo et al. 2000). Word-based integer linear programming formulation. They methods also suffer from the problem of produc- also proposed an approach for determining statis- ing too many spurious motifs. tical significance of uncovered motifs. Zhang and Zaki (2006a) proposed the algorithm SMOTIF to solve the structured motif search problem. Given Probabilistic Methods one or more sequences and a structured motif, SMOTIF searches the sequences for all occur- Probabilistic or randomized approaches make cer- rences of the motif. Further, they proposed another tain decisions randomly. This concept extends the algorithm, called EXMOTIF (Zhang and Zaki classical model of deterministic algorithms. The 2006b). On given some sequences and a struc- probabilistic approach involves representation of tured motif template, EXMOTIF extracts all fre- the motif model by a position weight matrix quent structured motifs that have quorum q. This (Bucher 1990). Probabilistic methods have the algorithm can also extract the composite regula- advantage of requiring few search parameters but tory binding sites in DNA sequences. rely on probabilistic models of the regulatory Pisanti et al. (2006) proposed a consensus- regions, which can be very sensitive with respect to based algorithm, called RISOTTO. This algorithm small changes in the input data. Many of the algo- exhaustively enumerates all motifs of a certain rithms developed from the probabilistic approach 8 Regulatory Motif Identification in Biological Sequences… 117 are designed to find longer or more general motifs First, subsequences that actually occur in the that are required for transcription factor binding biopolymer sequences are used as starting sites. Therefore, they are more appropriate for points for the EM algorithm to increase the motif finding in prokaryotes, where the motifs are probability of finding globally optimum motifs. generally longer than eukaryotes. However, these Second, the assumption that each sequence algorithms are not guaranteed to find globally opti- contains exactly one occurrence of the shared mal solutions, since they employ some form of motif is removed. Third, a method for probabi- local search, such as Gibbs sampling, expectation listically erasing shared motifs after they are maximization (EM), or greedy algorithms, that found is incorporated so that several distinct may converge to a locally optimal solution. motifs can be found in the same set of The Gibbs sampling (Lawrence et al. 1993) sequences. algorithm is one of the simplest Markov chain Further, Bailey et al. (2009) developed the Monte Carlo algorithms. By Gibbs sampling, the MEME suite. In MEME suite the MEME motif joint distribution of the parameters will converge discovery algorithm is complemented by the to the joint probability of the parameters in the GLAM2 algorithm which allows the discovery of given dataset. Gibbs sampling strategies claim to motifs containing gaps. They used three sequence be fast and sensitive. Roth et al. (1998) developed scanning algorithms, MAST, FIMO (Frith et al. a motif-finding tool AlignACE, which is based 2008), and GLAM2SCAN (Bailey and Gribskov on the Gibbs sampling algorithm. 1998), that allow scanning numerous DNA and EM for motif finding was introduced by protein sequence databases for motifs discovered Lawrence and Reilly (1990), and it was an by MEME and GLAM2. extension of the greedy algorithm for motif Li (2009) proposed the method GADEM, finding by Hertz et al. (1990). The EM algo- which combines spaced dyads and an expectation- rithm is used to estimate the probability density maximization (EM) algorithm. In this method of a given dataset by employing the Gaussian candidate words (four to six nucleotides) for con- mixture model. The probability density of a structing spaced dyads are prioritized by their dataset is modeled as the weighted sum of a degree of overrepresentation in the input sequence number of Gaussian distributions. No align- data. Spaced dyads are converted into starting ment of the sites is required, and the basic position weight matrices (PWMs). GADEM then model assumption is that each sequence must employs a genetic algorithm (GA), with an contain at least one common site. The uncer- embedded EM algorithm to improve starting tainty in the location of the sites is handled by PWMs, so as to guide the evolution of a popula- employing the missing information principle to tion of spaced dyads toward one whose entropy develop an EM algorithm. This approach allows scores are more statistically significant. Spaced for the simultaneous identification of the sites dyads whose entropy scores reach a prespecified and characterization of the binding motifs. The significance threshold are declared motifs. main advantage of EM is its fast speed, while Generally, PWM-based approaches assume the disadvantage is that it requires “appropri- independence between the base positions of the ate” starting values and is difficult to deal with sequence motif and suffer from high false- constrained parameters. positive rates. However, recent studies have The MEME algorithm by Bailey and shown that the independent assumption is not Elkan (1995b) extended the EM algorithm for true and modeling the dependencies in motifs identifying motifs in unaligned biopolymer could lead to better predictions (Bulyk et al. sequences. The aim of MEME is to discover 2002). Examples include feature-based method new motifs in a set of biopolymer sequences (Sharon et al. 2008; Chin and Leung 2008), where little is known in advance about any HMM-based method (Marinescu et al. 2005), motifs that may be present. MEME incorpo- and Markov chain-based method (Wang et al. rated three novel ideas for discovering motifs. 2006). 118 S. Vijayvargiya and P. Shukla

Phylogeny-Based Methods in the choice of motifs reported. Even if these methods were modified to weigh the input The major advantage of phylogenetic footprint- sequences unequally, this would still not capture ing over the coregulated genes approach is that the information in an arbitrary phylogenetic tree. the latter requires a reliable method for identify- Carmack et al. (2007) developed a scanning ing coregulated genes. Whereas in using phylo- algorithm, PhyloScan, which combines evidence genetic footprinting approach, it is possible to from matching sites found in orthologous data identify motifs specific to even a single gene, as from several related species with evidence from long as they are sufficiently conserved across the multiple sites within an intergenic region to bet- many orthologous sequences considered. The ter detect regulons. rapid accumulation of genomic sequences from a Some algorithms integrate two important wide variety of organisms makes it possible to aspects of a motif’s significance, i.e., overrepre- use the phylogenetic footprinting approach for sentation of motifs in promoter sequences of motif finding. The standard method used for phy- coregulated and cross-species conservation, into logenetic footprinting is to construct a global one probabilistic score. multiple alignment of the orthologous promoter Based on the Consensus algorithm (Hertz sequences and then identify conserved region in et al. 1990), Wang and Stormo (2003) developed the alignment using a tool such as CLUSTALW the motif-finding algorithm PhyloCon (Phylo- (Thompson et al. 1994). However, it has been genetic Consensus). Phylogenetic Consensus observed (Tompa 2001) that this approach to (PhyloCon) takes into account both conserved phylogenetic footprinting does not always work. orthologous genes and coregulated genes within The reason is that if the species are too closely a species. The key idea of PhyloCon is to com- related, the sequence alignment is obvious but pare aligned sequence profiles from orthologous uninformative, since the functional elements are genes or coregulated genes rather than unaligned not sufficiently better conserved than the sur- sequences. rounding nonfunctional sequence. On the other Sinha et al. (2004) developed the algorithm hand, if the species are too distantly related, it is PhyME based on a probabilistic approach that either difficult or impossible to find an accurate handles data from promoters of coregulated alignment. To overcome this problem, one of the genes and orthologous sequences. PhyME inte- several existing motif-finding algorithms such as grates two different axes of information content MEME, Consensus, and Gibbs sampler has been in evaluating the significance of candidate used for phylogenetic footprinting. Cliften et al. motifs. One axis is the overrepresentation that (2001) used AlignACE for motif finding by com- depends on the number of occurrences of motifs parative DNA sequence analysis of several spe- in each species. The other axis is the level of cies of Saccharomyces and reported some conservation of each motif instance across spe- successes where the global multiple alignment cies. Siddharthan et al. (2005) developed the tools failed. McCue et al. (2001) used Gibbs sam- algorithm PhyloGibbs that combines the motif- pler for motif finding using phylogenetic foot- finding strategies of phylogenetic footprinting printing in proteobacterial genomes. That the use and Gibbs sampling into one integrated Bayesian of such general motif discovery algorithms can framework. be problematic in phylogenetic footprinting Zhang et al. (2010) recently developed an has been pointed out by Blanchette and Tompa algorithm named GLECLUBS (Global Ensemble (2002). These motif-finding algorithms do not and Clustering of Binding Sites) for genome- take into account the phylogenetic relationship wide de novo prediction of motifs in a prokary- of the given sequences since these methods assume otic genome (Zhang et al. 2009). GLECLUBS the input sequences to be independent. Therefore, employs a phylogenetic footprinting technique to the datasets containing a mixture of some closely first identify all possible motifs, and then clusters related species would have an unduly high weight similar motifs. In order to harvest as many as 8 Regulatory Motif Identification in Biological Sequences… 119 possible true motifs by phylogenetic footprint- Chan et al. (2009) proposed a new generalized ing, GLECLUBS uses multiple complementary model, which tackles the width uncertainty in the motif-finding tools instead of using only a single motif widths by considering and evaluating a tool and considers multiple outputs of each tool. wide range of interests simultaneously. Moreover, Also, GLECLUBS assumes that only a small they also proposed a meta-convergence frame- portion of predicted motifs by phylogenetic foot- work for genetic algorithms to provide multiple printing are true motifs and that the vast majority overlapping optimal motifs simultaneously in an of them are spurious predictions. Therefore, the effective and flexible way. Incorporating Genetic clustering step of GLECLUBS aims to discrimi- Algorithm with Local Filtering (GALF) for nate true motifs from spurious ones using an searching, the new algorithm GALF-G (G for iterative filtering procedure. generalized) was proposed based on the generalized model and meta-convergence framework. The GA-based hybrid schemes have also been Machine Learning–Based Methods proposed. One such method is GARPS that combines GA and Random Projection Strategy (RPS) Liu et al. (2004) developed the algorithm FMGA to identify planted (l, d) motifs. In the method based on genetic algorithms (GAs) for finding GARPS, RPS is used to find good starting posi- motifs in the regulatory regions. The crossover is tions by introducing position-weighted function. implemented with specially designed gap penal- Then, GA is used to refine the initial population ties to produce the optimal child pattern. The obtained from RPS (Hongwei et al. 2010). mutation in GA is performed by using position Chengwei Lei and Jianhua Ruan (2010) devel- weight matrices to reserve the completely con- oped a motif-finding algorithm (PSO+) using the served positions. This algorithm also uses a particle swarm optimization (PSO) which is a pop- rearrangement method based on position weight ulation-based stochastic optimization technique. matrices to avoid the presence of a very stable They proposed a modification in the standard PSO local minimum, which may make it quite difficult algorithm to handle discrete values, such as charac- for the other operators to generate the optimal ters in DNA sequences. They used both consensus pattern. The authors reported that FMGA per- and position-specific weight matrix representations forms better in comparison to MEME and Gibbs in their algorithm to take advantage of the effi- sampler algorithms. ciency of the former and the accuracy of the latter. Liu et al. (2006) developed a self-organizing Many real motifs contain gaps; to address this neural network structure for motif finding in DNA issue, their method models gaps explicitly and pro- and protein sequences. The network contains sev- vides a solution to find gapped motifs without any eral layers with each layer performing classifica- detailed knowledge of gaps. tions at different levels. The authors maintained a Lee and Wang (2011) developed an SOM low computational complexity through the use of (Kohonen 2001)-based extraction algorithm layered structure so that each pattern’s classifica- (SOMEA) to discover overrepresented motifs in tion is performed with respect to a small subspace DNA datasets. SOMEA seek to use SOM to proj- of the whole input space. The authors also main- ect k-mers (i.e., a subsequence with length k of tain a high reliability of their search algorithm DNA sequences) onto a two-dimensional lattice using self-organizing neural network since it of nodes. Through this projection, input patterns will grow as needed to make sure that all input (i.e., k-mers) with closely related features are patterns are considered and are given the same projected onto the same or adjacent nodes on the amount of attention. From simulation results, the map. Hence, the complex similarity relationships authors reported that their algorithm outper- of the high-dimensional input sequence space formed the algorithms MEME and Gibbs sampler become apparent on the map. Analysis of selected in certain aspects and their algorithm also works nodes therefore can reveal potential patterns (i.e., well for long DNA sequences. motifs) in the dataset. 120 S. Vijayvargiya and P. Shukla

Other Methods defined two p-value scores for overrepresentation, one based on the total number of motif occur- Hu et al. (2005) introduced the ensemble rences and the other based on the number of approach for motif finding to improve the predic- sequences in a collection with at least one occur- tion accuracy of the motif-finding algorithms. rence. Third, their method exploits monotonic They developed a clustering-based ensemble properties of the compound Poisson approxima- algorithm named EMD (Hu et al. 2006) for motif tion and is by orders of magnitude faster than discovery by combining multiple predictions exhaustive enumeration of IUPAC strings. from multiple runs of one or more base component algorithms. The potential of an EMD algorithm lies in the fact that it could take Prediction Accuracy of Motif advantage of superb predictions of every compo- Identification Algorithms nent algorithm. The authors used five component algorithms, namely, AlignACE, BioProspector, There are several prediction accuracy measures MDScan (Liu et al. 2002), MEME, and for evaluating motif discovery algorithms (Sinha MotifSampler in their study. They tested their et al. 2004; Liu et al. 2002; Thijs et al. 2002). algorithm on a benchmark dataset generated Many of them are derived from the accuracy defi- from Escherichia coli RegulonDB. The EMD nitions for evaluating gene predictions (Burset algorithm achieved 22.4 % improvement in terms and Guigo 1996; Rogic et al. 2001). Most of the of the nucleotide level prediction accuracy over algorithm uses the parameters precision (speci- the best stand-alone component algorithm. ficity) and recall (sensitivity) as prediction accu- Chakravarty et al. (2007) proposed an ensemble racy measures. The accuracy of a prediction can learning method, SCOPE, that is based on the be measured by comparing the predicted motifs assumption that transcription factor binding with the true motifs. sites belong to one of three broad classes of To measure the accuracy of the algorithms at motifs: nondegenerate, degenerate, and gapped nucleotide level, the following values for calcu- motifs. SCOPE employs a unified scoring metric lating accuracy metrics at the nucleotide level are to combine the results from three motif-finding defined for each target binding site with overlap- algorithms each aimed at the discovery of one of ping predicted binding sites in an input sequence: these classes of motifs. 1. nTP (true positive) – the number of target Sandve et al. (2008) proposed a discrete binding site positions predicted as binding site approach to composite motif discovery (Compo) positions that supports rich modeling of composite motifs 2. nTN (true negative) – the number of nontarget and a realistic background model. Compo can binding site positions predicted as non- return either an ordered list of motifs, ranked binding site positions according to the general significance measure, or 3. nFP (false positive) – the number of nontarget a Pareto front corresponding to a multi-objective binding site positions predicted as binding site evaluation on sensitivity, specificity, and spatial positions clustering. 4. nFN (false negative) – the number of target Marschall and Rahmann (2009) proposed an binding site positions predicted as non- exact motif discovery method for a practically binding site positions relevant space of IUPAC-generalized string pat- Precision or specificity over a pair of target/ terns, using the p-value, with respect to a Markov predicted binding sites is defined as the number model as the measure of overrepresentation. of predicted sites that are true sites divided by the The key characteristics of their method can be total number of predicted sites: observed threefold. First, they used a compound nTP Poisson approximation for the null distribution of nS p = (8.1) the number of motif occurrences. Second, they nTPn+ FP 8 Regulatory Motif Identification in Biological Sequences… 121

TP TP TP nPC = + + nSn = nSp = TP FP FN TP + FN TP + FP Motif i Motif i+1 target motifs

Predicted motifs lnput FP TP FN sequence set 1

Fig. 8.3 Measures of prediction accuracy at the nucleotide levels. Accuracy scores over an input sequence set are the average accuracy scores over all its sequences (Hu et al. 2005)

Recall or sensitivity is the number of predicted References sites that are true sites divided by the total number of true sites: Bailey TL, Elkan C (1995a) The value of prior knowledge in discovering motifs with MEME. Proc ISMB nTP 1995:21–29 nSn = (8.2) Bailey TL, Elkan C (1995b) Unsupervised learning of nTPn+ FN multiple motifs in biopolymers using expectation In order to capture both specificity and sensi- maximization. Mach Learn 21:51–80 tivity in a single accuracy measurement, the Bailey TL, Gribskov M (1998) Combining evidence using p-values: application to sequence homology searches. nucleotide level performance coefficient (nPC) is Bioinformatics 14:48–54 used as follows (Pevzner and Sze 2000; Tompa Bailey TL, Williams N, Misleh C, Li WW (2006) MEME: et al. 2005): discovering and analyzing DNA and protein sequence motifs. Nucleic Acids Res 34:W369–W373 nTP Bailey TL et al (2009) MEME SUITE: tools for motif dis- nPC = (8.3) covery and searching. Nucleic Acids Res 37(Web nTPn++FP nFN Server issue):W202–W208. doi:10.1093/nar/gkp335 According to this definition, the nPC value Blanchette M, Tompa M (2002) Discovery of regulatory ranges over (0, 1) with the perfect prediction elements by a computational method for phylogenetic footprinting. Genome Res 12:739–748 being the value of 1. Compared with the cor- Blanco E, Farré D, Albà MM, Messeguer X, Guigó R relation coefficient (CC) (Burset and Guigo (2006) ABS: a database of Annotated regulatory 1996; Rogic et al. 2001), nPC has several ben- Binding Sites from orthologous promoters. Nucleic efits: it is straightforward to interpret, and Acids Res 34(1):D63–D67 Boucher C, Church P, Brown D (2007) A graph clustering practically it also tells the experimental biolo- approach to weak motif recognition. Proc WABI gists the probable range where the true binding 2007:149–160 sites are located around the predicted posi- Boyer LA, Lee TI et al (2005) Core transcriptional regulations. The F-score or harmonic mean is also tory circuitry in human embryonic stem cells. Cell 122(6):947–956 used as the overall accuracy measurement. Bucher P (1990) Weight matrix description for four Compared with geometric or arithmetic mean, eukaryotic RNA polymerase II promoter element it tends to penalize more the imbalance of sen- derived from 502 unrelated promoter sequences. J Mol sitivity and specificity. The F-score is defined Biol 212:563–578 Buhler J, Tompa M (2002) Finding motifs using random as (Fig. 8.3) projections. J Comput Biol 9:225–242 Bulyk ML, Johnson PL, Church GM (2002) Nucleotides 2´´SSnp of transcription factor binding sites exert interdepen- F = (8.4) SS+ dent effects on the binding affinities of transcription np factors. Nucleic Acids Res 30:1255–1261 122 S. Vijayvargiya and P. Shukla

Burset M, Guigo R (1996) Evaluation of gene structure Frith MC, Saunders NFW, Kobe B, Bailey TL (2008) prediction programs. Genomics 34:353–367 Discovering sequence motifs with arbitrary insertions Bussemaker HJ, Li H, Siggia ED (2001) Regulatory ele- and deletions. PLoS Comput Biol 4:e1000071 ment detection using correlation with expression. Nat Hertz GZ, Hartzell GW, Stormo GD (1990) Identification of Genet 27:167–171 consensus patterns in unaligned DNA sequences known Carmack CS, McCue LA, Newberg LA, Lawrence CE to be functionally related. Bioinformatics 6:81–92 (2007) PhyloScan: identification of transcription factor Hongwei H, Zhenhua Z, Vojislav S, Lifang L (2010) binding sites using cross-species evidence. Algorithms Optimizing genetic algorithm for motif discovery. Mol Biol 2:1 Math Comput Model 52(11–12):2011–2020 Carvalho AM, Oliveira AL (2011) GRISOTTO: a greedy Horak CE, Snyder M (2002) ChIP-chip: a genomic approach to improve combinatorial algorithms for approach for identifying transcription factor binding motif discovery with prior knowledge. Algorithms sites. Methods Enzymol 350(469–483):2002 Mol Biol 6:13 Hu J, Li B, Kihara D (2005) Limitations and potentials of Carvalho AM, Freitas AT, Oliveira AL, Sagot MF (2006) current motif discovery algorithms. Nucleic Acids Res An efficient algorithm for the identification of struc- 33:4899–4913 tured motifs in DNA promoter sequences. IEEE/ACM Hu J, Yang YD, Kihara D (2006) EMD an ensemble Trans Comput Biol Bioinform 3(2):126–140 algorithm for discovering regulatory motifs in DNA Chakravarty A, Carlson JM, Khetani RS, Gross RH sequences. BMC Bioinformatics 7:342 (2007) A novel ensemble learning method for de novo Sandve GK, Drablos F (2006) A survey of motif discov- computational identification of DNA binding sites. ery methods in an integrated framework. Biol Direct BMC Bioinformatics 8:249. doi:10.1186/1471- 1:11. doi:10.1186/1745-6150-1-11 2105-8-249 Kohonen T (2001) Self-organizing maps, vol 30, 3rd edn, Chan T, Li G, Leung K, Lee K (2009) Discovering multi- Springer series in information sciences. Springer, ple realistic TFBS motifs based on a generalized Berlin, Heidelberg, New York model BMC. Bioinformatics 10:321. doi:10.1186/1471 Kolchanov NA, Podkolodnaya OA et al (2000) -2105-10-321 Transcription Regulatory Regions Database (TRRD): Chengwei L, Jianhua R (2010) A particle swarm its status in 2000. Nucleic Acids Res 28(1):298–301 optimization-based algorithm for finding gapped motifs. Lawrence CE, Reilly AA (1990) An expectation maximi- BioData Min 3:9 zation (EM) algorithm for the identification and char- Chin FYL, Leung CM (2005) Voting algorithms for dis- acterization of common sites in unaligned biopolymer covering long motifs. Proc APBC 2005:261–271 sequences. Proteins 7:41–51 Chin FYL, Leung CM (2006) An efficient algorithm for Lawrence CE, Altschul SF, Boguski MS, Liu JS, Neuwald string motif discovery. Proc APBC 2006:79–88 AF et al (1993) Detecting subtle sequence signals: a Chin F, Leung HCM (2008) DNA motif representation Gibbs sampling strategy for multiple alignment. with nucleotide dependency. IEEE/ACM Trans Science 262:208–214 Comput Biol Bioinform 5:110–119 Lee NK, Wang D (2011) SOMEA: self-organizing map Cliften PF, Hillier LW, Fulton L, Graves T, Miner T, Gish based extraction algorithm for DNA motif identifica- WR, Waterston RH, Johnston M (2001) Surveying tion with heterogeneous model. BMC Bioinformatics Saccharomyces genomes to identify functional ele- 12(1):S16 ments by comparative DNA sequence analysis. Li L (2009) GADEM: a genetic algorithm guided forma- Genome Res 11:1175–1186 tion of spaced dyads coupled with an EM algorithm Das MK, Dai H-K (2007) A survey of DNA motif finding for motif discovery. J Comput Biol 16(2):317–329. algorithms. BMC Bioinformatics 8(7):S21 doi:10.1089/cmb.2008.16TT Davila J, Balla S, Rajasekaran S (2007) Fast and practical Liang S (2003) cWINNOWER algorithm for finding algorithms for planted (l, d) motif search. IEEE/ACM fuzzy DNA motifs. IEEE Compu Soc Bioinform Conf Trans Comput Biol Bioinform 4(4):544–552 2003:260–265 Elnitski L, Jin VX, Farnham PJ, Jones SJ (2006) Locating Liu XS, Brutlag DL, Liu JS (2002) An algorithm for mammalian transcription factor binding sites: a survey finding protein-DNA binding sites with applications of computational and experimental techniques. to chromatin-immunoprecipitation microarray experi- Genome Res 16(12):1455–1464 ments. Nat Biotechnol 20:835–839 Eskin E, Pevzner P (2002) Finding composite regulatory Liu F et al (2004) FMGA: finding motifs by genetic algo- patterns in DNA sequences. Bioinformatics 18(1): rithm. In proceedings of the fourth IEEE symposium S354–S363 on bioinformatics and bioengineering, Taichung, Ettwiller L, Paten B, Ramialison M, Birney E, Wittbrodt J Taiwan, pp 459–466 (2007) Trawler: de novo regulatory motif discovery Liu D, Xiong X, DasGupta B, Zhang H (2006) Motif pipeline for chromatin immunoprecipitation. Nat discoveries in unaligned molecular sequences using Methods 4:563–565 self-organizing neural network. IEEE Trans Neural Evans PA, Smith A, Wareham HT (2003) On the com- Netw 17:919–928 plexity of finding common approximate substrings. Lockhart D, Winzeler E (2000) Genomics, gene expression Theor Comput Sci 306(1–3):407–430 and DNA arrays. Nature 405:827–836 8 Regulatory Motif Identification in Biological Sequences… 123

Marinescu VD, Kohane IS, Riva A (2005) MAPPER: a Schneider TD, Stephens RM (1990) Sequence logos: a search engine for the computational identification of new way to display consensus sequences. Nucleic putative transcription factor binding sites in multiple Acids Res 18:6097–6100 genomes. BMC Bioinformatics 6:79 Sharon E, Lubliner S, Segal E (2008) A feature-based Marsan L, Sagot M (2000) Algorithms for extracting approach to modeling protein-DNA interactions. PLoS structured motifs using a suffix tree with an application Comput Biol 4:e1000154 to promoter and regulatory site consensus identifica- Siddharthan R, Siggia ED, Nimwegen E (2005) tion. J Comput Biol 7(3–4):345–362 PhyloGibbs: a Gibbs sampling motif finder that incor- Marschall T, Rahmann S (2009) Efficient exact motif porates phylogeny. PLoS Comput Biol 1:534–556 discovery. Bioinformatics 25:i356–i364. doi:10.1093/ Sinha S, Tompa M (2003) YMF: a program for discovery bioinformatics/btp188 of novel transcription factor binding sites by statisti- Matys V, Fricke E et al (2003) TRANSFAC: transcrip- cal overrepresentation. Nucleic Acids Res 31:3586– tional regulation, from patterns to profiles. Nucleic 3588 Acids Res 31(1):374–378 Sinha S, Blanchette M, Tompa M (2004) PhyME: a proba- McCue L, Thompson W, Carmack C, Ryan M, Liu J, bilistic algorithm for finding motifs in sets of ortholo- Derbyshire V, Lawrence C (2001) Phylogenetic gous sequences. BMC Bioinformatics 5:170 footprinting of transcription factor binding sites in Smith AD, Sumazin P, Das D, Zhang MQ (2005) proteobacterial genomes. Nucleic Acids Res 29: Mining ChIP-chip data for transcription factor and 774–782 cofactor binding sites. Bioinformatics 21(1):i403– Odom DT, Dowell RD et al (2006) Core transcriptional i412 regulatory circuitry in human hepatocytes. Mol Syst Stormo GD (2000) DNA binding sites: representation and Biol 2:0017 discovery. Bioinformatics 16:16–23 Pavesi G, Mauri G, Pesole G (2001) An algorithm for Sze S, Lu S, Chen J (2004) Integrating sample-driven and finding signals of unknown length in DNA sequences. pattern driven approaches in motif finding. Proc WABI Bioinformatics 17(1):S207–S214 2004:438–449 Pavesi G, Mereghetti P, Mauri G, Pesole G (2004) Weeder Thijs G, Marchal K, Lescot M, Rombauts S, DeMoor B, web: discovery of transcription factor binding sites in Rouze P, Moreau Y (2002) A Gibbs sampling method to a set of sequences from co-regulated genes. Nucleic detect overrepresented motifs in the upstream regions of Acids Res 32:W199–W203 coexpressed genes. J Comput Biol 9:447–464 Pevzner PA, Sze SH (2000) Combinatorial approaches to Thompson JD, Higgins DG, Gibson TJ (1994) finding subtle signals in DNA sequences. Proc Int CLUSTALW improving the sensitivity of progressive Conf Intell Syst Mol Biol 8:269–278 multiple sequence alignment through sequence Pisanti N, Carvalho AM, Marsan L, Sagot MF (2006) weighting, position specific gap penalties and weight RISOTTO: fast extraction of motifs with mismatches. matrix choice. Nucleic Acids Res 22:4673 In Proceedings of LATIN’06, Vol 3887 of LNCS. Tompa M (2001) Identifying functional elements by com- Springer, pp 757–768 parative DNA sequence analysis. Genome Res 11:1143– Rajasekaran S, Balla S, Huang CH (2005) Exact algo- 1144 rithms for planted motif problems. J Comput Biol Tompa M, Li N, Bailey TL, Church GM, De Moor B et al 12(8):1117–1128 (2005) Assessing computational tools for the discov- Ren B, Robert F et al (2000) Genome-wide location and ery of transcription factor binding sites. Nat Biotechnol function of DNA binding proteins. Science 290(5500): 23:137–144 2306–2309 Vavouri T, Elgar G (2005) Prediction of cisregulatory Rogic S, Mackworth AK, Ouellette FB (2001) Evaluation elements using binding site matrices—the successes, of gene-finding programs on mammalian sequences. the failures and the reasons for both. Curr Opin Genet Genome Res 11:817–832 Dev 15:395–402 Roth FP, Hughes JD, Estep PW, Church GM (1998) Vilo J, Brazma A, Jonassen I, Robinson A, Ukonnen E Finding DNA regulatory motifs within unaligned non- (2000) Mining for putative regulatory elements in the coding sequences clustered by whole-genome mRNA yeast genome using gene expression data. In: In pro- quantitation. Nat Biotechnol 16:939–945 ceedings of the eighth international conference on Sagot M (1998) Spelling approximate repeated or com- intelligent systems for molecular biology. AAAI mon motifs using a suffix tree. Lect Notes Comput Sci Press, San Diego, pp 384–394 1380:111–127 Vlieghe D, Sandelin A, Bleser P, Vleminckx K, Wasserman W, Sandelin A, Alkema W, Engstrom P, Wasserman WW, Roy F, Lenhard B (2006) A new generation of Lenhard B (2004) JASPAR: an open-access database JASPAR, the open-access repository for transcription for eukaryotic transcription factor binding profiles. factor binding site profiles. Nucleic Acids Res Nucleic Acids Res 32:D91–D94 34(Database issue):D95–D97 Sandve G, Abul O, Drablos F (2008) Compo: composite Wang T, Stormo GD (2003) Combining phylogenetic data motif discovery using discrete models. BMC with coregulated genes to identify regulatory motifs. Bioinformatics 9:527. doi:10.1186/1471-2105-9-527 Bioinformatics 19:2369–2380 124 S. Vijayvargiya and P. Shukla

Wang C, Xie J, Craig BA (2006) Context dependent Zhang Y, Zaki MJ (2006b) EXMOTIF: efficient struc- models for discovery of transcription factor binding tured motif extraction. Algorithms Mol Biol 1:21. sites. Stat Methodol 3:55–68 doi:10.1186/1748-7188-1-21 Wasserman WW, Sandelin A (2004) Applied bioinfor- Zhang S, Xu M, Li S, Su Z (2009) Genome-wide de novo matics for the identification of regulatory elements. prediction of cis-regulatory binding sites in prokary- Nat Rev Genet 5:276–287 otes. Nucleic Acids Res 37(10):e72 Wray GA, Hahn MW, Abouheif E, Balhoff JP, Pizer M, Zhang S, Li S et al (2010) Simultaneous prediction of Rockman MV, Romano LA (2003) The evolution of transcription factor binding sites in a group of pro- transcriptional regulation in eukaryotes. Mol Biol karyotic genomes. BMC Bioinformatics 11:397 Evol 20(9):1377–1419 Zhao F, Xuan Z, Liu L, Zhang MQ (2005) TRED: a Zaslavsky E, Singh M (2006) A combinatorial optimization Transcriptional Regulatory Element Database and a approach for diverse motif finding applications. platform for in silico gene regulation studies. Nucleic Algorithms Mol Biol 1:13. doi:10.1186/1748-7188-1-13 Acids Res 33(Database issue):D103–D107 Zhang Y, Zaki MJ (2006a) SMOTIF: efficient structured Zhu J, Zhang MQ (1999) SCPD: a promoter database of pattern and profile motif search. Algorithms Mol Biol the yeast Saccharomyces cerevisiae. Bioinformatics 1:22. doi:10.1186/1748-7188-1-22 15:607–611 Chitin Deacetylase: Characteristic Molecular Features and Functional 9 Aspects

Nidhi Pareek , V. Vivekanand , and R. P. Singh

Abstract Chitosan has a broad and impressive array of applications in diverse industrial sectors, like pharmaceutics (drug delivery), gene delivery, tissue engineering, food and cosmetics industry, water treatment, and agriculture. To date, majority of the chitosan is produced from thermo-alkaline deacetylation of chitin from crustacean shells. The process is incompatible as it leads to variability in the product properties, increased cost of production, and environmental concerns. Functional properties and in turn industrial applicability of chitosan depend on its degree of deacetylation; hence, a controlled biological process needs to be developed so as to realize the commercial value of the product. Chitin deacetylase (CDA) is the key enzyme employed for bioconversion of chitin to chitosan. It catalyzes deacetylation of N -acetyl-d -glucosamine residues under mild reaction conditions and results into production of novel superior-quality chitosan. The enzyme-aided production is a vital step towards the chitosan production in the green chemistry realm as the chemical process is engraved with a number of limitations and bottlenecks. Apart from being used in bioconversion reactions, CDA has a number of biological roles, namely, formation of spore wall in Saccharomyces cerevisiae and vegetative cell wall in Cryptococcus neoformans, responsible for pathogenesis of plant pathogenic fungi, and utilization of chitin in marine ecosystems .

N. Pareek Protein Engineering and Proteomics Group, Department of Biotechnology , Indian Institute Department of Chemistry, Biotechnology of Technology Roorkee , Roorkee 247667 , India and Food Sciences , Norwegian University of Life Sciences, Ås 1430 , Norway Department of Chemistry , Umeå University , SE-90187 Umeå , Sweden R. P. Singh () Department of Biotechnology , Indian Institute V. Vivekanand of Technology Roorkee , Roorkee 247667 , India Department of Biotechnology , Indian Institute e-mail: [email protected] of Technology Roorkee , Roorkee 247667 , India

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 125 DOI 10.1007/978-81-322-1094-8_9, © Springer India 2013 126 N. Pareek et al.

Keywords Chitin deacetylase • Chitin • Chitosan • Deacetylation • Bioconversion

Introduction Chitin Deacetylase

One of the major objectives of present age of S o u r c e biotechnology is headed towards production and application of a range of bio-based value- The enzyme has been detected in an array of organ- added products. Biopolymers, i.e., cellulose, isms including fungi, bacteria, and insects, among xylan, lignin, chitin, tannin, and pectin, are which fungal deacetylases are widely explored considered as the major resource materials for (Tsigos et al. 2000 ; Meens et al. 2001 ; Zhao et al. a number of industrial sectors due to their 2010a ) (Table 9.1 ). The physiological role of CDAs renewable and biodegradable nature. Among in microbes is primarily concerned with the pro- all the polymers, in spite of its abundance in duction of chitosan, a cell wall component, along nature, chitin is the most underutilized one due with the pathogenesis of plant pathogenic fungi. to its high degree of crystallinity and insolubility in most of thesolvents (Ruiz-Herrera 1978 ). Deacetylases of Fungal Origin Chitosan, the N -deacetylated derivative of chi- A number of attempts have been made for CDA tin has enormous commercial potential due to production from fungi mainly from Mucor rouxii its characteristically superior properties like (Davis and Bartnicki-Garcia 1984 ; Kafetzopoulos biodegradability, biocompatibility, solubility, et al. 1993a ), Colletotrichum lindemuthianum and non-toxicity (Kurita 2006 ). It is present in (Tsigos and Bouriotis 1995 ; Tokuyasu et al. 1996 ; nominal amounts in animal biomass, in the Shrestha et al. 2004 ), Absidia coerulea (Gao shells or cuticles of many crustaceans, and also et al. 1995), Aspergillus nidulans (Alfonso in the fungal cell wall; it is therefore mainly et al. 1995), Gongronella butleri (Maw et al. derived from chitin by chemical or biocatalytic 2002a , b ), Metarhizium anisopliae (Nahar et al. alkaline deacetylation process. Chemical route 2004 ), Rhizopus nigricans (Jeraj et al. 2006 ), Scopu- utilizes larger amounts of concentrated alkali lariopsis brevicaulis (Cai et al. 2006 ), Mortierella that in turn eventually leads into environmental sp. DY-52 (Kim et al. 2008 ), Rhizopus circinans deterioration (Chang et al. 1997). Bioconversion (Gauthier et al. 2008 ), Flammulina velutipes to chitosan is assisted by a member of carbohy- (Yamada et al. 2008 ), Absidia corymbifera (Zhao drate esterase family 4, i.e., chitin deacetylase et al. 2010b ), etc. Fungi are known to produce (CDA, EC 3.5.1.41). The enzyme catalyzes the both extra- and intracellular deacetylases during removal of acetyl groups from the nascent chi- different periods owing to their speciﬁ c biological tin chain via a multiple-attack mechanism, roles, namely, development of fungal cell wall, resulting into a polymer consisting of both glu- modifying the hyphal chitin of plant pathogenic cosamine and N -acetylglucosamine monomers fungi to aid in pathogenesis by protecting it from (Tsigos et al. 2000). CDA-assisted chitosan the plant resistance system. production had initially begun as a thought and now considered as a worthwhile, promising Deacetylases of Yeast Origin technology of the future. Intense research and Apart from these fungal strains, some yeast spe- developmental activities are under way to cies, namely, Saccharomyces cerevisiae (Martinou develop a robust enzyme preparation that can et al. 2002 ) and Schizosaccharomyces pombe be explored further as an eco-friendly approach (Matsuo et al. 2005 ), are also known to produce for production of chitosan with desired CDAs during sporulation, which may be involved properties. in the synthesis of spore wall. 9 Chitin Deacetylase: Characteristic Molecular Features and Functional Aspects 127

Table 9.1 Characteristic features of chitin deacetylase from microorganisms Opt. Metal ion MW temp Acetate activation/ Organism (kD) pI Opt. pH (°C) inhibition inhibition References Mucor rouxii 75 3 4.5 50 Yes – Kafetzopoulos ATCC 24905 et al. (1993a ) Aspergillus nidulans 27 2.7 7.0 50 Yes – Alfonso et al. (1995 ) CECT 2544 Colletotrichum 150 – 8.5 50 No Co+2 ( Ac. ) Tsigos and Bouriotis lindemuthianum (1995 ) DSM 63144 Colletotrichum 31.5 3.7 11.5–12.0 60 Yes Co +2 ( Ac. ) Tokuyasu et al. (1996 ) lindemuthianum ATCC 56676 Vibrio alginolyticus H-8 48, 46 3.3, 3.5 8.5, 8.0 45, 40 – Ag + , Hg+2 ( Ac. ) Ohishi et al. (2000 ) Thermus caldophilus 45 – 7.5 80 – Mn+2 , Co+2 , Fe+3 Shin et al. (1999 ) (Ac. ); Cu +2 ( In. ) Metarhizium anisopliae 70, 37, 26 2.6, 3.8, 4.1 8.5–8.8 – No – Nahar et al. ( 2004 ) Mucor circinelloides – – 4.5 50 – – Amorim et al. (2005 ) Scopulariopsis 55 – 7.5 55 – – Cai et al. ( 2006 ) brevicaulis Mortierella sp. DY-52 50, 59 – 5.5 60 – Co+2 , Ca+2 (Ac. ) Kim et al. (2008 ) Absidia corymbifera – – 6.5 55 Yes Co +2 Ca+2 , Zhao et al. (2010b ) DY-9 Mg +2 ( Ac. ) Ac activation, In inhibition

Deacetylases of Bacterial Origin Helicoverpa armigera (Campbell et al. 2008 ), Among the bacterial strains, members belonging and Mamestra confi gurata (Toprak et al. 2008 ) to family Vibrionaceae, widely distributed in all are reported to have CDAs in their midgut peri- oceanic and estuarine waters, are considered as trophic matrix. In Drosophila melanogaster two the major CDA producers (Ferguson and Gooday CDA-like proteins, CDA1 and CDA2 (serpentine 1996 ), where they are involved in the chitin and vermiform), have critical roles in shaping the metabolism (Hunt et al. 2008 ). Most of the Vibrio tracheal tubes as well as regulating the structural strains are known to produce chitin oligosaccha- properties of epidermal cuticle by infl uencing ride deacetylase (COD), which in turn produces a the structure and orientation of chitin fi brils characteristic inducer for chitinase production for (Luschnig et al. 2006 ; Wang et al. 2006 ). Apart chitin catabolism (Hirano et al. 2009 ). COD from from the above-mentioned species, Anopheles Vibrio parahaemolyticus and Vibrio cholerae gambiae , Apis mellifera, and Tribolium casta- was purifi ed and characterized by Kadokura neum (Dixit et al. 2008 ; Arakane et al. 2009 ; Noh et al. ( 2007a , b ) and Li et al. (2007 ), respectively. et al. 2011 ) are reported to produce CDAs. In spite Hyperthermophilic archaeon Thermococcus of the presence of CDAs in a large number of kodakaraensis KOD1 also possesses a deacetylase, insect species, their roles are still unresolved. involved in their chitin catabolic pathway (Tanaka et al. 2003 ). Bacillus amyloliquefaciens was also known to produce CDA (Zhou et al. 2010 ). Classifi cation

Deacetylases of Insect Origin CDAs (EC 3.5.1.41) are the members of carbohy- Occurrence of CDAs is not only restricted to drate esterase family 4 (CE-4s) as defi ned in the microbial population but also detected in insect CAZY database ( http://afmb.cnrs-mrs.fr/~cazy/ species. Trichoplusia ni (Guo et al. 2005 ), CAZY). According to the Henrissat classifi ca- 128 N. Pareek et al. tion, the family also includes rhizobial NodB Three CDA isoforms were also reported by chitooligosaccharide deacetylases, peptidogly- Trudel and Asselin ( 1990) in Mucor racemosus can N -acetylglucosamine deacetylases (EC with molecular mass of 26, 30, and 64 kDa. 3.5.1.104), acetyl xylan esterases (EC 3.1.1.72), Similarly, C. lindemuthianum, M. anisopliae , and xylanases A, C, D, E (EC 3.5.1.8) (Coutinho and R. nigricans are also known to produce and Henrissat 1999). Members of this family two, three, and four CDA isozymes, respectively. share a conserved region in their primary struc- Among bacterial species, Vibrio alginolyticus ture named the “NodB homology domain” or H-8 produced two extracellular CDA isoforms “polysaccharide deacetylase domain” (Caufrier (Ohishi et al. 2000). Genomes of Tribolium cas- et al. 2003 ; Gauthier et al. 2008 ). All fi ve members taneum , Drosophila melanogaster , Anopheles of this family catalyze the hydrolysis of either gambiae, and Apis mellifera contain 9, 6, 5, and N -linked acetyl groups from N -acetylglucosamine 5 genes, encoding proteins with a CDA motif. residues or O -linked acetyl groups from The presence of alternative exons in two of the O -acetylxylose residues of their substrates, Tribolium CDA genes, CDA2 and CDA5, leads namely, chitin, NodB factors, peptidoglycan, and into protein diversity further due to alternative acetyl xylan. Insect CDA-like proteins are classi- splicing. However, these CDA genes are quite fi ed into fi ve groups based on phylogenetic anal- different in terms of tissue specifi city and develop- ysis and the presence of additional motifs. Group mental patterns of expression. Possible reasons I include CDA1 and isoforms of CDA2, contain- for this multiplicity may be due to gene duplica- ing a polysaccharide deacetylase-like catalytic tion, differential m RNA processing, posttransla- domain, a chitin-binding peritrophin-A domain tional modifi cations such as glycosylation and (ChBD), and a low-density-lipoprotein receptor autoaggregation. Multiple CDAs can also be the class A domain (LDLa). Group II is composed of product from different alleles of the same gene, CDA3 proteins having similar domain organiza- i.e., allozymes. tion as group I CDAs, but with substantially different sequences. Group III includes CDA4s, which have only the ChBD domain. Group IV Substrate Specifi city comprises of CDA5s, which are the largest CDAs because of a very long intervening region sepa- Substrate specifi city of M. rouxii CDA was fi rst rating the ChBD and catalytic domains. Group V evaluated by Araki and Ito ( 1975 ) and they contains divergent group of proteins containing observed the enzyme activity for glycol chitin only a catalytic domain, lacking the N- terminal and chitooligomers with degree of polymeriza- ChBD and the LDLa domain (Dixit et al. 2008 ). tion greater than two. Deacetylase activity of the enzyme depends on the number of monosaccha- ride units in the substrate. The values of kinetic

Multiplicity parameters, namely, k cat, V max, and k cat / Km , were observed to increase along with the degree of Multiplicity corresponds to existence of multi- polymerization of chitin oligomers, while the ple forms of the similar enzymes or isozymes, values of Michaelis constant (K m ) decreased with which catalyzes the same reaction but differ in the increasing polymerization of glucosamine their amino acid sequence and physicochemical moieties. Caufrier et al. (2003 ) had analyzed the properties, such as molecular weight, isoelectric activity of CDA from M. rouxii and acetyl xylan point, and kinetic constants. Such multiple esterase from Streptomyces lividans on acetyl forms of CDAs were observed in many fungal xylan, peptidoglycan, and soluble chitin as the and bacterial species. In Uromyces viciae-fabae , substrates. The enzymes were observed to be ﬁ ve isoforms of CDA (12.7–48.1 kDa) were active on acetyl xylan and soluble chitin while produced during penetration of fungus through inactive on peptidoglycan. This might be attrib- leaf stomata (Deising and Siegrist 1995). uted to the difference in the sequence and 9 Chitin Deacetylase: Characteristic Molecular Features and Functional Aspects 129

structure of the catalytic domain of the enzymes. oligomers with a degree of polymerization less One such difference is the absence of the disul- than three. Only tetra-N- tetraacetylchitotetraose fi de bond, tethering the N -terminal and C -terminal and penta-N- pentaacetylchitopentaose were fully ends from the homologous peptidoglycan deacet- deacetylated by the enzyme while the reducing ylases from Streptococcus pneumoniae and end residues of N- acetylchitotriose, N- acetyl- Bacillus subtilis , while observed to be conserved chitohexaose, and N- acetylchitoheptaose always in CDA from M. rouxii and C. lindemuthianum remained intact. The enzyme initially removes an and also in S. lividans acetyl xylan esterase (Blair acetyl group from the nonreducing end residue et al. 2005 , 2006 ; Taylor et al. 2006 ). and further catalyzes the hydrolysis of the next acetamido group in progressive fashion. This report was in contrast with the fi ndings of Catalytic Mechanism Martinou et al. (1998 ) in which the enzyme was observed to be more active towards Mode of catalysis by CDAs on both chitin poly- reducing end. mers and oligomers had been studied by several C. lindemuthianum ATCC 56676 CDA was research groups. Catalytic action of CDAs observed to follow a multiple chain mechanism strongly depends on the degree of polymerization to remove the acetyl groups (Tokuyasu et al. of the substrate. Tokuyasu et al. (1997 ) studied 2000 ). The structural analysis of deacetylation the deacetylation of chitooligosaccharides products of (GlcNAc)4 suggested that the enzyme (degree of polymerization (DP), 2–4) by purifi ed has four subsites (−2, −1, 0, +1), in which sub- C. lindemuthianum CDA using FAB-MS (fast site 0 is the catalytic subsite. Reaction rate anal- atom bombardment mass spectrometry) and ysis of partially deacetylated substrates showed 1 H NMR spectroscopy and concluded that N, N′, that subsite −2 strongly recognizes the N -acetyl N″, N‴-tetraacetylchitotetraose and N, N′, group of the GlcNAc residue of the substrate. N″- triacetylchitotriose were deacetylated to Hekmat et al. ( 2003 ) had also suggested the corresponding chitosan oligomers. But N, presence of four enzyme subsites (−2 to +1) in N′-diacetylchitobiose was deacetylated on either the C. lindemuthianum CDA. Steady-state of the glucosamine residues to yield a unique kinetic analyses for the initial deacetylation compound, i.e., 2-acetamido-4-O -(2-amino-2- reaction of (GlcNAc) 2–6 by C. lindemuthianum deoxy- β -d -glucopyranosyl)-2-deoxy- d -glucose ATCC 56676 elucidated that the kinetic parame-

[GlcN-GlcNAc]. M. rouxii CDA deacetylates ters, i.e., K m and K cat /K m, depends on the degree water-soluble partially deacetylated chitosans of polymerization of the substrate, while K cat is (DP, 30) following a multiple-attack mechanism independent. A more detailed insight of the cata- with a degree of multiple attack of at least three lytic action of C. lindemuthianum was outlined (Martinou et al. 1998 ). The polarity of CDA was by Blair et al. (2006 ) via coupled structural and preferentially towards the reducing end and it fol- biochemical analysis. The reaction proceeds via lows an endo-type mechanism with no preferen- generation of a tetrahedral oxyanion intermedi- tial attack at any sequence in the chitosan chain. ate following a nucleophilic attack to the car- Deacetylation reaction was not detected at the bonyl carbon of the substrate, the charge of nonreducing end of the chain. The relative rate of which is stabilized by the oxyanion hole gener- enzymatic deacetylation increased linearly with ated by the backbone nitrogen of Tyr145 and the increasing fraction of acetylated units. zinc. Transfer of a proton from the water mole- Deacetylation of N -acetylchitooligosaccharides cule to the catalytic base Asp49 leads to the gen- (DP 2–7) by Mucor rouxii ATCC 24905 CDA eration of a nucleophile to attack the substrate was studied by Tsigos et al. (1999 ) employing an carbonyl carbon. Further, His206 protonates the exo-splitting system. The extent of deacetylation reaction intermediate on the nitrogen as it breaks depends on the length of the substrate. The down, generating a free amine and the acetate as enzyme could not effectively deacetylate chitin the product. 130 N. Pareek et al.

2007). Predicted three-dimensional structure of Cloning of Chitin Deacetylase Gene the gene had outlined the whole CDA functional domain and a polysaccharide deacetylase Till date CDA genes from various sources have domain. The cloned 1,506 bp CDA gene from been isolated, cloned, and expressed into suit- M. racemosus included a 67 bp 5′-untranslated able homologous as well as heterologous hosts. region, an open reading frame of 1,344 bp, and Kafetzopoulos et al. (1993b ) isolated and char- a 95 bp 3′-untranslated region including a tail- acterized a cDNA for CDA from M. rouxii ATCC ing site AATAAA. The gene coded for a 24905. Two CDA genes (CDA1 and CDA2) of 448-amino-acid protein and consisted of core S. cerevisiae were cloned and expressed into the nucleotides encoding a polysaccharide deacety- plasmid pSK and sequenced (Mishra et al. 1997 ). lase conserved domain, which had 144 amino Both of these deacetylases are expressed consti- acids and covered 32 % of the entire sequence in tutively during sporulation, since spore wall of the middle part. The structural domain of CDA the yeast contains chitosan, which along with the deduced from the primary sequence was very dityrosine layer have a role in the spore protec- similar to that of other species. A 75 kDa CDA tion in extreme conditions. CDA1- and CDA2- from R. circinans was cloned in P. pastoris defi cient mutants were observed to be more expression system using cDNA library, and 4.8- sensitive towards chemical and environmental fold increment was achieved in the production challenges. CDA gene from C. lindemuthianum level (Gauthier et al. 2008 ). Fv-pda , a gene ATCC 56676 was overexpressed in Escherichia encoding CDA, was isolated from F. velutipes coli as a fusion protein using an expression vec- during fruiting body development and expressed tor pQE60 (Tokuyasu et al. 1999 ). The CDA in P. pastoris (Yamada et al. 2008 ). The fv-pda open reading frame (ORF) consists of two open reading frame comprises 250 amino acid regions: one from the start codon (ATG) encod- residues and is interrupted by 10 introns. The ing a deduced preprodomain of 27 amino acids recombinant FV-PDA was observed to effec- and the other encoding a mature CDA of 221 tively catalyze the deacetylation of chitin oligo- amino acids. Ohishi et al. (2000 ) cloned a mers (dimer to pentamer), glycol chitin and deacetylase encoding gene DA1 from V. algino- colloidal chitin. cDNA amplifi cation of nine lyticus H-8 and sequenced using a shotgun Tribolium CDA genes were carried out using approach. The ORF of the gene starts at base 256 gene-specifi c primers, designed from the avail- and ends at the base 1,536. This 1,281-nucleo- able expressed sequence tags (ESTs) or from tide-long sequence encoded a 427-amino-acid- ORFs in GLEAN predictions from the Tribolium long enzyme. A putative Shine-Dalgarno (SD) shotgun genome sequences. sequence, ACGA, was found upstream to the start codon, ATG. R. nigricans B 154 CDA was cloned by Jeraj Crystal Structure et al. ( 2006) using yeast expression vector pFL61 in E. coli DH5α. Sequence analysis The structure of CDA from C. lindemuthianum divulged an open reading frame of 1,341 nucle- ( Cl CDA, Blair et al. 2006 ) is the fi rst and till otides encoding a complete 447-amino-acid now the only structure among CDAs that is protein. CDA gene from C. lindemuthianum known. It is a metalloenzyme and consists of a UPS9 was isolated and cloned in Pichia pastoris single catalytic domain, similar to the deformed as a tagged protein with six added terminal his- (β/α)8 fold in other CE-4 family members (Blair tidine residues. The specifi c activity of the puri- and Van Aalten 2004 ; Blair et al. 2005 ). The fi ed protein was 72 Umg−1 (Shrestha et al. 2004 ). active site cleft of the protein is formed from the A complete CDA cDNA from M. racemosus C -terminal ends of β-strands 2,4,5,7, and 8 of was cloned and sequenced by RT-PCR and the (β/α)8 barrel and includes fi ve distinct RACE with conserved primers (Xia-Yun et al. sequence motifs (MT1-MT5) conserved in the 9 Chitin Deacetylase: Characteristic Molecular Features and Functional Aspects 131 family (Blair et al. 2006 ). The catalytic subsite Functional Aspects contains a zinc- binding triad consisting of two histidines (His104, His108) from motif 2 and an Bioconversion to Chitosan aspartic acid (Asp50) from motif 1. In addition, A major objective for developing CDA has been a loop is present between strand β3 and helix α2, to replace the harsh chemical process for conver- supporting Trp79 that protrudes into the cata- sion of chitin into chitosan, the process as consid- lytic cleft. Two intramolecular disulfi de link- ered is not an environment-friendly one. The ages (Cys38- Cys237, Cys148-Cys152) are enzymatic deacetylation would be able to provide known to stabilize the structure. Sequence chitosans with defi ned levels of deacetylation, as alignment suggests that one disulfi de (Cys38- like the chemical process, it does not proceed in a Cys237), tethering the N - and C -terminal ends random fashion and permit the partially deacety- of the structure, is conserved in the fungal CDA lated chitosans obtained from the chemical process from M. rouxii and other CE-4 family to undergo the desired degree of deacetylation. members. The enzymatic deacetylation of various chitinous substrates was investigated by Aye et al. (2006 ) using the CDA isolated from Rhizopus oryzae . Salient Structural Features Chitin was observed to be a poor substrate for the enzyme, but reprecipitated chitin was moderately Among the carbohydrate active enzymes, CDAs better. F. velutipes CDA catalyzes deacetylation are relatively less studied in terms of structural of N -acetyl- chitooligomers, from dimer to pen- features. Cl CDA is a member of the carbohy- tamer, glycol chitin and colloidal chitin (Yamada drate esterases family 4 (CE-4s) which include et al. 2008 ). Chitosan with lower degree of deacet- several members that share the “NodB homol- ylation (28 % and 42 %) can be further deacety- ogy domain” (Caufrier et al. 2003 ). CDA, ace- lated with M. rouxii deacetylase (Martinou et al. tyl xylan esterase, rhizobial NodB, and 1995 , 1997a , b ). Yield and rate of deacetylation peptidoglycan deacetylases are involved in was observed to be more with amorphous chitin deacetylating chitin, xylan, the nonreducing substrates. Martinou et al. (2002 ) had attempted end GlcNAc from short chitooligosaccharides deacetylation of glycol chitin, chitin-50, and for synthesis of Nod factors, and N -acetylchitooligosaccharides using cobalt- N-acetylmuramic acid (MurNAc) or GlcNAc activated CDA from S. cerevisiae and concluded residues of the disugar repeats to modify bacte- that the enzyme requires at least two N -acetyl-d - rial cell wall peptidoglycan, respectively. In glucosamine residues for catalysis, exhibiting spite of sharing a homologous catalytic domain, maximum activity on hexa-N -acetylchitohexaose. signifi cant topological differences are observed CDAs from C. lindemuthianum (Kauss and Bausch between the deacetylases group, i.e., Cl CDA, 1988 ; Tsigos and Bouriotis 1995 ; Tokuyasu et al. Sp PgdA (Streptococcus pneumoniae peptido- 1997 ), A. nidulans (Alfonso et al. 1995 ), S. brevi- glycan deacetylase A), and Bs PdaA (Bacillus caulis (Cai et al. 2006 ), and Mortierella sp. (Kim subtilis peptidoglycan deacetylase), via struc- et al. 2008 ) also exhibited similar deacetylation tural analysis. The intramolecular disulfi de kinetics towards chitin and its oligomers. Crude bonds and an extended loop between strand β3 CDA from C. lindemuthianum was active on and helix α2 are present in Cl CDA while absent partially deacetylated chitin (chitin with 50 %, from the SpPgdA and BsPdaA. The N/C termini 65 %, 70 %, and 82 % degree of deacetylation) are present on the same side of the barrel in (Shrestha et al. 2004 ). Cl CDA while they are located at opposite ends Crystallinity and insolubility of chitin makes in SpPgdA and BsPdaA. Apart from these, the it a poor substrate for enzymatic deacetylation. conserved catalytic subsite lined by the His- The degree of crystallinity of the chitin must be His-Asp residues has only two metal coordinat- reduced to enable enzymes to access the internal ing residues in Bs PdaA (Blair et al. 2006 ). polysaccharide structure. Pretreatment of chitin 132 N. Pareek et al.

by either physical or chemical means prior to deacetylase specifi c for (GlcNAc) 2, but inactive enzymatic reaction was necessary to increase the with higher oligosaccharides (Ohishi et al. 2000 ). substrate accessibility to the enzyme (Martinou COD is considered as an essential part of the chi- et al. 1997a , b; Win et al. 2000). Nearly 90 % tin catabolic cascade of marine bacteria (Jung deacetylation of superfi ne chitin generated et al. 2008 ) due to its ability to produce heterodi- using CaCl 2 .2H2 O/methanol solvent system was saccharide GlcNAc-GlcN, a unique inducer for achieved using CDA from A. coerulea and C. lin- chitinase production. Chitin utilization among demuthianum (Win and Stevens 2001 ). Beaney several bacterial species, namely, V. parahae- et al. ( 2007 ) also modifi ed chitin by physical or molyticus KN1699 (Kadokura et al. 2007a ), chemical methods followed by deacetylating the V. furnissii (Bassler et al. 1991 ), T. kodakaraensis same with extracellular deacetylase from C. lin- KOD1 (Tanaka et al. 2003 ), Serratia marcescens demuthianum. Modifi cations of the chitin led to (Watanabe et al. 1997 ), S. lividans (Miyashita decreased crystallinity of the substrate with et al. 2000), and Streptomyces coelicolor (Saito simultaneous increase in enzymatic deacety- et al. 2000 , 2007), is found to be associated with lation. It was observed that the dissolution and the expression of COD. drying methods used in modifying the chitin had signifi cant impact on the fi nal effi ciency of the Synthesis of Cell Wall/Spore Wall enzymatic deacetylation reaction. Jaworska et al. Expression of CDA among fungal strains is (2009 ) found that immobilization of CDA to thought to be related to various stages of growth diethylaminoethyl cellulose via divinyl sulfone to execute different functions. In M. rouxii and led to high activity and stability towards various A. coerulea , CDA was localized near the peri- chitin and chitosans. plasmic space in the mycelia and contributed to formation of chitosan in the cell wall from Chitin Catabolism nascent chitin synthesized by the action of chitin Marine ecosystems contain huge quantities of synthetase (Davis and Bartnicki-Garcia 1984 ; chitin, with an annual production of >1011 metric Ruiz- Herrera and Martinez-Espinoza 1999 ). tons. This bulk amount of chitin is recycled by Even though enzymology and cytology of chitin the members of the family Vibrionaceae, capable biosynthesis in fungi have been extensively stud- of using chitin as the sole carbon source. Chitin ied, very little information exists on the correla- sensing, attachment, and degradation are the tion between CDA and chitosan biosynthesis three steps involved in the chitin utilization by (Hunt et al. 2008 ). Vibrio furnissii (Bassler et al. 1989 , 1991 ). S. cerevisiae requires chitin as an essential Interactions between V. cholerae and chitin occur component for vegetative growth; however, for at multiple hierarchical levels in the environment spore wall formation both chitin synthesis and and include cell metabolic and physiological chitin deacetylation are necessary. Chitin is syn- responses, e.g., chemotaxis, cell multiplication, thesized by three chitin synthases, Chs1, Chs2, induction of competence, biofi lm formation, com- and Chs3, in S. cerevisiae , and its conversion to mensal and symbiotic relationship with higher chitosan by either Cda1 or Cda2 had allowed the organisms, cycling of nutrients, as well as patho- second layered structure of the spore wall next to genicity for humans and aquatic animals (Pruzzo the outer dityrosine layer to retain its structural et al. 2008 ). Three sets of differentially regulated rigidity and resistance to various stresses (Mishra genes, i.e., a (GlcNAc)2 catabolic operon, two et al. 1997 ; Christodoulidou et al. 1999 ). Cda2p extracellular chitinases, a chitoporin, make is the predominant deacetylase and performs V. cholerae to be able to utilize chitin (Meibom most of the deacetylation task, while Cda1p con- et al. 2004 ). Induction of V. cholerae with tributes to the proper ascospore wall assembly + (GlcNH2 )2 or crab shells resulted into the produc- (Martinou et al. 2002 , 2003 ). In addition, a cda1 tion of COD, hydrolyzes the N -acetyl group encoded CDA in a fi ssion yeast S. pombe was attached to the penultimate GlcNAc unit (Li et al. identifi ed and required for proper spore forma- 2007 ). V. alginolyticus was observed to produce a tion (Matsuo et al. 2005 ). Four CDAs, namely, 9 Chitin Deacetylase: Characteristic Molecular Features and Functional Aspects 133

Cda1, Cda2, Cda3, and Fpd1, have been identi- biological control of insect pests, CDAs were ﬁ ed from C. neoformans , responsible for chito- observed to have dual roles. The enzyme initiates san synthesis which is an important component the pathogenesis of the insect-pathogenic fungus of the vegetative cell wall and helps to maintain by softening the insect cuticle to aid the mycelial cell integrity and aids in bud separation (Baker penetration and it also alters the fungal cell wall et al. 2007 ). to protect it from insect chitinase.

Plant-Pathogen Interaction CDA aids pathogenesis of plant pathogenic fungi, Conclusion and Future Perspectives namely, wheat stem rust fungus Puccinia grami- nis f. sp. tritici and the broad bean rust fungus Chitosan is a biopolymer with immense commer- U. fabae , and the causative agents of anthracnose, cial potential. Its functional characteristics cor- Colletotrichum graminicola and C. lindemuthia- relate with its structural features. Enzyme-aided num, by performing dual roles, i.e., protection of production of chitosan implying CDA possesses penetrating fungal hyphae from being lysed by ecological and economic benefi ts over conven- secretory plant chitinases by transforming the tional production approaches. Various research superfi cial cell wall chitin into chitosan and groups are involved in studying the production decrease the activity of chitin oligomers. Plants, and biochemical and molecular characterization when attacked by fungal pathogens, activate an of the enzyme but implication of bioconversion at elaborate defense system consisting of chemi- large scale needs intensive screening of novel CDA cally and physically performed resistance factors hyper-producers that are able to produce the and have induced resistance reactions to evade enzyme with higher catalytic effi cacy and lower the pathogenesis. Chitin oligomers play a vital inhibition by end product, i.e., acetate. Attempts role in eliciting the plant-defense mechanisms, should be made to further improve the catalytic namely, callose formation, lignifi cation, and syn- activity and stability of the enzyme by cloning and thesis of coumarin derivatives. Plant endo-type characterizing the corresponding molecular chitinases produce the chitin oligomers by domains and then targeted mutagenesis of the degrading the fungal chitin. Fungal pathogens corresponding sequences in the CDA gene. Apart escape the plant hydrolases by partially deacety- from this, an effi cient pretreatment approach needs lating the exposed chitin polymers during the initial to be developed to decrystallize the chitin structure growth via action of CDA (Kauss et al. 1983 ; to improve not only the substrate accessibility but Walker-Simmons et al. 1984 ; Vander et al. 1998 ; also the rate of enzymatic deacetylation. Gueddari et al. 2002 ; Hekmat et al. 2003 ). Chitin- binding domain (CBM14) of avirulence protein Avr4 of Cladosporium fulvum contributes to its References pathogenicity by protecting the fungal cell wall against hydrolysis by plant chitinases during Alfonso C, Nuero OM, Santamaria F, Reyes F (1995) infection (van den Burg et al. 2006 ). Purifi cation of a heat-stable chitin deacetylase from Aspergillus nidulans and its role in cell wall degradation. Curr Microbiol 30:49–54 Biocontrol Agent Amorim RVS, Ledingham WM, Fukushima K, Campos- The role displayed by CDAs during pathogenesis Takaki GM (2005) Screening of chitin deacetylase of plant pathogenic fungi makes them a crucial from Mucoralean strains (Zygomycetes) and its rela- target in the biological control of plant patho- tionship to cell growth rate. J Ind Microbiol Biotechnol 32:19–23 genic fungi (Brosson et al. 2005 ; Das et al. 2006 ; Arakane Y, Dixit R, Begum K, Park Y, Specht CA, Baker et al. 2007) and insect pests (Nahar et al. Merzendorfer H, Kramer KJ, Muthukrishnan S, 2004 ). Inhibition of CDA would result into Beeman RW (2009) Analysis of functions of the chitin hydrolysis of fungal cell wall by plant chitinases; deacetylase gene family in Tribolium castaneum . Insect Biochem Mol Biol 39:355–365 thus, the control of the plant pathogenic fungi Araki Y, Ito E (1975) A pathway of chitosan formation in becomes feasible (Tokuyasu et al. 1996 ). In the Mucor rouxii . Eur J Biochem 55:71–78 134 N. Pareek et al.

Aye KN, Karuppuswamy R, Ahamed T, Stevens WF Das S, Dellen KV, Bulik D, Magnelli P, Cui J, Head J, (2006) Peripheral enzymatic deacetylation of chitin Robbins PW, Samuelson J (2006) The cyst wall of and reprecipitated chitin particles. Bioresour Technol Entamoeba invadens contains chitosan (deacetylated 97:577–582 chitin). Mol Biochem Parasitol 148:86–92 Baker LG, Specht CA, Donlin MJ, Lodge JK (2007) Davis LL, Bartnicki-Garcia S (1984) Chitosan synthesis Chitosan, the deacetylated form of chitin, is necessary by the tandem action of chitin synthetase and chitin for cell wall integrity in Cryptococcus neoformans . deacetylase from Mucor rouxii . Biochemistry Eukaryot Cell 6:855–867 23:1065–1073 Bassler BL, Gibbons PJ, Roseman S (1989) Chemotaxis Deising H, Siegrist J (1995) Chitin deacetylase activity of to chitin oligosaccharides by Vibrio furnissii, a chitini- the rust Uromyces viciae-fabae is controlled by fungal vorous marine bacterium. Biochem Biophys Res morphogenesis. FEMS Microbiol Lett 127:207–212 Commun 161:1172–1176 Dixit R, Arakane Y, Specht CA, Richard C, Kramer KJ, Bassler BL, Gibbons PJ, Yu C, Roseman S (1991) Chitin Beeman RW, Muthukrishnan S (2008) Domain orga- utilization by marine bacteria: chemotaxis to chitin nization and phylogenetic analysis of proteins from oligosaccharides by Vibrio furnissii . J Biol Chem the chitin deacetylase gene family of Tribolium casta- 266:24268–24275 neum and three other species of insects. Insect Beaney PD, Gan Q, Magee TRA, Flealy M, Lizardi- Biochem Mol Biol 38:440–451 Mendoza J (2007) Modifi cation of chitin properties Ferguson MJL, Gooday GW (1996) Environmental recy- for enzymatic deacetylation. J Chem Technol cling of chitin. In: Muzzarelli RAA (ed) Chitin enzy- Biotechnol 82:165–173 mology. Atec, Grottammare, pp 393–396 Blair DE, Van Aalten DM (2004) Structures of Bacillus Gao XD, Katsumoto T, Onodera K (1995) Purifi cation subtilis PdaA, a family 4 carbohydrate esterase, and a and characterization of chitin deacetylase from complex with N -acetyl-glucosamine. FEBS Lett Absidia coerulea . J Biochem 117:257–263 570:13–19 Gauthier C, Clerisse F, Dommes J, Versali MJ (2008) Blair DE, Schuttelkopf AW, Macrae JI, van Aalten DM Characterization and cloning of chitin deacetylases (2005) Structure and metal-dependent mechanism of from Rhizopus circinans . Protein Expr Purif peptidoglycan deacetylase, a streptococcal virulence 59:127–137 factor. Proc Natl Acad Sci USA 102:15429–15434 Gueddari NEE, Rauchhaus U, Moerschbacher BM, Blair DE, Hekmat O, Schuttelkopf AW, Shrestha B, Deising HB (2002) Developmentally regulated con- Tokuyasu K, Withers SG, van Aalten DMF (2006) version of surface-exposed chitin to chitosan in cell Structure and mechanism of chitin deacetylase from walls of plant pathogenic fungi. New Phytol the fungal pathogen Colletotrichum lindemuthianum . 156:103–112 Biochemistry 45:9416–9426 Guo W, Li G, Pang Y, Wang P (2005) A novel chitin- Brosson D, Kuhn L, Prensier G, Vivarès CP, Texier C binding protein identifi ed from the peritrophic mem- (2005) The putative chitin deacetylase of brane of the cabbage looper, Trichoplusia ni . Insect Encephalitozoon cuniculi: a surface protein implicated Biochem Mol Biol 35:1224–1234 in microsporidian spore-wall formation. FEMS Hekmat O, Tokuyasu K, Withers SG (2003) Subsite struc- Microbiol Lett 247:81–90 ture of the endo-type chitin deacetylase from a Cai J, Yang J, Du Y, Fan L, Qiu Y, Li J, Kennedy JF (2006) Deuteromycete, Colletotrichum lindemuthianum : an Purifi cation and characterization of chitin deacetylase investigation using steady-state kinetic analysis and from Scopulariopsis brevicaulis. Carbohydr Polym MS. Biochem J 374:369–380 65:211–217 Hirano T, Kadokura K, Ikegami T, Shigeta Y, Kumaki Y, Campbell PM, Cao AT, Hines ER, East PD, Gordon KHJ Hakamata W, Oku T, Nishio T (2009) Heterodisaccharide (2008) Proteomic analysis of the peritrophic matrix 4-O -(N -acetyl-beta-d -glucosaminyl)-d -glucosamine is from the gut of the caterpillar, Helicoverpa armigera . a specifi c inducer of chitinolytic enzyme production in Insect Biochem Mol Biol 38:950–958 Vibrios harboring chitin oligosaccharide deacetylase Caufrier F, Martinou A, Dupont C, Bouriotis V (2003) genes. Glycobiology 19:1046–1053 Carbohydrate esterase family 4 enzymes: substrate Hunt DE, Gevers D, Vahora NM, Polz MF (2008) specifi city. Carbohydr Res 338:687–692 Conservation of the chitin utilization pathway in the Chang KLB, Tsai G, Lee J, Fu W-R (1997) Heterogeneous Vibrionaceae. Appl Environ Microbiol 74:44–51 N-deacetylation of chitin in alkaline solution. Jaworska MM, Bryjak J, Liesiene J (2009) A search of an Carbohydr Res 303:327–332 optimal carrier for immobilization of chitin deacety- Christodoulidou A, Briza P, Ellinger A, Bouriotis V lase. Cellulose 16:261–270 (1999) Yeast ascospore wall assembly requires two Jeraj N, Kunic B, Lenasi H, Breskvar K (2006) Purifi cation chitin deacetylase isozymes. FEBS Lett 460:275–279 and molecular characterization of chitin deacetylase Coutinho PM, Henrissat B (1999) Carbohydrate-active from Rhizopus nigricans. Enzyme Microb Technol enzymes: an integrated database approach. In: Gilbert 39:1294–1299 HJ, Davis GJ, Henrissat B, Svensson B (eds) Recent Jung BO, Roseman S, Park JK (2008) The central concept advances in carbohydrate bioengineering. Royal for chitin catabolic cascade in marine bacterium, Society of Chemistry, Cambridge, UK, pp 3–12 Vibrios . Macromol Res 16:1–5 9 Chitin Deacetylase: Characteristic Molecular Features and Functional Aspects 135

Kadokura K, Rokutani A, Yamamoto M, Ikegami T, Sugita chitin deacetylase (Cda2p) from Saccharomyces cere- H, Itoi S, Hakamata W, Oku T, Nishio T (2007a) visiae. Protein Expr Purif 24:111–116 Purifi cation and characterization of Vibrio parahae- Martinou A, Koutsiouli D, Bouriotis V (2003) Cloning molyticus extracellular chitinase and chitin oligo- and expression of chitin deacetylase gene (chitin saccharide deacetylase involved in the production deacetylase 2) from Saccharomyces cerevisiae in of heterodisaccharide from chitin. Appl Microbiol Escherichia coli . Purifi cation and characterization of Biotechnol 75:357–365 the cobalt-dependent recombinant enzyme. Enzyme Kadokura K, Sakamoto Y, Saito K, Ikegami T, Hirano T, Microb Technol 32:757–763 Hakamata W, Oku T, Nishio T (2007b) Production of Matsuo Y, Tanaka K, Matsuda H, Kawamukai M (2005) a recombinant chitin oligosaccharide deacetylase from cda1 +, encoding chitin deacetylase is required for Vibrio parahaemolyticus in the culture medium of proper spore formation in Schizosaccharomyces Escherichia coli cells. Biotechnol Lett 29:1209–1215 pombe . FEBS Lett 579:2737–2743 Kafetzopoulos D, Martinou A, Bouriotis V (1993a) Maw T, Tan TK, Khor E, Wong SM (2002a) Complete Bioconversion of chitin to chitosan: purifi cation and cDNA sequence of chitin deacetylase from Gongronella characterization of chitin deacetylase from Mucor butleri and its phylogenetic analysis revealed clusters rouxii . Proc Natl Acad Sci USA 90:2564–2568 corresponding to taxonomic classifi cation of fungi. J Kafetzopoulos D, Thireos G, Vournakis JN, Bouriotis V Biosci Bioeng 93:376–381 (1993b) The primary structure of a fungal chitin Maw T, Tan TK, Khor E, Wong SM (2002b) Selection of deacetylase reveals the function for two bacterial gene Gongronella butleri strains for enhanced chitosan products. Proc Natl Acad Sci USA 90:8005–8008 yield with UV mutagenesis. J Biotechnol 95:189–193 Kauss H, Bausch B (1988) Chitin deacetylase from Meens J, Schreiber C, Deshpande MV (2001) Screening Colletotrichum lindemuthianum . In: Woods WA, of marine fungi for new chitin deacetylase enzymes. Kellogg ST (eds) Methods in enzymology. Academic, In: Muzzarelli RAA (ed) Chitin enzymology. Atec, San Diego, pp 518–523 Grottammare, pp 533–540 Kauss H, Jeblick W, Young DH (1983) Chitin deacetylase Meibom KL, Li XBB, Nielsen AT, Wu CY, Roseman S, from the plant pathogen Colletotrichum lindemuthia- Schoolnik GK (2004) The Vibriocholerae chitin num . Plant Sci 28:231–236 utilization program. Proc Natl Acad Sci USA Kim YJ, Zhao Y, Oh KT, Nguyen VN, Park RD (2008) 101:2524–2529 Enzymatic deacetylation of chitin by extracellular chitin Mishra C, Semino CE, Mccreath KJ, De La Vega H, Jones deacetylase from a newly screened Mortierella sp. BJ, Specht CA, Robbins PW (1997) Cloning and expres- DY-52. J Microbiol Biotechnol 18:759–766 sion of two chitin deacetylase genes of Saccharomyces Kurita K (2006) Chitin and chitosan: functional biopoly- cerevisiae . Yeast 13:327–336 mers from marine crustaceans. Marine Biotechnol Miyashita K, Fujii T, Saito A (2000) Induction and repres- 8:203–226 sion of a Streptomyces lividans chitinase gene pro- Li XB, Wang LX, Wang XS, Roseman S (2007) The chi- moter in response to various sources. Biosci Biotechnol tin catabolic cascade in the marine bacterium Vibrio Biochem 64:39–43 cholerae: characterization of a unique chitin oligosac- Nahar P, Ghormade V, Deshpande MV (2004) The extra- charide deacetylase. Glycobiology 17:1377–1387 cellular constitutive production of chitin deacetylase Luschnig S, Bätz T, Armbruster K, Krasnow MA (2006) in Metarhizium anisopliae: possible edge to entomo- serpentine and vermiform encode matrix proteins with pathogenic fungi in the biological control of insect chitin binding and deacetylation domains that limit pests. J Invertebr Pathol 85:80–88 tracheal tube length in Drosophila . Curr Biol Noh MY, Muthukrishnan S, Kramer KJ, Beeman RW, 16:186–194 Arakane Y (2011) Expression profi les and functional Martinou A, Kafetzopoulos D, Bouriotis V (1995) Chitin analysis of genes encoding chitin deacetylases, deacetylation by enzymatic means: monitoring of extracellular matrix-modifying proteins in Tribolium deacetylation process. Carbohydr Res 273:235–242 castaneum. Entomol Res. doi: 10.1111/j.1748-5967. Martinou A, Tsigos I, Bouriotis V (1997a) Enzymatic 2011.00425.x deacetylation of chitooligosaccharides. In: Muzzarelli Ohishi H, Murase K, Ohta T, Etoh H (2000) Cloning and RAA, Peter MG (eds) Chitin handbook. Atec, sequencing of the deacetylase gene from Vibrio algi- Grottammare, pp 191–194 nolyticus H-8. J Biosci Bioeng 90:561–563 Martinou A, Tsigos I, Bouriotis V (1997b) Preparation of Pruzzo C, Vezzulli L, Colwell RR (2008) Global impact chitosan by enzymatic deacetylation. In: Muzzarelli of Vibrio cholerae interactions with chitin. Environ RAA, Peter MG (eds) Chitin handbook. Atec, Microbiol 10:1400–1410 Grottammare, pp 501–506 Ruiz-Herrera J (1978) The distribution and quantitative Martinou A, Bouriotis V, Stokke BT, Vårum KM (1998) importance of chitin in fungi. In: Muzzarelli RA, Mode of action of chitin deacetylase from Mucor Pariser ER (eds) In proceedings of the Ist international rouxii on partially N -acetylated chitosans. Carbohydr conference on chitin and chitosan. Massachusetts Res 311:71–78 Institute of Technology, Cambridge, MA, pp 11–21 Martinou A, Koutsioulis D, Bouriotis V (2002) Expression, Ruiz-Herrera J, Martinez-Espinoza AD (1999) Chitin bio- purifi cation, and characterization of a cobalt-activated synthesis and structural organization in vivo. In: Jolles 136 N. Pareek et al.

P, Muzzarelli RAA (eds) Chitin and chitinases. Tsigos I, Zydowicz N, Martinou A, Domard A, Bouriotis Birkhauser Verlag, Basel, pp 39–53 V (1999) Mode of action of chitin deacetylase from Saito A, Ishizaka M, Francisco PB, Fujii T, Miyashita K Mucor rouxii on N -acetylchitooligosaccharides. Eur J (2000) Transcriptional co-regulation of fi ve chitinase Biochem 261:698–705 genes scattered on the Streptomyces coelicolor A3(2) Tsigos I, Martinou A, Kafetzopoulos D, Bouriotis V chromosome. Microbiology 146:2937–2946 (2000) Chitin deacetylases: new, versatile tools in bio- Saito A, Shinya T, Miyamoto K, Yokoyama T, Kaku H, technology. Trends Biotechnol 18:305–312 Minami E, Shibuya N, Tsujibo H, Nagata Y, Ando A, Van den Burg HA, Harrison SJ, Joosten MHAJ, Vervoort J, Fujii T, Miyashita K (2007) The dasABC gene cluster, de Wit PJGM (2006) Cladosporium fulvum Avr4 pro- adjacent to dasR, encodes a novel ABC transporter for tects fungal cell walls against hydrolysis by plant chitin- the uptake of N , N ′-diacetylchitobiose in Strep- ases accumulating during infection. Am Phytopathol tomyces coelicolor A3(2). Appl Environ Microbiol Soc 19:1420–1430 73:3000–3008 Vander P, Varum KM, Domard A, Gueddari NEE, Shrestha B, Blondeau K, Stevens WF, Hegarat FL (2004) Moersschbacher BM (1998) Comparison of the ability Expression of chitin deacetylase from Colletotrichum of partially N-acetylated chitosans and chitooligosac- lindemuthianum in Pichia pastoris: purifi cation and charides to elicit resistance reactions in wheat leaves. characterization. Protein Expr Purif 38:196–204 Plant Physiol 118:1353–1359 Shin H-J, Kim M, Lee DS (1999) Purifi cation and charac- Walker-Simmons M, Jin D, West CA, Hadwiger CA, terization of N -acetylglucosamine 6-phosphate dea- Ryan CA (1984) Comparison of proteinase inhibitor- cetylase from Th ermus caldophilus . J Biosci Bioeng inducing activities and phytoalexin elicitor activities 88:319–322 of a pure fungal endopolygalacturonase, pectic frag- Tanaka T, Fukui T, Atomi H, Imanaka T (2003) ments, and chitosans. Plant Physiol 76:833–836 Characterization of an Exo-β- d -glucosaminidase Wang S, Jayaram SA, Hemphälä J, Senti KA, Tsarouhas Involved in a novel chitinolytic pathway from the V, Jin H, Samakovlis C (2006) Septate-junction- hyperthermophilic archaeon Thermococcus kodak- dependent luminal deposition of chitin deacetylases araensis KOD1. J Bacteriol 185:5175–5181 restricts tube elongation in the Drosophila Trachea. Taylor EJ, Gloster TM, Turkenburg JP, Vincent F, Brzozowski Curr Biol 16:180–185 AM, Dupont C, Shareck F, Centeno MSJ, Prates JAM, Watanabe T, Kimura K, Sumiya T, Nikaidou N, Suzuki K, Puchart V, Ferreira LMA, Fontes CMGA, Biely P, Davies Suzuki M, Taiyoji M, Ferrer S, Regue M (1997) GJ (2006) Structure and activity of two metal-ion Genetic analysis of the chitinase system of Serratia dependent acetyl xylan esterases involved in plant cell marcescens 2170. J Bacteriol 179:7111–7117 wall degradation reveals a close similarity to peptido- Win NN, Stevens WF (2001) Shrimp chitin as substrate for glycan deacetylase. J Biol Chem 281:10968–10975 fungal chitin deacetylase. Appl Microbiol Biotechnol Tokuyasu K, Ohnishi-Kameyama M, Hayashi K (1996) 57:334–341 Purifi cation and characterization of extracellular chitin Win NN, Pengju G, Stevens WF (2000) Deacetylation of deacetylase from Colletotrichum lindemuthianum . chitin by fungal enzymes. In: Peter MG, Domard A, Biosci Biotechol Biochem 60:1598–1603 Muzzarelli RAA (eds) Advances in chitin science, Tokuyasu K, Ono H, Ohnishi-Kameyama M, Hayashi K, vol 4. University of Postdam, Postdam, pp 55–62 Mori Y (1997) Deacetylation of chitin oligosaccharides Xia-Yun J, Shu-Ming Z, Pei-Gen Z (2007) Cloning and of dp 2–4 by chitin deacetylase from Colletotrichum sequence analysis of complete cDNA of chitin deacet- lindemuthianum . Carbohydr Res 303:353–358 ylase from Mucor racemosus . Chin J Agric Biotechnol Tokuyasu K, Ohnishi-Kameyama M, Hayashi K, Mori Y 4:167–172 (1999) Cloning and expression of chitin deacetylase Yamada M, Kurano M, Inatomi S, Taguchi G, Mitsuo M, gene from a deuteromycete, Colletotrichum lindemu- Shimosaka M (2008) Isolation and characterization of thianum . J Biosci Bioeng 87:418–423 a gene coding for chitin deacetylase specifi cally Tokuyasu K, Mitsutomi M, Yamaguchi I, Hayashi K, expressed during fruiting body development in the Mori Y (2000) Recognition of chitooligosaccharides basidiomycete Flammulina velutipes and its expres- and their N -acetyl groups by putative subsites of chitin sion in the yeast Pichia pastoris. FEMS Microbiol deacetylase from a deuteromycete, Colletotrichum lin- Lett 289:130–137 demuthianum . Biochemistry 39:8837–8843 Zhao Y, Park R-D, Muzzarelli AA (2010a) Chitin deacet- Toprak U, Baldwin D, Erlandson M, Gillott C, Hou X, ylases: properties and applications. Mar Drugs 8: Coutu C, Hegedus DD (2008) A chitin deacetylase 24–46 and putative insect intestinal lipases are components Zhao Y, Kim YJ, Oh KT, Van Nam Nguyen VN, Ro Dong of the Mamestra confi gurata (Lepidoptera: Noctuidae) Park RD (2010b) Production and characterization of peritrophic matrix. Insect Mol Biol 17:573–585 extracellular chitin deacetylase from Absidia Trudel J, Asselin A (1990) Detection of chitin deacetylase corymbifera DY-9. J Korean Soc Appl Biol Chem activity after polyacrylamide gel electrophoresis. Anal 53:119–126 Biochem 189:249–253 Zhou G, Zhang H, He Y, He L (2010) Identifi cation of a Tsigos I, Bouriotis V (1995) Purifi cation and character- chitin deacetylase producing bacteria isolated from ization of chitin deacetylase from Colletotrichum lin- soil and its fermentation optimization. Afr J Microbiol demuthianum . J Biol Chem 270:26286–26291 Res 4:2597–2603 Role of Enzymes and Proteins in Plant-Microbe Interaction: 1 0 A Study of M. oryzae Versus Rice

Jahangir Imam , Mukund Variar , and Pratyoosh Shukla

Abstract The wall interface between rice and Magnaporthe oryzae plays an important role in the outcome of their interactions, i.e., resistance or sus- ceptibility. A number of enzymes and proteins are involved in both external and internal interactions. The blast fungus secretes many enzymes which help in the plant cell wall degradation and the entry of fungus into the plant cell which results in the development of disease. To restrict the growth and development of blast fungus, the rice plants have also developed many defense mechanisms like generation of defense substances and hydrogen peroxide catalysis by the production of some enzymes in plant cells. These enzymes occur frequently in many isoforms and help in plant defense. Proteins also participate in the defense against blast fungus attack. These proteins are called as pathogenesis-related proteins (PRs). PR proteins have activities of both proteins and hydrolytic enzymes. Chitinase and β-1,3-glucanase are the most common PR proteins which can hydrolyze major components of blast fungal cell walls, chitin and β-1,3-glucan, respectively.

Keywords Rice • Magnaporthe oryzae • Xylanase • Cutinase • PR proteins • Resistance

J. Imam M. Variar Biotechnology Laboratory , Central Rainfed Upland Biotechnology Laboratory , Central Rainfed Upland Rice Research Station (CRRI), Hazaribagh 825301 , Rice Research Station (CRRI), Hazaribagh 825301 , Jharkhand , India Jharkhand , India Enzyme Technology and Protein Bioinformatics P. Shukla () Laboratory, Department of Microbiology , Enzyme Technology and Protein Bioinformatics Maharshi Dayanand University , NH10 , Laboratory, Department of Microbiology , Rohtak 124001 , Haryana , India Maharshi Dayanand University , NH10 , Rohtak 124001 , Haryana , India e-mail: [email protected]

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 137 DOI 10.1007/978-81-322-1094-8_10, © Springer India 2013 138 J. Imam et al.

Role of Enzymes in Magnaporthe Introduction oryzae-Rice Interaction

The interaction between plants and microbial The rice blast fungus secretes a battery of pathogens is among the most complex phenom- enzymes which facilitate its colonization in the ena in biology. Different aspects of interaction plant tissue. These enzymes are mainly cell wall- specifi city and defense mechanisms of plants degrading enzymes (CWDEs) like xylanases, against potential fungal pathogens have received cutinases, and other enzymes. Magnaporthe great attention in the last few years. Generally oryzae also secretes metabolic enzymes like tre- plants have two levels of defense mechanisms halase that helps in plant tissue colonization according to their function, structural (constitu- (Foster et al. 2003 ). On the other hand, rice tive) and biochemical (active). The structural plant synthesizes enzymes and proteins as their compounds present the fi rst line of defense defense against the blast fungus. This chapter against invading pathogens by forming mechani- focuses mainly on enzymes and proteins from cal barriers or by forming preformed chemical plant pathogens which have been extensively substances. Biochemical processes participate in studied and characterized in Magnaporthe oryzae - active defense reaction of plants against patho- rice interaction. gens (Lebeda et al. 1999 ). The primary walls of plant cells are pivotal battlegrounds between microbial pathogens and Xylanases their hosts. Microbial pathogens secrete an array of cell wall-degrading enzymes (CWDEs) and Xylan is the predominant hemicellulose compo- other enzymes capable of breaking the plant cell nent in plant cell walls and the second most abun- wall and causing infection (Wu et al. 2006 ). dant polysaccharide in nature (Subramaniyan and Interaction between cells of Magnaporthe oryzae Prema 2002 ). Xylan is a heteropolysaccharide and rice involves a complex of biological infl u- having a backbone of β-1,4-linked xylopyranose ences which lead to rice blast disease. Patho- units, with groups of acetyl, 4-O-methyl-d - genesis in general and the initial infection steps glucuronosyl, and α-arabinofuranosyl residues in particular may be viewed as a sequence of dis- linked to the backbone (Subramaniyan and Prema crete, critical events. In Magnaporthe oryzae- rice 2002 ). The complete degradation of xylan in pathogenesis, CWDEs as well as other enzymes plant requires the activity of complex hydrolytic play a crucial role and involve both external and enzymes with diverse mode of action (Beg et al. internal interactions. Proteins participating in 2001 ). Out of the many xylanolytic enzymes, defense mechanisms after the fungal attack are endo-β-1,4-xylanase is the most important, which generally called pathogenesis- related proteins is required to cleave the main xylan backbone (PR proteins). Initially, it was assumed that PR chain (Biely and Tenkanen 1998 ). The many proteins are devoid of any enzymatic activity, xylan-degrading enzymes secreted by fungi are but Legrand et al. (1987 ) detected chitinase activ- one of their components of offensive arsenal ity in four members of group 3 tobacco PRs and (Belien et al. 2006 ). Many plant pathogenic fungi later on established β-1,3- glucanase activity in secrete endoxylanases when grown in the pres- four members of group 2 tobacco PRs. PRs also ence of host cell walls (Cooper et al. 1988 ; have antifungal effect and show stronger accu- Lehtinen 1993 ; Ruiz et al. 1997 ; Wu et al. 1997 ; mulation in resistant than susceptible plants. Giesbert et al. 1998; Carlile et al. 2000 ; Hatsch There is also a high level of constitutive expres- et al. 2006 ). sion of PR proteins in naturally resistant plants The recently published Magnaporthe oryzae (Edreva 2005 ). genome sequence unveiled the possible presence 10 Role of Enzymes and Proteins in Plant-Microbe Interaction: A Study of M. oryzae Versus Rice 139 of as many as 20 xylanase genes, which encodes Cutinases six glycoside hydrolase family 10 (GH10), 5 GH11, and 9 GH43 members (Dean et al. 2005 ). The cuticle which is present over all parts of the The high level of redundancy is an indication that aerial plant presents the fi rst physical barrier to xylanase activity is essential for the vitality of pathogen entry and infection. The main structural Magnaporthe oryzae, either saprophytically or component of the plant cuticle is cutin which pathogenetically or both (Wu et al. 2006 ). occurs as a hydrophobic cutin network of esteri- M. oryzae secretes several isoforms of endo-β-1, fi ed hydroxyl and epoxy fatty acids which are

4-xylanase, and these isoforms act as pathogenicity n-C16 and n-C18 types intermingled with wax factors (Wu et al. 1997 ). Experiment proves that (Kolattukudy 2001 ; Lequeu et al. 2003 ; Nawrath deletion of one or two xylanase genes did not 2006 ). Magnaporthe oryzae uses direct method abolish endoxylanase activity (Wu et al. 1997 ) of penetration, i.e., through cuticle. The penetra- and no detectable effect on virulence is observed. tion through cuticle requires both physical pres- The presence of multiple endoxylanase and sure (Howard et al. 1991 ; Bechinger et al. 1999 ) β-xylosidase genes in M. oryzae may be the and enzymatic degradation by extracellular reason why mutants in individual xylanase genes cutinases (Skamnioti et al. 2007 ). Cutin mono- remain pathogenic (Apel-Birkhold and Walton mers promote germ tube and appressorium dif- 1996 ; Wegener et al. 1999 ; Gomez-Gomez et al. ferentiation on chemically inert surfaces in M. 2002). One evidence that supports an important oryzae (Gilbert et al. 1996 ; DeZwaan et al. 1999 ) role for xylanases in the pathogenicity of M. ory- and showed enhanced resistance to infection by zae is that when cultured rice cells were treated M. oryzae (Schweizer et al. 1994 ). Sweigard with commercial xylanase, it causes cell death et al. (1992a , b ) cloned and identifi ed cutin- (Ishii 1998 ). Many fungi produce glycanases degrading enzyme from M. oryzae . They named which facilitate colonization of plant tissue (eg. it as CUTINASE1 ( CUT1 ) gene. They showed galacturonases, xylanases and glucanases) that that this gene is expressed when cutin is the sole fragments plant cell wall polysaccharides that are carbon source but not when carbon source is generated by these glycanases, provide the fun- cutin and glucose together or glucose alone. Dean gus with a carbon source but also elicit the plant et al. (2005 ) revealed seven more members of the defence response (Wu et al. 1997 ). The purifi ca- cutinase family in M. oryzae genomes and 16 tion, cloning, and characterization of two xyla- putative cutinases in genome sequence release nases from M. oryzae are steps towards fi ve. Such large number of cutinases in M. oryzae analyzing the role of xylanase in the interaction genome refl ects functional redundancy or vary- of M. oryzae with its rice host. One early study ing specifi city of these enzymes (Skamnioti et al. reported two types of xylanases with different pH 2007 ). Appressoria formation in M. oryzae occurs optima from M. oryzae (Sumizu et al. 1961 ). Till either in hydrophobic surfaces or in the presence now at least 17 putative xylanases in the genome of soluble host cutin monomers but not on hydro- of M. oryzae have been identifi ed. Six of them philic surfaces (Lee and Dean 1994 ; Gilbert et al. (Xyl 1–6) have been partially characterized 1996), and the cutin monomers alone are suffi - (Wu et al. 1997 ). Xyl 1, Xyl 3, and Xyl 4 encode cient to induce appressorium differentiation class XI endo-β-xylanases, while Xyl 2, Xyl 5, (Choi and Dean 1997). The plasma membrane and Xyl 6 encode class 10 endo-β-xylanases (Wu protein Pth11p functions at the cell cortex as an et al. 1997 ). Knockout studies suggest that Xyl 1, upstream effector of appressorium differentia- Xyl 4, and Xyl 5 are pathogenicity factors, while tion in response to soluble plant cutin mono- Xyl 2 may have a role in initiating the host plant mers (DeZwaan et al. 1999 ). Skamnioti et al. defense responses. (2007 ) identifi ed a specifi c M. oryzae cutinase, 140 J. Imam et al.

CUTINASE2 (CUT2 ), which showed a dramatic kinase (PKA) and a putative Ca2+ binding site. uplift in transcription during appressorium matu- This reveals that NTH1 is a regulated protein. ration and penetration. They proposed that CUT2 The gene for trehalase (NTH1 ) in M. oryzae is is an upstream activator of the cAMP/PKA and expressed during its sporulation, plant infection, DAG/PKC signaling pathways that directs and in response to environmental stress (Foster appressorium formation and infection growth in et al. 2003 ). The mutant for ∆nth1 gene in M. oryzae . CUT2 mutant shows reduced extracel- M. oryzae showed slow proliferation of invasive lular serine esterase and plant cutin-degrading hyphae as compared to its wild type. This con- activity and attenuated pathogenicity on rice. cludes that NTH1 is required by M. oryzae to Exogenous application of synthetic cutin mono- generate severe blast symptoms (Foster et al. mers, cAMP and DAG, restores the morphologi- 2003 ). Three putative TRE1 products (trehalase cal and pathogenicity defects of the Cut2 mutant encoding) are 33 % similar to human and mouse to wild-type levels. CUT2 plays no part in spore TreA trehalase but are distinct from both acidic or appressorium adhesion or in appressorial tur- and neutral trehalases from fungi (Foster et al. gor generation, but mediates the formation of 2003 ). TRE1 -encoded trehalase is required both penetration peg (Skamnioti and Gurr 2008 ). The for growth on trehalose and mobilization of intra- cutin monomer ligand released by CUT2 is per- cellular trehalose in M. oryzae . ceived by one of the G-protein-coupled receptors (GPCRs) in M. oryzae (Kulkarni et al. 2005 ). Overall, CUT2 is required for surface sensing Catalase-Peroxidases leading to correct germ lining differentiation, penetration, and full virulence in M. oryzae The generation of reactive oxygen species (ROS) − (Skamnioti et al. 2007 ). such as superoxide anion radicals (O2 ), hydroxyl · radicals (OH ), and hydrogen peroxide (H2 O2 ) in plant cells is one of the most rapid and drastic Trehalases defense reactions activated following pathogen attack (Doke 1983 ; Lamb and Dixon 1997 ). Trehalase is a glycoside hydrolase enzyme that Catalase enzyme is present in peroxisomes and catalyzes the conversion of trehalose to glucose. catalyzes the hydrogen peroxide into water and Trehalose is a nonreducing disaccharide com- oxygen, while peroxidase enzyme catalyzes the monly found in all eukaryotic cells except mam- oxidation of substrates like phenol and its deriva- mals as storage carbohydrate (Arguelles 2000 ). tive with the help of hydrogen peroxide. Both the The disaccharide is hydrolyzed into two mole- enzymes are the competitor of each other because cules of glucose by the enzyme trehalase. they both use the same substrate. Class III peroxi- Trehalose mobilization may be involved by many dases ( POXs ) provide resistance to plants against virulence-associated functions in M. oryzae like blast disease infection, but there is no clear-cut germination of conidia, development of infected evidence of POX as self-defense for plants at the cells on the leaf surface, and subsequent plant molecular level (Sasaki et al. 2004 ). POX genes tissue colonization (Foster et al. 2003 ). In constitute a multigene family, and the redundant M. oryzae , breakdown of trehalose requires two expression of many POX genes against pathogen trehalases: a neutral trehalase encoded by a gene attack and environmental stresses may guarantee NTH1 and a novel trehalase encoded by TRE1 its necessities in self-defense (Sasaki et al. 2004 ). which is required for mobilization during The M. oryzae genome contains two true heme spore germination, but dispensable for pathoge- catalases, catalase A (CATA ) and catalase B nicity (Foster et al. 2003 ). Neutral trehalase (CATB ), plus two bifunctional catalase-peroxidase NTH1 is regulated by protein phosphorylation- genes, catalase-peroxidase A ( CPXA) and cata- dephosphorylation in M. oryzae . NTH1 has phos- lase-peroxidase B (CPXB ) (Skamnioti et al. 2007 ). phorylation site for cAMP-dependent protein In vitro, the CATA expression varied little with 10 Role of Enzymes and Proteins in Plant-Microbe Interaction: A Study of M. oryzae Versus Rice 141

time and H2 O2 concentration, whereas CATB a protein has to be newly expressed upon infection transcript abundance showed a moderate increase but not necessarily in all pathological conditions. with increasing H2 O2 concentration. In vivo also A unifying nomenclature for PRs was proposed there is upregulation of CATB gene at the time of based on their grouping into families sharing penetration of the host by M. oryzae . Skamnioti amino acid sequences, serological relationships, et al. (2007 ) showed that CATB plays a part in and enzymatic or biological activity (Van Loon strengthening the fungal cell wall and not in the et al. 1994 ; Van Loon and Van Strien 1999 ) detoxiﬁ cation of host-produced H2 O2 . Tanabe (Table 10.1 ). et al. (2011 ) in a gene knockout experiment The classiﬁ ed PR proteins are grouped into showed that CPXB is the major gene encoding two subclasses on the basis of acidic and basic the secretory catalase and confers resistance to subclass. The acidic subclass proteins are gener-

H2 O2 in M. oryzae hyphae. Their results suggest ally secreted to the extracellular spaces, and basic that CPXB plays a role in fungal defense against subclass proteins are transported to the vacuole

H 2 O2 accumulated in the epidermal cells of rice by C-terminal end signal sequence (Takeda et al. at the early stage of infection but not in pathoge- 1991 ; Koiwa et al. 1994 ; Sato et al. 1995 ). The nicity of M. oryzae (Tanabe et al. 2011 ). expression of basic PR proteins is constitutive and independent of pathogen infection in some organs like roots, seedling, and cultured cells Role of Proteins in Magnaporthe (Agrios 1997 ). There are two criteria on the basis oryzae-Rice Interaction of which new families are included in PR proteins. The fi rst is that the protein must be induced The small secreted proteins play an important by a pathogen in tissues that do not normally and decisive role in plant pathogenesis. Generally express it, and the second is that the induced these proteins are less than 200 amino acid resi- expression must occur in at least two different dues. These small secreted proteins from M. oryzae plant-pathogen combinations or expression in a into rice lead to disease symptom development. single plant-pathogen combination must be con- As a defense, the rice plant also produces profi rmed independently in different laboratories. teins against the fungus. Depending on their These are low molecular weight proteins function during the defense response, proteins (6–43KDa), stable at pH < 3, can be extracted can be grouped into three classes. The fi rst class biochemically, are thermostable and most impor- of proteins is structural proteins that participate tantly highly resistant to protease. Till now the in strengthening and repairing of the cell wall or presence of PR proteins is established in almost modifi cation of the properties of the extracellular all parts of plants like leaves, stems, roots, and matrix. The second class of proteins exhibits fl owers. Five to ten percent of total leaf proteins direct antimicrobial activities or catalyzes the account for PR proteins (Van Loon and Van synthesis of antimicrobial compounds (Lebeda Strien 1999 ). NMR reveals α-β-α sandwich struc- et al. 1999 ). The third class comprises of pro- ture which provides compactness to the struc- teins, which function in plant defense is not well ture of PR proteins and possibly helps in the known (Schoeltens et al. 1991 ) . resistance to protease (Fernandez et al. 1997 ). PR proteins have been well studied as a major defense response in several dicot plants, both in Pathogenesis-Related (PR) Proteins R gene-mediated resistance and in SAR. The roles of PR genes in disease resistance have Pathogenesis-related (PR) proteins are encoded been suggested by the tight correlation between by host plants in response to pathological or situ- expression levels of PR genes and disease resis- ations of nonpathogenic origin. Antoniw et al. tance and by the observation of enhanced disease (1980 ) coined the term “pathogenesis-related resistance in the transgenic plants overexpressing proteins” (PRs). To be included among the PRs, certain PR genes (Song and Goodman 2001 ). 142 J. Imam et al.

Table 10.1 Recommended classiﬁ cations and properties of families of Pathogenesis-Related Proteins (PRs ) (Sels et al. 2008 ) Proposed Typical microbial S.No. Family Type member size (KDa) Properties target Original reference 1 PR-1 Tobacco PR-1a 15 Antifungal Unknown Antoniw et al. (1980 ) 2 PR-2 Tobacco PR-2 30 β -1,3-Glucanase β-1,3-Glucan Antoniw et al. ( 1980 ) 3 PR-3 Tobacco P, Q 25–30 Chitinase (Class Chitin Van Loon (1982 ) I,II,IV,V,VI,VII) 4 PR-4 Tobacco R 15–20 Chitinase class I,II Chitin Van Loon (1982 ) 5 PR-5 Tobacco S 25 Thaumatin-like Membrane Van Loon (1982 ) 6 PR-6 Tomato inhibitor I 8 Proteinase-inhibitor _a Green and Ryan (1972 )

7 PR-7 Tomato P 69 75 Endoproteinase _a Vera and Conejero (1988 ) 8 PR-8 Cucumber 28 Chitinase class III Chitin Metraux et al. (1988 ) chitinase 9 PR-9 Lignin-forming 35 Peroxidase _a Lagrimini et al. (1987 ) peroxidase 10 PR-10 Parsley PR-1 17 ‘Ribonuclease-like’ _a Somssich et al. (1986 ) 11 PR-11 Tobacco class V 40 Chitinase class I Chitin Melchers et al. (1994 ) chitinase 12 PR-12 Radish Ps-AFP3 5 Defensin Membrane Terras et al. (1995 ) 13 PR-13 Arabidopsis 5 Thionin Membrane Epple et al. (1995 ) THI2.1 14 PR-14 Barley LTP4 9 Lipid-transfer protein Membrane Garcia-Olmedo et al. (1995 ) 15 PR-15 Barley OxOa 20 Oxalate oxidase _a Zhang et al. (1995 ) (germin) 16 PR-16 Barley OxOLP 20 ‘Oxalate _a Wei et al. (1998 ) oxidase-like’ 17 PR-17 Tobacco PRp27 27 Unknown _a Okushima et al. (2000 ) _a No in vitro antimicrobial activity reported

Some PR proteins have activities of hydrolytic Transgenic plants which constitutively expressed enzymes including chitinase and β-1,3-glucanase, a rice class I chitinase gene, Cht-2 or Cht-3 , which can hydrolyze major components of fun- showed signiﬁ cant resistance against two races of gal cell walls, chitin and β-1,3-glucan, respec- M. oryzae (Nishizawa et al. 1999 ). Interestingly, tively. Hydrolysis of these fungal cell wall hydrolytic enzymes of microbial origin have also constituents leads to the inhibition of the growth been demonstrated to be effective in engineering of several fungi in vitro (Punja 2006 ). Genes rice disease resistance against fungal pathogens. encoding chitinase or β-1,3-glucanase from rice Ninety percent of transgenic rice plants express- and microbes have been extensively used in gen- ing ChiC had higher resistance against M. oryzae eration of transgenic rice resistant to fungal than non-transgenic plants. Disease resistance in pathogens (Punja 2006 ). Transgenic plants con- the transgenic plants was correlated with the stitutively expressing the Gns1 gene, encoding a ChiC expression levels (Itoh et al. 2003 ). Three β-1,3-glucanase, accumulated Gns1 protein up to genes, ech42 , nag70 , and gluc78 , encoding 0.1 % of total soluble protein in leaves. The hydrolytic enzymes, from a biocontrol fungus Gns1 -overexpressing transgenic plants devel- Trichoderma atroviride, were introduced in single or oped many resistant-type lesions on the inocu- in combinations into rice. Gluc78 - overexpressing lated leaf, accompanying earlier activation of transgenic plants showed enhanced resistance to defense genes PR-1 and PBZ1 , when inoculated M. oryzae , while transgenic plants overexpressing with virulent M. oryzae (Nishizawa et al. 2003 ). the ech42 gene encoding for an endochitinase 10 Role of Enzymes and Proteins in Plant-Microbe Interaction: A Study of M. oryzae Versus Rice 143 increased resistance to Rhizoctonia solani , result- Acknowledgements This chapter is a part of NAIP-C4, ing in a reduction of 62 % in the sheath blight ICAR work. The authors thank the National Agriculture Innovative Project-C4 for supporting the work. The disease index (Liu et al. 2004 ; Shah et al. 2008 ). authors also express the sincere thanks to Miss Neha Nancy Toppo for her valuable inputs in the conclusion.

Conclusion References It can thus be inferred that enzymes and proteins play an integral role in the interaction of M. Agrios GN (1997) Plant pathology, 4th edn. Academic, Oryzae with rice and vice versa. A detailed study San Diego Antoniw JF, Ritter CE, Pierpoint WS, Van Loon LC of the interaction will further help us understand (1980) Comparison of three pathogenesis-related pro- the mechanism of fungal invasion and how plant teins from plants of two cultivars of tobacco infected defenses are activated in response to the attack with TMV. J Gen Virol 47:79–87 and give insight to subsequent changes in Apel-Birkhold PC, Walton JD (1996) Cloning, disruption, and expression of two endo-β-1,4Xylaanse genes, response in case of any deviation in the normal XYL2 and XYL3 from Cochliobolus carbonum . Appl mechanism/mode of fungal virulence. Research Environ Microbiol 62:4129–4135 till now has already elucidated on the fact that Arguelles JC (2000) Physiological roles of trehalose in infection initiation requires some prerequisites bacteria and yeasts: a comparative analysis. Arch Microbiol 174:217–224 (as in the case of cutinases) and that each enzyme Bechinger C, Giebel KF, Schnell M, Leiderer P, Deising is specifi c for a particular component of the plant HB, Bastmeyer M (1999) Optical measurements of with varying specifi city not only among the invasive forces exerted by appressoria of a plant patho- classes of enzyme but also between members of genic fungus. Science 285:1896–1899 Beg Q, Kapoor M, Mahajan L, Hoondal G (2001) the classes. There exist a lot of possibilities on Microbial xylanase and their industrial applications: a discovering the biochemical pathways and the review. Appl Microbiol Biotechnol 56:326–338 genes involved with the activity of each enzyme Belien T, Van Campenhout S, Robben J, Volckaert G and in turn comprehend the plant response. (2006) Microbial endoxylanases: effective weapons to breach the plant cell-wall barrier or, rather, triggers of Continued efforts in this research area will plant defense systems? Mol Plant Microbe Interact also help us understand better how the fungus 19:1072–1081 subverts the fi rst line of defense and further Biely P, Tenkanen M (1998) Enzymology of hemicellu- overcomes the different levels of defense. An loses degradation. In: Kubicek CP, Harman GE, Ondik KL (eds) Trichoderma and Gliocladium: basic biol- analytical approach to the interaction will also ogy, taxonomy and genetics. Taylor and Francis, throw light on the plant response elicited towards London, pp 25–47 pathogen invasion and why knowledge of the Carlile AJ, Bindschedler LV, Bailey AM, Bowyer P, plant defense mechanism is important, notwith- Clarkson JM, Cooper RM (2000) Characterization of SNP1, a cell wall-degrading trypsin, produced during standing the fungal defense evoked on plant infection by Stagonospora nodorum. Mol Plant response mechanisms. Microbe Interact 13:538–550 Resistant varieties as in the case of trans- Choi W, Dean R (1997) The adenylate cyclase gene genic rice or induction of overexpression of chi- MAC1 of Magnaporthe oryzae controls appressorium β formation and other aspects of growth and develop- tinases and -1,3-glucanases for increased plant ment. Plant Cell 9:1973–1983 defense in transgenic plants have already been Cooper R, Longman D, Campbell A, Henry M, Lees P developed. There still exists a lot of scope in the (1988) Enzymic adaptation of cereal pathogens to the vastly unexplored territory of transgenic plants. monocotyledonous primary wall. Physiol Mol Plant Pathol 32:33–47 One can also foray in the development of an Dean R et al (2005) The genome sequence of the rice blast antidote to the enzymes breaking the fi rst line of fungus Magnaporthe oryzae . Nature 434:980–986 defense and being responsible for invasion. The DeZwaan TM, Carroll AM, Valent B, Sweigard JA (1999) study of this plant-microbe interaction is thus of Magnaporthe oryzae Pth11p is a novel plasma membrane protein that mediates appressorium differentia- interest not only to the academician but also the tion in response to inductive substrate cues. Plant Cell researcher. 11:2013–2030 144 J. Imam et al.

Doke N (1983) Involvement of superoxide anion generation Kulkarni RD, Thon MR, Pan HQ, Dean RA (2005) Novel in the hypersensitive response of potato tuber tissues G-protein-coupled receptor-like proteins in the plant to infection with an incompatible race of Phytophthora pathogenic fungus Magnaporthe grisea. Genome infestans and to the hyphal wall components. Physiol Biol 6:R24. Plant Pathol 23:345–357 Lagrimini LM, Burkhart W, Moyer M, Rothstein M Edreva A (2005) Pathogenesis-related proteins: research (1987) Molecular cloning of complementary DNA progress in the last 15 years. Gen Appl Plant Physiol encoding the lignin-forming peroxidase from tobacco: 31:105–124 molecular analysis and tissue-specifi c expression. Epple P, Apel K, Bohlmann H (1995) An Arabidopsis Proc Natl Acad Sci USA 84:7542–7546 thaliana thionin gene is inducible via a signal trans- Lamb C, Dixon RA (1997) The oxidative burst in plant duction pathway different from that for pathogenesis- disease resistance. Annu Rev Plant Physiol Plant Mol related proteins. Plant Physiol 109:813–820 Biol 48:251–275 Fernandez C, Scyperski T, Bruyere T, Ramage P, Mosinger Lebeda A, Jancova D, Luhova L (1999) Enzymes in fun- E, Wutrich K (1997) NMR solution structure of the gal plant pathogenesis. Phyton (Austria) 39:51–56 pathogenesis-related protein p 14a. J Mol Biol Lee YH, Dean R (1994) Hydrophobicity of contact sur- 266:576–593 face induces appressorium formation in Magnaporthe Foster AJ, Jenkinson JM, Talboit NJ (2003) Trehalose oryzae . FEMS Microbiol Lett 115:71–76 synthesis and metabolism are required at different Legrand M, Kauffmann S, Geoffroy P, Fritig B (1987) stages of plant infection by Magnaporthe oryzae . Eur Biological function of pathogenesis-related proteins: Mol Biol Organ J 22:225–235 four tobacco pathogenesis related proteins are chitin- Garcia-Olmedo F, Molina A, Segura A, Moreno M (1995) ase. Proc Natl Acad Sci USA 84:6750–6754 The defensive role of nonspecifi c lipid-transfer pro- Lehtinen U (1993) Plant-cell wall degrading enzymes teins in plants. Trends Microbiol 3:72–74 of Septoria nodorum . Physiol Mol Plant Pathol Giesbert S, Lepping H, Tenberge K, Tudzynski P (1998) 43:121–134 The xylanolytic system of Claviceps purpurea : cyto- Lequeu J, Fauconnier ML, Chammai A, Bronner R, Blee E logical evidence for secretion of xylanases in infected (2003) Formation of plant cuticle: evidence for the occur- rye tissue and molecular characterization of two xyla- rence of the peroxygenase pathway. Plant J 36:155–164 nase genes. Phytopathology 88:1020–1030 Liu M, Sun ZX, Zhu J, Xu T, Harman GE, Lorito M Gilbert RD, Johnson AM, Dean RA (1996) Chemical sig- (2004) Enhancing rice resistance to fungal pathogens nal responsible for appressorium formation in the rice by transformation with cell wall degrading enzyme blast fungus Magnaporthe oryzae . Physiol Mol Plant genes from Trichoderma atroviride. J Zhejiang Univ Pathol 48:335–346 Sci 5:133–136 Gomez-Gomez E, Ruiz-Roldan MC, Di Pietro A, Roncero Melchers LS, Apotheker-de Groot M, Van der Knaap JA, MI, Hera C (2002) Role in pathogenesis of two endo- Ponstein AS, Sela-Buurlage MB, Bol JF, Cornelissen β-1,4-Xylanase genes from the vascular wilt fungus BJC, Van den Elzen PJM, Linthorst HJM (1994) A new Fusarium oxysporum . Fungal Genet Biol 35:213–222 class of tobacco chitinases homologous to bacterial exo- Green TR, Ryan CA (1972) Wound-induced proteinase chitinases displays antifungal activity. Plant J 5:469–480 inhibitor in plant leaves: a possible defense mecha- Metraux JP, Streit L, Staub T (1988) A pathogenesis- nism against insects. Science 175:776–777 related protein in cucumber is a chitinase. Physiol Mol Hatsch D, Phalip V, Petkovski E, Jeltsch JM (2006) Plant Pathol 33:1–9 Fusarium graminearum on plant cell wall: no fewer Nawrath C (2006) Unravelling the complex network of than 30 xylanase genes transcribed. Biochem Biophys cuticular structure and function. Curr Opin Plant Biol Res Commun 345:959–966 9:281–287 Howard RJ, Ferrari MA, Roach DH, Money NP (1991) Nishizawa Y, Nishio Z, Nakazono K, Soma M, Nakajima Penetration of hard substrates by a fungus employing E, Ugaki M, Hibi M (1999) Enhanced resistance to enormous turgor pressures. Proc Natl Acad Sci USA blast (Magnaporthe grisea ) in transgenic japonica rice 88:11281–11284 by constitutive expression of rice chitinase. Theor Ishii S (1998) Factors infl uencing protoplast viability of Appl Genet 99:383–390 suspension-cultured rice cells during isolation pro- Nishizawa Y, Saruta M, Nakazono K, Nishio Z, Soma M, cess. Plant Physiol 88:26–29 Yoshida T, Nakajima E, Hibi T (2003) Characterization Itoh Y, Takahashi K, Takizawa H, Nikaidou N, Tanaka H, of transgenic rice plants over-expressing the stress- Nishihashi H, Watanabe T, Nishizawa Y (2003) inducible beta-glucanase gene Gns1 . Plant Mol Biol Family 19 chitinase of Streptomyces griseus HUT6037 51:143–152 increases plant resistance to the fungal disease. Biosci Okushima Y, Koizumi N, Kusano T, Sano H (2000) Biotechnol Biochem 67:847–855 Secreted proteins of tobacco cultured BY2 cells: iden- Koiwa H, Sato F, Yamada Y (1994) Characterization of tifi cation of a new member of pathogenesis-related accumulation of tobacco PR-5 proteins by IEF- proteins. Plant Mol Biol 42:479–488 immunoblot analysis. Plant Cell Physiol 35:821–827 Punja ZK (2006) Recent developments toward achieving Kolattukudy PE (2001) Polyesters in higher plants. Adv fungal disease resistance in transgenic plants. Can J Biochem Eng Biotechnol 71:1–49 Plant Pathol 28:298–308 10 Role of Enzymes and Proteins in Plant-Microbe Interaction: A Study of M. oryzae Versus Rice 145

Ruiz MC, DiPietro A, Roncero MIG (1997) Purifi cation Sweigard JA, Chumley FG, Valent B (1992a) Disruption and characterization of an acidic endo-beta-1,4- of a Magnaporthe oryzae cutinase gene. Mol Gen xylanase from the tomato vascular pathogen Fusarium Genet 232:183–190 oxysporum f sp. lycopersici . FEMS Microbiol Lett Sweigard JA, Chumley FG, Valent B (1992b) Cloning and 148:75–82 analysis of Cut1, a cutinase gene from Magnaporthe Sasaki K, Iwai T, Hiraga S, Kuroda K, Seo S, oryzae . Mol Gen Genet 232:174–182 Mitsuhara I, Miyasaka A, Iwano M, Ito H, Matsui Takeda S, Sato F, Ida K, Yamada Y (1991) Nucleotide H, Ohashi Y (2004) Ten rice peroxidases redun- sequence of a cDNA for osmotin-like protein from dantly respond to multiple stresses including cultured tobacco cells. Plant Physiol 97:844–846 infection with rice blast fungus. Plant Cell Physiol Tanabe S, Ishii-Minami N, Saitoh KI, Otake Y, Kaku H, 45:1442–1452 Shibuya N, Nishizawa Y, Minami E (2011) The role of Sato F, Koiwa H, Sakai Y, Kato N, Yamada Y (1995) catalase-peroxidase secreted by Magnaporthe oryzae Synthesis and secretion of tobacco neutral PR-5 pro- during early infection of rice. Mol Plant Microbe tein by transgenic tobacco and yeast. Biochem Interact 24:163–171 Biophys Res Commum 211:909–913 Terras FR, Eggermont K, Kovaleva V, Raikhel NV, Schoeltens-Toma IMJ, Joosten MHA, De Wit PJGM Osborn RW, Kester A, Rees SB, Vanderleyden J, (1991) Appearance of pathogen-related proteins in Cammue BP, Broekaert WF (1995) Small cysteine- plant hosts: relationships between compatible and rich antifungal proteins from radish: their role in host incompatible interactions. In: Cole GT, Hoch HC defense. Plant Cell 7:573–588 (eds) The fungal spore and disease initiation in Van Loon LC (1982) Regulation of changes in proteins and plants and animals. Plenum Press, New York/ enzymes associated with active defense against virus London, pp 247–265 infection. In: Wood RKS (ed) Active defense mecha- Schweizer P, Jeanguenat A, Mdossinger E, Metraux JP nisms in plants. Plenum Press, New York, pp 247–273 (1994) Plant protection by free cutin monomers in two Van Loon LC, Van Strien EA (1999) The families of cereal pathosystems. In: Daniels MJ, Downie JA, pathogenesis-related proteins, their activities, and Osbourn AE (eds) Advances in molecular genetics of comparative analysis of PR-1 type proteins. Physiol plant-microbe interaction. Springer Publication, The Mol Plant Pathol 55:85–97 Netherlands, pp 371–374 Van Loon LC, Pierpoint WS, Boller T, Conejero V (1994) Sels J, Mathys J, De Coninck BM, Cammue BP, De Bolle Recommendations for naming plant pathogenesis- MF (2008) Plant pathogenesis-related (PR) proteins: related proteins. Plant Mol Biol Report 12:245–264 a focus on PR peptides. Plant Physiol Biochem Vera P, Conejero V (1988) Pathogenesis-related proteins 46:941–950 of tomato P-69 as an alkaline endoproteinase. Plant Shah JM, Raghupathy V, Veluthambi K (2008) Enhanced Physiol 87:58–63 sheath blight resistance in transgenic rice expressing Wegener S, Ransom RF, Walton JD (1999) A unique an endochitinase gene from Trichoderma virens . eukaryotic β-xylosidase gene from the phytopatho- Biotechnol Lett. doi: 10.1007/s10529-008-9856-5 genic fungus Cochliobolus carbonum . Microbiology Skamnioti P, Gurr SJ (2008) Cutinase and hydrophobin 145:1089–1095 interplay, a herald for pathogenesis? Plant Signal Wei Y, Zhang Z, Anderson CH, Schmelzer E, Gregersen Behav 3:248–250 PL, Collinge DB, Smedegaard-Petersen V, Thordal- Skamnioti P, Henderson C, Zhang Z, Robinson Z, Gurr SJ Christensen H (1998) An epidermis/papilla-specifi c (2007) A novel role of catalase B in the maintenance oxalate oxidase-like protein in the defence response of of fungal cell wall integrity during host invasion in the barley attacked by the powdery mildew fungus. Plant rice blast fungus Magnaporthe oryzae. Mol Plant Mol Biol 36:101–112 Microbe Interact 20:568–580 Wu SC, Ham KS, Darvill AG, Albersheim P (1997) Somssich IE, Schmelzer E, Bollmann J, Hahlbrock K Deletion of two endo-β-1,4-Xylanase genes reveals (1986) Rapid activation by fungal elicitor of genes additional isozymes secreted by the rice blast fungus. encoding “pathogenesis-related” proteins in cul- Mol Plant Microbe Interact 10:700–708 tured parsley cells. Proc Natl Acad Sci USA Wu SC, Helley JE, Luttig C, Fernekes LM, Gutierrez- 83:2427–2430 Sanchez G, Darvill AG, Albersheim P (2006) Song F, Goodman RM (2001) Molecular biology of Identifi cation of an endo - β-1,4- d -Xylanase from disease resistance in rice. Physiol Mol Plant pathol Magnaporthe oryzae by gene knockout analysis, 59:1–11 purifi cation, and heterologous expression. Am Soc Subramaniyan S, Prema P (2002) Biotechnology of Microbiol 72:986–993 microbial xylanases: enzymology, molecular biology, Zhang Z, Collinge DB, Thordal-Christensen H (1995)

and application. Crit Rev Biotechnol 22:33–64 Germin-like oxalate oxidase, a H2 O2 -producing Sumizu K, Yoshikawa M, Tanaka S (1961) Studies on xyla- enzyme, accumulates in barley attacked by the powdery nase of Pyricularia oryzae . J Biochem 50:538–543 mildew fungus. Plant J 8:139–145 Industrial Enzyme Applications in Bioreﬁ neries for Starchy 1 1 Materials

Vipul Gohel , Gang Duan , and Vimal Maisuria

Abstract This chapter reviews recent advances in technology developments in bioreﬁ nery industries through enzymatic approaches where various starchy materials have been used as feedstock for biofuel and various syrup productions. It further discusses the enzymes discovery, industrial challenges, and how enzymatic-based approaches help different industries to develop environmentally sustainable and cost-effective solutions by making industrial process into more simpliﬁ ed without compromising the product and by-product yields and their qualities.

Keywords Bioreﬁ nery • Enzymatic process • Ethanol • Speciality starch syrup • Corn wet milling • Starchy grain feedstock

Introduction policies/innovation/ﬁ les/lead-market-initiative/ bio_based_products_taksforce_report_en.pdf ). Recent biotechnological advances to enhance the Bio-based chemical products such as enzymes, manufacturing performance of bio-based chemical emulsiﬁ ers, and plastics provide an excellent products for a wide range of applications in many opportunity to reverse the trends through the industries have resulted in a surge of bioprospect- creation of a new generation of renewable, environ- ing approaches ( http://ec.europa.eu/enterprise/ mentally sustainable products (Chandel et al. 2007 ). (Advances in biotechnology have resulted in a revealing impact on bio-based industrial * V. Gohel ( ) • V. Maisuria enzymes.) A large number of commercial Genencor®, a Danisco Division, DuPont Industrial Biosciences, Danisco (India) Pvt. Ltd., enzymes used for various applications, from Plot No -46, Roz-Ka-Meo Industrial Area, Sohna, grain processing to the textile industry, are pro- Tehsil NUH , Sohna, Gurgaon 122 103 , India duced through large-scale microbial fermenta- e-mail: [email protected] tion process. Bio-based enzymes have found G. Duan extensive application in the food, feed and Genencor (China) Bio-Products Co. Ltd. , beverage, pharmaceutical, detergent, and textile 102, Mei Li Road , 214028 Wuxi , People’s Republic of China industries and in recent times also as analytical e-mail: [email protected] agents ( http://www.usda.gov/oce/reports/energy/

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 147 DOI 10.1007/978-81-322-1094-8_11, © Springer India 2013 148 V. Gohel et al.

BiobasedReport2008.pdf ). The largest market in 112062612.asp ). The major worldwide enzyme the enzyme sector, accounting for 59 % of sales, producers are Genencor (now a part of DuPont consists of industrial use enzymes, including Industrial Biosciences) and Novozymes A/S those used in the starch, food and animal feed, ( www.wiley-vch.de/books/biopoly/pdf_v07/ beverage, detergent, textile, leather, pulp, and vol07_04.pdf ) (Sutherland 2000 ). paper industries. The remaining 41 % of sales are Bio-based sustainable solutions are becoming accounted for by the personal care and pharma- important for food and energy security due to ceutical industries (Modilal et al. 2011 ). Food limited availability and increasing demand with enzymes, including enzymes that are employed ever-increasing population (Gohel et al. 2006 ). in the dairy, brewing, wine and juice, fats and oils, Grain processing is the biggest component in the and baking industries, account for the second organized food sector consuming over 40 % of largest segment with 17 % of the market share. the total value of all enzymes ( http://www.apind. Finally, feed enzymes that are used in animal feeds gov.in/Library/Note%20fp.pdf). Grain- processing account for approximately 10 % of the enzyme industries include milling of rice, wheat, maize, market share (Chandel et al. 2007 ; Maurer 2004 ; barley, millets, sorghum, fi nger millet, and pulses Gupta et al. 2002 ; Sivaramakrishnan et al. 2006 ; to grind them into fi ne fl our; malting by germi- Herrera 2004 ). nating seeds; and extracting soluble carbohy- Enzymes have signifi cant advantages over drates, proteins, vegetable oils, and fi bers for use chemical catalysts in that they are derived from in the food and livestock feed sectors ( http://www. natural resources (animal, plant, microbial) and ebrd.com/downloads/policies/environmental/ exhibit very high specificity under various grain.pdf ). At the same time, the demand for reaction conditions, such as pH, temperature, and these starchy feedstocks for ethanol production aqueous and nonaqueous environment. They are has increased as an alternative transportation easily biodegradable, thereby reducing the risk energy source and for use in recreational con- of environmental pollution and providing an sumption (Mussatto et al. 2010; Szulczyk et al. eco- friendly and sustainable solution for today’s 2010 ). To fulfi ll the competing demands of the industries in a variety of process applications food, feed, and energy sectors, the focus of the ( Buchholz et al. 2005 ; Aehle 2007; Polaina and starch-processing industries has been to develop MacCabe 2007 ; Olempska-Beer et al. 2006 ). effi cient processes to either maximize energy The industrial enzyme business is steadily (starch) availability in the grains using a biotech- growing worldwide due to enhanced production nological approach or improve starch utilization technologies, engineered enzyme with novel in value-added starch derivatives through sustain- properties, and discovery of new application able bio-based enzymatic solutions in order to fi elds. The global market for industrial enzymes, produce cheaper high-sweetening agents such which is estimated to be at about $3.3 billion as glucose, maltose, fructose, and specialty ( 14,904.4 crore), is attracting large investments syrups. Sweetener production is based on acid or throughout the world ( http://www.bccresearch. enzymatic hydrolysis of starch extracted mainly com/report/enzymes-industrial-applications- from corn through wet milling. The residual bio030f.html ; http://www.marketwire.com/press- cornstarch is used as feedstock for ethanol release/industrial-enzymes-market- estimated-at- production through yeast fermentation in which 33-billion-in-2010-1395348.htm). In India, a starch is converted into fermentable sugars using developing country, the bio-industrial market is industrial enzymes and the steep liquor generated estimated to be at about 625.94 crore in 2010– through wet milling as a fermentation booster. 2011, with a growth rate of 10.98 % in compari- Recent industrial trends show a shift to the pro- son to 2009–2010 ( 564 crore). This segment in duction of sweeteners and ethanol, both potable India is forecast to grow at a compounded annual and fuel ethanol from a variety of starch sources growth rate (CAGR) of 15 % until 2015 ( http:// such as rice, millet, sorghum, and wheat through biospectrumindia.ciol.com/content/BSTOP20/ dry-milling process. Currently, many different 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 149

Raw materials Grains (Corn, Wheat, Rice, Sorghum, Millet)

Physical separation Seeds Straw and husks

Decomposition Starch Gasification

Biotechnological Chemical Extrusion Cellulose Hemicellulose Lignin conversion conversion Grain Processing Glucose

Co-extrusion and Decomposition Glucose for Reduced plasticization of paste Fermentation Hydrogenation Esterification and amination Ether formation

Carboxymethyl Acetate starch Bio-plastic, Binders or starch Bio-products co- and mix- Adhesive or PHB Synags

polymerisate Sorbitol

Cement Methanol Ethanol Bio-isoprene n-butanol Glucosamine

Fig. 11.1 Grain-based biorefi nery processes and their bio-products starch sources have become key industrial raw conversion to ethanol via glucose fermentation materials apart from being the major ingredient in process (Fig. 11.1 ). the human diet over centuries as the major source With the rapid growth of biorefi neries, there is of daily caloric intake ( Tharanathan 2005 ). a pressing need for environmentally sustainable Ethanol from various sugar substrates such as and cost-effective enzyme-based solutions. A molasses and sugarcane juice has become a major plethora of technologies have been developed to biofuel, used to replace gasoline (Gough et al. 1997 ). screen and discover the potential robust enzymes India produces ethanol mostly from feedstock for application under stringent industrial condi- molasses, a byproduct of sugar manufacturing, tions. Technology has evolved from conventional unlike Brazil where ethanol is produced directly screening to the use of protein engineering and from sugarcane juice. The trend of replacing direct evaluation approaches. gasoline is expected to continue worldwide and increase at a rate almost one billion gallons annually (Gopinathan and Sudhakaran 2009 ). To Screening and Discovery maintain this accelerating momentum, continuous of Industrial Enzymes innovations in various technologies for ethanol production from different starch sources other To identify suitable enzymes for an application, than sugarcane juice and molasses are necessary, the traditional process has been to screen micro- in addition to bringing down enzyme cost with organisms either from naturally occurring envi- increasing production volumes. ronmental samples or from known cultures A whole-crop biorefi nery process has a unique ( Yeh et al. 2010 ; Warnecke and Hess 2009 ). With advantage, because it consumes the entire crop to recent advances in biotechnology, it is possible obtain useful bio-products. Several raw materials to isolate novel enzymes using bioprospecting such as wheat, rye, triticale, and maize can be uti- approaches. Bioprospecting, also known as lized as feedstock input in a whole-crop biorefi n- biodiversity prospecting, is the exploration of ery (Fernando et al. 2006 ). The process is initiated wild species of organisms for commercially by mechanical separation of biomass into various valuable biochemical and genetic resources components that are then treated separately. (Gohel et al. 2006 ). For example, seeds can be utilized directly after In general, bioprospecting is a search for grinding to meal or can be converted to starch, unique and robust bioactive compounds including followed by (i) extrusion, (ii) plasticization, (iii) novel enzymes existing in or produced by micro- chemical conversion, and (iv) biotechnological organisms, animal, and plant species found in 150 V. Gohel et al. extreme environments, such as hot springs, by the intrinsic instability of many enzymes rainforests, deserts, deep sea, and arctic regions particularly those with the complex structures (Gohel et al. 2006 ). The vast majority of micro- which challenge the production, processing, and bial species are yet to be explored for their genetic storage of such enzymes and lead to high produc- diversity, a key step for mining of bio-based tion costs (Iyer and Ananthanarayan 2008 ). industrial enzymes. Also recent biotechnology The activity and stability of the enzyme can be research has focused on enhancing enzyme modifi ed by chemical modifi cation, immobili- performance for a wide range of applications in zation, or use of a solvent (Marrs et al. 1999 ; many industries, concurrent with bioprospecting DeSantis and Jones 1999; Bull et al. 1999 ). approaches. Recently, enzyme engineering has yielded some Looking at the depth of microbial diversity, desired mutants using computer-aided molecular there is always a chance of fi nding microorgan- modeling and site-directed mutagenesis or with isms that produce novel enzymes with better directed (molecular) evolution techniques properties and suitable for commercial exploita- (Fig. 11.2 ). These techniques have helped in tion. The multitude of physicochemically diverse improving the catalytic effi ciency of certain habitats has challenged microbes to develop processes, reduce the formation of unusable appropriate molecular adaptations for survival by- products, or stabilize a biocatalyst or protein (Oberoi et al. 2001 ). Microbial diversity is a for prolonged activity under different process major resource for biotechnological products and conditions. Within a decade, directed evolution has processes (Bull et al. 1992 ). Natural biodiversity emerged as a standard methodology for protein among microbes is a vast but little-developed engineering, used either as a complementary resource for biotechnological innovation. The method or in combination with rational protein biosphere is dominated by microorganisms, design. Directed evolution technology for protein yet to date most microbes in nature have not designing has focused on meeting the demands been studied. This is mainly due to the fact that for industrially applicable biocatalysts with the historically the only way to reliably characterize desired level of chemoselectivity, regioselectivity, a microorganism was by isolation, purifi cation, and stereoselectivity, as well as the ability to per- and fermentation, to specify its biochemical and form under various process parameters (i.e., high physiological features on the “macroscopic” substrate concentrations, solvents, temperatures, level of a pure culture (Gupta et al. 2002 ). long-term stability) (Böttcher and Bornscheuer Many methodologies have been developed for 2010). This fascinating area of protein design will discovering enzymes including conventional no doubt be at the heart of future developments screening of environmental isolates and cultures, in the enzyme industry. genetic engineering of existing molecules, envi- In contrast, rational design generally requires ronmental gene screening or genomic database structural information regarding the enzyme and mining, protein engineering of an existing of the associations between sequence, structure, enzyme, gene shuffl ing, and metagenomics (Yeh and mechanism/function, a very information- et al. 2010 ; Martin et al. 2009 ; Warnecke and intensive effort. In past decades, it has been possible Hess 2009 ; Böttcher and Bornscheuer 2010 ). to predict how to increase enzyme activity, These search processes have greatly benefi tted substrate specifi city, and stability by employing enzyme manufacturing by providing innovative molecular modeling tools, even in the absence of bio-based sustainable solutions. Improvements structural data for an enzyme, using the structure in enzyme technology include reduction in of a homologous enzyme as a model in many manufacturing cost of industrial enzymes and cases. Amino acid substitutions are often selected their purifi cation, enzymes with high specifi city by comparisons of homologous sequences, (molecular and chiral), high turnover number, very depending on the purpose of the mutagenesis. high activity under mild conditions, and biode- However, the resulting molecules have to be gradability. These advantages are offset, however, carefully evaluated, because minor sequence 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 151

Fig. 11.2 An approach toward improvement of biocatalyst

changes by a single point mutation may cause the robust process. The next following section signiﬁ cant structural disturbance, resulting in explains how these new enzymes are applied in collateral negative mutations (Fig. 11.2 ). different industries to enable sustainable solutions Therefore, comparisons of the three-dimensional in the VUCA (volatility, uncertainty, complexity, structures of mutant and wild-type enzymes are and ambiguity) world. essential to verify that a single mutation was really site-directed. Genetic decoding of the enzyme(s) of interest, documenting a suitable Industrial Enzymes’ Application (usually microbial) expression system, and devel- in Ethanol Industries for Starchy oping a sensitive detection system are prerequi- Materials sites for both direct evolution and rational protein design (Bornscheuer and Pohl 2001 ). With these Worldwide, ethanol producers use different discovery and screening methods, enzymes can starchy grains as feedstock, such as rice, millet, be developed in short periods of time to meet the corn, sorghum, sweet sorghum, wheat, potato, needs of biotechnological industries in creating sweet potato, cassava, rye, triticale, barley, and 152 V. Gohel et al.

Fig. 11.3 Conventional and nonconventional processes for ethanol production using starchy materials

tapioca, for potable as well as fuel ethanol (Gohel process and no-cook process of ethanol production. and Duan 2012a ). The type of grain is selected by The low-temperature process works without jet industry based on local availability and its price cooking. In the no-cook process (Gohel and Duan (Kim and Dale 2004 ; Gopinathan and Sudhakaran 2012b), the liquefaction and saccharifi cation 2009 ). Industrial enzymes used to convert processes take place simultaneously, and it does starches into fermentable sugars include amy- not involve cooking of the starchy materials. The lases, glucoamylases, pullulanases, and proteases no-cook process eliminates steam consumption (Gohel and Duan 2012a ), mostly sourced from entirely, while the low- temperature process saves bacteria, fungi, and plants. about 50 % steam compared to conventional Most ethanol producers follow the conven- ethanol production. tional process for conversion of grain starches Today, most fuel ethanol is produced from into fermentable sugars and ethanol (Suresh et al. corn using either the dry-grinding (67 %) or 1999 ; Gibreel et al. 2009 ; Nikolić et al. 2010 ; the wet-milling (33 %) process (Bothast and Linko et al. 1983 ). The conventional process con- Schlicher 2005 ). The key distinction between sists of either of a four-step process with milling wet-mill and dry-grind facilities is the focus (wet or dry), liquefaction, saccharifi cation, and of resource use. In a dry-grind plant, the busi- fermentation (Fig. 11.3 ) or a three-step process ness focus is on maximizing the return on with the saccharifi cation step omitted by carrying capital exclusively based on the number of out simultaneous saccharifi cation and fermentation gallons of ethanol produced. In a wet-mill plant, in one tank (Fig. 11.3 ). Due to recent developments capital investments allow for the separation of in enzyme technology, higher temperatures are other valuable grain components in the grain no longer required to break down starch. These before fermentation to ethanol (Bothast and technologies are known as low-temperature Schlicher 2005 ). 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 153

Wet-Milling Process: Mainly Corn uses steeping at different temperatures ranging from ambient temperature to 50 °C to sustain Wet milling is a complex, capital- and energy- 45–50 % moisture in the corn grains. Steeping is intensive industrial process by which starch- a diffusion-driven process. During steeping, containing grain is hydrated and separated into chemicals and water diffuse into the corn kernel starch, fi ber, germ, and proteins or gluten. Corn is through the grain tip and cap and cross the tube the grain mainly used in the wet-milling process. cells of the pericarp into the kernel crown and The germ is separated from the kernel, and corn fi nally into the endosperm. Apart from softening oil extracted from the germ. The remaining germ the corn, SO 2 also plays a vital role in releasing meal is added to the fi ber and hull which results starch from the endosperm, breaking down the in corn gluten feed. Gluten is separated to produce disulfi de bonds holding the protein matrix that corn gluten meal, a high-protein animal feed. surrounds the starch particles. This increases the The residual starch is subjected to wet milling starch yield. Reducing steeping process time helps to produce ethanol, involving preparation of the manufacturer to increase plant capacity and starch solution, followed by fermentation of the reduce energy and SO 2 consumption associated fermentable sugars into ethanol. Wet-mill facilities with prolonged steeping time; these are the three are true “biorefi neries,” producing a number of leading challenges for wet millers to sustain high-value products (Shapouri et al. 2004 ; profi t margins (Johnston and Singh 2004 ). Bothast and Schlicher 2005 ). Many chemical approaches have been devel- These wet millers are engaged in producing a oped for reducing the steeping time, all of which variety of starch-based sweeteners to maximize either require major capital investments for existing profi tability apart from adding to their profi t process modifi cation or require pretreatment of margins through sale of by-products. However, corn kernel which increases energy consumption due to massive capital investments and operating and environmental safety concerns due to increasing costs of wet milling, new manufacturing units are pollution. To address these challenges, Johnston mainly based on dry-milling process. Recently, a and Singh ( 2004 ) demonstrated an enzymatic number of enzyme-based innovative technologies process in which maize kernels were soaked in have helped simplify the wet-milling process water with SO2 in the steeping tank for about (Singh and Johnston 2002 ; Ramírez et al. 2009 ). 8–10 h so that the germ gets fully hydrated and tensile enough to resist breakdown when the corn Enzymes in the Corn-Steeping Process is coarsely grounded and treated with acid prote- In the wet-milling process, steeping of corn is the ase in separate reactor. This process removes the most time-consuming step. The toxicity of sulfur diffusion barriers for protease penetration into the dioxide (SO2 ) causes an additional problem, as it corn endosperm to break down the protein sub- poses an environmental and health concern, and strate. After enzymatic treatment, the corn was there is growing pressure from environmental milled using the conventional wet- milling pro- agencies to find alternatives to SO 2 -based cess. Degermination milling was proposed by wet- milling processes (Johnston and Singh 2001 ; Johnston and Singh (2004 ) using coarse grinding Singh and Johnston 2002 ). It requires at least (also called fi rst grinding) with a larger gap

48–52 h of incubation of corn with SO2 (about between the grinding stones than what is currently 2,000–2,500 ppm) in the presence of 0.5–2 % used by wet millers, in order to signifi cantly lactic acid to soften the hard corn kernels which improve germ recovery. Apart from proteases, then swell up. This lactic acid is usually produced Johnston and Singh (2004 ) also studied various by bacteria belonging to the genus Lactobacillus carbohydrate enzymes including β-glucanase, during steeping. The steeping process is some- cellulase, and xylanase in the steeping process, times prolonged to 72 h because of poor corn which were found to produce signifi cantly lower quality or improper soaking with SO 2 . Wet milling starch yields compared to pepsin, acid protease, 154 V. Gohel et al. and bromelain. An increase in incubation time density differences (1.5 g/cm3 for starch vs. 1.1 g/cm3 with the three proteases resulted in reduction of for gluten particles) (Singh and Johnston 2004 ). total fi ber content. These proteases not only Primary centrifuges consist of a rotating bowl reduced the stepping process time but also reduced in which a stack of conical discs are separated by the residual protein content in the fi nal starch frac- a distance of 0.4–1.0 mm, depending on the den- tion. The benefi ts of this enzymatic wet-milling sity of particles to be separated. On the periphery process included a drastic reduction or elimina- of the rotating bowl, there are 6–12 nozzles. tion in the use of SO2 during steeping and also The mill starch enters the rotating bowl from reduced the steeping time by 70 % compared to the top or the bottom. Due to centrifugal force, the the conventional corn wet- milling steeping pro- heavier starch particles are forced toward the cess (Johnston and Singh 2001 , 2004 ; Ramírez periphery of the rotating bowl and exit through et al. 2009 ; Steinke and Johnson 1991 ). DuPont the nozzles as underfl ow. The starch slurry has a product for enzymatic steeping in the coming out from the primary separator has a wet-milling process, which it has marketed as protein content of 2–4 % and a specifi c gravity of PROSTEEP™ , a fungal protease. 1.160–1.198 (35–42 % dry solids). Lighter gluten Maize comprises phytate, which to a large particles move up between the discs and exit extent ends up in the corn steep liquor and consti- out as overfl ow. Gluten slurry from the primary tutes an undesirable component. Corn kernel is centrifuges comes out at a concentration of treated with phytase under steeping conditions in 15–30 g/L (2–4 oz/gal) and contains about the presence of SO2 to eliminate or greatly reduce 68–75 % protein (db). Routine maintenance of the phytate content in corn steep liquor. This further centrifuges is required to optimize performance helps in reducing the steeping time and also by (Blanchard 1992 ; Singh and Johnston 2004 ). facilitating the separation of starch from fi ber and The gluten slurry is concentrated from gluten, resulting in higher starch and gluten 15–30 g/L (2–4 oz/gal) to 150–165 g/L (20–22 oz/gal) yields as well as lower energy consumption by using another nozzle-bowl gluten thickener (Shetty et al. 2010 ). (GT) (Blanchard 1992 ). In this process, cellulase and hemicellulase enzymes are added to acceler- Enzymes in Gluten Filtration ate the dewatering of gluten with the prevailing In conventional wet-milling process, gluten process of rotary vacuum iterations (Fig. 11.4 ). dewatering and drying of corn consumes almost These enzymes are marketed by DuPont as 26 % of the total energy, next only to the starch- OPTIFLOW® RC 2.0, produced by controlled drying process (32 %) ( http://www.energystar. fermentation of Trichoderma reesei . The enzymes gov/ia/business/industry/LBNL-52307.pdf). hydrolyze the thin plate material and fi ne fi ber, To reduce this energy consumption and improve resulting in lower gluten cake moisture and fi ltration effi ciency, DuPont Genencor Science improved dewatering characteristics which has powered the production of cellulase and result in reduced drying energy cost by 5–25 %. hemicellulase enzymes. After germ and fi ber This process also increases the gluten fi ltration separation, the next stage is to separate gluten capacity from 20 % to 25 %, reduces fi ltration and starch with 5–6 % protein. This mixture is cloth wash, and decreases gluten cake recycle. known as mill starch, which is processed through These technologies, however, are sensitive to a de-gritting cyclone to remove any foreign multiple factors such as corn quality, upstream particles or sand to prevent damage or blocking processing, fi lter dryer operations, and design. of centrifuge nozzles. After de-gritting, the mill The cornstarch separated through the wet- starch having 6–8°Be (Baumé) is concentrated to milling process is used either for producing specifi c Be (Baumé) of 10–12° (Blanchard 1992 ). starch derivatives, sweeteners, or ethanol/biofuel The concentrated starch is passed through by enzymatic liquefaction and saccharifi cation, primary separators, a kind of centrifuge systems followed by yeast fermentation in case of ethanol/ to separate starch and gluten factions based on biofuel production. 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 155

Fig. 11.4 Enzymatic gluten separation processes used in wheat starch industries

Enzymes to Separate Gluten from Wheat Starch

Manufacture of starch and gluten from wheat is typically done in a dry-milling operation (Sugden 1997 ). The disintegration of milled wheat particles into A- or B-starch fractions and gluten, as well as their separation by hydrocyclones and decanters, is affected by the interaction with non-starch polysaccharides like arabinoxylans and β-glucans (Sayaslan 2004 ). Multicomponent blends of endo-β-1,4-glucanases and endo-β-1,4- xylanases derived from controlled fermentation of Trichoderma reesei (SPEZYME® CP and GC 220) improve the starch yield and gluten purity (Fig. 11.5 ). The enzymes are added at concentrations of 0.1–0.3 kg per ton of milled wheat. The enzymes used should be free from α-amylase or protease activity. The level of exo-beta- xylosidase should be low to limit the formation of monosaccharides, which might have a negative effect due to the reactivity of xylose. A novel enzyme preparation (GC 220) has been developed which exhibits high endo-β-glucanase and Fig. 11.5 Industrial enzymes used in corn gluten ﬁ ltration endo-xylanase activity but is virtually free of processes 156 V. Gohel et al. exo-xylosidase. These enzyme blends lower the viscosity of the ﬂ our suspension by hydrolysis of β-glucans in the wheat cell walls and by cleavage of soluble arabinoxylans. Viscosity measurements and microscopic evaluations of stained cell walls can be used to demonstrate the mode of action of these hydrolases. The impact of added enzymes on starch yield and gluten quality will be discussed in detail. In wheat starch processing, utilization of xylanase results in reduced viscosity of slurry along with improved starch-gluten separation and higher relative yields compared to the conventional process (Christophersen et al. 1997 ). Similar results have been reported in pilot plant experiments with hemicellulases (Weegels et al. 1992 ).

Enzyme for Cassava Starch Separation

Cassava ( Manihot esculenta Crantz) is a peren- nial tuber plant widely grown in many tropical countries including Nigeria and is one of the most important commercial crops. It is the third largest source of carbohydrates in the tropics after rice and corn and a staple for over 600 million people (FAO 2002 ). There are sweet and bitter cassavas. Sweet cassavas are normally used for human consumption. Bitter cassavas with higher starch content are used as animal feed or processed for industrial use (Vessia 2007 ). Dzogbefi a et al. (1999 ) used pectinase enzyme Fig. 11.6 Enzymatic processes for separation and processing from S. cerevisiae ATCC52712 for extraction of of cassava starch starch from cassava. The extraction of starch from cassava involves the combination of mechanical rasping and the et al. 2000; Daiuto et al. 2005). In a modifi ed use of hydrolytic enzymes such as pectinases that method, the combined application of pectinase, disintegrates the pulp releasing the starch. Usage xylanase, cellulase, and protease manufactured of enzyme saves the energy costs involved in by DuPont increased cassava starch extraction by starch extraction by alleviating the intensity of about 11 % (Fig. 11.6 ). In addition, more water mechanical rasping (Rahman and Rakshit 2003 ). was released from the mash after enzyme treat- The pectolytic enzymes break down the pectate- ment, which resulted in lower water content of composed cell wall leading to release of starch the pulp and fi ber residue after separating starch. granules. Therefore, the pulp and fi ber could be dried Commercial pectases have been documented easily while also reducing water consumption to enhance cassava starch extraction (Sriroth (Duan et al. 2010b ). 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 157

Dry Milling of Starchy Materials After dry milling, the grain fl our is subjected to liquefaction, saccharifi cation, series of fi ltrations Dry milling is cost-effective and also less capital (charcoal treatment followed by ion- exchange intensive. Due to this, most of the newly built process), and evaporation to concentrate the ethanol manufacturing units are coming up with syrup to more than 60 % Bx. (Brix). For ethanol dry-milling technology ( https://engineering.pur- production the grain fl our is subjected to lique- due.edu/~lorre/16/Midwest%20Consortium/ faction followed by sequential saccharifi cation DM%20DescManual%2042006-1.pdf). Starchy and fermentation, or simultaneous saccharifi cation grains such as rice, millet, corn, sorghum, and fermentation (SSF). cassava, and wheat are the most commonly used grains worldwide due to their ample availability Liquefaction and relatively low price (Gohel and Duan 2012a ). Liquefaction is the fi rst and most important step Milling is done to reduce grain kernel to small in starch processing for ethanol production. particle sizes for distribution in solutions. Two types of milling devices are used by industry, Conventional Liquefaction Process roller and hammer mills (Baron et al. 2005 ). The liquefaction process is variable, usually Roller mills are used in many potable ethanol processed in three to four tanks in the ethanol plants because of the use of grains with high husk industries, involving a slurry tank (55–65 °C), content, e.g., rye. Hammer mills are most com- primary liquefaction (85–95 °C), followed by jet monly used by industries producing non-potable cooking (105–120 °C), and secondary lique- ethanol. The grain is fed into a chamber with faction (85–98 °C). Alternatively, a three-step hammers rotating at high speed to grind the grain. process involves the use of slurry tank (55–65 °C), Screens between 6/64 and 3/16 in. are used to jet cooking (105–120 °C), and liquefaction control particle size. (85–98 °C) (Fig. 11.7 ). In the starch-processing Milling is an essential step for effi cient industries, liquefaction is carried out in a three- hydrolysis by enzymes since it directly impacts step process, slurry tank process (40–45 °C), jet the physicochemical reaction between insolu- cooking (105–108 °C), and dextrinization process ble grain fl our and enzyme. Large particle size (92–98 °C) (Fig. 11.8 ) (Gohel and Duan 2012b ). impedes gelatinization, resulting in lower In the slurry tank, vigorous agitation is enzyme effi ciency with corresponding loss in required to keep the mash mixed at pH 5.5–5.8 at yield up to 10 %. Hence, particle size should be a temperature of 60–65 °C for a holding period of such that 92–94 % of the fl our passes through 20–30 mins to produce 4–6 DE mash at tank out- US standard 20-mesh sieves, while very fi nely let. DE is equal to the percentage of bonds hydro- ground grain particles (<200 mesh) may lyzed in the starch. A DE of 100 indicates that all increase yield due to effi cient enzymatic action bonds between the glucosyl units have been on the substrate. However, it may hamper mixing hydrolyzed (Rong et al. 2009 ; Chaplin and Bucke of the fl our with water and cause problems in 1990). In jet cooking, the mash is heated by centrifugation, evaporation, and drying, with the mixing with high-pressure steam in a jet, to result of reduced DDGS yield (distilled dry gelatinize the starch and prepare it for easier grains with solubles), a by-product of dry millers. enzymatic breakdown. It is then agitated under Additionally, the milling process should have pressure for several minutes before fl ash cooling. temperature controls to limit the milled fl our Although some of the thermostable α-amylase temperature to 40 °C to avoid the formation of activity is reduced during this stage, it remains nonconvertible starch. To prevent the need for adequately effective in reducing the viscosity at aggressive dry milling, “DuPont Genencor Science” the high temperature. Many factors impact the jet has powered the development of a fungal protease, cooking process including temperature, pH, mash which supports by mashing the grain under the solids, retention time, and shear (back pressure, hydrolytic conditions. breaking up of granules). The initial and fi nal 158 V. Gohel et al.

Fig. 11.7 Industrial enzymatic processes for ethanol production

Temperature 42°C ;pH 5.5- 6.0; DS : 32-34%

Holding coils Flash Tank

Jet cooker Temperature 105 - 108°C

Dextrinization columns

DE 10 15%

Tank Saccharification Tank: Temperature: 60-63°C

PHE

Speciality syrup and powder

Fig. 11.8 Industrial enzyme-based carbohydrate processing for glucose syrup production liquefaction results in partial and complete glucose chains called dextrins and generating a solubilization of starch, respectively, by hydro- fi nal liquefact DE of 12–14 % with no precipitate lyzing the starch into low-molecular-weight in the mash coolers. DE is dextrose equivalent, a 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 159 measure of the reducing sugars in solution. Hence, In industry, a jet cooker is used to gelatinize starch it also measures the degree of starch hydrolysis by mixing starch slurry with steam under pressure (Rong et al. 2009 ; Chaplin and Bucke 1990 ). (Van der Maarel et al. 2002 ). The liquefaction process for about 1.5–2 h The thermostable α-amylase, an endo-acting results in complete gelatinization and dextriniza- enzyme, acts faster on starch granules during tion of pure starch or raw starchy grain substrate. swelling and randomly hydrolyzes α-1, A shortened liquefaction process time (0.5 or 1 h) 4- glucosidic bonds to reduce the peak viscosity produces ineffi cient hydrolysis into glucose of gelatinized starch, producing soluble dextrins polymers ( Montesinos and Navarro 2000; Neves and oligosaccharides (Tester et al. 2004). et al. 2006 ). An increase in the dosage of α-amylase Thermostable α-amylase also plays a key role as reduces the liquefaction process time but may a thinning agent (Aiyer 2005 ). After gelatiniza- not produce the proper substrate for saccharifi - tion, partial hydrolysis to produce dextrin has cation required for the production of various to be sustained without interruption by ensuring sweeteners or for ethanol production in which that the slurry does not cool down to avoid the glucoamylase and pullulanase enzymes produce formation of crystalline-stable starch molecules fermentable sugars in yeast fermentation. This is which resist further degradation by α-amylase. because α-amylase produces shorter 1-6 linkage The formation of such crystallized starch is called branches of glucose polymers which reduce retrograding (Aiyer 2005 ; Englyst et al. 1992 ; the yield of the desired end products (Gohel and Zeleznak and Hoseney 1986 ). For pure starch liq- Duan 2012a ; Roy and Gupta 2004 ; Kłosowski uefaction after gelatinization, the dextrinization et al. 2010 ). process temperature should be between 90 °C The key functional characteristics of starch and 98 °C in a closed dextrinization tank after that are critical for the biorefi nery industry are passing through the fl ash tank. granular swelling, gelatinization, and pasting. Granular swelling facilitates starch liquefaction, Higher Dry Solid (DS) Liquefaction which should be followed by gelatinization and Utility consumption involves energy, electricity, dextrinization. Gelatinization causes the intermo- water, and temperature changes, i.e., cooling and lecular bonds of starch molecules to hydrolyze in heating. Thus, water and energy (steam and the presence of water and heat. This irreversibly cooling generated with water) are highly used dissolves the starch granule. Penetration of water commodities in the processing industries. Water increases randomness in the general granule struc- scarcity and environmental regulations on water ture and decreases the number and size of crystalline effl uents are a major concern currently in these regions. Crystalline regions do not allow water industries (Mann and Liu 1999 ; Karuppiah entry. Heat causes such regions to become diffuse, et al. 2008 ). To become more sustainable and so that the chains begin to separate into an amor- cost-effective, industries constantly look for phous form (Jenkins and Donald 1998 ). ways to control the use of these inputs. Due to During gelatinization, water acts as a plasticizer. these limitations, industries are pushing their Water is fi rst absorbed in the amorphous space of plants to run on higher DS (dry solids) from starch leading to swelling during heating, and the 15–20 % to 25–35 % to save water and save up to water is then transmitted through connecting mol- 40–50 % energy consumption. In this process, ecules to crystalline regions (Cauvain and Young the maximum backset or thin slop is recycled 2001). Water enters tightly bound amorphous through liquefactions (Pohit et al. 2009 ; regions of double helical structures to swell the Grafelman and Meagher 1995; Gohel and amylopectin, thus causing crystalline structures to Duan 2012a). The use of concentrated dry sol- melt and break free. Stress caused by this swelling ids requires appropriate processes for plant phenomenon eventually interrupts the structure optimization depending on the feedstock. For and allows the leaching of amylose molecules example, whole wheat grain feedstock contains into the surrounding water (Englyst et al. 1992 ). fi bers, hemicellulase, and pentosans other than 160 V. Gohel et al. starch, increasing the dry solid concentration which reducing negative impact of tannin in the grain-based increases viscosity. Thermostable α-amylase ethanol process (Gohel and Duan 2012b ). alone cannot reduce the viscosity generated DuPont-Genencor Science has developed a through non-starch polysaccharides. To reduce thermostable phytase enzyme (marketed as GC such viscosity, it requires a mixture of cellulase, 980) to be used with α-amylase to enhance its hemicellulase, and xylanase added with thermostability during liquefaction of grains α-amylase before or during liquefaction (Crabb containing phytic content, such as sorghum, rice, and Shetty 1999 ). millet, and corn. Other thermostable α-amylases In industrial processes, the concentration of produced from different genetically modifi ed dry solids in the starch suspension subjected to Bacillus licheniformis are marketed as liquefaction is usually 25–35 %, which results in SPEZYME® XTRA, SPEZYME® FRED, very high viscosity following gelatinization SPEZYME® ALPHA, and CLEARFLOW® AA, (Aiyer 2005; De Cordt et al. 1994). Many ther- which function within the pH range of 5.0–7.0 mostable α-amylases are commercially produced and at temperatures above 85 °C. Liu et al. (2008 ) from different microorganisms (Klibanov 1983 ). have reported that α-amylase is sensitive to acidic The effi cacy of α-amylase is measured by the medium and loses its hydrolytic activity. speed and uniformity with which it reduces the α-Amylase functions optimally at 90 °C and pH peak viscosity, when used at a suitable concentra- 6.0 (Liu et al. 2008 ). Previously, the liquefaction in the slurry at 105–107 °C. This translates tion step in cornstarch hydrolysis was performed into fewer problems in the jet cooker or lique- at 85 °C and pH 6.0 (Mojovic ́ et al. 2006 ). faction tank for mixing slurry and, in turn, lowers Optimum α-amylase action to produce reducing power consumption and more rapid liquefaction. sugar in continuous enzymatic hydrolysis was Use of split dosages of α-amylase is a common obtained at pH 6.0 and 30 °C. The amount of practice among ethanol producers to generate reducing sugar produced from sago starch was adequate fermentable sugars for yeast propagation 0.464 g/L. It is documented that 5.9 % (w/v) in the SSF process to keep up a steady production fermentable sugar is produced from 25 % DS of ethanol. Many ethanol producers also use an corn with RSH (raw starch hydrolyzing) added at online dosing pump or periodic enzyme feeding 2.5 kg per MT of corn during liquefaction at in the continuous liquefaction process. 48 °C for 2 h, while 3.70 % (w/v) fermentable The thermostability of α-amylase varies with sugar produced from 30 % DS cassava starch the starchy substrate used and the quality of feed- treated with SPEZYME® XTRA at 90 °C applied stock used in liquefaction. Reduced thermo- at 0.66 kg per MT of starch (Shanavas et al. stability of α-amylase in prolonged liquefaction 2011 ). After liquefaction, SPEZYME® ALPHA may result in poor starch digestion and therefore treated 30 % and 35 % Indian broken rice pro- interfere with fi ltration of the resultant liquefact duced about 4–5 % and 6–7 % fermentable sugars, due to turbidity (Rosendal et al. 1979 ). To over- respectively. The same enzyme in the presence of come this, researchers have identifi ed many factors 30 % and 35 % pearl millet produced 3–4 % and that reduce the digestibility of grain protein and 5–6 % fermentable sugars, respectively (Gohel starches. Grain containing phytic acid and tannin and Duan 2012a ). reduce the thermostability of α-amylase enzyme, resulting in reduced starch digestibility under Liquefaction Without Jet Cooking high-temperature liquefaction (due to reacting Steam generation is expensive due to limited with proteins, including hydrolyzing enzymes). availability of fossil fuel (Mussatto et al. 2010 ; However, at lower temperatures, the impact of Szulczyk et al. 2010 ). Hence, many ethanol pro- tannin can be minimized (Cawley and Mitchell ducers have already started looking at alterna- 1968; Yan et al. 2009 ). DuPont Genencor Science tives to reduce or omit the jet cooker step which has pioneered the development of low-temperature cooks the starchy materials at high temperature and no-cook processes, which are effective in and pressure. For this, manufacturers use 30–40 % 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 161 higher doses of α-amylase in liquefaction with DuPont has commercialized glucoamylases the rest of the process remaining the same that are marketed as GA-L NEW and (Fig. 11.7 ). This process reduces about 30–50 % DISTILLASE® ASP. Both the products are effec- of steam consumption. However, research is con- tive in sequential saccharification and fermen- tinuing to develop an effi cient α-amylase with tation, as well as in the SSF process. GA-L NEW persistent activity while cooking at or above is intended for glucose production from liquefi ed gelatinization temperature. starch using controlled fermentation with a selected strain of Aspergillus niger. GA-L NEW Saccharifi cation and Fermentation: performs best at a pH of 4.0–5.0 and a tempera- Sequential or Concurrent ture of 55–60 °C. After liquefaction, the liquefact is subjected to DISTILLASE® ASP is a blend of enzymes, saccharifi cation at 55–65 °C (Figs. 11.3 and predominantly containing glucoamylase supple- 11.7 ). Saccharifi cation is a widely used process mented with pullulanase and protease produced in the production of almost all sweeteners and of by controlled fermentation of genetically ethanol. Saccharifi cation yields about 96 % modifi ed strains of Bacillus licheniformis and glucose and 4 % by-product from the starch Trichoderma reesei , respectively. This enzyme substrate. Saccharifi cation of cornstarch is reported blend was designed for ethanol production to as being widely performed at 55–60 °C and pH perform under conditions of variable pH during 5.0 (Mojovic ́ et al. 2006 ), although optimum sac- saccharifi cation. This enables the complete charifi cation is documented to take place at 60 °C process (the liquefaction and saccharifi cation and pH 4.5 ( http://umpir.ump.edu.my/863/1/ and SSF) to be independent of the pH conditions, Siti_Nor_Shadila_Alias.pdf). In another study, which saves chemicals, less uncertainty about however, Aggarwal et al. (2001 ) found that high product yield, and less acidity in thin stillage temperatures retard saccharifi cation and that the water. The process also ensures additional nutrients optimum conditions were 45 °C and pH of 5.0. to yeasts having a subsidiary protease activity to Recent technological improvements have produce FAN (free amino nitrogen) (Duan et al. eliminated one enzymatic step, the separate sac- 2011a ). When used in the sequential saccharifi ca- charifi cation for ethanol production. Elimination tion/fermentation process, DISTILLASE® ASP of this step avoids high osmolarity stress in the is thermostable up to 70 °C and active in a broader initial stage of yeast fermentation and reduces the pH range, 4.0–5.5. Due to these unique advantages, risk of contamination during the fermentation it also supports the goal of reducing microbial process (Nikolić et al. 2010 ; Grafelman and contamination. It is reported that pullulanase Meagher 1995 ). This process is also known as activity enhances the breakdown of long-chain simultaneous saccharifi cation and fermentation branches of α-1,6-glycoside linkages to produce (SSF) (Figs. 11.3 and 11.7 ). In SSF, the sacchari- linear dextrins during SSF, which in turn enhances fying enzymes hydrolyze the liquefi ed starch into fermentable sugar production by the action of fermentable sugars, while concurrently yeast glucoamylase. The net result is a shortening of fermentation is used to ferment the sugars into hydrolysis time by as much as 37 % (Gantelet ethanol. The ethanol industries mainly use and Duchiron 1999 ). commercially available glucoamylase, acid fungal These industries are highly water and energy α-amylase, and pullulanase as the saccharifi cation intensive. Increasing prices of crude oil and other enzymes. Glucoamylase is an exo-acting enzyme fossil fuels have stoked worldwide interest in that hydrolyzes starch liquefact at the nonreducing alternative fuel sources (Mann and Liu 1999 ; end of the α-1,4-glycosidic bond. To accelerate Karuppiah et al. 2008 ). Energy security is a critical glucoamylase activity in the liquefact slurry, priority for all countries because of the volatility, pullulanase, a debranching enzyme, is used for uncertainty, complexity, and ambiguity (VUCA) its unique role in hydrolyzing the 1-6 linkages of of the global fossil fuel market, due to high the amylopectin branch to produce a linear, free prices, declining production, and unstable geopo- dextrin chain. litical acts of war and terrorism. These issues 162 V. Gohel et al. underscore the vulnerability of currently dominant whole starch-containing feedstock. This product global energy needs to supply disruptions is commercially available as FERMGENTM . This (Gopinathan and Sudhakaran 2009 ). Hence, fuel protease hydrolyzes grain proteins into amino alcohol from starch needs constant process acids, peptides, and free amino nitrogen (FAN) improvement in the biomass conversion process essential for yeast growth. Furthermore, it plays a to make it economically viable. key role in hydrolyzing the protein matrices in The emerging very high gravity (VHG) the grain kernel that bind the various fractions, fermentation technology is one such measure to releasing “hard-to-hydrolyze” starch. The prote- increase fermentation rates and ethanol concen- ase also accelerates ethanol production and tration and to minimize waste effl uent (Bvochora enables higher ethanol yield from grain-based et al. 2000 ). VHG technology is now widely substrates than processes that do not use protease used to increase the concentration of starchy (Duan et al. 2011a ). materials in the feedstock (≥30% w/w DS) Protease also shortens the fermentation cycle and to increase plant throughput (Devantier by 8–12 h and has improved the fermentation et al. 2005). Concurrently, VHG technology effi ciency of VHG fermentation without compro- lowers energy cost per liter of alcohol, bacterial mising ethanol recovery. VHG fermentation is contamination risk, and capital costs. In addition, enhanced when hydrolyzed wheat protein is used increased harvest of high-protein spent yeast as a source of FAN (Thomas and Ingledew 1990 ). residue is obtained with VHG fermentation In barley mash fermentation, FAN enhances (Bvochora et al. 2000 ). VHG requires selection of yeast growth and multiplication (Kim et al. the right yeast species (Saccharomyces cerevisiae ) 2008). It is reported that expression of aspartyl that can produce high ethanol concentrations protease on the cell surface of industrial ethanol- and not be retarded by high sugar and ethanol producing yeast results in higher yeast cell counts concentrations. VHG requires the maintenance with increased rate of growth. A recombinant of appropriate environmental and nutritional con- strain of S. cerevisiae exhibiting a higher yield ditions for optimum performance. The drawback of ethanol and lower residual sugar compared of VHG fermentation is that the increased etha- with the parental strain has been reported by nol recovered does not compensate for its higher Guo et al. 2010 . cost relative to other processes (Gohel and Gohel and Duan (2012a ) documented a three- Duan 2012a ). step conventional liquefaction process followed A 6 % yeast inoculum size was found to be by SSF under yeast fermentation conditions critical for reducing the fermentation time during using 30 % and 35 % dry solid concentrations of VHG fermentation for ethanol production Indian broken rice (starch # 68.23 %) and pearl (Breisha 2010 ). More rapid yeast cell growth millet (starch # 60.64 %) feedstocks. These stud- was associated with shorter lag phase in the ies were performed with and without addition of growth cycle (Breisha 2010 ). At low (<5 %) yeast acid fungal protease platform (FERMGENTM ). inoculum sizes, longer lag phases and lower At the end of the SSF process, ethanol yield with ethanol production rates during the initial stage 30 % and 35 % DS of Indian broken rice feed- of yeast fermentation were observed along with stock was observed to be ~15.14 % and 16.23 % higher contamination ( Akin et al. 2008 ; Sharma (v/v at 20 °C) without FERMGEN TM versus et al. 2007 ). 16.03 % ± 0.02 % and 16.41 % ± 0.08 % (v/v at Urea is widely used by ethanol producers 20 °C), respectively, with the addition of worldwide as an organic nitrogen source for FERMGENTM . The corresponding yields with yeast. The downside is that ethanol reacts with pearl millet feedstock were ~12.14 % and urea to produce ethyl carbamate which is carci- 12.67 % (v/v at 20 °C) at 55 h in 30 % and nogenic to humans (Ough et al. 1988 ; Pretorius 35 % DS concentrations, respectively, without 2000 ; Kitamoto et al. 1991 ). To replace urea, FERMGENTM and 14.13 % and 14.56 % (v/v at DuPont has produced acid fungal proteases for 20 °C) with FERMGENTM . These results showed 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 163 that the addition of protease plays a key role in ethanol due to minimal sugar loss that is inevitably producing more ethanol. Wu et al. (2006 ) used a incurred in a high-temperature cooking process three-step conventional process of ethanol pro- due to heat-catalyzed Maillard reaction between duction from US pearl millet with 65.30 % starch amino acids and reducing sugars. Thus, the and 25 % dry solid concentration. Their process GSHE process yields less biomass (1.95 kg per involved liquefaction at 95 °C for 45 min fol- 100 kg starch) due to reduced yeast stress. In the lowed by 80 °C for 30 min, saccharifi cation at conventional jet cooking process, the biomass of 60 °C for 30 min, and fi nally yeast fermentation, yeast produced is 3.88 kg per 100 kg starch with an ethanol yield of ~11 % v/v at 20 °C, a (Shetty et al. 2007 ). This is because the Maillard fermentation effi ciency of 90 %, and residual reaction products cannot be utilized by yeast for starch concentration of 3.45 %. Zhan et al. (2006 ) fermentation into ethanol (Göğüş et al. 1998 ). used the conventional process for US sorghum The free soluble sugars that are spared from the containing 68.8 % starch and 25 % DS concen- Maillard reaction by eliminating the cooking tration and obtained 10.72 % v/v ethanol yield. process additionally play a role in ethanol production from yeast fermentation. The conventional process involving higher No-Cook Process in Ethanol liquefaction temperatures has about 81–90 % fer- Production mentation effi ciency, compared to 97–98 % with the GSHE process using Indian broken rice and Following the success of VHG fermentation in pearl millet feedstocks (Gohel and Duan 2012b ). ethanol manufacture, the ethanol-producing In the conventional process, Indian ethanol industries have aggressively searched for solutions producers with a plant capacity of 110–130 MT to reduce steam consumption used in the lique- of Indian broken rice (68 % starch, 28 % dry solids) faction process. However, steam consumption for feedstock per day consume 49.5 MT steam for distillation cannot be circumvented. DuPont- the liquefaction process, followed by SSF with Genencor Science has powered the development yeast fermentation, producing 10 % v/v at 20 °C, of STARGENTM technology, a granular starch 410 L ethanol per MT of Indian broken rice with hydrolyzing enzyme (GSHE) to hydrolyze no- 86 % fermentation effi ciency ( http://www.pcbas- cook starch directly into fermentable sugars sam.org/EIAREPORT/EIA_Radiant/2%20 under yeast fermentation conditions without Chapter%20The%20Project.pdf ). In the GSHE using steam (Fig. 11.3 ). It is an optimized blend process, the daily consumption of 49.5 MT of amylases that hydrolyze raw (granular) starch steam in liquefaction is saved, which directly in a no-cook process. These amylases are derived impacts the overall ethanol production cost. The from genetically modifi ed strains of Trichoderma STARGENTM technology works effectively with reesei . The no-cook process enables all these bio- all starchy feedstocks (Gohel and Duan 2012b ). logical processes in a single step without requiring any steam to cook the starchy materials (Bellissimi and Ingledew 2005 ; Bothast and Specialty Syrup Production Schlicher 2005 ; Gohel and Duan 2012b ). The single step is performed in a fermenter under Most of the corn-milling processes use pure yeast fermentation conditions, reducing capital starch for the production of specialty syrups, in investment and physical space used for separate addition to wheat- and cassava-extracted starch. tanks for the liquefaction and saccharifi cation Some syrup producers also use dry-mill process processes. The space can be utilized to install when they use grain like broken rice and millet, more fermentation tanks instead to produce more as the starch within the grains is diffi cult to sepa- ethanol within the existing plant facility. This rate. The grade of specialty syrups produced GSHE process has the additional advantages depends on the method used to hydrolyze the of higher effi ciency of starch conversion into starch and how long the hydrolysis is permitted. 164 V. Gohel et al.

Different grades vary by physical and functional The differences in structure of maltodextrins properties, such as sweetness, compressibility, derived from different starchy materials also and viscosity. The extent of starch hydrolysis is determine their physicochemical properties, and characterized by DE determination. Maltodextrin the characterization of maltodextrins and source and 32, 42, and 60 DE glucose syrups are widely identifi cation assume importance to match each used specialty syrups used in food and confec- application (Wang and Wang 2000; Moore tionery industries (Tharanathan 2005 ). These et al. 2005 ). syrups are produced through enzymatic as well The traditional industrial method of malto- as acid hydrolysis methods (Underkofl er et al. dextrin production involves conventional high- 1965 ; Aiyer 2005 ). temperature liquefaction of whole rice grain fl our, corn, or wheat starch using thermostable α-amylase (Fig. 11.8 ) to produce 8–10 % DE Maltodextrin Production liquefact, followed by saccharifi cation using maltogenic enzymes such as acid α-amylase,

Maltodextrin [(C6 H10 O5 )n H2 O] is a polymer of beta-amylase, pullulanase, and glucoamylase to saccharides, nutritive and not sweet, and consists achieve a DE level of 14–30 % maltodextrin with of glucose units primarily linked by α-1,4- specifi c properties as desired in the current mar- glucosidic bonds, with DE (dextrose equivalent) ket. Maltodextrins of 20–35 % DE are widely values between 8 % and 30 % (Storz and Steffens produced in powder as well as syrup forms with 2004 ; Dokic-Baucal et al. 2004 ; Gibiński 2008 ; more than 78 % Bx. (Brix). Muntean et al. 2010). Maltodextrin is more Various commercially available thermostable soluble in water than native starches and is α-amylases such as SPEZYME® FRED, cheaper than other major edible hydrocolloids, CLEARFLOW® AA, and SPEZYME® ALPHA and its solutions have a bland fl avor and smooth from DuPont are used by the starch industries in texture in the mouth (Dokic-Baucal et al. 2004 ). the liquefaction process, followed by β-amylases It is used for multifaceted purposes in food for manufacturing of a variety of maltodextrin products, including bulking, caking resistance, syrups or powder. The beta-amylases used are texture and body improvement, fi lm formation, OPTIMALT® BBA (produced from barley); acid binding of fl avor and fat, serving as an oxygen fungal amylase, CLERASE® L (produced by a barrier, imparting a surface sheen, aiding disper- selected strain of Aspergillus oryzae var); pul- sion and solubility, increasing soluble solids, lulanase, OPTIMAX® L 1000 (produced from crystallization inhibition, controlling the freezing controlled fermentation of a genetically modifi ed point, fi llings, and as product extenders. Malto- strain of Bacillus licheniformis ); and glucoamy- dextrin can be used as an anticaking agent and lase, GA-L NEW (produced from controlled suitable fi ller in the production of spray-dried fermentation of a selected strain of Aspergillus foods (Chronakis 1998 ; Setser and Racette 1992 ; niger ). Alexander 1992 ). Maltodextrin is produced by enzymatic as well as acid hydrolysis methods (Wang and Wang Specialty Glucose Syrup 2000 ). The enzymatic method has distinct advantages, circumventing the need to remove Currently, glucose syrup produced by acid hydro- salts formed during acid neutralization, operating lysis of starch of 30 % and 42 % DE is widely in a wider pH range and at lower temperatures used in many industries because of its special (with obvious energy savings), with easier process composition, although enzyme hydrolysis is control (Haki and Rakshit 2003 ). It is diffi cult to becoming popular and common in the starch produce syrup or powder of less than 30 % DE sugar industry. The industrial process of acid through acid hydrolysis process due to the forma- hydrolysis of starch at high pressures uses hydro- tion of crystalline-stable nondegradable starch. chloric acid as a catalyst. After achieving 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 165

Table 11.1 Specialty glucose syrup production by acid and enzymatic hydrolysis Degree of polymerization (% DP) Processes Syrup types DP1 DP2 DP3 DP4 DP4+ Acid hydrolysis DE30 9.50 9.00 8.50 8.00 65.00 DE42 17.50 14.00 11.50 10.00 47.00 DE55 30.00 18.00 13.50 9.50 29.00 Enzyme hydrolysis DE34 9.404 9.029 15.17 11.41 54.99 DE43 17.58 14.46 20.49 10.56 36.91 DE54 31.49 13.42 16.87 6.618 31.61

30–42 % DE, the starch hydrolysate is neutralized DuPont Genencor Science has powered the to a pH of 4.6–4.8 using sodium carbonate which development of enzymatic methods to produce also precipitates out the lipids and proteins 32 %, 42 %, and 60 % DE glucose syrup with the (Howling 1992 ). These impurities are removed desired DE and DP profi les that were best pro- through skimming or centrifugation. Residual duced by the acid hydrolysis method (Table 11.1 ). solid impurities of fat and protein fl akes are In the enzymatic processes, any of the DuPont’s removed from the acid hydrolysate liquor by α-amylases, as discussed earlier, are used in sieving it through a series of pressure cloth fi lters. liquefaction to produce 12–14 % DE liquefact, The remaining impurities such as color precur- followed by two enzymatic saccharifi cation sors, protein or protein hydrolysate, peptides, reactions with acid α-amylases, GC 626 (derived amino acids, and undesired fl avors are eliminated from a strain of Aspergillus kawachii expressed by refi ning with powdered or granular activated in Trichoderma reesei) and glucoamylase mar- carbon techniques followed by ion-exchange res- keted as GA-L NEW (produced by controlled ins (Cotillon 1992 ). After refi ning, the syrup fermentation using a selected strain of Aspergillus liquor is adjusted to ~80 % dry solids and cooled. niger). Various dosages of this combination are The fi nal product called glucose syrup is packed used to achieve the desired DE and DP profi les in containers. Acid hydrolysis has the disadvan- formerly produced using acid hydrolysis. This tage that acid is corrosive and challenging to technology has been found successful with any handle. Furthermore, it produces undesirable by- source of starch or starchy materials. Enzymatic products such as 5-hydroxymethylfurfural ( http:// technology is environmentally friendly, with umpir.ump.edu.my/863/1/Siti_Nor_Shadila_ much less acid water produced; it also does not Alias.pdf ) and tints the product with a brownish cause corrosion of the process tanks, produces color if dextrose reversion is carried out for high yields, and is amenable for control of DE extended acid hydrolysis reaction to produce and DP profi le targeted through inactivating the 55 % DE syrup (Bozell 2001 ). Starch processing enzymes by reducing pH and/or increasing the by acid hydrolysis has additional disadvantages temperature. including poor product quality, environmental pollution, complex process handling, high capital expenditure periodically due to corrosion of the High-Glucose Syrup (>95 DE) process tank, high steam pressure, and lower Production product yields. However, industries continue to use acid hydrolysis to achieve the desired DP High-glucose syrup is widely used as a raw (degree of polymerization, number of dextrose material in sorbitol production through chemical molecules linked together) profi le (Table 11.1 ), catalytic hydrogenation process (Ahmed et al. to meet the specifi c property profi le needs of the 2009); high-fructose syrup production through confectionary industry engaged in chocolate or enzymatic glucose isomerase process (Lee et al. candy production. 1990 ); biochemicals such as lactic acid, lysine, 166 V. Gohel et al. and MSG fermentation (Duan 2009 ); and ethanol fructose and 10 % glucose, used in specialty production through yeast fermentation process applications (Marshall and Kooi 1957 ). Among (Dombek and Ingram 1987 ). It is also used for all three, HFCS-42 and HFCS-55 are most widely dextrose monohydrate production through used to replace sugar because of having more crystallization of >95 DE glucose syrup (Hull than 40 % of sweetening value relative to the 2009 ). The process of crystallization allows only caloric value. HFCS is so sweet that it is cost- dextrose to crystallize leaving behind other effective for companies to use small quantities of sugars dissolved in mother liquor. The dextrose HCFS in place of other more expensive sweet- crystals are recovered and washed using a eners or fl avorings. High-fructose-containing centrifuge and dried to produce a very pure syrups are prepared by enzymatic isomerization product. Dextrose is less sweet than sucrose, of dextrose with glucose isomerase (Bhosale which is useful in food processing industries et al. 1996 ). The starch is fi rst converted into where less sweetness is desired. A >95 DE is the dextrose by enzymatic liquefaction and sacchari- ultimate product of starch hydrolysis using an fi cation. The dextrose syrup feed is processed ideal enzymatic process. A liquefact with through immobilized glucose isomerase (GI) 12–14 % DE is suitable for saccharifi cation to columns in a continuous process to produce produce such syrup (Hull 2009). Liquefaction HFCS-42 (Gromada et al. 2008; Illanes et al. is ideally a continuous process, whereas 1992 ). Syrup with 55 % fructose is blended using saccharifi cation is most often conducted as a enriched fructose syrup with fructose of more batch process. Saccharifi cation is followed by a than 90 % together with 42 % fructose syrup. treatment with a blend of various concentrations More than 90 % concentrate is produced using of bacterial pullulanase and fungal glucoamylase simulated moving bed chromatography (Ching marketed as products marketed under and Ruthven 1985 ). OPTIMAX® brand. These enzymes accelerate Immobilization of glucose isomerase (IGI) the reaction and can produce higher glucose offers several advantages for industrial and bio- yields (>95 %) at 38 % DS (dry solids). technological applications, including repeated Saccharifying at higher solid levels substantially use, ease of separation of reaction products reduces evaporation costs at the plant level, in from the biocatalyst, improvement of enzyme addition to enabling increased throughput stability, continuous operation in a packed-bed without loss in yield to meet seasonal demands. reactor, and ready alteration of the properties of These enzymes produce a better substrate for the enzyme ( Seyhan and Dilek 2008 ). GI obtained isomerization into fructose or for hydrogenation from different sources such as Flavobacte- into sorbitol. The enzymes also reduce refi ning rium , Bacillus , and some Streptomyces and costs and permit saccharifi cation at high Arthrobacter species is immobilized on different concentrations of dissolved solids. support materials such as DEAE cellulose (Chen and Anderson 1979 ; Huitron and Limon-Lason 1978 ), polyacrylamide gel (Demirel et al. 2006 ; High-Fructose Syrup Strandberg and Smiley 1971 ), and alginate beads (Rhimi et al. 2007 ). High-fructose syrup is also called as glucose- GENSWEET™ IGI is an immobilized glucose fructose syrup. A variety of high-fructose syrups isomerase [EC 5.3.1.5, D -xylose ketol isomerase] such as HFCS-42, HFCS-55, and HFCS-90 are from DuPont, produced by the controlled fer- produced for various applications (Parker et al. mentation of a selected strain of Streptomyces 2010 ). HFCS-42 has approximately 42 % fruc- rubiginosus . This enzyme is cross-linked using tose and 53 % glucose and is mostly used in polyethylenimine and glutaraldehyde, and granular beverages, processed foods, cereals, and baked particles are produced by extrusion/marumerization goods; HFCS-55 has approximately 55 % fruc- technology, followed by drying. This immobilized tose and 42 % glucose mostly used in soft drinks; enzyme offers unique physical and functional and HFCS-90 contains approximately 90 % properties primarily designed to offer predictable, 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 167 consistent performance and tolerance to process canning, confectionery, and other food and variations, including variation in substrate quality. beverage industries. Ultrapure maltose is used This enzyme requires Mg2+ (25–100 ppm) and as an intravenous nutrient. Catalytic reduction of metabisulfi te (50–175 ppm) as an activator. Prior maltose results in maltitol, a low-calorie sweet- to loading this enzyme into the column, it requires ener. Recently, high-maltose syrup has become a a hydration process, achieved by suspending the key raw material for industrial production of a new GENSWEETTM IGI enzyme in isofeed (substrate) class of sugars, i.e., isomalto-oligosaccharides (IMO) at pH 7.6–8.0 and 54–60 °C at a ratio of 1 kg dry (Duan et al. 2011b ). These sugars are receiving enzyme per 1.5 gal of syrup and mildly agitating increased attention as health (Bifi dobacterium for 1–2 h. The hydrated enzyme is transferred to growth factors) and functional food ingredients. a column, preferably with a diaphragm pump Corn, potato, sweet potato, and cassava starches to avoid excess abrasion of the particles. The col- as well as whole rice fl our are known raw mate- umn upfl ow is initiated with isofeed, gradually rials for maltose manufacture. In enzymatic increasing the feed rate to about 0.9–1.0 bed manufacturing of syrups, the fi rst step of starch volumes per hour over a 2 h period. The column liquefaction is common to all, but it is important upfl ow is continued at this rate for 2 h or until to get it right in order to achieve the right DE froth generated by excessive agitation or pump- and DP profi le for the next saccharifi cation ing is removed. The upfl ow is gradually decreased enzyme which is used to manufacture a variety of to zero over 30 min. Then the column downfl ow maltose syrups or specialty syrups ( http://www. is initiated with isofeed by gradually increasing agfdt.de/loads/st07/gangabb.pdf ). This is because fl ow over 4 h from zero to desired operational of the variety of maltogenic enzymes used in fl ow. The benefi ts of using GENSWEET TM IGI in saccharifi cation based on the target sugar compo- this process include well-controlled and consis- sition desired. To produce 40–50 % maltose tent performance, more rapid hydration and less syrup, using a single maltogenic enzyme, discoloration of the glucose produced, fl exibility β-amylase or fungal α-amylase, the liquefact DE of plant operation, less reduction in upfl ow pressure, should be in the range of 12–14 %. Higher and reduced channeling. concentration maltose syrup (50–60 %) can be produced either by using β-amylase alone or with pullulanase with a liquefact DE in the range of Maltose Syrup Production 10–12 %. High concentration maltose syrup (>80 %) is produced using β-amylase, acid Maltose is a naturally occurring disaccharide, α-amylase, and pullulanase with a liquefact DE consisting of two glucosyl residues linked by an of 4–5 %. DuPont has marketed several liquefac- α-1,4-glucosidic linkage, and is the smallest in tion enzymes such as discussed earlier to achieve the family of oligosaccharides. It is the main desired DE liquefacts that can be saccharifi ed component of maltosugar syrup (Sugimoto with the same maltogenic saccharifi cation enzymes, 1977 ). Maltose is the main component of high- such as β-amylase, OPTIMALT® BBA; acid fun- maltose syrup. The syrup is classifi ed based on gal amylase, CLERASE® L; and pullulanase, the content of maltose. Maltose syrup, containing OPTIMAX® L 1000, for producing a range of different levels of maltose, can be produced from maltose syrups. liquefi ed starch using enzymatic processes. Maltose syrups are produced on a large scale in syrup, powder, and crystal form with several Functional Oligosaccharides grades of purity. Various maltose syrups are drawing considerable interest for commercial Oligosaccharides are an important group of applications, because it is less susceptible to polymeric carbohydrates with 2–10° of polymer- crystallization and is relatively nonhygroscopic. ization that are found either free or in combined Commercial applications for maltose syrups forms in all living organisms. Structurally, oligo- are possible in the brewing, baking, soft drink, saccharides are composed of 2–10 monosaccha- 168 V. Gohel et al. ride residues linked by glycosidic bonds that are sidase. IMO is commercially one of the most readily hydrolyzed by acids or enzymes to release important polysaccharide categories with an the constituent monosaccharides (Nakakuki estimated market demand of about 200,000 t per 1993). Functional oligosaccharides have recently year worldwide (van Dokkum et al. 1999 ). received more attention in recent years because The conventional method of producing IMO from of their role in the microecology of intestinal fl ora starch involves a three-step enzymatic process, and their potential application in health sector liquefaction using thermostable α-amylase, (Zivkovic and Barile 2011 ; Tuohy et al. 2005 ; followed by saccharifi cation using pullulanase Qiang et al. 2009 ). The major functional oligo- and beta-amylase to produce maltose, which in saccharides are xylo-oligosaccharides, fructo- the third step is used as the substrate for the trans- oligosaccharides, and isomalto-oligosaccharides glucosidase enzyme to produce IMO (Pan and (Lai et al. 2011 ; Oku and Nakamura 2003 ). Lee 2005). The transglucosidase catalyzes both Xylo- oligosaccharide (XO) is a kind of func- hydrolytic and transfer reactions on incubation tional oligosaccharide that is considered a safe with α- D -gluco-oligosaccharides. Transfer occurs health additive. Xylo-oligosaccharide is composed most frequently to HO-6 (hydroxyl group 6 of the of 2-7 xylopyranoses linked with β-1,4-glycosidic glucose molecule), producing isomaltose from bonds. The xylopyranoses include xylobiose, D -glucose and panose from maltose. The enzyme trisaccharide, and other oligosaccharides (Zhou can also transfer to HO-2 or HO-3 of D -glucose et al. 2009 ). Fructo-oligosaccharides are pro- to form kojibiose or nigerose or back to HO-4 to duced from inulin by endoinulinases, used as form maltose (McCleary et al. 1989 ). The action potent prebiotics and dietary fi bers, and also on maltose produces equimolar concentration of possess other benefi cial functionalities (Guiraud panose and glucose. As a result of transglucosi- et al. 1987 ; Sangeetha et al. 2005 ; Singh and dase reactions, the malto-oligosaccharides are Singh 2010 ). Furthermore, the completely hydro- converted to isomalto-oligosaccharides, a new class lyzed product, i.e., fructose, produced with exoi- of polysaccharides containing high proportions of nulinase is emerging as a safe sweetener in the glucosyl residues linked by α- D -1,6 linkages from food industry. Recently, inulin has emerged as a the nonreducing end. Being a non-fermentable promising substrate for the enzymatic synthesis sugar, IMO is widely used as a bulking agent in of fructo-oligosaccharides and high-fructose animal feed to increased body weight, in dental syrup (Singh and Singh 2010 ). care due to anti-cariogenic activity, and in baking due to anti-spoiling (preventing staleness) properties. Isomalto-Oligosaccharides DuPont has commercialized the transglucosidase Isomalto-oligosaccharides also known as IMO con- enzyme in the form of purifi ed D -glucosyltrans- tain 40 % α-1,6-glucosidic linkages. IMOs include ferase (transglucosidase, EC 2.4.1.24) free from isomaltose, panose, isomaltotriose, and higher glucoamylase activity, produced through controlled branched sugars. Isomalto-oligosaccharides fermentation using a selected strain of Aspergillus (IMOs) are receiving growing attention due to (Li et al. 2005). In molasses, non-fermentable their biological functions/role as prebiotics that sugars including raffi nose and stachyose are can enhance the growth of Bifidobacteria in converted to sucrose, galactose, glucose, and the large intestine of humans and animals and fructose, which can subsequently be fermented reduce the cariogenic effect (causing dental caries) into alcohol. of sucrose (Kaneko et al. 1995 ). Isomalto- oligosaccharides have been produced by using the transglucosylation activity of enzymes Maltotetraose Syrup obtained from microorganisms (Kuriki et al. 1993 ). The IMO-producing enzyme that cata- Maltotetraose, a linear tetramer of α- D -glucose, lyzes the transglucosylation of maltose to form has many uses in the food and pharmaceutical isomalto-oligosaccharides is called transgluco- industries because of its uniquely low sweetness 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 169

(equivalent to 20 % sucrose), resistance to Several factors will contribute to this growth: retrogradation, retention of desired levels of (1) protein engineering and direct evaluation moisture in foods, and high viscosity compared with high-throughput screening, (2) improved to sucrose, thus improving the texture of processed knowledge of enzyme mechanisms, (3) reduc- foods. It reduces sweetness without affecting the tion in the production costs of industrial enzymes, inherent taste and fl avor of foods. G4 syrup and (4) improved means for enzyme immobilization (high-maltotetraose syrup) exhibits a lower rate and bioprocess engineering. of Millard reaction as it has less glucose and maltose content. This syrup does not lower the freezing point of water as much as sucrose or References high-fructose syrup. Hence, it can be used to alter the freezing points of frozen foods. G4 syrup Aehle W (2007) Enzymes in industry: production and imparts gloss and can be used in industries such applications, 3rd edn. Wiley-VCH Verlag GmbH and as a paper sizer (Aiyer 2005 ). In addition to Co. KGaA, Weinheim Aggarwal NK, Nigam P, Singh D et al (2001) Process nutritional and taste properties, its antimicrobial optimization for the production of sugar for the bio- property was also discovered. Feeding a ethanol industry from sorghum, a nonconventional maltotetraose-rich corn syrup inhibits the growth source of starch. World J Microbiol Biotechnol of intestinal putrefactive bacteria such as C. per- 17:411–415 Ahmed MJ, Khadom AA et al (2009) Optimization hydro- fringens and Enterobacteriaceae (Kimura and genation process of D-glucose to D-sorbitol over Nakakuki 1990 ). Commercial G4-forming amy- Raney nickel catalyst. Eur J Sci Res 30:294–304 lase produced by Pseudomonas saccharophila Aiyer PV (2005) Amylases and their applications. Afr J was expressed in Bacillus licheniformis that Biotechnol 4:1525–1529 Akin H, Brandam C, Meyer XM et al (2008) A model for performs effi ciently in the presence of pullulanase pH determination during alcoholic fermentation of a at 60–65 °C and pH 5.0–5.5 to produce >45 % grape must by Saccharomyces cerevisiae . Chem Eng DP4 G4 syrups (Duan et al. 2010a ). Process 47:1986–1993 Alexander RJ (1992) Maltodextrins. Production, properties and applications. In: Schenk EW, Hebeda RE (eds) Starch hydrolysis products. VCH, New York Summary and Conclusions Baron M, Chamayou A, Marchioro L et al (2005) Radicalar probes to measure the action of energy on Industrial enzymes provide green and sustain- granular materials. Adv Powder Technol 16:199–211 Bellissimi E, Ingledew WM (2005) Analysis of commer- able solutions for various starch industries in cially available active dry yeast used for industrial the midst of growing environmental anxiety. fuel ethanol production. J Am Soc Brew Chem Commercialization of industrial enzymes calls 63:107–112 for continuous technological innovation to Bhosale SH, Rao MB, Deshpande VV (1996) Molecular and industrial aspects of glucose isomerase. Microbiol identify and characterize new catalysts from Mol Biol Rev 60:280–300 natural sources as well as directed evolution Blanchard PH (1992) Technology of corn wet milling with optimal performance for selected applica- and associated processes. Elsevier, Amsterdam, tions, further modifi cation for enhanced perfor- chaps. 1 and 3 Bornscheuer UT, Pohl M (2001) Improved biocatalysts by mance, and increased expression in suitable directed evolution and rational protein design. Curr model systems. The enzyme bio-industry sector Opin Chem Biol 5:137–143 has played a signifi cant role in the current com- Bothast RJ, Schlicher MA (2005) Biotechnological mercial status of biotechnology at global scale. processes for conversion of corn into ethanol. Appl Microbiol Biotechnol 67:19–25 The future will witness more novel applications Böttcher D, Bornscheuer UT (2010) Protein engineering of industrial enzymes in far more arenas than of microbial enzymes. Curr Opin Microbiol 13:274–282 anticipated today. Bozell JJ (2001) Chemicals and materials from renewable The global market for industrial enzymes will resources, ACS symposium series. American Chemical Society, Washington, DC continue to expand as new uses for enzymes are Breisha GZ (2010) Production of 16% ethanol from 35% discovered in the chemical industry at large. sucrose. Biomass Bioenerg 34:1243–1249 170 V. Gohel et al.

Buchholz K, Kasche V, Bornscheuer UT (2005) Biocata- Effect of glucoamylase dosage, pre-saccharifi cation lysts and enzyme technology. Wiley-VCH, Weinheim and yeast strain. Appl Microbiol Biotechnol Bull AT, Goodfellow M, Slater JH (1992) Biodiversity as 68:622–629 a source of innovation in biotechnology. Annu Rev Dokic-Baucal L, Dokic P, Jakovljevic J (2004) Infl uence Microbiol 46:219–252 of different maltodextrins on properties of O/W Bull AT, Bunch AW, Robinson GK (1999) Biocatalyst for emulsions. Food Hydrocoll 18:233–239 clean industrial products and processes. Curr Opin Dombek KM, Ingram LO (1987) Ethanol production Microbiol 2:246–251 during batch fermentation with Saccharomyces Bvochora JM, Read JS, Zvauya R (2000) Application of cerevisiae: changes in glycolytic enzymes and internal very high gravity technology to the cofermentation of pH. Appl Environ Microbiol 53:1286–1291 sweet stem sorghum juice and sorghum grain. Ind Duan G (2009) Impact of the industrial enzyme progress Crop Prod 11:11–17 on the production of chemicals. Chin J Biotechnol Cauvain SP, Young LS (2001) Baking problems solved. 25:1808–1818 Woodhead Publishing Ltd., Cambridge/England, Duan G, Qian Y, Rafael SF, Shetty J (2010a) Improved pp 25–26 production of maltotetraose syrup using a Cawley RW, Mitchell TA (1968) Inhibition of wheat Pseudomonas saccharophila moltoteraohydrolase α-amylase by bran phytic acid. J Sci Food Agric variant and a debranching enzyme. WO/2010/132157 19:106–108 Duan G, Qian Y, Shetty J et al (2010b) Starch separation Chandel AK, Rudravaram R, Rao LV et al (2007) process. PCT WO2010/129648 Industrial enzymes in bioindustrial sector development: Duan G, Dunn-Coleman N, Lantero O et al (2011a) Acid an Indian perspective. J Comm Biotechnol 13:283–291 fungal protease in fermentation of insoluble starch Chaplin MF, Bucke C (1990) The large-scale use of substrates. US8075694 B2, December 13 2011 enzymes in solution. In: Chaplin MF, Bucke C (eds) Duan G, Shetty J, Vadakoot J et al (2011b) Grain compo- Enzyme technology. Cambridge University Press, sition containing prebiotic iso-maltooligosaccharides Cambridge, UK and method of making and using same. US Patent Chen WP, Anderson AW (1979) Purifi cation, immobiliza- 7,993,689 B2 tion and some properties of glucose isomerase from Dzogbefi a VP, Buamah R, Oldham JH (1999) The con- Streptomyces fl avogriseus. Appl Environ Microbiol trolled fermentation of cocoa ( Theobroma cacao L): 38:1111–1119 enzymatic process and associated physicochemical Ching CB, Ruthven DM (1985) An experimental study of changes in cocoa sweatings. Food Biotechnol counter-current adsorption system. Chem Eng Sci 13:1–12 40:877–891 Englyst HN, Kingman SM, Cummings JH (1992) Christophersen C, Andersen E, Jakobsen TS et al (1997) Classifi cation and measurement of nutritionally Xylanase in wheat separation. Starch-Stärke 49:5–12 important starch fractions. Eur J Clin Nutr 46:33–50 Chronakis LS (1998) On the molecular characteristics, FAO (2002) Partnership formed to improve cassava, sta- compositional properties and structural-functional ple food for 600 million people. Food and Agriculture mechanisms of maltodextrins. Crit Rev Food Sci Nutr Organization of the United Nations, Rome, http:// 38:599–637 www.fao.org/english/newsroom/news/2002/10541- Cotillon M (1992) Refi ning: carbon treatment, reactive en.html precoats and ion exchange. In: Schenck F, Hebeda R Fernando S, Adhikari S, Chandrapal C et al (2006) (eds) Starch hydrolysis products. VCH, New York Biorefi neries: current status, challenges, and future Crabb WD, Shetty JK (1999) Commodity scale produc- direction. Energy Fuel 20:1727–1737 tion of sugars from starches. Curr Opin Microbiol Gantelet H, Duchiron F (1999) A new pullulanase from a 2:252–256 hyperthermophilic archaeon for starch hydrolysis. Daiuto E, Cereda M, Sarmento S et al (2005) Effects of Biotechnol Lett 21:71–75 extraction methods on yam ( Dioscorea alata ) starch Gibiński M (2008) Production of oat hydrolysates with a characteristics. Starch-Stärke 57:153–160 low degree of starch saccharifi cation. Pol J Food Nutr De Cordt S, Hendrickx M, Maesmans G et al (1994) The Sci 58:295–300 infl uence of polyalcohols and carbohydrates on the Gibreel JR, Sandercock J, Lan LA et al (2009) thermostability of α-amylase. Biotechnol Bioeng Fermentation of barley by using Saccharomyces 43:107–114 cerevisiae : examination of barley as a feedstocks for Demirel G, Ozcetin G, Sahin F et al (2006) Semiin- bioethanol production and value-added products. Appl terpenetrating polymer networks (IPNs) for entrap- Environ Microbiol 75:1363–1372 ment of glucose isomerase. React Funct Polym G ö ğüş F, Bozkurt H, Eren S (1998) Kinetics of Maillard 66:389–394 reactions between the major sugars and amino acids of DeSantis G, Jones JB (1999) Chemical modifi cation of boiled grape juice. LWT- Food Sci Technol 31:196–200 enzymes for enhanced functionality. Curr Opin Gohel V, Duan G (2012a) Conventional process for Microbiol 10:324–330 ethanol production from Indian broken rice and Devantier R, Pedersen S, Olsson L (2005) Characterization pearl millet. Bioprocess Biosyst Eng. doi: 10.1007/ of very high gravity ethanol fermentation of corn mash. s00449-012-0717-1 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 171

Gohel V, Duan G (2012b) No-cook process for ethanol comparison with several saccharides using the rat production using Indian broken rice and pearl millet. jejunum loop method. Biosci Biotechnol Biochem Int J Microbiol. doi: 10.1155/2012/680232 59:1190–1194 Gohel V, Singh A, Maisuria V, Phadnis A, Chhatpar HS Karuppiah R, Peschel A, Grossmann IE et al (2008) (2006) Bioprospecting and antifungal potential of Energy optimization of an ethanol plant. AICHE J chitinolytic microorganisms. Afr J Biotechnol 5:54–72 54:1499–1525 Gopinathan MC, Sudhakaran R (2009) Biofuels: opportu- Kim S, Dale BE (2004) Global potential bioethanol pro- nities and challenges in India. In Vitro Cell Dev duction from wasted crops and crop residues. Biomass Biol- Plant 45:350–371 Bioenerg 26:361–375 Gough S, Brady D, Nigam P et al (1997) Production Kim TH, Taylor F, Hicks KB (2008) Bioethanol production of ethanol from molasses at 45°C using alginate- from barley hull using SAA (soaking in aqueous immobilized Kluyveromyces marxianus IMB3. ammonia) pretreatment. Bioresour Technol Bioprocess Eng 16:389–392 99:5695–5702 Grafelman DD, Meagher MM (1995) Liquefaction of Kimura T, Nakakuki T (1990) Maltotetraose, a new starch by a single-screw extruder and post-extrusion saccharide of tertiary property. Starch-Stärke static-mixer reactor. J Food Eng 24:529–542 42:151–157 Gromada A, Fiedurek J, Szczodrak J (2008) Isoglucose Kitamoto K, Oda K, Gomi K et al (1991) Genetic engi- production from raw starchy materials based on a two neering of a sake yeast producing no urea by successive stage enzymatic system. Pol J Microbiol 57:141–148 disruption of arginase gene. Appl Environ Microbiol Guiraud JP, Bourgi J, Stervinou M et al (1987) Isolation 57:306–306 of a respiratory-deﬁ cient Kluyveromyces fragilis Klibanov AM (1983) Stabilization of enzymes against mutant for the production of ethanol from Jerusalem thermal inactivation. Adv Appl Microbiol 29:1–28 artichoke. Biotechnol Bioeng 29:850–858 Kłosowski G, Mikulski D, Czupryński B et al (2010) Guo ZP, Zhang L, Ding ZY et al (2010) Improving the Characterisation of fermentation of high-gravity performance of industrial ethanol-producing yeast by maize mashes with the application of pullulanase, expressing the aspartyl protease on the cell surface. proteolytic enzymes and enzymes degrading non- Yeast 27:1017–1027 starch polysaccharides. J Biosci Bioeng 109: Gupta R, Beg QK, Lorentz P (2002) Bacterial alkaline 466–471 proteases: molecular approaches and industrial Kuriki T, Yanase M, Takata H et al (1993) A new way of application. Appl Microbiol Biotechnol 59:15–32 producing isomalto-oligosaccharide syrup by using Haki GD, Rakshit SK (2003) Developments in industrially the transglycosylation reaction of neopullulanase. important thermostable enzymes: a review. Bioresour Appl Environ Microbiol 59:953–959 Technol 89:17–34 Lai P, Chi-Juan S, Chiun-CR W (2011) Effects of oligo- Herrera S (2004) Industrial biotechnology- a chance at saccharides on phase transition temperatures and rheo- redemption. Nat Biotechnol 22:671–675 logical characteristics of waxy rice starch dispersion. J Howling D (1992) Glucose syrup: production, properties Sci Food Agric. doi: 10.1002/jsfa.4712 and applications. In: Schenck F, Hebeda R (eds) Starch Lee C, Saha BC, Zeikus JG (1990) Characterization of hydrolysis products. VCH, New York thermoanaerobacter glucose isomerase in relation to Huitron C, Limon-Lason J (1978) Immobilization of glu- saccharidase synthesis and development of single-step cose isomerase to ion change materials. Biotechnol processes for sweetener production. Appl Environ Bioeng 20:1377–1391 Microbiol 56:2895–2901 Hull P (2009) Glucose syrups: technology and applications. Li F, Vadakoot J, Duan G, Shetty JK (2005) Grain Wiley, West Sussex composi-tions containing prebiotic isomalto- Illanes A, Zúñiga ME, Contreras S et al (1992) Reactor oligosaccharides and methods of making and using design for the enzymatic isomerization of glucose to same. Genencor International Inc. EP1601699 A2. fructose. Bioprocess Biosyst Eng 7:5199–5204 Accessed 7 Dec 2005 Iyer PV, Ananthanarayan L (2008) Enzyme stability and Linko P, Hakulin S, Linko YY (1983) Extrusion cooking stabilization—aqueous and non-aqueous environment. of barley starch for the production of glucose syrup Process Biochem 43:1019–1032 and ethanol. J Cereal Sci 1:275–284 Jenkins PJ, Donald AM (1998) Gelatinisation of starch: a Liu YH, Lu FP, Li Y et al (2008) Characterisation of combined Saxs/waxs/dsc and Sans study. Carbohydr mutagenised acid-resistant alpha-amylase expressed Res 308:133–147 in Bacillus subtilis WB600. Appl Microbiol Johnston DB, Singh V (2001) Use of protease to reduce Biotechnol 78:85–94 steep time and SO2 requirements in a corn wet-milling Mann JG, Liu YA (1999) Industrial water reuse and process. Cereal Chem 78:405–411 wastewater minimization. McGraw-Hill, New York Johnston DB, Singh V (2004) Enzymatic milling of corn: Marrs B, Delagrave S, Murphy D (1999) Novel approaches optimization of soaking, grinding, and enzyme incu- for discovering industrial enzymes. Curr Opin bation steps. Cereal Chem 81:626–632 Microbiol 2:241–245 Kaneko T, Yokoyama A, Suzuki M (1995) Digestibility Marshall RO, Kooi ER (1957) The enzymatic conversion characteristics of isomaltooligosaccharides in of d-glucose to d-fructose. Science 125:648–649 172 V. Gohel et al.

Martin CH, Nielsen DR, Solomon KV et al (2009) Parker K, Salas M, Nwosu VC (2010) High fructose corn Synthetic metabolism: engineering biology at the syrup: production, uses and public health concerns. protein and pathway scales. Chem Biol 16:277–286 Biotechnol Mol Biol Rev 5:71–78 Maurer KH (2004) Detergent proteases. Curr Opin Pohit S, Biswas PK, Kumar R et al (2009) International Biotechnol 15:330–334 experiences of ethanol as transport fuel: policy impli- McCleary BV, Gibson TS, Sheehan H et al (1989) cations for India. Energy Policy 37:4540–4548 Purifi cation, properties, and industrial signifi cance of Polaina J, MacCabe AP (2007) Industrial enzymes, structure, transglucosidase from Aspergillus niger . Carbohydr functions and applications. Springer, Dordrecht Res 185:147–162 Pretorius IS (2000) Tailoring wine yeast for the new Modilal MRD, Shenoi RR, Sebastian B et al (2011) millennium: novel approaches to the ancient art of Isolation, purifi cation and characterization proteases winemaking. Yeast 16:675–729 from vegetable waste. Int J Pharm Life Sci 1:1–11 Qiang X, YongLie C, QianBing W (2009) Health benefi t Mojovic ́ L, Nikolic ́ S, Rakin M et al (2006) Production of application of functional oligosaccharides. Carbohydr bioethanol from corn meal hydrolyzates. Fuel Polym 77:435–441 85:1750–1755 Rahman MM, Rakshit KS (2003) Improved extractability Montesinos T, Navarro JM (2000) Production of alcohol of sweet potato starch using commercial hydrolytic from raw wheat fl our by Amyloglucosidase and enzymes. American society agricultural biological Saccharomyces cerevisiae. Enzyme Microb Technol engineers publication paper no. 036070. ASAAE 27:362–370 annual meeting, Michigan Moore GRP, Luciana RC, Amante ER (2005) Cassava and Ramírez EC, Johnston DB, McAloon AJ et al (2009) corn starch in maltodextrin production. Quim Nova Enzymatic corn wet milling: engineering process and 28:596–600 cost model. Biotechnol Biofuel 2:2. doi: 10.1186/1754- Muntean E, Muntean N, Paizs C et al (2010) Evaluation of 6834-2-2 effective parameters on enzymatic hydrolysis at low Rhimi M, Messaud EB, Borgi MA et al (2007) starch concentrations. Bull UASVM Agric:322–327. Co-expression of L-arabinose isomerase and doi: 10.1590/S0100-40422005000400008 D-glucose isomerase in E. coli and development of Mussatto SI, Dragone G, Guimarães PMR et al (2010) an effi cient process producing simultaneously Technological trends, global market, and challenges of D-tagatose and D-fructose. Enzyme Microb Technol bio-ethanol production. Biotechnol Adv 28:817–830 40:1531–1537 Nakakuki T (1993) Oligosaccharides: production, properties Rong Y, Sillick M, Gregson CM (2009) Determination of and applications, vol 3, Japanese technology reviews. dextrose equivalent value and number average molec- Gordon and Breach, Switzerland ular weight of maltodextrin by osmometry. J Food Sci Neves MA, Kimura T, Shimizu N et al (2006) Production 74:C33–C40 of alcohol by simultaneous saccharification and Rosendal P, Nielsen BH, Lange NK (1979) Stability of fermentation of low-grade wheat fl our. Braz Arch Biol bacterial alpha-amylase in the starch liquefaction pro- Technol 49:481–490 cess. Starch-Stärke 31:368–372 Nikolić S, Mojović L, Rakin M et al (2010) Ultrasound- Roy I, Gupta MN (2004) Hydrolysis of starch by a mix- assisted production of bioethanol by simultaneous ture of glucoamylase and pullulanase entrapped indi- saccharifi cation and fermentation of corn meal. Food vidually in calcium alginate beads. Enzyme Microb Chem 122:216–222 Technol 34:26–32 Oberoi R, Beg QK, Puri S et al (2001) Characterization Sangeetha PT, Ramesh MN, Prapulla SG (2005) Recent and wash performance analysis of an SDS-stable alka- trends in microbial production, analysis, and applica- line protease from a Bacillus sp. World J Microbiol tions of fructooligosaccharides. Trends Food Sci Biotechnol 17:493–497 Technol 16:442–457 Oku T, Nakamura S (2003) Comparison of digestibility Sayaslan A (2004) Wet-milling of wheat fl our: industrial and breath hydrogen gas excretion of fructo- processes and small-scale test methods LWT. Food Sci oligosaccharide, galactosyl-sucrose, and isomalto- Technol 37:499–515 oligosaccharide in healthy human subjects. Eur J Clin Setser CS, Racette WL (1992) Macromolecule replac- Nutr 57:1150–1156 ers in food products. Crit Rev Food Sci Nutr Olempska-Beer ZS, Merker RI, Ditto MD et al (2006) 32:275–297 Food-processing enzymes from recombinant Seyhan S, Dilek TA (2008) Catalytic effi ciency of immo- microorganisms – a review. Regul Toxicol Pharm bilized glucose isomerase in isomerization of glucose 45:144–158 to fructose. Food Chem 111:658–662 Ough CS, Crowell EA, Gutlove BR (1988) Carbamyl Shanavas S, Padmaja G, Moorthy SN et al (2011) Process compound reactions with ethanol. Am J Enol Vitic optimization for bioethanol production from cassava 39:303–307 starch using novel eco-friendly enzymes. Biomass Pan YC, Lee WC (2005) Production of high-purity Bioenerg 35:901–909 isomalto- oligosaccharides syrup by the enzymatic Shapouri HJ, Duffield J, McAloon A et al (2004) conversion of transglucosidase and fermentation of The 2001 net energy balance of corn-ethanol. In: yeast cells. Biotechnol Bioeng 89:797–804 Tumbleson M (ed) Corn: feedstock of the future (Corn 11 Industrial Enzyme Applications in Biorefi neries for Starchy Materials 173

utilization and technology conference proceedings). Thomas KC, Ingledew WM (1990) Fuel alcohol produc- Corn Utilization and Technology, Indianapolis tion: effects of free amino nitrogen on fermentation of Sharma V, Rausch KD, Tumbleson ME et al (2007) very-high-gravity wheat mashes. Appl Microbiol Comparison between granular starch hydrolyzing Biotechnol 56:2046–2050 enzyme and conventional enzymes for ethanol produc- Tuohy KM, Rouzaud GCM, Brück WM, Gibson GR tion from maize starch with different amylase: amylo- (2005) Modulation of the human gut microfl ora pectin ratios. Starch-Stärke 59:549–556 towards improved health using prebiotics – assess- Shetty J, Chotani G, Duan G et al (2007) Cassava as an ment of effi cacy. Curr Pharm Des 11:75–90 alternative feedstocks in the production of renewable Underkofl er LA, Danault LJ, Hou EF (1965) Enzymes in transportation fuel. Int Sugar J 109:3–11 starch industry. Starch-Stärke 17:179–184 Shetty J, Strohm BA, Singh V (2010) Phytase in enzymatic Van der Maarel MJEC, Van der Veen B, Uitdehaag JCM, wet milling process. WO2010027846 A1. Accessed 11 Leemhuis H, Dijkhuizen L (2002) Properties and May 2010 applications of starch-converting enzymes of the Singh V, Johnston DB (2002) Pasting properties and surface α-amylase family. J Biotechnol 94:137–155 characteristics of starch obtained from an enzymatic van Dokkum W, Wezendonk B, Srikumar TS et al (1999) corn wet-milling process. Cereal Chem 79:523–527 Effect of nondigestible oligosaccharides on large Singh V, Johnston D (2004) An enzymatic process for bowel functions, blood lipid concentrations & glucose corn wet milling. Adv Food Nutr Res 48:151–171 absorption in young healthy male subjects. Eur J Clin Singh RS, Singh RP (2010) Production of fructooligosac- Nutr 53:1–7 charides from inulin by endoinulinases and their prebi- Vessia A (2007) Cassava: the food of the poor for future otic potential. Food Technol Biotechnol 48:435–450 food security. Le Scienze Web News. http://www. Sivaramakrishnan S, Gangadharan D, Nampoothiri KM lswn.it/en/nutrition/news/cassava_the_food_of_the_ et al (2006) Alpha-amylase from microbial sources-an poor_for_future_food . Accessed 25 Jan 2010 overview on recent developments. Food Technol Wang Y-J, Wang L (2000) Structure and properties of Biotechnol 44:173–184 commercial maltodextrins from corn, potato, and rice Sriroth K, Chollakup R, Chotineeranat S et al (2000) starches. Starch-Stärke 52:296–304 Processing of cassava waste for improved biomass Warnecke F, Hess MA (2009) A perspective: metatran- utilization. Bioresour Technol 71:63–69 scriptomics as a tool for the discovery of novel bio- Steinke JD, Johnson LA (1991) Steeping maize in the catalysts. J Biotechnol 142:91–95 presence of multiple enzymes. I. Static batch wise Weegels PL, Marseille JP, Hamer RJ (1992) Enzymes as a steeping. Cereal Chem 68:7–12 processing aid in the separation of wheat fl our into Storz E, Steffens K (2004) Feasibility study for determi- starch and gluten. Starch-Stärke 44:44–48 nation of the dextrose equivalent (DE) of starch Wu X, Wang D, Bean SR, Wilson JP (2006) Ethanol hydrolysis products with near-infrared spectroscopy production from pearl millet using Saccharomyces (NIRS). Starch-Stärke 56:58–62 cerevisiae . American society agricultural biological Strandberg GW, Smiley KL (1971) Free and immobilized engineer meeting: 057077 glucose isomerase from Streptomyces phaechromo- Yan S, Wu X, MacRitchie F, Wang D (2009) Germination- genes. Appl Microbiol 21:588–593 improved ethanol fermentation performance of high- Sugden D (1997) Wheat starch and gluten manufacturing. tannin sorghum in a laboratory dry-grind process. World Grain 3:20–23 Cereal Chem 86:597–600 Sugimoto K (1977) Denpunkagaku handbook. Asakura, Yeh W-K, Yang H-C, McCarthy JR (eds) (2010) Enzyme Tokyo, pp 450–461 technologies: metagenomics, evolution, biocatalysis Suresh K, Sree NK, Rao LV (1999) Utilization of dam- and biosynthesis. Wiley, Hoboken aged sorghum and rice grains for ethanol production Zeleznak KJ, Hoseney RC (1986) The role of water in the by simultaneous saccharifi cation and fermentation. retrogradation of wheat starch gels and bread crumb. Bioresour Technol 68:301–304 Cereal Chem 63:407–411 Sutherland JD (2000) Evolutionary optimisation of Zhan X, Wang D, Bean SR, Mo X, Sun XS, Boyl D enzymes. Curr Opin Chem Biol 4:263–269 (2006) Ethanol production from supercritical- Szulczyk KR, McCarl BA, Cornforth G (2010) Market fl uid-extrusion cooked sorghum. Ind Crops Prod 23: penetration of ethanol. Renew Sustain Energy Rev 304–310 14:394–403 Zhou E, Xiaoliang P, Xiuzhi T (2009) Application study Tester RF, Karkalas J, Qi X (2004) Starch structure and of xylo-oligosaccharide in layer production. Mod digestibility enzyme-substrate relationship. World’s Appl Sci 3:103–107 Poult Sci J 60:186–195 Zivkovic AM, Barile D (2011) Bovine milk as a source Tharanathan RN (2005) Starch – value addition by of functional oligosaccharides for improving human modifi cation. Crit Rev Food Sci Nutr 45:371–384 health. Adv Nutr 2:284–289 About the Editors

Dr. Pratyoosh Shukla is working as Head, India Award in Probiotics & Enzyme Technology Department of Microbiology, at Maharshi (2010), and was also selected as Scientist in the Dayanand University, Rohtak, India. Southern Ocean Antarctica Expedition (2011), His research interests are in the ﬁ elds of DST-Fast Track Young Scientist (2012), etc. enzyme technology, protein bioinformatics, and microbial biotechnology. He has more than 12 Prof. Brett I. Pletschke is currently a Professor years of research and teaching experience in and Head of Biochemistry in the Department of well-reputed universities of India and abroad. Biochemistry, Microbiology and Biotechnology He is the author of four book chapters and one at Rhodes University, Grahamstown, South patent, has edited 3 Biotech (Springer journal), Africa. Prof. Pletschke has served as the Vice and published more than 30 peer-reviewed President and President of SASBMB (South international papers in highly reputed interna- African Society for Biochemistry and Molecular tional journals and more than 60 conference Biology) and is currently the Immediate Past technical papers in biotechnology-related ﬁ elds. President of SASBMB. In this capacity, he has He has also served as the technique committee acted as a voting member for SASBMB at member in some international and national con- IUBMB on occasions. Prof. Pletschke’s research ferences. He has successfully carried out four interest is focused on the phenomenon of enzyme- R&D projects as Principal investigator and/or enzyme synergy, using lignocellulose as a suit- Coinvestigator. able model substrate. He was awarded a Rhodes He received several awards, including Prof. University Alty Teaching Award in 2006 and was S.B. Saksena, F.N.A., Award in life sciences nominated for the Vice Chancellor’s Distinguished (1999); Best Presentation Award (Senior Category, Research Award in 2008. He has delivered several 2006) by the National Council for Science and plenary talks at international conferences, pub- Technology Communication (NCSTC), India; lished more than 50 papers in peer-reviewed inter- NRF-DUT Post-doctoral Fellowship Award in national journals or books, and has supervised or Enzyme Biotechnology (2008); and Danisco graduated 29 M.Sc./Ph.D. students.

P. Shukla and B.I. Pletschke (eds.), Advances in Enzyme Biotechnology, 175 DOI 10.1007/978-81-322-1094-8, © Springer India 2013