Dietary Carbohydrates Influence the Structure and Function of the Intestinal Alpha-Glucosidases Mohammad Chegeni

Home , Invertase, Lactase, Maltase, Octose

Purdue University Purdue e-Pubs

Open Access Dissertations Theses and Dissertations

Fall 2014 Dietary carbohydrates influence the structure and function of the intestinal alpha-glucosidases Mohammad Chegeni

Follow this and additional works at: https://docs.lib.purdue.edu/open_access_dissertations Part of the Biochemistry Commons, Biology Commons, Food Science Commons, and the Molecular Biology Commons

Recommended Citation Chegeni, Mohammad, "Dietary carbohydrates influence the structure and function of the intestinal alpha-glucosidases" (2014). Open Access Dissertations. 433. https://docs.lib.purdue.edu/open_access_dissertations/433

This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. Please contact [email protected] for additional information.

DIETARY CARBOHYDRATES INFLUENCE THE STRUCTURE AND FUNCTION

OF THE INTESTINAL α-GLUCOSIDASES

A Dissertation

Submitted to the Faculty

Purdue University

Mohammad Chegeni

In Partial Fulfillment of the

Requirements of the Degree

Doctor of Philosophy

May 2015

Purdue University

West Lafayette, Indiana ii

I dedicate this work to my wife to honor her love, patience, and support.

iii

ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my advisor, Dr. Bruce Hamaker, who has supported me throughout my thesis with his inspiring and delicate guidance, advice, and expertise. I sincerely thank my advisory committee members, Dr. Mario

Ferruzzi, Dr. Buford Nichols and Dr. Kee-Hong Kim. I am much grateful to them for their valuable advice and using their precious times to read this thesis. I would like also to thank our collaborator Dr. Hassan Naim. Without his expertise, this research would not have been as developed as it is.

I would also like to acknowledge the help and support of my labmates, for always providing a positive work atmosphere. Much appreciation goes to my special friends Dr.

Deepak Bhopatkar, Dr. Amandeep Kaur, Dr. Madhuvanti Kale, Dr. Choon Young Kim,

Dr. Chih-Yu Chen, and Beth Pletsch. Their wisdom and endless help in research and

Ph.D life were much appreciated. My very special and warm recognition goes to my very best friend Dr. Daniel Erickson. He made my Ph.D. life enjoyable at Purdue University. I really appreciated all his time helping me. The time we have shared together is unforgettable.

My family and my in-laws have provided invaluable support to me through this journey, and for that and much more, I express my deepest appreciation. And last, but not the least, I owe my loving thanks to my dear wife, Fariba Tayyari. Words fail me to

iv convey my gratitude to her dedication and love. I owe her for generously letting her intelligence, passions, and ambitions collide with mine.

TABLE OF CONTENTS

Page

LIST OF TABLES ...... viii

LIST OF FIGURES ...... ix

ABSTRACT ...... xii

CHAPTER 1. INTRODUCTION ...... 1

1.1 Research Hypothesis and Specific Aims ...... 4 1.2 Significance of the Study ...... 5 1.3 Thesis Organization ...... 6 1.4 References ...... 8

CHAPTER 2. REVIEW OF LITERATURE ...... 10

2.1 Carbohydrates ...... 10 2.2 Carbohydrate Digestion ...... 12 2.2.1 α-Amylase ...... 13 2.2.2 Intestinal α-Glucosidases ...... 14 2.2.3 Sucrase-Isomaltase ...... 16 2.3 Carbohydrate Absorption ...... 18 2.3.1 Monosaccharide Absorption ...... 18 2.3.1.1 Movement of Monosaccharides Across the Brush Border Membrane ...... 18 2.3.1.2 Movement of Monosaccharides Across the Basolateral Membrane ....19 2.4 Carbohydrate Sensing ...... 20 2.4.1 Sweet Taste Receptor ...... 21 2.4.1.1 Carbohydrate Sensing by the Sweet Taste Receptor ...... 22 2.4.2 Dietary Carbohydrate Sensing and α-Glucosidase Enzymes ...... 23 2.5 References ...... 32

Page

CHAPTER 3. DIETARY CARBOHYDRATES INFLUENCE THE STRUCTURE AND FUNCTION OF SUCRASE-ISOMALTASE ...... 41

3.1 Introduction ...... 41 3.2 Materials and Methods ...... 45 3.2.1 Immunochemical Reagents ...... 45 3.2.2 Cell Culture Procedure ...... 46 3.2.3 Cell Treatment ...... 46 3.2.4 Total Protein Extraction ...... 47 3.2.5 Western Blot Analysis ...... 47 3.2.6 Membrane Protein Extraction ...... 48 3.2.7 Preparation of Brush Border Membranes ...... 49 3.2.8 Pulse-chase and Metabolic Labeling ...... 49 3.2.9 Immunoprecipitation and SDS-PAGE ...... 50 3.2.10 Isolation of Detergent Resistant Membranes (DRMs) ...... 51 3.2.11 Immunoassays ...... 52 3.2.12 Enzyme Activity Measurement ...... 52 3.2.13 Total RNA Isolation ...... 53 3.2.14 cDNA Synthesis and Real-Time PCR Analysis ...... 53 3.2.15 Quantification of Band Intensities ...... 55 3.2.16 Statistical Analysis ...... 55 3.3 Result ...... 55 3.3.1 Intracellular Distribution and Compartmentalization of SI ...... 55 3.3.2 The Kinetics of Glycosylation Alterations ...... 56 3.3.3 Association of SI with Detergent-Resistant Microdomains ...... 57 3.3.4 Sucrase-Isomaltase Enzyme Activity ...... 58 3.3.5 Deglycosylation Treatment of Treated Cells ...... 59 3.3.6 Gene Expression of the Sweet Taste Receptor with Maltose Exposure ...... 60 3.4 Discussion ...... 62 3.3 References ...... 98

CHAPTER 4. INDUCTION OF DIFFERENTIATION OF SMALL INTESTINAL ENTEROCYTE CELLS BY MALTOOLIGOSACCHARIDE TREATMENT USING A METABOLOMICS APPROACH ...... 103

4.1 Introduction ...... 103 4.1.1 Sample Preparation ...... 104 4.1.1.1 Cell Samples ...... 105 4.1.1.2 Cell Sample Preparation ...... 105 4.1.2 Analytical Instrument Platforms ...... 106 4.1.3 Data Analysis ...... 107 4.1.3.1 Pre-Data Processing ...... 107 4.1.3.2 Statistical Methods ...... 109 4.1.4 Metabolomic Application in the Present Study ...... 111

vii

Page

4.2 Materials and Methods ...... 113 4.2.1 Chemicals and Samples ...... 113 4.2.1.1 Chemicals ...... 113 4.2.1.2 Cell Culture Procedure ...... 114 4.2.1.3 Cell Treatment ...... 114 4.2.1.4 Measurement of Epithelial Monolayer Electrical Resistance ...... 115 4.2.1.5 Determination of Paracellular Permeability ...... 115 4.2.1.6 Sample Preparation for NMR Experiments ...... 116 4.3 NMR Experiments and Data Processing ...... 117 4.3.1 NMR Experiments ...... 117 4.3.2 Data Processing ...... 118 4.3.2.1 Statistical Analysis ...... 119 4.4 Results ...... 119 4.4.1 Different Source of Carbohydrates Alters Intestinal Epithelial Tight Junction Barrier ...... 119 4.4.2. Different Source of carbohydrate Alters Metabolite Profiles in Caco-2 Monolayers ...... 120 4.5 Discussion ...... 122 4.6 References ...... 143

CHAPTER 5. OVERALL CONCLUSIONS AND FUTURE RECOMMENDATIONS ...... 150

5.1 References ...... 154

VITA ...... 155

viii

LIST OF TABLES

Tables Page

2.1 Classification of monosaccharides (Fennema, 2005)...... 25

2.2 Substrates and products of small intestinal glycosidase enzyme...... 26

3.1 Relative protein amount (P2/P1) of intercellular distribution of SI in present of different carbohydrate ...... 72

3.2 Association of SI with detergent-resistant microdomains in present of glucose and maltose ...... 73

3.3 Sucrase activity of SI in cells fed with different carbohydrates...... 74

3.4 Maltase activity of SI in cells fed with different carbohydrates ...... 75

3.5 The kinetic parameters of sucrase and maltase in the cells fed with glucose or maltose ...... 76

4.1 Metabolite assignments and chemical shifts of identified peaks ...... 127

LIST OF FIGURES

Figure Page

2.1 The structures of common dietary disaccharids lactose (a), sucrose (b), and maltose (c) (Fennema, 2005)...... 27

2.2 Schematic diagram of the action of salivary and pancreatic α-amylases on starch ...... 28

2.3 Position and protein organization of the MGAM and SI complex. TMD: transmembrane domain; O-link: O-glycosylated linker (adapted from Sim et al., 2008)...... 29

2.4 The structure of the sucrase-isomaltase complex (Lieberman, 2013) ...... 30

2.5 The absorption and transfer out of sugars by the enterocyte. G, glucose; Ga, galactose; F, fructose (adapted from Paulev, 1999) ...... 31

3.1 Biochemical analysis of the distribution of SI. Brush border membranes were isolated by the calcium chloride precipitation method...... 77

3.2 Band intensities from P1 and P2 fractions ...... 78

3.3 Trafficking kinetics of SI to the cell surface ...... 79

3.4 Graphical representation of the pulse-chase experiment ...... 80

3.5 Transport kinetics of SI to the cell surface ...... 81

3.6 Distribution of SI in Tween-20 detergent resistant membranes ...... 82

3.7 Graphical representation of the distribution of SI in Tween-20 detergent resistant membrane ...... 83

3.8 Comparison of sucrase activity of SI at the brush border membrane of Caco-2 cells treated with different concentrations of glucose and maltose ...... 84

3.9 Comparison of maltase activity of SI at the brush border membrane of Caco-2 cells treated with different concentrations of glucose and maltose ...... 85

Figure Page

3.10 Michaelis-Menten plots for SI. The enzymatic activity of sucrase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3) ...... 86

3.11 Lineweaver-Burk plots for SI. The enzymatic activity of sucrase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3) ...... 87

3.12 Michaelis-Menten plots for SI. The enzymatic activity of maltase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3) ...... 88

3.13 Lineweaver-Burk plots for SI. The enzymatic activity of maltase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3) ...... 89

3.14 Deglycosylation (Endo F) of SI in cells treated with glucose and maltose ...... 90

3.15 Deglycosylation (Endo H) of SI in cells treated with glucose and maltose ...... 91

3.16 T1R3 gene transcription in Caco-2 cells in the presence of glucose and maltose...... 92

3.17 T1R2 gene transcription in Caco-2 cells in the presence of glucose and maltose...... 93

3.18 α-Gustducin gene transcription in Caco-2 cells in the presence of glucose and maltose...... 94

3.19 SI gene transcription in Caco-2 cells in the presence of glucose and maltose ..... 95

3.20 GLUT5 gene transcription in Caco-2 cells in the presence of glucose and maltose ...... 96

3.21 GLUT2 gene transcription in Caco-2 cells in the presence of glucose and maltose ...... 97

4.1 Intestinal epithelial tight junction barrier function in Caco-2 cells in presence of different types of carbohydrates ...... 128

Figure Page

4.2 Intestinal epithelial tight junction barrier function in Caco-2 cells in presence of different types of carbohydrates...... 129

4.3 Epithelial paracellular permeability in Caco-2 cells in presence of different types of carbohydrates...... 130

4.4 Epithelial paracellular permeability in Caco-2 cells in presence of different types of carbohydrates...... 131

4.5 Representative 1D 1H NMR noesygppr1d spectrum of metabolites extracted from Caco-2 cells fed with glucose...... 132

4.6 Aliphatic region of a representative 1D 1H NMR noesygppr1d spectrum of metabolites extracted from Caco-2 cells fed with glucose. Identified metabolites have been labeled with numbers, according to Table 4.1 ...... 133

4.7 Aromatic region of a representative 1D 1H NMR noesygppr1d spectrum of metabolites extracted from Caco-2 cells fed with glucose. Identified metabolites have been labeled with numbers, according to Table 4.1 ...... 134

4.8 Relative population of taurine in Caco-2 cells in presence of different types of carbohydrates...... 135

4.9 Relative population of taurine in Caco-2 cells in presence of different types of carbohydrates...... 136

4.10 Relative population of taurine in Caco-2 cells in presence of different types of carbohydrates...... 137

4.11 Relative population of myo-inositol in Caco-2 cells in presence of different types of carbohydrates...... 138

4.12 Relative population of phosphorylcholine in Caco-2 cells in presence of different types of carbohydrates...... 139

4.13 Relative population of phosphorylcholine in Caco-2 cells in presence of different types of carbohydrates...... 140

4.14 Relative population of glycerophosphocholine in Caco-2 cells in presence of different types of carbohydrates...... 141

4.15 Relative population of glycerophosphocholine in Caco-2 cells in presence of different types of carbohydrates...... 142

xii

ABSTRACT

Chegeni, Mohammad. Ph.D., Purdue University, May 2015. Dietary Carbohydrates Influence the Structure and Function of the Intestinal α-Glucosidases. Major Professor: Bruce R. Hamaker.

As the primary products of starch digestion by pancreatic α-amylase, maltooligosaccharides (including maltose) are the main substrates for the α-glucosidases at the intestinal brush border. Here, maltose was shown to induce the formation of a higher molecular weight (HMW) sucrase-isomaltase (SI) species in Caco-2 cells that sorts more quickly to the enterocyte surface to act as a digestive enzyme. As this finding suggested a maltose sensing ability of small intestinal enterocytes, molecular mechanisms associated with the maturation and trafficking of HMW SI were further investigated. A pulse-chase experiment using [35S]-methionine revealed a higher rate of early trafficking and maturation of the HMW SI species in cells treated with maltose. Endoglycosidase treatment of immunoprecipitated SI showed that increased molecular weight is a consequence of additional N- and O-glycosylation of the enzyme. In comparison to the control, the HMW SI was found to be more associated with lipid rafts at the membrane surface which was related to the higher apical sorting of these species. Thus, maltose sensing of small intestinal enterocytes triggers the intracellular processing of HMW SI, speculatively to enhance its digestive property. Study on the sweet taste receptor subunits (T1R2 and T1R3) by qRT-PCR showed that the expression of T1R2 and T1R3 increased in presence of maltose compared to glucose. It appeared also that

xiii maltooligosaccharides may signal other events in small intestine enterocytes. Culture- related conditions (e.g. glucose concentration) are known to alter physical barrier properties of Caco-2 cell monolayers, which affect transepithelial transport of solutes permeating the cell monolayer barrier. A transepithelial electrical resistance experiment was conducted to measure the integrity of the Caco-2 monolayer in response to maltooligosaccharides. Maltose promoted higher tight junction formation and permeability compared to glucose and sucrose at 12 hours. Contrary to this finding, paracellular permeability increased with maltose. In addition to barrier function of small intestinal enterocytes, a metabolomics study was conducted using high-resolution 1H

NMR. Results showed that concentration of metabolites (taurine, phosphorycholine, and glycerophosphocholine) which are known to be markers for cell differentiation increased in Caco-2 cells treated with different types of maltooligosaccharides compared to glucose and sucrose. Overall, these findings are indicative of a maltooligosaccharide sensing ability by enterocytes that trigger a number of important events in the cell including a higher level of SI processing and enzyme activation for digestion purpose, and an increased cell differentiation and tight junction barrier function. From the broader perspective of a desire to control of postprandial glycemic response, as well as to elicit the gut-brain axis and ileal brake mechanisms for appetite control, this work suggests a new potential point of controlling glucose release and absorption in the small intestine

(i.e., a putative maltooligosaccharide enterocyte receptor).

CHAPTER 1. INTRODUCTION

Food intake provides essential nutrients for growth, repair, and maintenance of the body. Glycemic carbohydrates are a main source of energy for many animal species.

Different homeostatic systems (e.g., brain and gut hormones) effectively regulate the digestion and absorption of dietary carbohydrates through various physiological responses (e.g., expression/secretion of hydrolytic enzymes and transporters, gastric emptying).

Dietary carbohydrates can be absorbed only in the form of monosaccharides

(glucose, fructose, and galactose), with the major source being glucose from starch, and glucose and fructose from sucrose and glucose and galactose from lactose. The glycemic response of starchy foods is heavily dependent on the rate and amount of glucose released during starch digestion. A slow glucose release has been shown to be beneficial for maintaining glucose homeostasis and is linked to reduction of the incidence of certain diet-related metabolic diseases (Atkinson et al., 2008). Low glycemic index foods and slowly digestible starches are two possible ways to alleviate the detrimental effect caused by long-term exposure of postprandial glycemic spikes from consumption of foods with high glycemic index (Jenkins et al., 2002). However, more mechanistic studies on the digestion and absorption of carbohydrates in the gastrointestinal tract are needed to fully understand the role of postprandial glucose release rate and health. Investigation of the

2 nutrition property of glycemic carbohydrates and metabolism is important for human health.

Starch, the main dietary carbohydrate in foods, is digested first by α-amylase, consisting of salivary and pancreatic isoforms, that cleave the α-1,4 glycosidic linkages between internal glucose units that are the basic building blocks of starch molecules. The degradation products, small linear oligosaccharides and branched α-limit dextrins, are generally composed of maltose and somewhat larger oligosaccharides with different chemical structures (size and linkages), termed together “maltooligosaccharides”. The α- limit dextrins are then hydrolyzed by brush border-anchored α-glucosidases, maltase- glucoamylase (MGAM) and sucrase-isomaltase (SI) to liberate glucose for absorption

(Holmes, 1971; Freeze, 1999). Although pancreatic α-amylase is generally thought to be the dominant enzyme in carbohydrate digestion (Englyst et al., 1992), SI and MGAM are the ultimate enzymes for glucose liberation, and their relative activities can, and probably are, related glycemic control in everyday life. They also are the control point of glucose generation in the gut lumen, and consequently offer another potential opportunity for glucose control. After starch is fully digested, glucose is transported into small intestinal enterocytes through the energy and sodium-dependent glucose transporter (SGLT1) and transported from enterocytes into the circulation system through the facilitated glucose transporter type 2 (GLUT2) (Levin, 1994).

As the primary product of starch digestion by pancreatic α-amylase, maltose and maltooligosaccharides are presented to the α-glucosidases at the apical surface of the enterocytes. Based on our initial studies with Caco-2 cells as an enterocyte model system, maltose treatment appeared to induce the maturation of SI, while other sugars including

3 fructose, sucrose, isomaltose, and glucose do not show this effect (Cheng et al., 2014).

This was surprising, as it suggested that a receptor for maltose/maltooligosaccharides exists that has not yet been reported in literature. So we hypothesized that there is a maltose sensing mechanism in the enterocytes that triggers the maturation and trafficking of SI, and that may have other effects related to enterocyte function. Further, an attenuation of this pathway might be used to control glucose production rate and the glycemic response.

From what has been known so far, sugar sensing is rather complex process involving their sensing by receptors on cell membranes, their further digestion, if necessary, and transport through the small intestinal enterocytes, and metabolism in the body. GLUT2 is a glucose sensor resulting in high expression of SI, which is termed the

GLUT2-dependent signaling pathway (Le Gall et al., 2007). Another important sugar sensor is the sweet taste receptor composed of T1R2 and T1R3 subunits on the enteroendocrine cells of gastrointestinal tract. The receptor can bind sucrose, other sugars

(e.g. glucose and sucrose), and artificial sweeteners (e.g. sucralose) to stimulate enterocyte SGLT1 expression through the function of hormones like gastric inhibitory polypeptide (GIP) or glucagon-like peptide-1 (GLP-1) secreted by enteroendocrine cells

(Margolskee et al., 2007). In behavioral studies on rodents that sense starch taste, including starch-derived maltose and Polycose™ (a maltodextrin), starch taste was found distinct from sugar (mainly sucrose) taste (Sclafani, 1987). Recent studies (Zukerman et al., 2009) also showed that the T1R3 subunit of the sweet taste receptor is critical for sucrose sensing, but that starch-derived maltodextrins showed no stimulus of the receptor. Dietary glucose and fructose have been shown to up-regulate the transcription

4 of SI by binding of the dimeric nuclear protein (Cdx-2) on the SI promoter, while glucose can down-regulate SI expression (Goda, 2000) by affecting the binding of hepatocyte nuclear factors (HNF)-1a to its promoter (Gu et al., 2007).

As mentioned above, a rapid digestion of starch and absorption of glucose is detrimental to health due to a glycemic spike that is a stress to the glucose homeostasis regulatory system. If the mechanism of SI maturation triggered by maltose is understood, there may be ways to manipulate the activity of brush border anchored α-glucosidases so as to modulate the postprandial glycemic response for glycemic control and improved health. Perhaps this approach could be used by way of a dietary intervention to manage, or even prevent, type 2 diabetes.

1.1 Research Hypothesis and Specific Aims

Based on our initial study using Caco-2 cell as the enterocyte model system,

Cheng et al. (2014) showed that maltose treatment induced apparent maturation of SI, leading to the hypothesis that maltooligosaccharides generated by α-amylase degradation of dietary starch regulate the maturation, trafficking, and activation of HMW SI to the apical membrane of intestinal enterocytes.

Our long term goal is to develop ways to control delivery of glucose to the body, both in amount and rate, for glycemic control and release of certain small intestinal hormones through the ileal brake and gut-brain axis mechanisms that relate to relevant metabolic diseases. The aim, based our initial studies, was to characterize SI maturation

5 and activation as regulated by maltose or maltooligosaccharides using the Caco-2 cell model.

Specific Aim 1: Demonstrate the influence of maltooligosaccharides with different chemical structures on the maturation of SI by showing SI post-translational glycosylation event including trafficking, apical sorting, and activation in lipid rafts.

Specific Aim 2: Investigation of the molecular mechanism of the maturation of SI triggered by maltose and maltooligosaccharides. Genomic and metabolomics techniques were used to reveal the genes associated with the glycosylation of SI.

Research results showed the influence and mechanism involved in maturation and activation of the brush border α-glucosidases for glucose generation of dietary glycemic carbohydrates at the small intestinal enterocyte level, and could bring forth the development of novel technologies to modulate SI glucogenic enzyme activity and postprandial glycemia.

1.2 Significance of the Study

The hypothesize of this study was to investigate the mechanism of the maturation, trafficking, and activation of HMW SI induced by maltose sensing on the apical membrane of intestinal enterocytes. This hypothesis is indirectly supported by studies on rodents that “taste” starch and that this sensing (including starch-derived maltooligosaccharide) is distinct from sugar (mainly sucrose) taste (Sclafani 1987).

Maltose, as an important digestion product of starch by α-amylase, and symbolizes the presence of glucose-composed starch which is the main glycemic carbohydrate on which

6 humans evolved. On the other hand, sucrose is a source of carbohydrate nutrition from fruits and there is a known receptor for sucrose, the sweet taste receptor. From the viewpoint of evolution, there could be different receptors for maltose and sucrose representing different sources of carbohydrate in human diets.

Rapid digestion of glycemic carbohydrates and absorption of glucose can be detrimental to health, particularly for pre-diabetics and diabetics, due to glycemic spikes that stress the glucose homeostasis regulatory system. If the mechanism of recently observed SI maturation triggered by maltose is understood, an approach may be developed to manipulate the activity of brush border anchored enzymes so as to modulate postprandial glycemic response for improved health.

1.3 Thesis Organization

Chapter 1 is a brief introductory part that introduces background of carbohydrate digestion, and the research purpose and its significance.

Chapter 2 reviews literature on carbohydrates and mucosal α-glucosidases. This chapter has a review of protein structure, expression at the cell level, enzyme three dimensional properties, and distribution in the small intestine of the mucosal α- glucosidases. In addition, structure and function of the sweet taste receptor is reviewed.

Chapter 3 is the main experimental chapter. It investigates influence of dietary carbohydrates on structure and function of the intestinal α-glucosidases.

Chapter 4 is the second experimental chapter. It presents the effects of dietary carbohydrates on the physiological state of small intestinal enterocyte cells.

Chapter 5 provides an overall summary of the thesis, the major findings, and suggestions for future research.

1.4 References

Atkinson, F. S., Foster-Powell, K., & Brand-Miller, J.C. (2008). International tables of glycemic index and glycemic load values: 2008. Diabetes Care, 31(12), 2281-2283.

Cheng, M. W., Chegeni, M., Kim, K. H., Zhang, G. Y., Benmoussa, M., Quesada- Calvillo, R., Nichols, B. L., & Hamaker, B. R. (2014). Different sucrose-isomaltase response of Caco-2 cells to glucose and maltose suggests dietary maltose sensing. Journal of Clinical Biochemistry and Nutrition, 54(1), 55-60.

Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification and measurement of nutritionally important starch fractions. European Journal of Clinical Nutrition, 46, S33-S50.

Freeze, H. H. (1999). Monosaccharide Metabolism. Essentials of Glycobiology, 69-84.

Goda, T. (2000). Regulation of the expression of carbohydrate digestion/absorption- related genes. British Journal of Nutrition, 84(S2), S245-S248.

Gu, N., Adachi, T., Matsunaga, T., Tsujimoto, G., Ishihara, A., Yasda, K., & Tsuda, K. (2007). HNF-1α participates in glucose regulation of sucrase-isomaltase gene expression in epithelial intestinal cells. Biochemical and Biophysical Research Communications, 353(3), 617-622.

Holmes, R. (1971). Carbohydrate digestion and absorption. Journal of Clinical Pathology, 5, 10-13.

Jenkins, D. J., & Kendall, C. W. (2002). Glycemic index: overview of implications in health and disease. American Journal of Clinical Nutrition, 76(1), 266S- 273S.

Le Gall, M. L., Tobin, V., Stolarczyk, E., Dalet, V., Leturque, A., & Brot-Laroche, E. (2007). Sugar sensing by enterocytes combines polarity, membrane bound detectors and sugar metabolism. Journal of Cellular Physiology, 213(3), 834-843.

Levin, R. (1994). Digestion and absorption of carbohydrates-from molecules and membranes to humans. American Journal of Clinical Nutrition, 59(3), 690S-698S.

Margolskee, Robert F., Jane Dyer, Zaza Kokrashvili, Kieron SH Salmon, Erwin Ilegems, Kristian Daly, Emeline L. Maillet, Yuzo Ninomiya, Bedrich Mosinger, and Soraya P. Shirazi-Beechey. (2007). T1R3 and gustducin in gut sense sugars to regulate expression of Na+-glucose cotransporter 1. Proceedings of the National Academy of Sciences USA, 104(38), 15075-15080.

Sclafani, A., & Mann, S. (1987). Carbohydrate taste preferences in rats: glucose, sucrose, maltose, fructose and polycose compared. Physiology & Behavior, 40(5), 563-568.

Zukerman, S., Glendinning, J. I., Margolskee, R. F., & Sclafani, A. (2009). T1R3 taste receptor is critical for sucrose but not polycose taste. American Journal of Physiology- Regulatory, Integrative and Comparative Physiology, 296(4), R866-R876.

CHAPTER 2. REVIEW OF LITERATURE

2.1 Carbohydrates

In order to understand the regulation of carbohydrate digestion, it is important to briefly review carbohydrate structure and terminology. Carbohydrates are organic compounds consisting of carbon, hydrogen, and oxygen with a general composition of

Cx(H2O)y. They constitute more than 90% of the dry matter in nature and are important in living organisms as structural components of plant cell walls, backbone molecules of

DNA and RNA, linked to some lipids and proteins (Berg, 2002), and as energy sources.

Starch, lactose, and sucrose are the main dietary carbohydrates used by the body for energy, providing 70-80% of the calories in the human diet worldwide.

Carbohydrates are classified into three major catagories: monosaccharides, oligosaccharides, and polysaccharides (Fennema, 2005). Carbohydrates also are classifield based on dietary purpose. Monosaccharides and disacchardies are called simple sugar and oligosaccharides are called complex carbohydrates. Monosaccharides are the simplest form of carbohydrate with the formula (CH2O)n. Glucose, for example, is one of the monosaccharides with n=6. The classification of monosaccharides scheme is shown in Table 1.1 (Fennema, 2005).

Carbohyrates which contain of two to twenty monosaccharides units, are called oligosacchardies. Oligosaccharides are named based on number of monosaccharides units

11 such as disaccharides, trisaccharides, tetrasaccharides, and so forth. Sucrose, maltose, and lactose are simple oligosaccharides woth two monosaccharide units and they are the most important dietary disaccharides. Figure 1 illustrates examples of common disaccharides.

Sucrose is the most abundant disaccharide found in nature (also called simply sugar or table sugar) composed of α-D-glucopyranosyl and a β-D-fructofuranosyl units.

Maltose, maltotriose, maltotetraose, and maltopentose are α-1,4-linked glucose- containing oligosaccharides, or maltooligosaccharides produced in the gastrointestinal tract from starch hydrolysis by α-amylase. Maltose rarely occurs in nature and only in plants as a result of partial hydrolysis of starch. It is produced during malting of grains, such as of barley, and commercially by the specific enzyme-catalyzed hydrolysis of starch using β-amylase from Bacillus species, although the β-amylases from barley seed, soybeans, and sweet potatoes may also be used. Maltose is used sparingly as a mild sweetener in foods and can be reduced to the maltitol, which is used in sugarless chocolate. Oligosaccharides play several roles in living organisms (e.g. protein post- translational modification).

Polysaccharides are polymers of monosaccharides composed of glycosyl units, usually referring to linear or branched arrangements with more than 20 monosaccharide units. Polysaccharides are categorized in two groups base on their monosaccharides compositions. Polysaccharide with more than one kind of monosaccharide is called a heteroglycan or heteropolysaccharide. When a polysaccharide contains of only one type of monosaccharide, it is called a homoglycan or homopolysaccharide. Examples of homoglycans are starch amylose and amylopectin. Polysaccharides play two important functions for living organisms: they form the supporting structures and protection of

12 living organisms and they act as the storage energy source. In living cells, maintaining proper osmotic pressure is necessary to keep cells functioning well.

Starch is the main dietary available carbohydrate in foods which is present in grains, tubers, and some fruits and vegetables as storage energy polysaccharides. The structure and composition of amylose and amylopectin is different among different plant

(Kennedy et al., 1995). Starches contain 10,000 to over 1 million glucosyl units.

Carbohydrates are common components of foods, both naturally and as added ingredients. They can be chemically, enzymatically, and physically modified to improve their properties and functions. Sweeteners, in the form of sucrose and high-fructose corn syrup (hydrolyzed starch partially isomerized to fructose), also appear in the diet as additives to processed foods. In starch hydrolysates, starch molecules are partly hydrolyzed with acid and/or enzymes to produce blends containing different amounts of glucose, maltose, and maltooligosaccharides (Strickler, 1982). Most of these hydrolyzate products have high amount of sugar and can be used as sweeteners, while others are designed to have a low sugar content and are used or their bulking property. The variation in the composition of carbohydrate components and how they are metabolized in the body makes it important to understand the mechanisms driving their fate in the digestive system.

2.2 Carbohydrate Digestion

According to the Dietary Guidelines for Americans 2010 (USDA), dietary carbohydrates should provide 45-65% of total caloric intake. For the available

13 carbohydrates that are digested to glucose in the small intestine and utilized by the body for energy, there exist variation concerning the rate and extent of carbohydrate digestion and absorption (Jenkins et al., 1981; Englyst et al., 2003). Chronic consumption of foods with a fast rate of glucose production and absorption may be detrimental to health

(Ludwig, 2002). Under prolonged hyperglycemic conditions, increased intracellular glucose levels elevate the production of mitochondrial superoxides. These superoxides elevated the hexosamine pathway favoring energy storage by lipid synthesis (Brownlee,

2001). On the contrary, glucose released at a slow rate is likely associated with beneficial health effects by reducing the risk factors for chronic diseases (Jenkins et al, 2002). How to make glucose release more slowly along the gastrointestinal tract is one of the important keys to the carbohydrate nutrition and human health. Although there are some techniques that have been developed to make slowly digestible starches by focusing on the food itself, the regulation of carbohydrate digestion and absorption into the body at the small intestine enterocyte is also an important side to consider. A deep understanding of the carbohydrate digestion process in the gastrointestinal tract and the regulation mechanisms involved will be another basis for developing novel techniques for glycemic control.

2.2.1 α-Amylase

Food starches, commercial starch-based products, and the disaccharides of sucrose and lactose are the important nutritional carbohydrates for mammalian species, which have to be digested to their respective monosaccharides and absorbed first in order to be utilized by the body’s cells or tissues. Carbohydrate digestion in the human involves

14 the function of several enzymes. Salivary α-amylase is secreted in the buccal cavity to start the carbohydrate digestion (Gropper et al., 2004). α-amylase is an endoamylase with ability to hydrolyze the α-1,4 glucosidic linkages of amylose and amylopectin. The product of α-amylase is shorter linear chains form, i.e. maltose, maltotriose, maltotetraose, maltopentose, and α-limit dextrins (Semenza et al., 2001).

The pancreas also produces α-amylase in the small intestine. Pancreatic α- amylase plays important role in starch digestion (Lebenthal, 1987). This enzyme continues to hydrolyze the starches (and glycogen), forming the same collection of linear maltooligosaccharides and α-limit dextrins (Figure 2).

The branched α-limit dextrins are usually four to nine glucosyl units long and contain one or more α-1,6 branches. The two glucosyl residues that contain the α-1,6- glycosidic bond eventually become in digestion the disaccharide isomaltose, but α- amylase itself does not cleave these branched oligosaccharides all the way down to isomaltose. α-Amylase has no activity toward sugar-containing polymers other than glucose linked by α-1,4-bonds. α-Amylase also has little activity for the α-1,4-bond at the non-reducing end of a chain. Digestion of these branched linkage structures, as well as the linear maltooligosaccharides produced from α-amylase digestion of starch, is at the level of the intestinal glucosidases in the small intestine.

2.2.2 Intestinal α-Glucosidases

After the initial hydrolysis of starch by α-amylase, the products, and sucrose and lactose, are further digested by integral glycosidases found in the brush border

15 membranes of small intestinal enterocytes. These enzymes are MGAM, SI, and trehalase.

The substrates for each enzyme and their products are shown in Table 1.2 (Gray et al.,

1979; Gropper et al., 2004; Levin, 1994; Quezada-Calvillo et al., 2007b).

MGAM and SI each contain two catalytic subunits (Figure 1.3), an N-terminal subunit (maltase and isomaltase) bordering with the membrane and a C-terminal luminally-oriented subunit (glucoamylase and sucrase).

The MGAM and SI subunits are 60% identical in amino acid sequence. In addition, MGAM and SI are within complexes 40% identical. It has been suggested that

MGAM and SI evolved through duplication of an ancestral gene (Nichols et al. 2003).

Further, the catalytic site signature sequence WiDMNE is characteristic of GH31 subgroup 4 members (Ernst et al. 2006). Both MGAM and SI comprise a small cytosolic domain, a transmembrane domain (TMD), an O-glycosylated linker (O-link), and two homologous catalytic subunits (maltase, glucoamylase, isomaltase, sucrase) (Jones et al.

2011).

Galand (1989) reported that MGAM contributes about 20% of the maltase activity and SI contributes 80% in the small intestine. He demonstrated that sucrase plays an important role in hydrolyzing intermediate α-amylolyzed products of starch digestion in the small intestine. This is because mammals express SI more abundantly than MGAM

(Quezada-Calvillo et al., 2007).

Both MGAM and SI have exo-hydrolytic activity on α-1,4 (Gray et al., 1979;

Heymann et al., 1995; Robayo-Torres et al., 2006) and α-1,6 glucosidic linkages (Sim et al., 2010). However, isomaltase has much higher hydrolytic activity on the branch

16 linkage than maltase (Lin et al., 2012). The glucoamylase subunit of MGAM has much higher maltase activity than does maltase itself (Lee at al., 2014). These enzymes also have different hydrolysis properties on different chain-length oligosaccharides (Heymann et al., 1994) and α-glucosidic linkages (Lin et al., 2012). Quezada-Calvillo et al. (2007a) observed that MGAM become the dominant enzymes when substrate concentration is low.

2.2.3 Sucrase-Isomaltase

Because this thesis focuses on SI processing, and not that of MGAM, more detail about SI is provided. The SI mRNA in human contains 5481 nucleotide bases which translates a 1827 amino acid SI polypeptide (Hunziker et al., 1986). The primary structure of SI contains five different domains, including cytoplasmic segment, a single membrane-spanning anchor, a glycosylated stalk, isomaltase subunit and sucrase subunit

(Semenza, 1979) (Figure 1.4).

SI is synthesized in a single chain precursor form. The N-terminal subunit of this precursor form includes the isomaltase domain and the C-terminal subunit includes the sucrase subunit. After SI precursor is anchored to the brush border membrane with its N- terminal subunit, SI will be cleaved into the two catalytic subunits denoting the mature form enzyme complex (Hauri et al., 1982; Naim et al., 1988a; Naim et al., 1988b). The isomaltase subunits of SI represent the brush border membrane-bound ends. Figure 1.4 shows the way SI are anchored to the brush border (Sim et al., 2008).

Both subunits of SI are heavily N- and O-glycosylated with the N-glycosylation occurring at 16 potential N- and O-glycosylation sites (Naim et al., 1988). In addition, the stalk domain contains more Ser/Thr-rich sites for glycosylation in the vicinity of the membrane (Hauri et al., 1982; Naim et al., 1988b; Gorvel et al., 1991). The fully glycosylated precursor (pro-Slc, 245 kDa) is transported to the microvillus apical membrane where it is cleaved by trypsin (Naim et al., 1988) into sucrase and isomaltase which remain strongly associated with each other through non-covalent ionic interactions

(Semenza, 1979).

A high mannose intermediate (O-linked glycon) of SI or pro-SIh has a molecular weight of 210 kDa produced by rapid processing by rough endoplasmic reticulum membrane-bound glycosidases. SI is then glycosylated in the Golgi apparatus by additional O-linked and N-linked glycons to its mature form, or pro-SIc, (Mw=245 kDa)

(Kornfeld and Kornfeld, 1985). A study on individual subunits of SI shows that the glycosylated isomaltase subunit has a molecular weight of 145 kDa and glycosylated sucrase subunit of 130 kDa. Both sucrase and isomaltase contain eight N-linked glycan units. Moreover, the mature form of sucrase comprises at least four populations varying in their content of O-linked glycans (Naim et al., 1988).

N- and O-Glycosylation are essential co- and posttranslational modifications that are associated with the folding, trafficking, and biological function of SI (Varki, 1993).

N- and O-linked glycosylation are involved in trafficking and polarized sorting events beyond the endoplasmic reticulum. SI is associated with membrane micro-domains at the enterocyte apical membrane (detergent-resistant membranes or lipid rafts) and

18 glycosylation is a key factor in this pathway mechanism (Scheiffele et al., 1995; Hein et al., 2009). In addition to their role in trafficking mechanisms,

O-glycosylation is also implicated in enhancement of the enzymatic activity of SI in lipid rafts (Wetzel et al., 2009).

The catalytic centers of isomaltase and sucrase are the aspartate residues 505 and

1394 within the motif WDMNE (Hunziker et al., 1986). These residues and the entire catalytic regions are surrounded by several potential N- and O-glycosylation sites that likely affect the final conformation of Sl and, thus, would play a role in modulating the glycosyl hydrolase activities (Hoefsloot et al., 1988).

2.3 Carbohydrate Absorption

The absorption of monosaccharides in the small intestines can be devided into two phases, the movement of monosaccharides across the brush border membrane and the movement of monosaccharides across the basolateral membrane.

2.3.1 Monosaccharide Absorption

2.3.1.1 Movement of Monosaccharides Across the Brush Border Membrane

There are three mechanisms for monosaccharide absorption from the lumen across the brush border membrane of the small intestinal enterocytes: active transport, facilitated diffusion, and passive diffusion (Levin, 1994).

In active transport, molecules transfer across a membrane against a concentration gradient using a carrier protein. The active transport requires adenosine triphosphate as energy source. Glucose and galactose use sodium-dependent glucose transporter

(SGLT1) for transportation across intestinal enterocytes membrane.

Facilitated diffusion requires a carrier to transport molecules across a membrane without energy expenditure. This diffusion is triggered by concentration difference of the transported molecules. Glucose transporter 5 (GLUT5) is an example of facilated diffusion which transport fructose from the lumen of the small intestine into the small intestinal enterocyte cells (Dehnam and Levin, 1975; Dawson et al., 1987).

Passive diffusion is another transport system which is also caused by a concentration difference. These concentration gradients in molecules permeate the movement from high concentration to low concentration side of membrane. Passive diffusion does not require carrier proteins (Levin, 1994).

2.3.1.2 Movement of Monosaccharides Across the Basolateral Membrane

After monosaccharides transport into small intestinal enterocytes cells, another facilitative glucose transporter, GLUT2, transports these sugars across the basolateral membrane into the circulation system going to the liver. GLUT2 transfers all available glucose, fructose, and galactose out of the enterocyte into the intracellular space and into circulation (Thorens et al., 1992). The classical model of monosaccharide absorption is shown in Figure 1.5.

Another potential diffusive mechanism of glucose has been suggested. This involves intercellular movment of nutrients and ions in the small intestine where diffusion occurs into circulation system through tight junctions between enterocytes

(Kellett et al., 2008). The mechanism is that glucose transport via SGLT1 which induces a cytoskeletal rearrangement of the enterocyte involving restructuring tight junction and openning of the intercellular spaces. It is believed that this mechanism of glucose absorption could contribute 75% of all glucose absorption, but it is not able to be detected in vivo (Pappenheimer and Reiss, 1987). In this model, GLUT2 can be inserted into apical membrane of the enterocyte when there is high luminal sugar concentration, and has a role in sugar absorption (AU et al., 2002; Kellett and Brot-Laroche, 2005).

2.4 Carbohydrate Sensing

The sense of taste is important in assessing foods’ nutritive quality. Mammals can detect and differentiate between five taste qualities: sweet, bitter, sour, salty, and umami.

While all tastes are important, sweet and bitter tastes are particularly significance in obtaining the carbohydrates, and avoiding potentially hurmful substrates like alkaloids and other bitter tasting toxins (Breslin, 2006).

The sweet taste receptor is present throughout GI tract and has the ability to sense carbohydrates by a single heterodimeric receptor, T1R2/T1R3 (Hoon et al., 1999; Adler et al., 2000). Another protein which has the sugar sensing ability is GLUT2. As a basal- lateral transporter for monosaccharide to the blood stream, GLUT2 is a sugar sensor which causes increase in the expression of SI in the presence of glucose and fructose (Le

Gall et al., 2007). Recently, there has been work on the sweet taste receptor and its

21 importance in regulation of carbohydrate digestion and absorption (Treesukoso et al.,

2011). In the next section this receptor will be described in more detail.

2.4.1 Sweet Taste Receptor

The sweet taste receptor belongs to the G protein-coupled receptors (GPCR) family characterized by a large extracellular amino terminal domain, which is linked by a cysteine-rich domain to the 7TM domain (Xu et al., 2004).

Sweet taste receptor has two subunits (T1R2 and T1R3) which form a heterodimer and can detect wide range of compounds that humans describe as “sweet”, including natural and artificial sweeteners and some proteins (Li et al., 2002). Thus, the

T1R2/T1R3 receptor is assumed to be involved in sweet sensing. Key components of the lingual sweet taste transduction mechanism are also expressed in the small intestinal epithelium and, thereby, constitute a putative molecular mechanism for detection of intestinal carbohydrates (Nelson et al., 2001). These findings postulate a potential mechanism through which carbohydrates trigger vagal and hormonal feedback to alter gastric motor function.

There have been different mechanisms suggested for sugar sensing by the sweet taste receptor. Margolskee et al. (2007) proposed a mechanism by which sweet taste receptors in the small intestine act through T1R2 and T1R3 subunits. The main components of the sweet taste receptor pathways are the α, β, and γ subunits of gustducin

(Damak et al., 2003), phospholipase Cβ2 (PLCβ2), and transient receptor potential channel type M5, a Ca2+-activated cation channel (Mace et al., 2007). Sweet compounds

22 likely bind to heterodimer T1R2/T1R3 and activate one or more GPCR. α-Gustducin activates a specific second messenger cascades phospholipase C (PLCβ2) which increases intracellular Ca2+ concentration. Increased Ca2+ triggers the transient receptor potential (TRP) family TRPM5 and causes significant expression of gastrointestinal peptides (i.e., GLP-1 and GIP) into the basolateral portion of the cell (Damak et al.,

2003).

2.4.1.1 Carbohydrate Sensing by the Sweet Taste Receptor

In appetite studies, maltose sensing was recently reported at the site of the sweet taste receptor T1R2 and T1R3 subunits. This suggestion is based on the observation that

T1R2 and/or T1R3 knockout mice show no preference for maltose compared to wild type mice (Treesukoso et al., 2011). In addition, rodents showed preferences for flavored solutions that were paired with concurrent intragastric infusions of sugars or maltodextrins (Allen et al., 1966; Catonguay et al., 1985; Collier et al., 1968).

The sweet taste receptor can bind to different carbohydrates and this, in turn, has some effect on regulation of carbohydrate digestion and absorption. The expression of glucose transporters SGLT1 and GLUT2 in the small intestine increase sugar absorption during a meal (Booth et al., 1985; Cabanac and Johnson, 1983) Importantly, the up- regulation of sugar transporters can be increased by intragastric infusions of both nutritive and non-nutritive sugars (Cangan and Maller, 1974; Cantor and Eichler, 1977).

Gut sweet receptors also cause the release of incretin hormones (i.e., GLP-1 and GIP) in mice. This is based on a study that T1R3 knockout mice show decrease in GLP-1 release

(Kokrashvili et al., 2009).

Sugar sensing by the sweet taste receptor is supported by behavioral studies of rodents showing that starch taste (including starch-derived maltooligosaccharide) is distinct from sugar (mainly sucrose) taste (Sclafani, 1987). Recent studies (Zukerman et al., 2009) showed that T1R3, one of the subunits of the sweet taste receptor, is critical for sucrose sensing, but was not related to maltodextrin (Polycose™) sensing. The sensing receptor candidates are GPCRs composed of T1R2 and T1R3 (Bachmanov et al., 2001;

Max et al., 2001; Nelson et al., 2001). Additionally, genetically engineered mice missing one of the two subunits of the sweet taste receptor in the gut (either T1R2 or T1R3) have displayed very weak concentration-dependent responses to maltose, maltotriose, and maltotetraose, as measured by brief-access taste tests and electrophysiological methods

(Treesukoso et al., 2011).

2.4.2 Dietary Carbohydrate Sensing and α-Glucosidase Enzymes

Recent studies have been focused on the effects of different dietary carbohydrates on α-glucosidase expression and activity. Among the monosaccharides, fructose influenced expression and rate of small intestine enterocyte enzymes and transporters

(Goda, 2000). Sugar sensing is a complex process involving sugar transport, nutrient metabolism, and receptors on cell membranes throughout gastrointestinal track. Dietary sucrose and fructose have been shown to up-regulate the transcription of SI, while glucose down-regulates SI expression (Goda, 2000) by affecting the binding of hepatocyte nuclear factors (HNF)-1a to its promoter (Gu et al., 2007). The transcription factor Cdx-2 is also important for the expression of the SI gene, and a change of histone

H3 modification for Cdx-2 binding is observed during the transition of enterocytes from

24 the crypt to the villi (Suzuki et al., 2008). The expression of the SI gene induced by a high carbohydrate, low fat diet was related to acetylation of histone H3 and H4 as well as binding of HNF-1 and Cdx-2 (Honma et al., 2007).

Recently, maltose was shown to induce the formation of a higher molecular weight (HMW) SI species in Caco-2 cells (Cheng et al., 2014). Immunoblotting showed that most of the sugar treatments had a positive result on SI protein synthesis, but only maltose treatment showed synthesis of a second higher Mw protein band that was concentration-dependent. Protein sequencing of HMW SI using MALDI-TOF confirm that the HMW SI has the same sequence as SI in other carbohydrate treatments. Since SI is a glycoprotein that requires post-translational modification to add glycans in the Golgi apparatus for full processing and trafficking of the enzyme complex (with a higher molecular weight), the second band is speculated from this work to be the mature SI enzyme, to coincide with the digestion of the disaccharide of maltose.

Table 2.1 Classification of monosaccharides (Fennema, 2005).

Number of Kind of carbonyl group

carbon atoms Aldehyde Ketone

3 Triose Triulose

4 Tetrose Tetulose

5 Pentose Pentulose

6 Hexose Hexulose

7 Heptose Heptulose

8 Octose Octulose

9 Nonose Nonulose

Table 2.2 Substrates and products of small intestinal glycosidase enzyme. Hydrolysis Enzyme Substrate(s) Product(s) linkage(s) Lactase-phlorizin Lactose β-1,4 linkage Glucose, Galactose hydrolase

Trehalase Trehalose α-1,1 linkage Glucose

Sucrase-isomaltase Sucrose α-1,2 linkage of Glucose, Fructose sucrose

Linear and branched α-1,4 & α-1,6 Glucose oligosaccharides linkages of starch

Maltase- Linear or branched Mainly α-1,4 Glucose glucoamylase oligosaccharides linkage

(a) (b)

(c)

Figure 1.1. The disaccharides lactose (a), sucrose (b), and Figure 2.1 The structures maltose of common (c) ( Fennemadietary disaccharids, 2005) lactose (a), sucrose (b), and maltose (c) (Fennema, 2005).

Starch Linear Oligosaccharides Maltase α-amylase α-1,4 Branched Glucoamylase Oligosaccharides

Isomaltase α-1,4 Sucrase α-1,4

α-1,4 α-1,6 α-1,2

Glucose ! Figure'1)1:'Schematic'diagram'of'the'process'of'starch'digestion'by'α)amylase,'maltase) glucoamylase,'and'sucrase'isomataseFigure 2.2 Schematic diagram of the.' action of salivary and pancreatic α-amylases on starch. SI!demonstrates!hydrolytic!activity!on!branched!α@1,6!linkages,!which!is!complemented!by! the!hydrolytic!activity!of!both!SI!and!MGAM!on!α@1,4!linkages!(Quezada@Calvillo!et!al.!2008);! however,!the!main!substrate!of!SI!is!the!α@1,4!linkage !(Nichols!et!al.!2003).!The!complementary! activities!of!the!human!enzymes!allow!for!digestion!of!the!spectrum!of!food!starches!from!more! than!400!plants!that!comprise!two@thirds!of!most!hu man!diets;!SI!is!more!abundant!but!MGAM! displays!higher!hydrolytic!activities!(Heymann!and!Günther!1994;!Heymann,!Breitmeier,!and! Günther!1995;!Robayo!Torres,!Quezada!Calvillo,!and!Nichols!2006) .!In!studies!of!severely! malnourished!infants,!it!has!been!demonstrated!that!the!terminal!steps!of!MGAM!α@1,4!and!SI!α@

1,6!starch!digestion!are!decreased.!The!loss!of!MGAM!message!parallels!the!reduction!in!villin! message!and!degree!of!villous!atrophy,!and!also!parallels! reduction!in!the!SI!message.!It!has!also! been!shown!that!the!correlations!between!the!two!enzyme!activities!(SI!and!MGAM)!exist!at!the! mRNA!level!(Nichols!et!al.!2000).!Recent!studies!demonstrate!that!feeding!rats!a!high! fat/carbohydrate!ratio!diet!reduces!the!jejunal!sucrase/isomaltase!(S/I)!activity!ratio.!In! addition,!the!SI@linked!oligosaccharides!show!modifications!of!unsialylate d!galactose!as!well!as! the!mannose@linked!form!when!the!S/I!activity!ratio!changes,!and!thus!is!thought!to!be! regulated!by!altering!the!glycosylation!of!SI!(Samulitis@Dos!Santos,!Goda,!and!Koldovsky!1992;!

! 2!

TMD O-link Maltase Glucoamylase

Small intestinal Cytosol lumen

TMD O-link Isomaltase Sucrase

Figure 2.3 Position and protein organization of the MGAM and SI complex. TMD: transmembrane domain; O-link: O-glycosylated linker (adapted from Sim et al., 2008).

30 CHAPTER 27 ■ DIGESTION, ABSORPTION, AND TRANSPORT OF CARBOHYDRATES 499

C Can the glycosidic bonds of the structure shown hereafter be hydrolyzed by ␣-amylose?

Sucrase CH2OH CH2OH O O

O O N OH OH C

OH OH n

Isomaltase

Connecting segment (stalk) Acarbose is an FDA-approved drug that blocks the activities of pancre- Transmembrane atic ␣-amylase and brush border segment ␣-glucosidases (with specificity for glucose). The drug is produced from a microorganism N Sucrase– Cytoplasmic and is a unique tetrasaccharide. Acarbose is isomaltase domain given to patients with type 2 diabetes, with the purpose of reducing the rate at which ingested FIG. 27.5. The major portion of the sucrase–isomaltase complex, containing the catalytic carbohydrate reaches the bloodstream after Figuresites, 2 .4protrudes The structure from the of absorptive the sucrase cells-isomaltase into the lumen complex of the intestine.(Lieberman, Other 2013) domains. of a meal. This is one approach to better control the protein form a connecting segment (stalk) and an anchoring segment that extends through the membrane into the cell. The complex is synthesized as a single polypeptide chain that is blood glucose levels in such patients. Weight split into its two enzyme subunits extracellularly. Each subunit is a domain with a catalytic loss has not been associated with use of site (distinct sucrase–maltase and isomaltase–maltase sites). In spite of their maltase activity, this drug, but flatulence and diarrhea (due to these catalytic sites are often called just sucrase and isomaltase. colonic bacterial metabolism of the sugars) are side effects of taking this drug.

polypeptide chain forms two globular domains, each with a catalytic site. In glucoamylase, the two catalytic sites have similar activities, with only small differences in substrate specificity. The protein is heavily glycosylated with oligosaccharides that protect it from digestive proteases. Glucoamylase is an exoglucosidase that is specific for the ␣-1,4-bonds between Maltose glucosyl residues (Fig. 27.6). It begins at the nonreducing end of a polysaccharide or limit dextrin and sequentially hydrolyzes the bonds to release glucose α-1,4 bond monosaccharides. It will digest a limit dextrin down to isomaltose, the glucosyl OO disaccharide with an ␣-1,6-branch, which is subsequently hydrolyzed principally HO O OH by the isomaltase activity in the sucrase–isomaltase complex. maltase activity 2. SUCRASE–ISOMALTASE COMPLEX 12

The structure of the sucrase–isomaltase complex is very similar to that of O O O reducing glucoamylase, and these two proteins have a high degree of sequence homology. HO O O end However, after the single polypeptide chain of sucrase–isomaltase is inserted Maltotriose through the membrane and the protein protrudes into the intestinal lumen, an intestinal protease clips it into two separate subunits that remain attached to each FIG. 27.6. Glucoamylase activity. Glucoamy- other. Each subunit has a catalytic site that differs in substrate specificity from lase is an ␣-1,4-exoglycosidase that initiates cleavage at the nonreducing end of the sugar. the other through noncovalent interactions. The sucrase–maltase site accounts for Thus, for maltotriose, the bond labeled 1 is approximately 100% of the intestine’s ability to hydrolyze sucrose in addition hydrolyzed first, which then allows the bond at to maltase activity; the isomaltase–maltase site accounts for almost all of the position 2 to be the next one hydrolyzed.

G Ga F

SGLT1 GLUT5

Ga G F

GLUT2 GLUT2

G Ga F Blood

Figure 2.5 The absorption and transfer out of sugars by the enterocyte. G, glucose; Ga, galactose; F, fructose (adapted from Paulev, 1999).

2.5 References

Adler, E., Hoon, M. A., Mueller, K. L., Chandrashekar, J., Ryba, N. J., & Zuker, C. S. (2000). A novel family of mammalian taste receptors. Cell, 100(6), 693-702.

Allen, R. J. L., & Leahy J. S. (1966). Some effects of dietary dextrose, fructose, liquid glucose and sucrose in the adult male rat. British Journal of Nutrition, 20(02), 339-347.

Atkinson, F. S., Foster-Powell, K., & Brand-Miller, J.C. (2008). International tables of glycemic index and glycemic load values: 2008. Diabetes Care, 31(12), 2281-2283.

Au, A., Gupta, A., Schembri, P., & Chesseman, C. I. (2002). Rapid insertion of GLUT2 into the rat jejunal brush-border membrane promoted by glucagon-like peptide 2. Biochemical Journal, 367, 247-254.

Bachmanov, A. A., Li X., Reed, D. R., Ohmen, J. D., Li S, Chen, Z., Tordoff, M. G., de Jong, P. J, Wu C, West, D. B., Chatterjee, A., Ross, D. A., & Beauchamp, G. K. (2001). Positional cloning of the mouse saccharin preference (Sac) locus. Chemical Senses, 26(7), 925-933.

Beaulieu, J.-F., Nichols, B., & Quaroni, A. (1989). Posttranslational regulation of sucrase-isomaltase expression in intestinal crypt and villus cells. Journal of Biological Chemistry, 264(33), 20000-20011.

Berg, J. M., Tymoczko, J. L., & Stryer, L. (2002). Biochemistry 5th edition. W. H. Freeman and Company. New York. 295 – 303.

Booth, D. A., Lovett, D., & McSherry, G. M. (1972). Postingestive modulation of the sweetness preference gradient in the rat. Journal of Comparative and Physiological Psychology, 78(3), 485-512.

Breslin, P. A., & Huang, L. (2006). Human taste: peripheral anatomy, taste transduction, and coding. Adv. Otorhinolaryngol, 63,152-90.

Brownlee, M. (2001). Biochemistry and molecular cell biology of diabetic complications. Nature, 414(6865), 813-20.

Cabanac, M., & Johnson, K. G. (1983). Analysis of a conflict between palatability and cold exposure in rats. Physiology & Behavior, 31, 249-253.

Cagan, R. H., & Mailer, O. (1974). Taste of sugars: Brief exposure single-stimulus behavioral method. Journal of Comparative and Physiological Psychology, 87(1), 47-55.

Cantor, M. B., & Eichler, R. J. (1977). Sweetness: A supernormal reinforcer. Chemical Technology, 7, 214-216.

Castonguay, T. W., Philips, S., & Collier, G. H. (1985). Sucrose procurement cost and dietary selection. Nutrition & Behavior, 2(4), 201- 21.

Collier, G., & Bolles, R. (1968). Some determinants of intake of sucrose solutions. Journal of Comparative and Physiological Psychology, 65(3p1), 379-383.

Conklin, K. A., Yamashiro, K. M., & Gray, G. M. (1975). Human intestinal sucrase- isomaltase. Identification of free sucrase and isomaltase and cleavage of the hybrid into active distinct subunits. Journal of Biological Chemistry, 250(15), 5735-5741.

Dahlqvist, A., & Borgstrom, B. (1959). Characterization of intestinal invertase as a glucosido-invertase. II: Studies on transglycosylation by intestinal invertase. Acta Chemica Scandinavica, 13(3), 1659-1667.

Damak, S., Rong, M., Yasumatsu, K., Kokrashvili, Z., Varadarajan, V., Zou, S., Jiang, P., Ninomiya, Y., & Margolskee, R. F. (2003). Detection of sweet and umami taste in the absence of taste receptor T1r3. Science, 301, 850-853.

Dawson, D. J., Burrows, P. C., Lobley, R. W., & Holmes, R. (1987). The kinetics of monosaccharide absorption by human jejunal biopsies: evidence for active and passive processes. Digestion, 38(2), 124-132.

Dehnam, E. S., & Levin, R. J. (1975). An experimental method of identifying and quantifying the active transfer of electrogenic component from the diffusive component during sugar absorption measured in vivo. Journal of Physiology, 246(1), 181-196.

Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification and measurement of nutritionally important starch fractions. European Journal of Clinical Nutrition, 46, S33-S50.

Englyst, K. N., Vinoy, S., Englyst, H. N., & Lang, V. (2003). Glycaemic index of cereal products explained by their content of rapidly and slowly available glucose. British Journal of Nutrition, 89(3), 329-339.

Ernst, H. A., & Lo Leggio L. (2006). Structure of the Sulfolobus solfataricus α- Glucosidase: Implications for Domain Conservation and Substrate Recognition in GH31. Journal of Molecular Biology, 358(4), 1106-1124.

Fennema, O.R. (2007). Fennema’s Food Chemistry, (fourth ed.), CRC Press, 1170p. ISBN:0849392721.

Freeze, H. H. (1999). Monosaccharide Metabolism. Essentials of Glycobiology, 69-84.

Goda, T. (2000). Regulation of the expression of carbohydrate digestion/absorption- related genes. British Journal of Nutrition, 84(S2), S245-S248.

Gorvel, J. P., Ferrero, A., Chambraud, L., Rigal, A., Bonicel, J., & Maroux, S. (1991). Expression of Sucrase-Isomaltase and Dipeptidylpeptidase-Iv in Human Small-Intestine and Colon. Gastroenterology, 101(3), 618-625.

Grand, R. J., Watkins, J. B., & Torti, F. M. (1976). Development of the human gastrointestinal tract. Gastroenterology, 70(5), 790-810.

Gray, G. M., Lally, B. C., & Conklin, K. A. (1979). Action of intestinal sucrase- isomaltase and its free monomers on an α-Limit Dextrin. Journal of Biological Chemistry, 254(13), 6038-6043.

Gropper, S. S., Smith, J. L., & Groff, J. L. (2004). The digestive system: Mechanism for nourishing the body. In: Advanced Nutrition and Human Metabolism. Thomson/Wadsworth, California. 4th ed. 24-53.

Hauri, H. P., Wacker, H., Rickli, E. E., Bigler-Meier, B., Quaroni, A., & Semena, G. (1982). Biosynthesis of sucrase-isomaltase Purification and NH2-terminal amino acid sequence of the rat sucrase-isomaltase precursor (pro-sucrase-isomaltase) from fetal intestinal transplants. Journal of Biological Chemistry, 257(8), 4522-4528.

Heine, M., Rohwedder, A., & Naim, H. Y. (2009). Impact of glycosylation and detergent resistant membranes on the function of intestinal sucrase-isomaltase. Biological Chemistry, 390(7), 545-549.

Heymann, H., & Günther, S. (1994). Calculation of subsite affinities of human small intestinal glucoamylase–maltase. Biological Chemistry Hoppe-Seyler, 375(7), 451–455.

Heymann, H., Breitmeier, D., & Günther, S. (1995). Human small intestinal sucrase– isomaltase: different binding patterns for malto- and isomaltooligosaccharides.Biological Chemistry Hoppe-Seyler, 376(4), 249–253.

Hoefsloot, L. H., Hoogeveenwesterveld, M., Kroos, M. A., Vanbeeumen, J., Reuser, A. J. J., & Oostra, B. A. (1988). Primary Structure and Processing of Lysosomal Alpha- Glucosidase - Homology with the Intestinal Sucrase Isomaltase Complex. Embo Journal 7(6), 1697-1704.

Hoffman, L. R., & Chang, E. B. (1991). Determinants of regional sucrase-isomaltase expression in adult rat small intestine. Journal of Biological Chemistry, 266(32), 21815- 21820.

Hoffman, L. R., & Chang, E. B. (1993). Regional Expression and Regulation of Intestinal Sucrase-Isomaltase. Journal of Nutritional Biochemistry, 4(3), 130-142.

Holmes, R. (1971). Carbohydrate digestion and absorption. Journal of Clinical Pathology, 5, 10-13.

Honma, K., Mochizuki, K., & Goda, T. (2007). Carbohydrate/fat ratio in the diet alters histone acetylation on the sucrase-isomaltase gene and its expression in mouse small intestine. Biochemical and Biophysical Research Communications, 357(4), 1124-1129.

Hoon, M. A., Adler, E., Lindemeier, J., Battey, J. F., Ryba, N. J., & Zuker, C. S. (1999). Putative mammalian taste receptors: a class of taste-specific GPCRs with distinct topographic selectivity. Cell, 96(4), 541-551.

Hunziker, W., Spiess, M., Semenza, G., & Lodish, H. (1986). The sucrase-isomaltase complex: primary structure, membrane-orientation, and evolution of a stalked, intrinsic brush border protein. Cell, 46, 227-234.

Jenkins, D., & Wolever, T. (1981). Glycemic index of foods: a physiological basis for carbohydrate exchange. American Journal of Clinical Nutrition, 34(3), 362- 366.

Jenkins, D. J., & Kendall, C. W. (2002). Glycemic index: overview of implications in health and disease. American Journal of Clinical Nutrition, 76(1), 266S- 273S.

Jones, K., & Sim, L. (2011). Mapping the intestinal alpha-glucogenic enzyme specificities of starch digesting maltase-glucoamylase and sucrase-isomaltase. Bioorganic & Medicinal Chemistry, 19(13), 3929-3934.

Kellett, G. L., & Brot-Laroche, E. (2005). Apical GLUT2: a major pathway of intestinal sugar absorption. Diabetes, 54(10), 3056-3062.

Kellett, G. L., Brot-Laroche, E., Mace, O. J., & Leturque, A. (2008). Sugar absorption in the intestine: the role of GLUT2. Annual Review of Nutrition, 28, 35- 54.

Kennedy, J. F., Knill, C. J., Taylor, D. W. (1995). Handbook of Starch Hydrolysis Products and their Derivatives. Blackie Academic., London. 65–82.

Kitagawa, M., Kusakabe, Y., Miura, H., Ninomiya, Y., & Hino, A. (2001). Molecular genetic identification of a candidate receptor gene for sweet taste. Biochemical and Biophysical Research Communications, 283(1), 236-242.

Kokrashvili, Z., Mosinger, B., & Margolskee, R. F. (2009). T1r3 and alpha-gustducin in gut regulate secretion of glucagon-like peptide-1. Annals of the New York Academy of Sciences, 1170(1), 91-94.

Kornfeld, R., & Kornfeld, S. (1985). Assembly of asparagine-linked oligosaccharides. Annual Review of Biochemistry, 54(1), 631-664.

Lebenthal, E. (1987). Role of salivary amylase in gastric and intestinal digestion of starch. Digestive Disease and Science, 32(10), 1155-1157.

Lee, M. F., Russell, R. M., Montgomery, R. K., & Krasinski, S. D. (1997). Total intestinal lactase and sucrase activities are reduced in aged rats. Journal of Nutrition, 127(7), 1382-1387.

Lee, B. H., Lin, A. H., Nichols, B. L., Jones, K., Rose, D. R., Quezada-Calvillo, R., & Hamaker, B. R. (2014). Mucosal C-terminal maltase-glucoamylase hydrolyzes large size starch digestion products that may contribute to rapid postprandial glucose generation. Molecular Nutrition & Food Research, 58(5), 1111-1121.

Levin, R. (1994). Digestion and absorption of carbohydrates-from molecules and membranes to humans. American Journal of Clinical Nutrition, 59(3), 690S-698S.

Li, X., Staszewski, L., Xu, H., Durick, K., Zoller, M., & Adler, E. (2002). Human receptors for sweet and umami taste. Proceedings of the National Academy of Sciences USA, 99(7), 4692-4696.

Lin, A. H.-M., Lee, B.-H., Nichols, B. L., Quezada-Calvillo, R., & Hamaker, B. R. (2012). Starch source influences dietary glucose generation at the mucosal α-glucosidase level. Journal of Biological Chemistry, 287(44), 36917–36921.

Ludwig, D. S. (2002). The glycemic index. Journal of the American Medical Association, 287(18), 2414-2423.

Mace, O. J., Affleck, J., Patel, N., Kellett, G. L. (2007). Sweet taste receptors in rat small intestine stimulate glucose absorption through apical GLUT2. Journal of Physiology, 582(1), 379-392.

Max, M., Shanker, Y. G., Huang, L., Rong, M., Liu, Z., Campagne, F., Weinstein, H., Damak, S., & Margolskee, R. F. (2001). Tas1r3, encoding a new candidate taste receptor, is allelic to the sweet responsiveness locus Sac. Nature Genetics, 28(1), 58-63.

Naim, H. Y., Sterchi, E. E., & Lentze, M. J. (1988a). Biosynthesis of the human sucrase- isomaltase complex. Differential O-glycosylation of the sucrase subunit correlates with its position within the enzyme complex. Journal of Biological Chemistry, 263(15), 7242- 7253.

Naim, H. Y., Sterchi, E. E., & Lentze, M. J. (1988b). Structure, biosynthesis, and glycosylation of human small intestinal maltase-glucoamylase. Journal of Biological Chemistry, 263(36), 19709-19717.

Nelson, G., Hoon, M. A., Chandrashekar, J., Zhang ,Y., Ryba, N. J., & Zuker, C. S. (2001). Mammalian sweet taste receptors. Cell, 106(3), 381-390.

Nichols, B. L., Avery, S., Sen, P., Swallow, D. M., Hahn, D., & Sterchi, E. (2003). The maltase-glucoamylase gene: common ancestry to sucrase-isomaltase with complementary starch digestion activities. Proceedings of the National Academy of Sciences USA, 100(3), 1432-1437.

Pappenheimer, J.R., & Reiss, K. Z. (1987). Contribution of solvent drag through intercellular junctions to absorption of nutrients by the small intestine of the rat. Journal of Membrane Biology, 100(1), 123-136.

Paulev, P. E. (1999-2000). Gastrointestinal function and disorders. In: Medical physiology and pathophysiology - Essentials and Clinical Problems. Copenhagen Medical Publishers.

Quezada-Calvillo, R., Robayo-Torres, C. C., Ao, Z., Hamaker, B. R., Quaroni, A., Brayer, G. D., Sterchi, E. E., Baker, S. S., & Nichols, B. L. (2007a). Luminal substrate "brake" on mucosal maltase-glucoamylase activity regulates total rate of starch digestion to glucose. Journal of Pediatric Gastroenterology and Nutrition, 45(1), 32-43.

Quezada-Calvillo, R. Robayo-Torres, C. C., Opekun, A. R., Sen, P., Ao, Z., Hamaker, B. R., Quaroni, A., Brayer, G. D., Wattler, S., Nehls, M. C., Sterchi, E. E., & Nichols, B. L. (2007b). Contribution of mucosal maltase-glucoamylase activities to mouse small intestinal starch α-glucogenesis. Journal of Nutrition, 137(7), 1725-1733.

Robayo-Torres, C. C., Quezada-Calvillo, R., & Nichols, B. L. (2006). Disaccharide digestion: clinical and molecular aspects. Clinical Gastroenterology and Hepatology, 4(3), 276-287.

Scheiffele, P., Peranen, J., & Simons, K. (1995). N-Glycans as Apical Sorting Signals in Epithelial-Cells. Nature, 378(6552), 96-98.

Sclafani, A. (2007). Sweet taste signaling in the gut. Proceedings of the National Academy of Sciences USA, 104(38), 14887-14888.

Sclafani, A., & Mann, S. (1987). Carbohydrate taste preferences in rats: glucose, sucrose, maltose, fructose and polycose compared. Physiology & Behavior, 40(5), 563-568.

Semenza, G., Auricchio, S., & Mantei, N. (2001). The Metabolic Basis of Inherited Disease. McGraw- Hill, New York. 8th ed. 1623-1650.

Semenza, G. (1979). Mode of insertion of the sucrase-isomaltase complex in the brush border membrane: Implications for the biosynthesis of the stalked intrinsic membrane protein. In Ciba Foundation Symposium 70-Development of Mammalian Absorptive Processes. John Wiley & Sons, Ltd. 133-146.

Sim, L., Willemsma, C., Mohan, S., Naim, H. Y., & Hamaker, B. R. (2010). Structural basis for substrate selectivity in human maltase- glucoamylase and sucrase-isomaltase N- terminal domains. Journal of Biological Chemistry, 285(23), 17763-17770.

Sim, L., Quezada-Calvillo, R., Sterchi, E. E., Nichols, B. L., & Rose, D. R. (2008). Human intestinal maltase–glucoamylase: crystal structure of the N-terminal catalytic subunit and basis of inhibition and substrate specificity. Journal of Molecular Biology, 375(3), 782-792.

Strickler, A. J. (1982). Corn syrup selections in food applications. In: Food Carbohydrates. Edited by D. R. Lineback and G. E. Inglett. Westport, CT: Avi Publishing Co. Inc. 12-24.

Suzuki, T., Mochizuki, K., & Goda, T. (2008). Histone H3 modifications and Cdx-2 binding to the sucrase-isomaltase (SI) gene is involved in induction of the gene in the transition from the crypt to villus in the small intestine of rats. Biochemical and Biophysical Research Communications, 369(2), 788-793.

Thorens, B. (1992). Molecular and cellular physiology of GLUT2, a high-Km facilitated diffusion glucose transporter. International Review of Cytology, 137(A), 209–238.

Treesukosol, Y., & Spector, A. C. (2011). The functional role of the T1R family of receptors in sweet taste and feeding. Physiology & Behavior, 105(1), 14-26.

Trugnan, G., Rousset, M., Chantret, I., Barbat, A., & Zweibaum, A. (1987). The posttranslational processing of sucrase-isomaltase in HT-29 cells is a function of their state of enterocytic differentiation. Journal of Cell Biology, 104(5), 1199-1205.

Varki, A. (1993). Biological Roles of Oligosaccharides: All of the Theories Are Correct. Glycobiology, 3(2), 97-130.

Wetzel, G., Heine, M., Rohwedder, A., & Naim, H. Y. (2009). Impact of glycosylation and detergent-resistant membranes on the function of intestinal sucrase-isomaltase. Biological Chemistry, 390(7), 545-549.

Xu, H., Staszewski, L., Tang, H., Adler, E., Zoller, M., & Li, X. (2004). Different functional roles of T1R subunits in the heteromeric taste receptors. Proceedings of the National Academy of Sciences of the United States of America USA, 101(39), 14258- 14263.

CHAPTER 3. DIETARY CARBOHYDRATES INFLUENCE THE STRUCTURE AND FUNCTION OF SUCRASE-ISOMALTASE

3.1 Introduction

Based on the Dietary Guidelines for Americans 2010 (USDA), dietary carbohydrates should provide 45-65% of total caloric intake. Available carbohydrates, however, are not all equal and in particular there exist significant variations in rate and extent of carbohydrate digestion and absorption that relate to its nutritional quality

(Jenkins et al., 1981; Englyst et al., 2003). Chronic consumption of foods with fast and high rate of glucose production and absorption may be detrimental to health (Ludwig,

2002) through stresses on glucose homeostasis controls, advanced glycation end-product formation, and favoring energy storage by hexosamine pathway which leads to lipid synthesis (Brownlee, 2001). This is particularly relevant for diabetics and pre-diabetics, where control of blood glucose level is a problem. On the other hand, glucose released at a slow rate is likely associated with health benefits through reduction of risk factors for some diet-associated metabolic diseases (Jenkins et al., 2002). How to make glucose release and absorption slower is one of the keys to improving carbohydrate nutrition and its relation to human health. Although there are some techniques to make low glycemic index (GI) foods, such as slowly digestible starches, by focusing on food factors; the results seem to be variable. Another possible and viable strategy is to control postprandial

42 glycemic excursions through control of the carbohydrate digestion and absorption processes. This might be through natural inhibitors to the α-glucosidases and SGLT-1, or through another mechanism that reduces rate of glucose delivery to the body. Thus, a fundamental and mechanistic understanding of the carbohydrate digestion and absorption processes could be the basis for developing novel techniques for glycemic control.

Starch is the main dietary glycemic carbohydrate in foods. It is digested first by α- amylase which hydrolyzes the α-1,4 glycosidic linkages between glucosyl units of linear chains. The degradation products are composed of maltose and other oligosaccharides with different structures features (size and both linear and branched oligomers). These degradation products are then hydrolyzed by the brush border-anchored α-glucosidases,

MGAM and SI, to glucose for absorption. Although pancreatic α-amylase is generally thought of as the dominant enzyme in starch digestion (Englyst et al., 1992), MGAM and

SI are the ultimate control point for glucose liberation and their activities and amount are critical to glycemic control, as evidenced by the study of Nichols et al. (2009) showing a decrease of 40% glucose absorption for MGAM null mice, that had only SI, compared to the wild-type control. The mucosal α-glucosidases, and their control, are thus critical for the health outcome of dietary glycemic carbohydrates.

Sucrase-isomaltase is a common marker for differentiated polarized small intestinal cell cultured enterocytes. Its biosynthesis and structure have been extensively studied. Two different subunits of SI, which are associated through ionic interaction after protease cleavage, are present in the apical membrane of absorptive brush border villus cells, or enterocytes, and the isomaltase subunit at the N-terminal is the site for membrane anchoring (Hunziker et al., 1986). SI is synthesized by co-translational

glycosylation as a single chain precursor (‘high mannose precursor’, pro-SIh, Mw 210 kDa) that has low enzyme activity. After the precursor is processed by further glycosylation in the Golgi apparatus, a mature ‘complex precursor’ (pro-SIc , Mw 245 kDa) containing additional N- and O-linked oligosaccharide chains is generated with full enzyme activity (Beaulieu et al., 1989), and then it is inserted into the apical membrane of the small intestine enterocytes through apical sorting and trafficking (Hauri, 1979).

The cleavage by trypsin, the small intestine luminal protease, generates two subunits with different substrate specificities. The sucrase subunit (C-terminal) is known for sucrose digestion though has other activities, and the N-terminal subunit can cleave both α-1,6 and α-1,4-linked glucans (Gray et al., 1979; Heymann et al., 1995; Robayo-Torres et al.,

2006). The integrity of the SI structure is important for its function as it was shown that disruption of the detergent resistant membrane (DRM) with Triton X-100, the anchoring location, substantially reduces its activity, and modification of its glycosylation pattern decreases its association with DRMs, and also decreases enzyme activity (Wetzel et al.,

2009).

Composition of the diet (carbohydrate) influences the expression of SI, which is termed dietary adaptation. Dietary sucrose, fructose, and starch all have been shown to up-regulate the transcription of SI, while glucose down-regulates SI expression (Goda et al., 2000) by increasing the binding of hepatocyte nuclear factors (HNF)-1a to its promoter (Gu et al., 2007). In addition to gene transcription regulation, SI glycosylation occurs at its translational and post-translational stages and is essential to the maturation of the SI enzyme active form. Beaulieu et al (1989) studied SI expression regulation in

44 small intestine intestinal crypt and villus cells, and showed that post-translational regulation is critical for the enzyme activity of SI in differentiated villus cells.

Sugar sensing is essential to understand the molecular mechanism of SI expression. GLUT2, as a basolateral transport for monosaccharides to the blood stream, is also a sugar sensor for high expression of SI (at the mRNA level). This is termed the

GLUT2-dependent signaling pathway (Le Gall et al., 2007). Another important sugar sensor is the sweet taste receptor composed of two subunits T1R2, T1R3 and the G protein gustducin (Gαgust) on the enteroendocrine cells of gastrointestinal tract. The receptor can bind sucrose, artificial sweeteners (sucralose) and other sugars to stimulate absorptive enterocyte SGLT1 expression through the function of hormones like GIP or

GLP-1secreated by enteroendocrine cells (Margolskee et al., 2007). Another study showed the existence of T1R2, T1R3, and G protein gustducin in the enterocytes that can sense sugar molecules, although no effect on SI expression was reported (Mace et al.,

2007). However, no receptor for maltose, one of the products from starch by α-amylase digestion, has been reported in literature.

Recently, our laboratory showed that maltose induces the formation of a higher molecular weight (HMW) SI species in Caco-2 cells (Cheng et al., 2014).

Immunoblotting showed that most of the sugar treatments had a positive result on SI protein synthesis, but only maltose treatment showed synthesis of a second higher Mw protein band that was concentration-dependent. Protein sequencing of HMW SI using

MALDI-TOF confirmed that the HMW SI has the same sequence as SI in other carbohydrate treatments. Since SI is a glycoprotein that requires post-translational modification to add glycans in the Golgi apparatus for full processing and trafficking of

45 the enzyme complex (with a higher molecular weight), the second band is speculated from this work to be the mature SI enzyme, and coinciding with the digestion of the disaccharide of maltose. Thus, the objectivw of this study was to investigate the mechanism of the maturation and trafficking of HMW SI caused by maltose sensing on apical membrane of intestinal enterocytes. If the mechanism of SI maturation triggered by maltose is understood, an approach may be developed to manipulate the activity of brush border anchored enzymes so as to modulate postprandial glycemic response for improved health.

3.2 Materials and Methods

3.2.1 Immunochemical Reagents

Monoclonal mouse anti-SI antibodies (mAb) HSI2, HBB1/691/79, HBB2/614/88,

HBB2/219/20, and HBB3/705/60, to recognize different conformations and glycoforms of SI, were provided by Dr. H.P. Hauri (University of Bern, Bern, Switzerland). A mixture of these antibodies was used for immunoprecipitation of the different glycoforms of SI. Horseradish peroxidase-coupled goat anti-rabbit and goat anti-mouse and rabbit anti-goat antibodies were from Dako (Glostrup, Denmark). HybondTM-P PVDF membrane was obtained from Amersham Biosciences (Freiburg, Germany), and enhanced chemiluminescence (ECL) from Pierce (Thermo Fisher Scientific (Bonn,

Germany). Cell culture media, fetal calf serum, and antibiotics were purchased from PAA

(Pasching, Austria). Secondary antibodies coupled to Alexa Fluor dyes were obtained

46 from Invitrogen (Karlsruhe, Germany) and horseradish peroxidase-conjugated secondary antibodies were from Thermo Fisher Scientific (Bonn, Germany).

3.2.2 Cell Culture Procedure

Caco-2 cells HTB37 (ATCC, Manassas, VA) (passage 30 to 45) were used for this study. Cells were grown in Dulbecco’s modified Eagle’s medium (DMEM)

(Corning, Lowell, MA) supplied with 10% fetal bovine serum (FBS) (Lonza,

Walkersville, MD), 50 µg/ml gentamycin sulfate (J R Scientific Inc., Woodland, CA), 10 mM HEPES, 100 µg/ml streptomycin and 100 U/ml penicillin (Lonza, Walkersville,

MD), and 100 µM non-essential amino acid (Corning, Lowell, MA). Cells were incubated at 37°C with 5% CO2, 95% air atmosphere, and at constant humidity. Cells were grown to confluence for the complete differentiation of the cells according to the procedures of Mahraoui et al. (1994) and Le Gall et al. (2007).

3.2.3 Cell Treatment

Cells on transwells were washed twice with 1 ml PBS after they reached 100% confluence and were fully differentiated. Cells were then fed with glucose-free DMEM

(Gibco, Carlsbad, CA). Test carbohydrates of maltose, maltooligosaccharides

(maltotriose), and glucose (as control) were added to the glucose-free DMEM (Gibco,

Carlsbad, CA). Different carbohydrate concentrations were used with mannitol to adjust the osmolarity of the media and the treatment time points were adjusted according to each experiment (Cheng, 2009).

3.2.4 Total Protein Extraction

Cells were washed twice with 1X PBS and harvested in 1 ml of 1X ice cold PBS followed by centrifugation at 14,000 rpm for 3-5 min at 4°C. After discarding the PBS supernatant, cell pellets were lysed on ice for 20 min using 50- 60 µl of lysis buffer consisting of 1% Nonidet P-40 (NP-40), 0.5% sodium deoxycholate, 150 mM NaCl, 20 mM Tris - HCl pH 8.0, 5% SDS, 10 mM DTT, and 1mM PMSF. During the lysis process, cells were vigorously shaken every 5 min for 20 min. Cells were then centrifuged at 14,000 rpm for 10 min at 4oC to remove cell debris. Protein concentration was measured using the BCA protein assay kit according to the manufacturer’s protocol

(Pierce, Rockford, IL). Cell lysates were denatured and stabilized by adding sample solvent (final concentration after adding together was 10% glycerol, 0.1% bromophenol blue, 2% SDS, 50 mM Tris-Cl pH 6.8, and 10 mM DTT) and then placed in a boiling water bath for 5 min. Samples were then directly used for SDS-PAGE or stored at -80°C

(Cheng, 2009).

3.2.5 Western Blot Analysis

For Western blot analysis, equal amounts of protein-containing lysates (35 µg) were loaded onto 8% SDS-PAGE mini-gels and separated by electrophoresing with 35 mA/gel constant current for 1 hour. The separated proteins were then electrotransferred at

100 V to PVDF membranes (BioRad, Hercules, CA) at room temperature with ice in the transfer cassette for 1.5 h. The membranes were blocked with phosphate buffered saline containing 5% skim milk (BioRad, Hercules, CA) and 0.1% Tween 20 (BioRad,

Hercules, CA) at room temperature for 1-2 hours and then allowed to react with 1/1500

48 dilution of mouse anti-SI antibodies (B. Nichols, Baylor College of Medicine) (mixed with equal volume of HSI 1/691/79, HIS 3/190, and HIS 3/42/1/2; Quezada-Calvillo et al

2007). After incubation with horseradish peroxidase-conjugated secondary antibody

(1/2000 dilution, anti-mouse and anti-goat, from Santa Cruz Biotechnology, Santa Cruz,

CA; anti-rabbit from Amersham, GE Healthcare Biosciences, Piscataway, NJ), proteins were visualized using enhanced chemiluminescence (ECL) (Pierce, Rockford, IL or

Amersham of GE Healthcare Biosciences, Piscataway, NJ) followed by exposure to

Amersham Hyperfilm (Cheng, 2009).

3.2.6 Membrane Protein Extraction

Apical (brush border) membranes fraction were obtained as follows. Cells (5x106 cells) were washed 3 times with cold PBS and homogenized using glass/Teflon Potter-

Elvehjem homogenizer in 2 mM-Tris/50mM-mannitol.

Homogenates were centrifuged at 2 700 g for 10 min and supernatants were spun at 207 000 g for 30 min. Resulting pellets were designated 'crude membrane fractions '. A mixture of protease inhibitors was added to all buffers and solutions used for homogenization and membrane purification.

3.2.7 Preparation of Brush Border Membranes

Brush border membranes of Caco-2 cells were isolated by the modified (Sterchi and Woodley, 1980) divalent cation precipitation method (Schmitz et al., 1973). Briefly, the cells were homogenized using a Potter-Elvehjem homogenizer in a hypertonic

49 homogenization buffer (2 mM Tris-HCl, 50 mM mannitol, pH 7.1) supplemented with a mixture of protease inhibitors. The homogenates were passed through a Luer-21 gauge needle and CaCl2 was added to a final concentration of 10 mM. Following 30 minutes of gentle agitation at 4°C, the homogenates were centrifuged at 2,000 g for 20 min to obtain basolateral and intracellular membranes. Finally the supernatant was centrifuged at

25,000 g for 30 min to yield the brush border membranes. Where appropriate, solubilization of the membranes was performed with 1% Triton X-100 in cold PBS for 2 hours followed by high-speed centrifugation and the supernatant was further used for biochemical analyses.

3.2.8 Pulse-chase and Metabolic Labeling

In brief, treated cells were starved for 2 h and labeled with [35S]methionine (ICN

Biomedicals, Meckenheim, Germany). The dishes were rinsed once with 80 µCi L-

[35S]methionine in 5 ml of methionine-free (Met-free) DMEM containing 10% dialyzed fetal calf serum, 50 units/ml penicillin, and 50 µg/ml streptomycin (Met-free medium).

The cells were incubated in 10 ml of Met-free medium 2 hours at 37oC to deplete their methionine pool. Thereafter, the cells were labeled with 80 µCi of [35S]methionine in 10 ml of Met-free medium for 30 min, followed by a chase with nonlabeled methionine for

1, 2, 4, 8, 12, 24, and 48 hours. The cells were placed on ice and rinsed three times with cold PBS, and cells corresponding to one dish were solubilized with 1 ml of cold lysis buffer containing 25 mM Tris-HCI, pH 8.0, 50 mM NaC1, 0.5% sodium deoxycholate,

0.5% Triton X-100, 1 mM phenylmethanesulfonyl fluoride, 1 g/ml pepstatin, 5 µg/ml leupeptin, 5 µg/ml antipain, and 1 µg/ml aprotinin. The cell extracts were centrifuged for

15 min at 1,000 g to remove nuclei and debris. The supernatant was retained and spun at

100,000 g for 1 hour at 4°C and immunoprecipitated with anti-SI antibodies. SI was also immunoprecipitated from purified human intestinal brush border membranes. In this case,

1 mg/ml membranes in 25 mM Tris-HC1, pH 8.0, supplemented with 50 mM NaCl were solubilized with Triton X-100 and sodium deoxycholate (0.5% final concentrations) by stirring on ice for 30 min. The solubilized material was spun at 100,000 g, and the supernatant was processed further for immunoprecipitation with anti-SI antibodies as described below (Naim et al., 1987).

3.2.9 Immunoprecipitation and SDS-PAGE

Metabolically-labeled Caco-2 cells were lysed at 4°C for 1 hour in lysis buffer (25 mM Tris-HCl, pH 8.0, 50 mM NaCl, 0.5% Triton X-100, 0.5% sodium deoxycholate and a mixture of protease inhibitors containing 1 mM PMSF, 1 µg/ml pepstatin, 5 µg/ml leupeptin, 80 µg/ml soybean trypsin inhibitor, and 1 µg/ml aprotinin, all from Sigma Co.,

Deisenhofen, Germany). Ice-cold lysis buffer (1 ml) was used for each 100 mm culture dish (about 2-4×106 cells). Detergent extracts of cells were centrifuged for 1 hour at

100,000 g at 4°C and the supernatants were immunoprecipitated. For immunoprecipitation, first the lysate was pre-cleared with protein A-Sepharose alone and then the preconjugated immunobeads were incubated with the sample with gentle mixing at 4°C for 1 hour. Finally, the immunoprecipitants were washed three times with each wash buffer I (0.5 % Triton X-100 and 0.05 % deoxycholate in PBS) and wash buffer II

(125 mM Tris, 10 mM EDTA, 500 mM NaCl, 0.5 % Triton X-100, pH 8.0), sequentially.

Usually, 0.2 µl mAb in the form of ascites was used for each immunoprecipitation.

Human α-glucosidase was immunoprecipitated using 0.5 µl of a rabbit polyclonal anti- human α-glucosidase serum (Reuser et al., 1985). SDS-PAGE was performed according to the method of Laemmli (1970) and the apparent molecular masses were assessed by comparison with high molecular mass markers (Bio-Rad Laboratories GmbH, München,

Germany) run on the same gel. In some experiments, deglycosylation of the immunoprecipitates with Endo-N-acetylglucosaminidase H (Endo H) or Endo-N- acetylglucosaminidase F/gly- copeptidase F (Endo F/GF, also known as PNGase F) (both from New England Biolabs GmbH, Schwalbach/Taunus, Germany) was performed prior to SDS-PAGE analysis (Naim et al., 1987).

3.2.10 Isolation of Detergent Resistant Membranes (DRMs)

Caco-2 cells were collected in PBS and solubilized with 1 % Triton X-100 on ice, briefly homogenized by pipetting, and incubated at 4°C for 2 hours with gentle shaking.

The lysates were then centrifuged at 5,000 g for 10 min at 4°C and the supernatants were subjected to stepwise sucrose gradient ultracentrifugation at 100,000 g for 18 hours at

4°C. Finally, 1 ml fractions were collected from the top and examined by immunoblotting for the target proteins.

3.2.11 Immunoassays

In immunoprecipitation experiments, Caco-2 cells were solubilized with Triton X-

100 (1% w/v in PBS) for 2 hours at 4°C and a post-nuclear supernatant (lysate) was obtained by a short high speed centrifugation. The supernatant was subjected to

52 ultracentrifugation (100,000 g for 1 hour). Supernatant (S1) was used in immunoprecipitation and the pellet (P1) (DRM) was solubilized further with deoxycholate (0.5 % in PBS) at 4°C for 1 hour. This lysate was centrifuged briefly to separate insoluble particles (P2) and the antibody for immunoprecipitation was added to its supernatant (S2).

3.2.12 Enzyme Activity Measurement

After cells were cultured and treated with glucose (25 mM) and maltose (12.5 mM) as described above, brush border membrane (BBM) containing mostly the mature form of SI was prepared as described in Section 3.2.7. The enzyme assay was performed with 20 µl of BBM in PBS, pH 6.5, and incubated with 20 µl of different substrate concentrations of sucrose (0, 10, 15, 20, 30, 50, 100 mM) or maltose (0, 2, 4, 6, 8, 10, 20,

40 mM) for 1 hour at 37°C. Activities were calculated based on detection of the released glucose by adding 200 µl GOPOD mono-reagent (Megazyme, Ireland) using a microplate reader at 550 nm and glucose as a standard. Each sample was compared to a similar zero time sample for the mean activity. The reported unit of activity was µM.h-1.mg-1.

3.2.13 Total RNA Isolation

Total RNA of cells were stabilized by storing in RNAlater solution (Ambion,

Austin, TX) at 4°C for overnight. Equipment used for RNA extraction was soaked in

DEPC-H2O (diethyl dicarbonate, 0.1% v/v; Sigma Co., St. Louis, MO) for at least 1 hour at 37°C followed by autoclaving at 121°C for 30 min to inactivate RNases. Total RNA

53 was extracted using the SV Total RNA Isolation System (Promega, Madison, WI) according to manufacturer’s protocol. RNA samples were diluted and measured at 260 and 280 nm. RNA concentration was calculated by A260 according to the Beer-Lambert law and RNA purity was determined by the ratio of A260/A280 (Cheng, 2009).

3.2.14 cDNA Synthesis and Real-Time PCR Analysis

Relative levels of mRNA expressed in the treated Caco-2 cells of T1R2, T1R3, α- gustducin, SI, GLUT2, GLUT5, and 18S rRNA were quantified by qRT-PCR. Oligo dT

(500 µg/ml, Promega, Madison, WI) was used as primer for first-strand cDNA synthesis of total RNA. qRT-PCR was performed using two systems. Some samples were reverse transcripted by using the Access qRT-PCR System (Promega, Madison, WI) according to manufacturer’s instructions. qRT-PCR of other samples (20 µl) were performed by heating a mixture [1 µl of Oligo dT, 1-2 µg of total RNA, 1 µl of 10 mM dNTP mix

o (Invitrogen, Grand Island, NY) and 16 µl of DEPC-H2O] at 65 C for 5 min and then by adding M-MuLV Reverse Transcriptase Reaction Buffer (BioLabs, Ipswich, MA) and 25 units of M-Mulv reverse transcriptase (BioLabs, Ipswich, MA) to react at 42°C for 1 hour. The reaction was inactivated by heating at 95°C for 5 min. qRT-PCR analysis was performed by using the MX3000P QPCR system (Stratagene, La Jolla, CA) in the presence of SYBR-green (Cheng, 2009). The primers used were:

• T1R2 primers, 5’- TGG CAT TTA TCA CGG TAC TCA AA-3’ and 5’- AGT

ACG GGT GGT GGG ACT GA-3’

• T1R3 primers, 5´-CAC ACG CTC CTG CAG CTT CG-3´ and 5´-CCA TGC

CCA GGA CAC AGA G-3´

• α-gustducin primers, 5´- TTG TGC TGC ACT TAG TGC CTA TG -3´ and 5´-

GAA GGC TTT CAT GCA TTC TAT TCA-3´

• SI primers, 5´-CAT CCT ACC ATG TCA AGA GCC AG-3´ and 5´-GCT TGT

TAA GGT GGT CTG GTT TAA ATT-3´

• GLUT2 primers, 5’-GTC CAG AAA GCC CCA GAT ACC-3’ and 5 ́-GTG ACA

TCC TCA GTT CCT CTT AG-3 ́

• GLUT5 primers, 5’-TCT CCT TGC AAA CGT AGA TGG-3’ and 5’-GAA GAA

GGG CAG CAG AAG G-3’

• 18S rRNA primers, 5´-ATC AGA TAC CGT CGT AGT TC-3 and 5-CCA GAG

TCT CGT TCG TTA T-3´

Partial sequences were matched to corresponding human full-length sequences by

GenBank BLAST searches. The qRT-PCR reaction was performed using Brilliant II

QPCR Master Mix (Stratagene, La Jolla, CA) scaled down to 25 µl per reaction. The reaction followed standard conditions and the melting temperature was 55°C. qRT-PCR results were expressed as fold change of relative amount of mRNA (Schmittgen and

Livak, 2008; User Bulletin of ABI Prism 7700 Sequence Detection System, 1997).

3.2.15 Quantification of Band Intensities

Specific bands on PVDF membranes detected by antibodies were visualised using the ChemiDoc XRS Molecular Imager (BioRad, Hercules, CA) device. Digital images obtained from the imager were quantified by image processing and analysis software

(ImageJ, 1.37v).

3.2.16 Statistical Analysis

Data are shown as means ± S.E.M. Statistical analysis was performed using SAS

9.4 software. T-test was used to determine the significance of treatment effect.

Much of these studies were done with our collaborator, Professor Hassan Naim and Dr. Mahdi Amiri at the University of Veterinary Medicine in Hannover, Germany.

3.3 Results

3.3.1 Intracellular Distribution and Compartmentalization of SI

Intracellular sorting of SI in Caco-2 cell fed with glucose or maltose was examined to gain insights into the cell compartment distribution of HMW SI induced by maltose and to compare it to SI in cells fed with glucose. For this, cell-surface immunoblotting of pro-SI was performed with Caco-2 cells treated with maltose (12.5 and 25 mM) or glucose (25 and 50 mM). Results showed that more pro-SIc appeared predominantly at the apical membrane in Caco-2 cells fed with maltose after 12 hours

(Figure 3.1). This increase in sorting of SI into the apical membrane was concentration- dependent. More HMW SI was observed in in cells fed with 25 mM maltose compared to 12.5 mM maltose. In addition, the results show that HMW SI distributed more to apical membrane and that higher glycosylation on HMW SI did not impair SI reaching the cell surface. Relative protein amount (P2/P1 ratio) showed that enrichment of SI in the brush border fraction increased substantially (about 6 fold) in Caco-2 cells fed with maltose compared to cells fed with glucose (Table 3.1). The greater intensity of SI bands

56 in cells fed with maltose suggests that maltose induced more and faster trafficking of

HMW SI from intracellular fraction (P1) to brush border membrane (P2) (Figure 3.2).

3.3.2 The Kinetics of Glycosylation Alterations

A pulse-chase experiment was conducted to understand the glycosylation process of SI and identify and assess the size of the precursor and mature forms of HMW SI.

After differentiation, Caco-2 cells were fed with glucose or maltose, labeled with

[35S]methionine for 15 minutes and were collected in different times (0, 1, 2, 4, 8, 12, 24, and 48 hours). SI was immunoprecipitated with monoclonal antibodies and SDS-PAGE used to visualize SI bands (Figure 3.3). The intensity of each band was measured to compare the amount of SI in different time points (Figure 3.4). A single high mannose polypeptide MW≅ 210 kDa (SIh) was detected after 15 minutes of labeling. SI in cells fed with glucose did not show any molecular weight difference from SI in cells fed with maltose. After 1hour, a second band which indicates the mature form of SI MW≅ 245 kDa was revealed. SI in Caco-2 cells fed with maltose appeared faster with higher intensity and higher molecular weight (≅265kDa) indicating faster formation of HMW SI. At 2 hours and 4 hours, a population of SI reached the Golgi apparatus and acquired more O- and N-glycosylation to develop the mature SIc form. In the cells fed with maltose, the intensity and molecular weight of SIh stayed high compare to cells fed with glucose. After

8 hours to 24 hours, a smear population of SIh with different high molecular weights was observed only in cells fed with maltose which disappeared and became a single band at

48 hours.

To determine the glycosylation nature of the SI population with smear effect after

8 hours, the experiment was conducted using the endo F/GF deglycosylation enzyme.

After feeding differentiated cells with glucose and maltose for 12 hours, cells were labeled with [35S]methionine for 15 minutes and collected in different times (2, 4, 8, and

24 hours). Then SI was immunoprecipitated with monoclonal antibodies and subjected to endo F/GF enzyme. SDS-PAGE was used to visualize SI bands (Figure 3.5). With the endo F/GF enzyme, glycan of some portion of HMW SI was cleaved and reduced in molecular weight. However, the majority of HMW SI remained of a similar molecular weight, perhaps indicating the diversity of HMW SI glycan species.

3.3.3 Association of SI with Detergent-Resistant Microdomains

It has been shown that SI is associated with lipid-raft on apical membrane of intestinal epithelial cells (Danielsen and Van Deurs, 1997). To assess if the higher molecular weight pro-SIc was associated with lipid raft microdomains at the apical membrane, a detergent-resistant microdomain experiment (DRM experiment) was performed with Caco-2 cells fed with DMEM contained 25 mM glucose (as control) compared to 12.5 mM maltose for 12 hours. Cells then were solubilised with 1% Triton

X-100 and the supernatants and pellets were further processed for immunoblotting.

Pellets contained the SI in DRM portion and the supernatant contained the non-lipid raft portion of SI. Within 12 hours, in comparison to the control, HMW SI was found to be more associated with Triton X-100 lipid rafts (Figure 3.6), indicating higher apical sorting of these species, as further supported by quantification of the brush border

58 membrane of the Caco-2 cells (Table 3.2). The intensity of SI was measured and showed significantly higher amount of SI in DRM (Figure 3.7).

3.3.4 Sucrase-Isomaltase Enzyme Activity

The activity of SI was measured in Caco-2 cells fed with maltose compared to cells fed with glucose. For this analysis, brush border membranes were separated from intracellular and basolateral membrane through the discriminatory effect of divalent ions

(Ca2+ or Mg2+). Brush border membranes do not precipitate with these divalent ions and can be separated in a relatively pure form from other membranes by centrifugation.

Usually the enrichment factor of the brush border membranes in the P2 fraction (see

Materials and Methods) correlates with a well developed brush border membrane

(Chantret et al., 1988; Schmitz et al., 1973). This factor was assessed by comparison of the activities of the intestinal differentiation of SI in the P2 fraction versus P1 and the total cellular homogenates.

Brush border membranes were prepared in Caco-2 cells fed with glucose (0, 25, and 50 mM) or maltose (12.5 and 25 mM). Sucrose (200 mM) and maltose (50 mM) were used as substrates for measuring SI activity. Here, the sucrase activity of SI in the P2 fraction versus P1 and the total cellular homogenates was lower in cell fed with maltose compared to cells fed glucose (Figure 3.8). The activity of SI in different cell fractions decreased when sucrose was used as a substrate; there were substantially lower SI activity levels when sucrose was used as the substrate (Table 3.3).

The maltase activity of SI in the P2 fraction versus P1 and the total cellular homogenates was same in cells fed with maltose compared to cells fed with glucose

(Figure 3.9). The maltase activity of SI in different cell fractions decreased, but not significantly, when maltose was used as the substrate (Table 3.4).

For further investigation of the enzyme kinetics of digestion of the disaccharides, maltase and sucrase activity of SI was measured using different concentrations of both substrates and cells fed with glucose or maltose. The data were analyzed using

Michaelis–Menten kinetics (Figures 3.10, 3.12) and presented as Lineweaver-Burk plots

(Figure 3.11, 3.13). Vmax and Km parameters were determined for each sample. There was no significant change in Vmax and Km of maltase and sucrase activities in cells fed with maltose or glucose (Table 3.5).

3.3.5 Deglycosylation Treatment of Treated Cells

To determine the size, nature (high mannose/complex type), and type of linkage

(N-linked/O-linked) of the glycan units present on SIh and SIc, enzymic deglycosylation was performed. For this objective, differentiated cells were fed with glucose or maltose and then labeled with [35S]methionine. After cells were collected and lysed, the immunoprecipitated SI was treated with endo F/GF, SI and endo H (Figure 3.15) deglycosylation enzymes (see Materials and Methods). Endo F/GF removes almost all N- linked (Asn-linked) glycosylated moieties. Endo H, on the other hand, removes only high mannose and some hybrid types of N-linked carbohydrate moieties.

As shown in Figure 3.14, SI immunoprecipitated from cells fed with glucose and treated with endo F/GF revealed predominately pro-SIC which converted upon treatment with endo F/GF to a MW≅205 and 185 kDa species. Endo F/GF treated SI immunoprecipitated from cells fed with maltose produced the similar digested SI with both MW≅205 and 185 kDa species showing that HMW SI contains N-link glycan.

The endo H treatment of immunoprecipitated SI to remove high mannose glycans is shown in Figure 3.15. SI immunoprecipitated from cells fed with glucose and treated with endo H lowered the MW of pro-SIc to 210 kDa and the high mannose band (SIh) to

134 KD. Endo H-treated SI immunoprecipitated from cells fed with maltose produced the same digested SIh at 134 kDa, but the complex form of SI (pro-SIc) had a higher molecular weight band for pro-SIc (MW≅265).

3.3.6 Gene Expression of the Sweet Taste Receptor with Maltose Exposure

In appetite studies, mice and rats were reported to have different carbohydrates preference (including maltose) (Faller et al., 2004; Treesukosol, 2009). These results suggest a stimulation of sweet taste receptor subunits (T1R2 and T1R3) by different carbohydrates. Sweet taste receptor is a G protein-coupled receptors (GPCRs) expressed through gastrointestinal tract (GI) (Hoon et al. 1999; Adler et al. 2000; Margolskee et al.,

2007). To investigate if maltose induces sweet taste receptor expression, the relative quantity of mRNA for different genes (SI, T1R2, T1R3, α-gustducin, GLUT2, and

GLUT5) were measured in Caco-2 cells fed with glucose or maltose. Since GLUT5 has no function in glucose transportation and GLUT2 is positively affected by glucose, these

61 proteins were used as controls. For this study, confluenced Caco-2 cells were fed with glucose or maltose, and mRNA was extracted. qRT-PCR was used to measure mRNA expression of the genes encoding these proteins.

Results from qRT-PCR revealed a higher gene expression of both sweet taste receptor subunits in the presence of maltose compared to Caco-2 cells fed with glucose.

This increase in T1R3 mRNA level in cells fed with maltose was significantly higher

(compared to cells fed with glucose) in the first 2 hours and after 24 h (Figure 3.16). The increase was greater in the T1R2 subunit in the first 2 hours (Figure 3.17). There was no significant difference in mRNA level of α-gustducin between treatments (Figure 3.18).

Expression of the SI gene in cells fed with maltose increased over time (Figure 3.19). The highest transcriptional level of SI mRNA was observed at 12 hours followed by degradation of SI mRNA after 24 hours suggesting a dynamic change in SI gene transcription and degradation of SI mRNA.

Cells appeared to accumulate more GLUT5 mRNA initially at 2 hours in the presence of glucose and maltose (Figure 3.20). However the amount of expression was greater in cells treated with glucose. In general, cells displayed higher GLUT5 mRNA levels in the first 6 hours of treatment with glucose and maltose. In the presence of both glucose and maltose, a constant decreasing trend of GLUT5 mRNA was observed after

12 hours. Both carbohydrates seemed to have a positive effect on GLUT2 gene expression. Caco-2 cells fed with maltose showed increase in level of GLUT2 mRNA compared to the glucose (Figure 3.21) over the 24 hours treatment period.

3.4 Discussion

It has been known that the presence of monosaccharides (e.g., glucose, fructose) in small intestinal enterocytes alter intracellular events, including SI gene expression, and trigger gut hormone responses (Kishi et al., 1999; Goda et al., 2000; Le Gall et al., 2007;

Margolskee et al., 2007). However, sensing of α-amylase starch degradation products

(maltooligosaccharides) by gut enterocytes with effect on SI had not been reported until recently with the publication of an article by Cheng et al. (2014) from our laboratory suggesting such a relationship. In appetite studies, carbohydrate sensing (including maltose) was reported at the site of the sweet taste receptor T1R2 and T1R3 subunits

(Sclafani, 2007; Treesukosol et al., 2009), however it was not related to SI processing or starch maltooligosaccharide product digestion.

In Cheng et al. (2014), we reported that maltose appears to induce the transformation of SI to a HMW SI species in Caco-2 cells. Analyzing the HMW SI by a combination of MALDI-TOF and MS/MS suggested that this maltose-induced event was due to SI glycosylation. They proposed that maltose, in contrast to other sugars tested

(i.e., glucose, fructose, sucrose, isomaltose), acts as a signal molecule to trigger HMW SI processing and trafficking to the apical membrane. The data presented in the current study provide a better understanding of the sensing mechanism, and maturation and trafficking of this HMW SI species in Caco-2 cells in the presence of maltose.

SI is a highly glycosylated protein with N- and O-glycosylation sites that are added during co-translational and post-translational processing before the final insertion

63 of the enzyme in the microvillar membrane of the small intestinal enterocytes (Hunziker et al., 1986). Polypeptides can be modified in the rough endoplasmic reticulum (RER) either during their synthesis (co-translational) or after the initial synthesis has been completed (post-translational). N- and O-glycosylation of SI are essential modifications that are associated with folding and intracellular trafficking (Naim et al., 1988). SI is sorted to the apical membrane via O-linked glycans that mediate its association with lipid rafts (Wetzel, 2009) which is associated with increase of SI activity (Heine et al., 2009).

Here, we report on changes in SI glycosylation and sorting as induced by maltose, a surrogate for the array of maltooligosaccharides that are presented to small intestine enterocytes from the digestion of starch by α-amylase.

Deglycosylation of HMW SI was done using endo F and endo H enzymes to investigate the structure of HMW SI. The endo F treatment cleaves the N-linked glycosylation sites and endo H treatment cleaves most of (but not all) the O-linked glycosylation sites. Endo F treatment resulted in reduced molecular weight of HMW SI

(induced by maltose) to the original molecular weight of SI associated with normal feeding of glucose to the cells. Endo H treatment of HMW SI (induced by maltose) somewhat reduced the molecular weight, but it was still higher than the original SI molecular weight. The deglycosylation experiments showed that there is a higher degree of N-glycosylation than O-glycosylation on the HMW SI. N- and O-glycosylation are involved in processing and trafficking of SI (Naim et al., 1999; Wetzel et al., 2009), therefore we can conclude that SI in presence of maltose gets both N- and O-linked glycans, though more N-, which produces faster trafficking of SI.

In the DRM experiment, regions of apical membrane which are resistance to detergent were isolated. These regions, also called the lipid rafts, have a unique structure.

Lipid rafts are enriched in sphingolipids such as sphingomyelin and cholesterol (Pike,

2009). Some proteins such as SI are associated with these regions, meaning that more SI is found at the enterocyte apical membrane in lipid raft regions (Le Bivic et al., 1990). In addition, SI has more enzyme activity when they in lipid raft region (Alfalah et al., 1999).

In the DRM experiment, we assessed the association of SI with the lipid raft region on the enterocyte apical membrane. SI is sorted to the apical membrane via O-linked glycans that mediate its association with DRMs (Jacob et al., 2000), and the deglycosylation experiment described above showed that O-glycosylation occurred with maltose exposure. There was a higher amount of HMW SI in membrane in cells fed with maltose compare to SI to cells fed with glucose, and HMW SI induced by maltose was more associated with the DRM fraction. The association of SI with DRMs requires an intact transmembrane domain and minimal stretch of the Ser/Thr-rich stalk O-glycosylated domain (Jacob et al., 2000). Jacob et al. (2000) showed that alteration of O-glycosylation in the stalk region damaged the trafficking and maturation of SI. It is, therefore, likely that the additional glycosylation caused by the presence of maltose on SI did not occur on stalk region. Since Jacob et al. (2000) showed that O-glycosylation of SI is the main factor in trafficking of SI to lipid raft, it is also probable that a different O-glycosylation pattern on HMW SI could have caused more association of HMW SI with DRMs (Wetzel et al., 2009), as was found in the present experiment.

The pulse-chase experiment clearly shows that the SI in cells fed with maltose was processed at a faster rate compared to SI in cells fed with glucose. We observed the

formation of pro-SIc with greater band intensity after 1 h. The molecular weight of the pro-SIc induced by maltose was also greater (≅ 265 kDa) than SI in cells induced with glucose (245 kDa). After 8 h, SI in cells with maltose treatment began to become more glycosylated and formed a smear; this population existed until 24 h then was reduced to one single HMW SI. When this smeared higher molecular weight population of SI was treated with endo F enzyme, the molecular weight of HMW SI population decreased but the smear still existed. Since endo F only cleaves N-linked glycans, the smear effect might be due to a variety of O-glycosylation types on SI in presence of maltose. The reason for higher O-glycosylation on HMW SI is not completely clear to us. Base on

DRMs and pulse-chase results, it seems these additional N- and O-glycosylations have a direct impact on maturation and trafficking of HMW SI, and that there may be various forms that could have different trafficking or other (e.g., enzyme activity) functions.

This finding was confirmed by biochemical analysis of the distribution of SI in cells fed with glucose or maltose. The results from this experiment showed that a greater proportion of SI appeared at the apical membrane when maltose was fed to the cells. This shows that the trafficking of SI in cells fed with maltose increased with the new N- and

O-glycosylations on HMW SI. A remarkable characteristic of intestinal SI is its heterogeneous O-glycosylation pattern revealed by the expression of multiple differently

O-glycosylated molecules (Naim et al., 1988). This higher N- and O-glycosylation of SI in the presence of maltose correlates with the high amount of SI at apical membrane.

We asked the question of whether the new glycosylation pattern of SI induced by maltose, affected its enzyme activity. It has been shown that disruption of the DRM with

Triton X-100 substantially reduces its activity (Wetzel et al., 2009). The results from

66 enzyme activity experiment showed that sucrase activity significantly decreased HMW SI of cells induced by maltose, however, the total maltase activity was similar to SI of cell induced by glucose. Naim et al. (1988) showed that the sucrase subunit in SI contains at least four populations varying in their content of O-linked glycans. Taken together with the higher O-glycosylation of HMW SI noted above, this may be due to different O- glycosylation events on the sucrase subunit. There was no change in maltase activity in

HMW SI. The next step in evaluating enzyme activity of HMW SI was to check if the enzyme followed Michealis-Menten kinetics. The Michaelis–Menten and Lineweaver-

Burk shows that Vmax and Km decreases slightly in HMW SI induced by maltose. Since the maltase activity was not changed, we conclude that HMW SI form induced with maltose has better binding to maltose with similar enzyme activity for digesting maltose.

Another question posed in the current study was related to the investigation of a potential receptor for maltose sensing which could trigger the processing of HMW SI. In an appetite study, maltose preference was reported in mice and rats (Faller et al., 2004;

Treesukosol, 2009). These studies suggested a stimulation of sweet taste receptor subunits (T1R2 and T1R3) by maltose related to appetite. Since these studies showed potential binding of maltose to sweet taste receptor, we tested the gene expression of sweet taste receptor subunits (T1R2/T1R3) and SI in presence of maltose.

The sweet taste receptor signaling pathway might be involved in maltose sensing which induces HMW SI expression. Results from qRT-PCR of sweet taste subunits

(T1R2 andT1R3) revealed a significant increase in gene expression of both sweet taste receptor subunits in cells fed by maltose compared to glucose. mRNA expression of

T1R2 subunit increased in the first two hours in the presence of maltose but not in

67 presence of glucose. Also, mRNA expression of T1R3 increased at 2 and 24 hours.

However, no significant change was observed in α-gustducin expression level. Measuring mRNA expression of the SI gene showed significant increase of SI mRNA level in the first 12 hours following by decrease in SI expression after 24 hours. This suggests a dynamic change in SI gene transcription and degradation of SI mRNA. Cheng et al.

(2012) showed that maltose increases expression of SI after 6 hours, but here the results indicated that this higher expression starts even at one hour of feeding the cells with maltose.

Gene expression of monosaccharide transporters (GLUT5 and GLUT2) were measured to monitor their change in presence of maltose. GLUT5 is facilitative fructose transporter (Bell et al., 1993; Corpe et al., 2002; Wasserman et al., 1996). In the small intestine, fructose was shown to significantly increase levels of GLUT5 mRNA and enhance the expression of GLUT5 protein (Cui et al., 2003; David et al., 1995; Jiang et al., 2001; Jiang and Ferraris, 2001; Kirchner et al., 2008; Kishi et al., 1999; Miyamoto et al., 1993). This may explain the high expression of GLUT5 in both glucose and maltose fed cells. GLUT2 is known to be enhanced in high glucose diet (Miyamoto et al., 1993), but the expression of GLUT2 was greater in cells fed with maltose. Cui et al. (2003) presented evidence that glucose enhance GLUT2 mRNA levels by transcriptional regulation. It possible that maltose may enhance the expression of GLUT2 more than glucose.

Maltooligosaccharides, important digestion products of starch by α-amylase, symbolize the presence of starch, which is the main glycemic carbohydrate on which humans evolved. Rapid digestion of glycemic carbohydrates and absorption of glucose

68 can be detrimental to health, particularly for diabetics and pre-diabetics, due to glycemic spikes that stress the glucose homeostasis regulatory system. This study showed that the sweet taste receptor subunits sense the existence of maltooligosaccharides via gastrointestinal tract and signal the cell to express higher molecular weight species of SI.

This increase in molecular weight in SI is due to post-translational events which are critical to the maturation of the SI enzyme active form. These new N- and O-linked glycans promoted different events of maturation and trafficking of SI. The change in glycosylation of SI induced by maltose also altered the distribution of SI in cell compartments through higher apical sorting (intercellular/apical membrane).

The pulse-chase study clearly revealed that maltose not only induces the formation of higher N- and O-glycosylated SI, but also that the mature form of SI traffics faster to the apical membrane for digestion of maltooligosaccharides in the small intestine. This novel finding led us to postulate that the sweet taste receptor on the small intestine enterocytes senses maltose and triggers signaling events leading to the processing and trafficking of SI. This expression of HMW SI promotes faster digestion of maltooligosaccharides. One might postulate that the function of the receptor for maltooligosaccharides could be manipulated through receptor blocking or through a signaling pathway to modulate glucose production for glycemic control and improved health.

Table 3.1 Relative protein amount (P2/P1) of intercellular distribution of SI in present of different carbohydrate. Carbohydrate treatments Relative protein amount P2/P1

Without glucose 1.72±1.098

Glucose 25mM 1.00

Glucose 50mM 1.80±0.41

Maltose 12.5mM 6.12±0.73

Maltose 25mM 8.35±1.65

Table 3.2 Association of SI with detergent-resistant microdomains in present of glucose and maltose. Carbohydrate treatments Lysate Supernatant Pellet

Glucose 1.28±0.014 0.67± 0.011 1±0.013

Maltose 1.35±0.018 0.65±0.015 1.18± 0.013

Table 3.3 Sucrase activity of SI in cells fed with different carbohydrates. Sucrase activity Homogenate P1 P2

Without glucose 0.85±0.12 0.62±0.09 0.78±0.05

Glucose 25mM 1 1 1

Glucose 50mM 1.33±0.06 1.23±0.16 1.11±0.16

Maltose 12.5mM 0.76±0.10 0.72±0.04 0.68±0.07

Maltose 25mM 0.70±0.10 0.63±0.03 0.57±0.06

Table 3.4 Maltase activity of SI in cells fed with different carbohydrates. Maltase activity Homogenate P1 P2

Without glucose 1.01±0.1 0.94±0.05 0.9±0.06

Glucose 25mM 1 1 1

Glucose 50mM 0.96±0.06 1±0.07 0.96±0.08

Maltose 12.5mM 1.05±0.08 0.98±0.03 0.85±0.05

Maltose 25mM 1.02±0.07 0.96±0.06 0.88±0.05

Table 3.5 The kinetic parameters of sucrase and maltase in the cells fed with glucose or maltose. Sucrase Activity Glucose-treated Maltose-treated

Km 14.949 ± 1.771 13.050 ± 1.667

Vmax 0.0194 ± 0.001 0.0168 ± 0.001

Maltase Activity

Km 1.4554 ± 0.241 1.3898 ± 0.114

Vmax 0.0582 ± 0.002 0.0545 ± 0.001

Protein Fraction P2 P1 Treatments: Glc(mM) 0 25 50 0 0 0 25 50 0 0

Mal (mM) 0 0 0 12.5 25 0 0 0 12.5 25

Figure 3.1 Biochemical analysis of the distribution of SI. Brush border membranes were isolated by the calcium chloride precipitation method. Protein amounts in homogenate (H) (total membranes and cytosol, microsomal, internal membranes (P1) (membrane bound intracellular components and basolateral membranes) and brush border membranes (P2) were determined and equal amounts were loaded on SDS-PAGE gels for each protein investigated (SI, 75µg).

4 Relative protein amount P2/P1 amount P2/P1 protein Relative

0 -Glc +Glc 25mM +Glc 50mM +Mal 12.5mM +Mal 25mM

Figure 3.2 Band intensities from P1 and P2 fractions were quantified and mean intensity of P2 was divided by mean intensity of P1 for SI (means ± SEM, n=3).

Treatment Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal

Time 0h 0h 1h 1h 2h 2h 4h 4h 8h 8h 12h 12h 24h 24h 48h 48h

Figure 3.3 Trafficking kinetics of SI to the cell surface. Caco-2 cells were labelled with [35S]-methionine for 30 minutes and chased for the indicated times. SI was immunoisolated from cell lysate. After cells were lysed with Triton X-100, SI was detected by immunoprecipitation. Samples were resolved on SDS-PAGE and visualised by phosphorimaging.

SIh 2.5 SIc

1.5

0.5 Intensity of SI bands unit) (arbitrary

0 Glc 0 Mal 0 Glc 1 Mal 1 Glc 2 Mal 2 Glc 4 Mal 4 Glc 8 Mal 8 Glc 12 Mal 12 Glc 24 Mal 24 Glc 48 Mal 48

Figure 3.4 Graphical representation of the pulse-chase experiment of Figure 3.3 by quantification of the SDS-PAGE band intensities from 3 independent experiments (means ±SEM, n=3). SIh is high mannose form of SI and SIc is complex form of SI.

Endo F treatment Untreated Treated

Treatment Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal Glc Mal

Time 2h 2h 4h 4h 8h 8h 24h 24h 2h 2h 4h 4h 8h 8h 24 24h

Figure 3.5 Transport kinetics of SI to the cell surface. Caco-2 cells were labelled with [35S]-methionine for 30 minutes and pulse chased for the indicated times. SI was immunoisolated at the cell surface. After cells were lysed with Triton X-100, SI was detected by immunoprecipitation. Enzymatic deglycosylation was carried out using Endo F/GF for 1 hour at 37 °C. Samples were resolved on SDS-PAGE and visualised by phosphorimaging.

Treatment Glc (25mM) Mal (12.5mM)

Tween-20 DRMs Lys Sup Pellet Lys Sup Pellet

Figure 3.6 Distribution of SI in Tween-20 detergent resistant membranes. Cells were treated with glucose 25 mM and maltose 12.5 mM for 12 hours. Lysis was performed in 1% Tween-20 in PBS for 2 hours at 4 °C. Detergent resistant membrane were pelleted by ultracentrifugation.

1.6

Lys 1.2 Sup Pellet

0.8 (arbitrary unit) (arbitrary Intensity of SI bands

0.4

0 Glucose Maltose

Figure 3.7 Graphical representation of the distribution of SI in Tween-20 detergent resistant membrane as shown in Figure 3.6, by quantification of the SDS-PAGE band intensities from 3 independent experiments (means ±SEM, n=3).

1.8

1.6

1.4 Homogenate P1 1.2 P2 1

0.8 Specific Activity Activity Specific 0.6

0.4

0.2

0 -Glu +Glu +50 Glu +12.5Malt. +25Mlat.

Figure 3.8 Comparison of sucrase activity of SI at the brush border membrane of Caco-2 cells treated with different concentrations of glucose and maltose. Brush border membranes were isolated by the calcium chloride precipitation method. Protein amount in homogenate (H) (total membranes and cytosol, microsomal, internal membranes (P1) (membrane bound intracellular components and basolateral membranes) and brush border membranes (P2) were determined. Results represent the mean ± SEM n=6 (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

1.4

1.2 Homogenate P1 P2 1

0.8

0.6 Specific Activity Activity Specific

0.4

0.2

0 -Glu +Glu +50 Glu +12.5Malt. +25Mlat.

Figure 3.9 Comparison of maltase activity of SI at the brush border membrane of Caco-2 cells treated with different concentrations of glucose and maltose. Brush border membranes were isolated by the calcium chloride precipitation method. Protein amounts in homogenate (H)(total membranes and cytosol, microsomal, internal membranes (P1) (membrane bound intracellular components and basolateral membranes) and brush border membranes (P2) were determined. Results represent the mean ± SEM n=6 (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

Michaelis-Menten

Glucose treated cells V (unit) V Maltose treated cells

[Sucrose] (mM)

Figure 3.10 Michaelis-Menten plots for SI. The enzymatic activity of sucrase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3).

Lineweaver-Burk

Glucose treated cells

Maltose treated cells 1/V

1/[Sucrose] (mM) Figure 3.11 Lineweaver-Burk plots for SI. The enzymatic activity of sucrase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3).

Michaelis-Menten

V (unit) V Glucose treated cells Maltose treated cells

[Maltose] (mM)

Figure 3.12 Michaelis-Menten plots for SI. The enzymatic activity of maltase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3).

Lineweaver-Burk

Glucose treated cells

Maltose treated cells 1/V

1/[Maltose] (mM)

Figure 3.13 Lineweaver-Burk plots for SI. The enzymatic activity of maltase was measured in the presence of different concentrations of substrate at the brush border membrane of Caco-2 cells treated with glucose and maltose (means ±SEM, n=3).

Figure 3.14 Deglycosylation (Endo F) of SI in cells treated with glucose or maltose. Caco-2 cells were biosynthetically labeled with [35S]methionine for 1.5 hours. The cells were homogenized and solubilized. Homogenates and medium were immunoprecipitated with SI antibodies. The immunoprecipitates were treated with Endo F/GF or left untreated. They were subjected to SDS-PAGE using 7% slab gels. Gels were analyzed by fluorography.

Endo H Untreated Treated

Treatments Glc (mM) 0 25 0 0 0 25 0 0

Mal (mM) 0 0 12.5 25 0 0 12.5 25

Figure 3.15 Deglycosylation (Endo H ) of SI in cells treated with glucose or maltose. Caco-2 cells were biosynthetically labeled with [35S]methionine for 1.5 hours. The cells were homogenized and solubilized. Homogenates and medium were immunoprecipitated with SI antibodies. The immunoprecipitates were treated with Endo H or left untreated. They were subjected to SDS-PAGE using 7% slab gels. Gels were analyzed by fluorography.

10 **

8 ***

6 Glc

T1R3 mRNA T1R3 mRNA Mal Relative amount) Relative ( 4

0 1 2 6 12 24 Hours

Figure 3.16 T1R3 gene transcription in Caco-2 cells in the presence of glucose or maltose. Cells were fed with 12.5 and 25 mM glucose (Glc) for 1, 2, 6, 12, and 24 hours. The data are presented as relative amount of T1R3 mRNA levels. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

*** 40

Glc R2 mRNA R2 mRNA 20 T1 Mal (Relative amount) (Relative

*** 10

0 1 2 6 12 24 Hours

Figure 3.17 T1R2 gene transcription in Caco-2 cells in the presence of glucose or maltose. Cells were fed with 12.5 and 25 mM glucose (Glc) for 1, 2, 6, 12, or 24 hours. The data are presented as relative amount of T1R2 mRNA levels. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

6 mRNA mRNA

Glc 4 Mal gustducin - (Relative amount) (Relative α

0 1 2 6 12 24 Hours

Figure 3.18 α-Gustducin gene transcription in Caco-2 cells in the presence of glucose or maltose. Cells were fed with 12.5 and 25 mM glucose (Glc) for 1, 2, 6, 12, or 24 hours. The data are presented as relative amount of α-Gustducin mRNA levels. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

120

100 **

80 **

60 ** * Glc SI mRNA SI mRNA *** Mal (Relative amount) (Relative 40

0 1 2 6 12 24 Hours

Figure 3.19 SI gene transcription in Caco-2 cells in the presence of glucose or maltose. Cells were fed with 12.5 and 25 mM glucose (Glc) for 1, 2, 6, 12, or 24 hours. The data are presented as relative amount of SI mRNA levels. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

60 ** *** 50

30 Glc Mal GLUT5 mRNA GLUT5 mRNA (Relative amount) (Relative 20

0 1 2 6 12 24 Hours

Figure 3.20 GLUT5 gene transcription in Caco-2 cells in the presence of glucose or maltose. Cells were fed with 12.5 and 25 mM glucose (Glc) for 1, 2, 6, 12, or 24 hours. The data are presented as of GLUT5 mRNA levels. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

** 40

* 30

*** Glc 20 Mal GLUT2 mRNA GLUT2 mRNA

(Relative amount) (Relative *

0 1 2 6 12 24 Hours

Figure 3.21 GLUT2 gene transcription in Caco-2 cells in the presence of glucose or maltose. Cells were fed with 12.5 and 25 mM glucose (Glc) for 1, 2, 6, 12, or 24 hours. The data are presented as relative amount of GLUT2 mRNA levels. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

3.5 References

Alfalah, M., Jacob, R., Preuss, U., Zimmer, K. P., Naim, H., & Naim, H. Y. (1999). O- linked glycans mediate apical sorting of human intestinal sucrase-isomaltase through association with lipid rafts. Current Biology, 9(11), 593-S2.

Beaulieu, J. F., Nichols, B., & Quaroni, A. (1989). Posttranslational regulation of sucrase-isomaltase expression in intestinal crypt and villus cells. The Journal of Biological Chemistry, 264: 20000-20011.

Bell, G. I., Burant, C. F., Takeda, J., & Gould, G. W. (1993). Structure and function of mammalian facilitative sugar transporters. Journal of Biological Chemistry, 268(26), 19161-19164.

Brownlee M. (2001). Biochemistry and molecular cell biology of diabetic complications. Nature, 414(6865): 813-20.

Chantret, I., Rodolosse, A., Barbat, A., Dussaulx, E., Brotlaroche, E., Zweibaum, A., & Rousset, M. (1994). Differential expression of sucrase-isomaltase in clones isolated from early and late passages of the cell line Caco-2: evidence for glucose-dependent negative regulation. Journal of Cell Science, 107, 213-225.

Corpe, C. P., Bovelander, F. J., Munoz, C. M., Hoekstra, J. H., Simpson, I. A., Kwon, O., Levine, M., & Burant, C. F. (2002). Cloning and functional characterization of the mouse fructose transporter, GLUT5. Biochimica Et Biophysica Acta-Gene Structure and Expression, 1576(1-2), 191-197.

Cui, X. L., Jiang, L., & Ferraris, R. P. (2003). Regulation of rat intestinal GLUT2 mRNA abundance by luminal and systemic factors. Biochimica Et Biophysica Acta- Biomembranes, 1612(2), 178-185.

David, E. S., Cingari, D. S., & Ferraris, R. P. (1995). Dietary induction of intestinal fructose absorption in weaning rats. Pediatric Research, 37(6), 777-782.

Dudley, M.A., Nichols B.L., Rosenberger J., Perkinson J.S., & Reeds P.J. (1992). Feeding status affects in vivo prosucrase.isomaltase processing in rat jejunum. Journal of Nutrition, 122:528-34.

Dudley M.A., Hachey D.L., Quaroni A., Hutchens T.W., Nichols B.L., Rosenberger J., Perkinson J.S., Cook G., & Reeds P.J. (1993). In vivo sucrase-isomaltase and lactase- phlorizin hydrolase turnover in the fed adult rats. Journal of Biological Chemistry, 268:13609-13616.

Englyst, H. N., Kingman, S. M., & Cummings, J. H. (1992). Classification and measurement of nutritionally important starch fractions. European Journal of Clinical Nutrition, 46: S33-S50.

Faller, G., Dimmler, A., Rau, T., Spaderna, M., Hlubek, F., Jung, A., Kirchner, T., & Brabletz, T. (2004). Evidence for acid-induced loss of Cdx2 expression in duodenal gastric metaplasia. Journal of Pathology, 203(4), 904-908.

Goda, T. (2000). Regulation of the expression of carbohydrate digestion/absorption- related genes. British Journal of Nutrition, 84: S245-8.

Hauri, H. P., Quaroni, A., & Isselbacher, K. J. (1979). Biogenesis of intestinal plasma membrane: Posttranslational route and cleavage of sucrase-isomaltase. Proceedings of the National Academy of Sciences USA, 76(10), 5183-5186.

Heine, M., Rohwedder, A., & Naim, H. Y. (2009). Impact of glycosylation and detergent resistant membranes on the function of intestinal sucrase-isomaltase. Journal of Biological Chemistry, 390: 545-549.

Hunziker W, Spiess M, Semenza G, & Lodish H. (1985). The sucrase-isomaltase complex: primary structure, membrane-orientation, and evolution of a stalked, intrinsic brush border protein. Cell, 46: 227-234.

100

Jacob, R., Weiner, J. R., Stadge, R., & Naim, H. Y. (2000). Additional N-glycosylation and its impact on the folding of intestinal lactase-phlorizin hydrolase, Journal of Biological Chemistry, 275: 10630–10637.

Jenkins, D., & Wolever, T. (1981). Glycemic index of foods: a physiological basis for carbohydrate exchange. The American Journal of Clinical Nutrition, 34(3): 362- 366.

Jenkins, D. J., & Kendall, C. W. (2002). Glycemic index: overview of implications in health and disease. The American Journal of Clinical Nutrition, 76(1): 266S- 273S.

Jiang, L., & Ferraris, R. P. (2001). Developmental reprogramming of rat GLUT-5 requires de novo mRNA and protein synthesis. American Journal of Physiology- Gastrointestinal and Liver Physiology, 280(1), G113-G120.

Jiang, L., David, E. S., Espina, N., & Ferraris, R. P. (2001). GLUT-5 expression in neonatal rats: crypt-villus location and age-dependent regulation. American Journal of Physiology-Gastrointestinal and Liver Physiology, 281(3), G666-G674.

Kirchner, S., Muduli, A., Casirola, D., Prum, K., Douard, V., & Ferraris, R. P. (2008). Luminal fructose inhibits rat intestinal sodium-phosphate cotransporter gene expression and phosphate uptake. American Journal of Clinical Nutrition, 87(4), 1028-1038.

Kishi, K., Tanaka, T., Igawa, M., Takase, S., & Goda, T. (1999). Sucraseisomaltase and hexose transporter gene expressions are coordinately enhanced by dietary fructose in rat jejunum. Journal of Nutrition, 129(5), 953-956.

Le Bivic, A., Quaroni, A., Nichols, B., & Rodriguez-Boulan, E. (1990). Biogenetic pathways of plasma membrane proteins in Caco-2, a human intestinal epithelial cell line. The Journal of Cell Biology, 111(4), 1351-1361.

Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227: 680–685.

Ludwig, D. S. (2002). The glycemic index. The Journal of the American Medical Association, 287(18): 2414-2423.

101

Mace, O. J., Affleck, J., Patel, N., & Kellett, G. L. (2007). Sweet taste receptors in rat small intestine stimulate glucose absorption through apical GLUT2. The Journal of Physiology, 582: 379–92.

Mahraoui, L., Rodolosse, A., Barbat, A., Dussaulx, E., Zweibaum, A., Rousset, M., & Brotlaroche, E. (1994). Presence and differential expression of SGLT1, GLUT1, GLUT2, GLUT3 and GLUT5 hexose-Transporter messenger-RNAs in Caco-2 cell clones in relation to cell growth and glucose consumption. Biochemical Journal, 298, 629-633.

Margolskee, R. F., Dyer, J., Kokrashvili, Z., & Salmon, K. S. H. (2007). T1R3 and gustducin in gut sense sugars to regulate expression of Na+-glucose cotransporter 1. Proceedings of the National Academy of Sciences USA. 104:15075-15080.

Miyamoto, K., Hase, K., Takagi, T., Fujii, T., Taketani, Y., Minami, H., Oka, T., & Nakabou, Y. (1993). Differential responses of intestinal glucose-transporter messenger- RNA transcripts to levels of dietary sugars. Biochemical Journal, 295, 211-215.

Naim, H. Y., Sterchi, E. E., & Lentze, M. J. (1987). Biosynthesis and maturation of lactase-phlorizin hydrolase in the human small intestinal epithelial cells, Biochemistry Journal, 241; 427–434.

Naim, H. Y., Sterchi, E. E., & Lentze, M. J. (1988). Biosynthesis of the human sucrase- isomaltase complex. Differential O-glycosylation of the sucrase subunit correlates with its position within the enzyme complex. Journal of Biological Chemistry, 263: 7242- 7253.

Naim, H. Y., Joberty, G., Alfalah, M., & Jacob, R. (1999). Temporal Association of the N-andO-Linked Glycosylation Events and Their Implication in the Polarized Sorting of Intestinal Brush Border Sucrase-Isomaltase, Aminopeptidase N, and Dipeptidyl Peptidase IV. Journal of Biological Chemistry, 274(25), 17961-17967.

Nichols, B. L., Quezada-Calvillo, R., Robayo-Torres, C. C., Ao, Z., Hamaker, B. R., Butte, N., Marini, J., Jahoor, F., & Sterchi, E. E. (2009). Mucosal maltase-glucoamylase plays a crucial role in starch digestion and prandial glucose homeostasis of mice. Journal of Nutrition, 139:684–690.

Pike, L. J. (2009). The challenge of lipid rafts. Journal of Lipid Research, 50, S323-S328.

102

Quezada-Calvillo, R., Robayo-Torres, C. C., Ao, Z., Hamaker, B. R., Quaroni, A., Brayer, G. D., Sterchi, E. E., Baker, S. S., & Nichols, B. L. (2007). Luminal substrate "brake" on mucosal maltase-glucoamylase activity regulates total rate of starch digestion to glucose. Journal of Pediatric Gastroenterology and Nutrition, 45: 32-43.

Reuser, A. J., Kroos, M., Oude Elferink, R. P., & Tager, J. M. (1985). Defects in synthesis, phosphorylation, and maturation of acid alpha- glucosidase in glycogenosis type II. Journal of Biological Chemistry, 260: 8336-8341.

Schmittgen, T. D., & Livak, K. J. (2008). Analyzing real-time PCR data by the comparative C-T method. Nature Protocols, 3(6), 1101-1108.

Schmitz, J., Preiseer, H., Maeshacci, D., Ghosh, B. K., Cerda, J. J., & Crane, R. K. (1973). Purification of the human intestinal brush border membrane. Biochimica et Biophysica Acta, 323: 98-112.

Sclafani, A. (2007). Fat and sugar flavor preference and acceptance in C57BL/6J and 129 mice: Experience attenuates strain differences. Physiology & Behavior, 90: 602–611.

Sim L., Willemsma C., Mohan S., Naim H. Y., Pinto B.M., & Rose D. R. (2010). Structural basis for substrate selectivity in human maltase-glucoamylase and sucrase- isomaltase N-terminal domains. Journal of Biological Chemistry, 285: 17763–17770.

Sterchi, E. E., & Woodley, J. F. (1980). Peptide hydrolases of the human small intestinal mucosa: distribution of activities between brush border membranes and cytosol. Clinica Chimica Acta, 102: 49-56.

Treesukosol Y., Blonde G. D., & Spector A. C. (2009). T1R2 and T1R3 subunits are individually unnecessary for normal affective licking response to polycose: implication for saccharide taste receptors in mice. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology, 296: R855–R865.

Wasserman, D., Hoekstra, J. H., Tolia, V., Taylor, C. J., Kirschner, B. S., Takeda, J., Bell, G. I., Taub, R., & Rand, E. B. (1996). Molecular analysis of the fructose transporter gene (GLUT5) in isolated fructose malabsorption. Journal of Clinical Investigation, 98(10); 2398-2402.

Wetzel, G., Heine, M., Rohwedder, A., & Naim, H. Y. (2009). Impact of glycosylation and detergent-resistant membranes on the function of intestinal sucrase-isomaltase. Journal of Biological Chemistry, 390: 545-9.

103

CHAPTER 4. INDUCTION OF DIFFERENTIATION OF SMALL INTESTINAL ENTEROCYTE CELLS BY MALTOOLIGOSACCHARIDE TREATMENT USING A METABOLOMICS APPROACH

4.1 Introduction

While genomic and proteomic techniques have been greatly improved for studying biological system, an additional high-throughput method is desirable to investigate the end-products of cellular processes. In 1998 for the first time, the term

”metabolome” was introduced to convey the concept of the complete pool of metabolites synthesized by an organism (Oliver, et al., 1998). Although the metabolite profiling concept has been used for decades for human disease diagnosis (Gates, et al., 1978), the term “metabolic profile” was introduced by Horning et al. (1971). The term of metabonomics is defined as “the quantitative measurement of the dynamic multiparametric metabolic response of living systems to pathophysiological stimuli or genetic modification” (Nicholson, et al., 1999) and metabolomics as defined by Fiehn

(2002) is “a comprehensive and quantitative analysis of all the metabolites”. However, often these two terms are used interchangeably, metabolomics aim to measure the whole metabolome, whereas metabonomics intend to measure the changes across the metabolome which is a smaller set of metabolites.

Metabolites are small molecules produced during metabolism including intermediates and products. These small molecules are the final downstream products of

104 gene and protein expression, therefore, they can provide a bigger picture of what is happening in biological systems than other “omics” techniques. Studying the metabolome provides valuable information to monitor alterations of metabolites caused by diseases, drug treatments, or any other environmental stress. The information from metabolomics can be combined with genomic and proteomic data to open up new perspectives for understanding biological systems.

Metabolomic data are classified into two categorizes, non-targeted (also known as global metabolite profiling) and targeted metabolomics. Untargeted metabolomics includes a comprehensive analysis of all the measurable metabolites in a biological system which requires extensive data analysis. On the other hand, targeted metabolomics considers a subset of metabolites with chemically similar characteristics.

In the following sections, steps involved in performing a metabolomics study will be discussed. Sample preparation, analytical instrument platform, and data analysis are the main steps in performing a metabolomics study.

4.1.1 Sample Preparation

Variety of matrices has been used for metabolomics studies including biofluids, tissues, cells, plants, and foods. It is important to choose an appropriate sample preparation method to avoid inaccurate data interpretation.

Methods are designed based on purpose of the studies, type of the samples (e.g. blood, tissue, cell), and the analytical techniques for data collection. In general, the sample preparation step should be reliable, reproducible, and time efficient. If the non- targeted metabolomics is the aim of the study, the extraction method should be as non-

105 selective as possible. Appropriate storage at cold temperature should be applied between sample collection and preparation. Depending on the experimental design and strategy to be employed, additional preparation steps such as separation, purification, and derivatization may be applied.

4.1.1.1 Cell Samples

Numerous metabolomics studies have been conducted on different cell lines which provide valuable information to quantitatively measure molecular biomarkers and physiological consequences in response to a treatment (Tate et al., 1996; Tate et al., 1998,

31; Florian et .al, 1995). Cell culture media also can be used to assess the impact of the medium on the cell culture performance. However, they are not as common as direct analysis on the cells.

4.1.1.2 Cell Sample Preparation

Cell sample collection for a metabolomics study is a very sensitive procedure. It is important to maintain the integrity of the cells and to limit leakage of intracellular metabolites, particularly for cell lines that are highly sensitive to osmotic changes of the medium (Faijes et al., 2007; Bolten et al., 2008). The medium has to be removed and cells should be collected as quickly as possible. Then, cells and/or medium have to be quenched and stored, or the metabolites extracted right after quenching. Quenching is the process used to inhibit enzyme activity and stop metabolic reactions by freezing samples.

Therefore, quenching gives a snapshot of a metabolic state at a given time point. The

106 common strategies to quench cell metabolism are immersion of cells in liquid nitrogen, or a dry ice-ethanol or methanol slurry (Viant et al., 2005; Castrillo et al., 2003; Koning et al., 1992).

The extraction of metabolites is an essential step in sample preparation. Several protocols have been developed for extracting of intracellular metabolites such as methanol/water solution, chloroform/methanol/water solution, and a freeze-thaw technique (Le Belle et al., 2002; Dettmer et al., 2011). Different extracting methods can be selected based on targeted or non-targeted metabolomics analysis. The non-targeted metabolomics extraction method should be non-selective to collect all metabolites in cell, whereas targeting metabolomics is limited to a group of metabolites with known chemical properties.

4.1.2 Analytical Instrument Platforms

The two most common techniques in metabolomics studies are nuclear magnetic resonance spectroscopy (NMR) and mass spectrometry (MS). NMR is highly reproducible, non-destructive, quantitative, and non-selective which gives the ability to collect the data of all the detectible metabolites in a one snap-shot, with minimal sample preparation, and the ability to detect isomers. However, it has limitations including sensitivity, complexity, and spectral overlap in NMR spectra. Although MS is more sensitive than NMR, it requires a pre-separation of the metabolic components using liquid chromatography or gas chromatography (GC), and to use GC metabolites have to be first derivatized.

107

4.1.3 Data Analysis

4.1.3.1 Pre-Data Processing

1H NMR spectra are recorded on high field NMR spectrometers and run on NMR processing softwares packages such as Topspin (BrukerBiospin, Billerica, MA), VnmrJ

(Varian, Palo Alto, CA), NMRPipe (Delaglio et al., 1995), and MestReNova (Mestrelab

Research, Santiago de Compostela, Spain). The collected NMR data are processed by

Fourier transformation following by apodization and zero filling of the free induction decay. After that, the phase of each spectrum has to be corrected to obtain absorption line shapes followed by a subsequent baseline correction, if it is necessary. Spectra are referenced to chemical shift of an internal standard peak such as the singlet of trimethylsilyl propanoic acid or 4,4-dimethyl-4-silapentane-1-sulfonic acid (DSS), or one of the metabolites such as alanine. This step is also known as global alignment. Phasing, baseline correcting, and referencing correction are done using one of the available NMR processing software packages mentioned above.

Most of the time global alignment is not sufficient and has to be followed by alignments of individual peaks performed manually or using particular algorithms such as the recursive segment-wise peak alignment (Veselkov et al., 2009).

Other processing steps are binning, normalization, and scaling. Binning, also known as bucketing, is often used after spectral alignment to perform multivariate data analysis. The spectra are divided into buckets or segments. The intensities within each bucket are summed and this results in data reduction. However, the drawback for this

108 method is the spectral complexity of peaks with overlapping data. Binning can introduce errors from inaccurate spectral intensities when neighbor peaks contribute to a bucket. To avoid or eliminate these types of errors, analysis of full resolution spectra (Cloarec et al.,

2005) or using variable binning (Davis et al., 2007; Meyer et al., 2008) has been suggested. The alternative to bucketing is the direct integration of spectral peaks. This can be more accurate and is used to quantify absolute concentrations of specific metabolites.

Normalization applies for variations of the overall concentrations of samples. The goal of normalization is to remove unwanted variation (e.g., dilution factor) between samples and make the profiles comparable to each other. There are several technical factors causing variation in data. Some of these factors are data recording with different devices, different number of scans, or the absolute signal intensities data analysis.

Without normalizing data for these variations, the data analysis is less meaningful.

(Dieterle et al., 2006).

A number of normalization methods exist, including Probabilistic Quotient

Normalization (PQN) (Dieterle et al., 2006) and Constant Sum Normalization (CSN)

(Craig et al., 2007). Scaling has to be done before applying multivariate statistical methods, as metabolite concentrations vary over many orders of magnitude. In the absence of scaling, large and variable signals can be falsely picked as biomarker candidates, and some low concentration metabolites that are true biomarkers can be ignored. A number of scaling methods exist, such as variance scaling, scaling factor as the standard deviation of peak intensities across the spectra, and Pareto scaling, where the scaling factor is the square root of the standard deviation (Eriksson et al., 2004).

109

4.1.3.2 Statistical Methods

NMR and MS both generate huge and complex data sets. To interpret the data, multivariate statistical analysis methods, also known as chemometrics, has been extensively used (Cuperlovic-Culf et al., 2009; Khoo and Al-Rubeai, 2009, Van Vliet et al., 2007).

Two approaches used to analyze metabolomics data sets are, 1) exploratory

(unsupervised) analysis and 2) confirmatory (unsupervised) analysis. Unsupervised data analysis is used to find patterns in the data sets without prior knowledge of the groups.

Two examples of unsupervised data analysis are principal component analysis (PCA) and hierarchical clustering analysis (HCA) (Mitchell et al., 2002). Supervised data analysis is applied by assigning of spectra to classes. Examples of supervised analysis are the univariate Student’s t-test, and multivariate methods that include, though are not limited to, partial least square-discriminant analysis (PLS-DA), orthogonal signal correction-

PLS-DA (O-PLS-DA), and logistic regression (Gottschalk et al., 2008).

PCA is one of the most popular data analysis techniques and has become a norm for data visualization in the field of metabolomics (Khoo et al., 2007). PCA reduces the large number of variables down to a smaller number of factors or principal components.

Each principal component (PC) is a linear combination of the original data parameters and explains the maximum amount of variance possible that has not been accounted for by the previous PCs. The PCA model is built on the basis of orthogonal vectors

(Eigenvectors), therefore principal components are independent from each other. The first

PC (PC1) describes the direction of the largest variations generated from the set of

110 spectra, the second PC (PC2) describes the direction for the largest portion of the remaining variation, and likewise for the rest of the PCs.

Conversion of the original data to PCs produces, in two matrices, scores and loadings. Scores and loadings are presented as scores and loading plots. Each point represents a single sample spectrum in a score plot. Loadings explain how the old variables are linearly combined and form the new variables. Loadings also indicate the variables carrying the greatest weight and show the contribution of each variable in the original data in the scores. Each point in the loading plots represents a different spectral intensity. Therefore, loadings can be used to interpret relationships among variables that cause any cluster separation in the score plot. From the output of PCA analysis, compounds that are of great importance in classifying different samples are highlighted and considered as the initial biomarker candidates.

PLS-DA has been used extensively as a powerful tool for the metabolomics field.

PLS-DA is the combination of partial least square (PLS) and discriminant analysis (DA).

PLS is one of the common supervised methods used in metabolomics data analysis. PLS fits the data matrix of predictors X and class matrix of responses Y, and is used to find a linear regression model of the new coordinate system. Each orthogonal axis in PLS is named a latent variable (LV). DA is a statistical method for determination of a linear combination of features to predict to which class a sample belongs (Wold et al., 2003;

Baker et al., 2003). PLS-DA create linear mathematical models from training sets of spectral data within known classes. PLS-DA can be used to find the discriminant between pre-assigned sample groups and recognize variables which best separates classes.

111

A drawback of PLS-DA is data over-fitting, therefore it is necessary and important to perform the validation of the results (Westerhuis, 2008). Cross-validation is the most common method for validation, in this method results are applied to a new set of observations that was not used to build the model (Anderssen, et al. 2006). An example to the cross-validation is leave-one-out cross-validation (LOOCV). This method is one of the most commonly used methods for PLS-DA data validation in the field of metabolomics (Stretch et al., 2012; Gu et al., 2011; Guan et al., 2009). To predict the accuracy of the PLS-DA model, one single observation from the original sample set is used. The rest of the samples are used as the training data in order to build the model.

This process is repeated until all the samples are used in the validation data once.

4.1.4 Metabolomic Application in the Present Study

Metabolomics is an emerging field with wide range of application in different biological research areas that focuses on high-throughput identification of metabolites in biological systems. Intrinsically, metabolomics is well positioned to be used in many areas of food and nutrition science, where scientists are challenged to find better ways of preventing or treating diseases such as obesity, diabetes, and cardiovascular disease caused by unbalanced diets (German et al., 2005). A typical approached used to address these problems is to first implement epidemiological studies to relate long-term food consumption with health status, and then to design dietary interventions. This is based on the hope to find associations between diet and health status (German et al., 2005; Zeisel,

2007), however correlative relationships do not necessarily indicate causation of diseases.

In order to relate diet effect to normalcy or to understand biochemical alterations in the

112 body due to diet, metabolomics techniques can be used to measure and compare small molecule biomarkers (metabolites) in the presence and absence of diet or dietary factors of interest (Gibney et al., 2005; Zeisel, 2007).

Based on the Dietary Guidelines for Americans 2010 (USDA), dietary carbohydrates should provide 45-65% of total caloric intake. Carbohydrates are among the macronutrients that provide energy and can, thus, contribute to excess energy intake.

However, the available carbohydrates differ in their rate of digestion and glucose

(mainly) liberation and absorption (e.g., for starch, rapidly or slowly digestible starch or resistance starch), and healthiness. Pertinent to our discussion here, slow glucose release has been shown to be beneficial for maintaining glucose homeostasis and may be associated with health benefits through reduction of risk factors for some diet-associated metabolic diseases (Jenkins et al., 2002).

Maltooligosaccharides are the primary product of starch digestion by salivary and pancreatic α-amylases and are substrates for the small intestinal brush border-bound α- glucosidases. We have recently shown that maltose acts as a signal molecule that changes the molecular size of SI, which was postulated to stimulate the maturation and trafficking of this enzyme complex to the enterocyte apical surface for digestion of the maltooligosaccharides to glucose (Cheng et al., 2014). We view this action as another potential point of control for starch-originated glucose release and absorption into the body. While this finding is indicative of a maltooligosaccharides sensing ability by enterocytes, it also suggests alteration in other events in small intestine enterocytes leading to such things as secretion of gut hormones and production of tight junction proteins. It is worth noting that such cellular events are stimulated by short chain fatty

113 acids (SCFAs) which are products of bacterial fermentation of undigested carbohydrates in the intestine (Bond & Levitt, 1976; Sakata, 1987). They can enhance intestinal barrier function as measured by increases in TEER (Schumann et al., 2005), and also affect upper gut motility and satiety through triggering release of certain gut hormones

(Cherbut, 2006). The enteroendocrine L-cells secrete the gut hormones such as GLP-1,

PYY, and oxyntomodulin which are involved in appetite regulation (Delzenne et al.,

2006). In several studies it has been shown that fermentable carbohydrates such as inulin

(Delzenne et al., 2006), lactitol (Gee & Johnson, 2005), and FOS (Cani et al., 2005;

Delmee et al., 2006), increase satiety and decrease weight gain. From preliminary study of the effect of maltooligosaccharides on these cell responses, a full study was conducted to examine whether maltose and other maltooligosaccharides affect enterocyte tight junctions and more generally cell differentiation. The purpose of this study was to investigate the effect of maltooligosaccharides on cell differentiation biomarkers and to identify other potential intracellular change by maltooligosaccharides using a metabolomics approach.

4.2 Materials and Methods

4.2.1 Chemicals and Samples

4.2.1.1 Chemicals

Sodium phosphate dibasic anhydrous (Na2HPO4), sodium phosphate monobasic

anhydrous (NaH2PO4), sodium hydroxide (NaOH), and sodium chloride (NaCl) were

114

obtained from Fisher Scientific. Sodium 2,2-dimethyl-2-silapentane-5-sulfonate (DSS,

D6, 98%), and D2O (99.9% D) were purchased from Cambridge Isotope Laboratories,

Inc.

4.2.1.2 Cell Culture Procedure

Caco-2 cells (passage 30 to 45) were used for this study. Cells were seeded as 6.4

× 104 cells/well on a 6-well solid support (Corning, Lowell, MA). Cells were grown in

Dulbecco’s modified Eagle’s medium (25 mM, or equal to 4.5 g/l glucose DMEM,

Corning, Lowell, MA) supplied with 10% heat-inactivated (56°C, 30 min) fetal bovine serum (FBS) (Lonza, Walkersville, MD), 50 µg/ml gentamycin sulfate (J R Scientific

Inc., Woodland, CA), 10 mM HEPES, 100 µg/ml streptomycin and 100 U/ml penicillin

(Lonza, Walkersville, MD), and 100 µM non-essential amino acid (Corning, Lowell,

MA). Cells were incubated at 37°C with 5% CO2, 95% air atmosphere, and at constant humidity. Cells were grown to confluence for the complete differentiation of the cells and maintained according to the procedures of Mahraoui et al. (1994) and Le Gall et al.

(2007).

4.2.1.3 Cell Treatment

Cells on transwells were washed twice with 1 ml PBS after they reached 100% confluence and were fully differentiated. Cells were then fed with glucose-free DMEM

(Gibco, Carlsbad, CA). Test carbohydrates were supplied to the glucose-free DMEM

(Gibco, Carlsbad, CA). Different carbohydrate concentrations were used with mannitol to

115 adjust the osmolarity of the media and the treatment time points were according to each experiment (Dudley et al., 1993).

4.2.1.4 Measurement of Epithelial Monolayer Electrical Resistance

In order to determine the integrity of the monolayer of Caco-2 cells, transepithelial electrical resistance (TEER) of monolayers was measured by an epithelial voltohmmeter (Millicell ERS-2 system, Millipore, Billerica, MA). Caco-2 monolayers were grown to confluence on 12-well transwell inserts (0.4 µm pore size, polycarbonate membrane; Corning, NY). After cells differentiated (15 days after seeding cells), high glucose DMEM was switched to a modified DMEM supply with different added carbohydrates (6 mM glucose, 25 mM glucose, 12.5 mM maltose, 8.33 mM maltotriose, mixture of 6 mM glucose + 9.5 mM maltose, and 12.5 mM sucrose). TEER was measured at 0, 12, 24, and 48 hours; and calculated and normalized to the TEER measured at the day of differentiation (0 h).

4.2.1.5 Determination of Paracellular Permeability

Paracellular permeability was determined measuring the flux of fluorescein isothiocyanate-dextran (FITC-dextran) concentration through cell monolayers. Cells were seeded and fed as described in TEER experiment above. FITC-dextran was added to the apical compartment and media was collected from the basolateral chambers at different incubation times. Fluorescence intensity of FITC-dextran was measured at excitation wavelength of 485 nm and emission wavelength of 538 nm using a fluorescence plate

116 reader (Molecular Devices, Chicago, IL). Paracellular permeability was calculated and normalized to the FITC measured at day of differentiation (0 hour).

4.2.1.6 Sample Preparation for NMR Experiments

Caco-2 monolayers were grown to confluence on tissue-culture treated culture dishes (100 mm x 20 mm; Corning, NY). After cells differentiated (20 days after seeding cells), high glucose DMEM was switched to a modified DMEM supply with different carbohydrates (6 mM glucose, 25mM glucose, 12.5 mM maltose, 8.33 mM maltotriose, mixture of 5.5 mM glucose + 10 mM maltose, and 12.5 mM sucrose).

After treatment, cells were washed five times with 10 mL of ice-cold PBS with the plates sitting on ice to remove media and other residues. Then cells were quickly scrapped off and pelleted in a 15 mL tube using a refrigerated centrifuge. The cell pellet was snap-frozen using dry ice and stored at -80 C until metabolite extraction. To isolate metabolites, cell pellets were thawed on ice and resuspended in 1500 µL of 80% methanol solution and vigorously mixed for every 5 min for 20 min. Then cells were homogenized using FastPrep®-24 homogenizer (MP Biomedicals, Santa Ana, CA). After that, homogenized cells were centerifuged at full speed (13,000 rpm) at 4°C for 20 min.

The supernatant was collected and methanol was evaporated using a SpeedVac concentrator (Thermo Fisher Scientific, Waltham, MA). Phosphate buffer (150 µl, 0.1M, pH=7.4) containing NaN3 (3 mM), and DSS-D6 (0.11 mM) was added to each sample, vortexed, and the mixture transferred to a 2.5 mm NMR tube (Norell, Landisville, USA).

Twenty microliters of pellet was used for protein concentration measurements.

117

4.3 NMR Experiments and Data Processing

4.3.1 NMR Experiments

The spectra were collected on a 600 MHz proton frequency Avance II NMR

(Bruker Biospin, Germany) spectrometer equipped with a cryogenically cooled probe or on a Bruker Avance-III-800 (Bruker Biospin, Germany) spectrometer equipped with a

TXI probe. All NMR data were acquired at 300oK and carried out in a random order. The probe was locked to H2O+D2O (%90+%10), shimmed, tuned, and matched for each sample. A 90° pulse length, the offset of the water signal, water suppression and receiver gain for a data set were also determined.

One-dimensional (1D) nuclear Overhauser enhancement spectroscopy (NOESY)- presat (noesypr1d) pulse sequence was used to acquire 1D proton spectra; 128 scans, 4 second relaxation delay, spectral width of 20 ppm and 65k data points were set to acquire the data. Total acquisition time was 15 min.

Two-dimensional (2D) Heteronuclear Single Quantum Coherence (HSQC) and

Total Correlation Spectroscopy (TOCSY) were applied to one of the samples in each group. All of the 2D spectra were collected on the 600 MHz proton frequency Avance II

NMR. Metabolites were assigned using Complex Mixture Analysis by NMR (COLMAR) using 13C-1H HSQC. This information was combined and validated with 2D TOCSY data.

The 2D 13C-1H HSQC spectra were acquired with 1024 and 2048 complex data points in t1 and t2, respectively. The number of scans per t1 increment was 8 with a corresponding spectral width of 24904.8 Hz along the indirect dimension and 7211.5 Hz

118 along the direct dimension. The transmitter frequency offsets were 70 ppm and 4.7 ppm in the 13C and the 1H dimensions, respectively. Total acquisition time was 3 h and 49 min.

The 2D 1H-1H TOCSY spectra were acquired with 1024 and 2048 complex data points in t1and t2, respectively. The number of scans per t1 increment was 8 with a corresponding spectral width of 9014.4 Hz along the indirect dimension and 9003.4 Hz along the direct dimension. The transmitter frequency offset was 4.7 ppm in both 1H dimensions. Total acquisition time was 5 h and 50 min.

4.3.2 Data Processing

All 1D NMR spectra were phased and baseline corrected using Topspin 3.2. The

Bruker 1r files were imported into MATLAB (R2013B, the Mathworks, Inc., Natick,

MA). Spectra were all referenced to DSS peak at 0.0 ppm. Water regions (4.65- 4.80 ppm) and end points before -0.2 ppm and after 10 ppm were removed and the spectra were aligned, normalized and scaled for further analysis. Ccow, PQN, and logoff algorithms were used for alignment, normalization and scaling, respectively. PCA model was performed to initially examine intrinsic variations in the data set.

All the 2D spectra were zero-filled, Fourier transformed, and phase and baseline corrected by NMRPipe (Delaglio et al., 1995). Peak pickings were done using

NMRViewJ (9.0.0-b127, One Moon Scientific, Inc., Westfield, NJ).

119

NMR metabolomics data was collected at the Southeast Center for Integrated

Metabolomics at the University of Florida using the National High Magnetic Field

Laboratory’s AMRIS Facility.

4.3.2.1 Statistical Analysis

Data are shown as means ± S.E.M. Statistical analysis was performed using SAS

9.4 software. Two-way ANOVA was used to determine significance of treatment effect and interactions. Significant differences between group means were assessed by Tukey method which was accepted at p < 0.05.

4.4 Results

4.4.1 Different Source of Carbohydrates Alters Intestinal Epithelial Tight Junction

Barrier

The effect of different carbohydrates on tight junction barrier function was measured using an epithelial voltohmmeter. Cells were fed with different carbohydrates in two different ways. First, cells were seeded and fed high glucose Dulbecco's Modified

Eagle's medium (DMEM). After cells were differentiated (15 days after seeding), they were fed with carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and

5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours.

120

In addition to tight junction barrier function, the effect of different carbohydrates on epithelial paracellular permeability was measured using FITC-labeled dextran. In parallel with the TEER experiment, cells were fed the same two ways as in the tight junction barrier experiment.

Results from ANOVA analysis of the TEER experiment in cells fed with different carbohydrates showed that only glucose and maltose maintained the highest level of tight junction barrier over the 72 hours period (Figure 4.1). In addition, maltose promoted better tight junction at 12 hours which was significantly higher than glucose at 12 hours

(Figure 4.2). The other carbohydrates showed lower tight junction formation compared to glucose and maltose (Figure 4.1). ANOVA analysis also was performed on epithelial paracellular permeability in Caco-2 cells fed with different carbohydrates (Figure 4.3).

Result showed that glucose, maltose, maltose+low glucose, and maltotriose promoted similar permeability. However, sucrose and low glucose were significantly higher and significantly lower in permeability levels compared to other carbohydrates. In addition, permeability of cells fed with maltose and sucrose was significantly higher at 12 hours compared to cells fed with other carbohydrates (Figure 4.4). The low glucose fed cells had significantly lower permeability compared to the other treatments.

4.4.2 Different Source of Carbohydrate Alters Metabolite Profiles in Caco-2 Monolayers

In this study, high-resolution 1H NMR was used to monitor the metabolites involved in cell differentiation in Caco-2 cells fed with different carbohydrates.

Metabolites were extracted using 80% methanol-water which was found to provide the best extraction condition in preliminarily experiments run on the same cells. The first

121 aspect noticed from the NMR spectra is the richness of the spectra showing that a wide variety of metabolites were detected. Thus, the extraction method yielded a variety of metabolites. Figure 4.5 shows the full spectrum of a representative glucose sample.

Identified peaks were assigned and listed in Table 4.1, the annotated metabolites are shown in Figures 4.6 and 4.7. In the 1H NMR spectra, the aliphatic regions from 1 to 4.5 ppm are dominated by peaks from isoleucine, leucine, valine, 3-hydroxybutyrate, threonine, lactate, acetate, alanine, arginine, glutamate, glutamine, succinate, creatine, phosphorylcholine, taurine, myo-inositol, glycine, proline, and glycerophosphocholine

(Figure 4.6). These metabolites were identified by COLMAR 13C-1H HSQC query database and by comparison of the NMR spectra with literature reports (Beckonert,

2007). The NMR spectrum from 5 to 10 ppm represents chemical shifts of the aromatic nucleoside and ribose signals as well as shows the aromatic amino acids (nicotinamide dinucleotide (NAD), tyrosine, histidine, tryptophan, phenylalanine, UDP-derivate, AMP- derivate, fumarate, and formate (Figure 4.7). The results from the spectra of metabolites in cells fed with different carbohydrates were used to monitor differences in the metabolic profiles in different time points.

Alteration in cell contents corresponds to a change in metabolite levels and can be considered as the metabolic fingerprint of the cells. Lee at al. (2009) identified the metabolites involved in differentiation of Caco-2 cells. They showed that taurine, creatine and glycerophocholine increased and myo-inositol decreased during cell differentiation.

In this study, we monitored the alteration in these metabolites in presence of different carbohydrates. Taurine levels were significantly different in cells fed with maltose and glucose compared to cells fed with other carbohydrates over 72 hours (Figure 4.8). Cells

122 fed with maltose showed significantly higher taurine level at 12 hours compare to cells fed with other carbohydrates at 12 hours. (Figure 4.9). Also, level of myo-inositol metabolites was significantly higher in cells fed with maltose compared to other carbohydrates (Figure 4.10). Phosphorycholine (Figure 4.12 and Figure 4.13) and glycerophosphocholine (Figure 4.14 and Figure 4.15) levels were significantly increased in cells fed with maltooligosaccharides (maltose, mixture of maltose and glucose, maltotriose). Results showed no change in creatine level over the different treatments.

4.5 Discussion

In the present study, the effect of maltooligosaccharides (maltose and maltotriose), glucose (high and low concentration), and sucrose on two important characters of the Caco-2 cell line (intestinal permeability and cell differentiation) were evaluated.

While the main purpose of the small intestine is digestion and absorption of nutrients, it also functions as a barrier. This prevents harmful substances (such as antigens) from entering the body. Increase in intestinal permeability is a factor in several autoimmune conditions such as Crohn's disease, type 1 diabetes, and inflammatory bowel disease (Fasano, 2011; Johnson, 2013; Watts et al., 2005). Tight junctions are important in limiting intestinal permeability and maintaining the barrier integrity by sealing the paracellular space between epithelial cells. Tight junctions also regulate the entry of nutrients, ions, and water while restricting pathogen entry, and determine the barrier function of the epithelium (Farquhar et al., 1963). The structures of tight junctions are

123 dynamic and change due to interactions with external stimuli, such as food (Hashimoto et al., 1994; Jensen-Jarolim et al., 1998).

Cell proliferation, polarization, and differentiation has a significant role in creating and maintaining epithelial barrier integrity through tight junctions (Matter et al.,

2005). Cell differentiation is the process in which a less specialized cell becomes a more specialized cell type. The Caco-2 cells differentiation process occurs from homogeneously undifferentiated to heterogeneously polarized and differentiated to homogeneously polarized and differentiated (Vachon & Beaulieu, 1992). Cells in cell culture can also go the other way and lose properties they originally had, such as protein expression, physiological, morphological, and/or biochemical properties. This process is termed dedifferentiation (Schnabel et al., 2002). Culture-related conditions (e.g. glucose concentration) can trigger dedifferentiation in cells (Lu et al., 1996).

Recently, we reported that maltose may act as a signaling molecule that changes the molecular size of SI (Cheng et al., 2014). As this finding is indicative of a maltose sensing ability by enterocytes, it also suggests alteration in other events in small intestine enterocytes leading to such things as secretion of gut hormones and production of tight junction proteins. In this study, we investigated the effect of maltooligosaccharides on cell differentiation biomarkers and identifying other potential intracellular changes using the metabolomics method, as well as their effect on permeability of intestinal epithelia using the Caco-2 cells model.

First, we assessed intestinal epithelial tight junction barrier and epithelial paracellular permeability function in the Caco-2 monolayer when exposed to different carbohydrates. Results from TEER showed that that maltose and glucose significantly

124 increased tight junction formation compared to other carbohydrates (Figure 4.1). Maltose induced a significant increase in cells at 12 hours when it compared to other carbohydrates (Figure 4.2). Sucrose caused the lowest tight junction barrier formation in

Caco-2 cells. High glucose concentration is known to increase tight junctions (Madara &

Pappenheimer, 1987; Pappenheimer, 1987). Maltose had a similar effect to glucose compare to other carbohydrates which decreased tight junction barrier over the 72 hour treatment period. Although SI has the function of digesting maltose to glucose, in our previous study (Cheng et al., 2014) using the same concentration of maltose in the medium, we found a conversion at 12 hours of only about 2 mM glucose. This is far below the 25 mM glucose in the high glucose treatment and well above the 5.5 mM glucose of the low glucose treatment which had a low 12 hour TEER value. Hence, maltose itself was implicated in the high tight junction barrier value, particularly at the 12 hour time point.

The results from the paracellular permeability experiment were contrary to

TEER for maltose and high glucose treatments at 12 hours with maltose showing higher permeability than high glucose. The maltose+low glucose and maltotriose treatments had similar effects to maltose on cell permeability (Figure 4.3). Sucrose promoted the highest permeability and low glucose caused the lowest permeability in Caco-2 cells.

1H NMR spectroscopy was used to examine the intracellular metabolic level of

Caco-2 cells in the presence of different small molecule carbohydrates over a 72 h period and to identify cell biomarkers that are involved in cell differentiation. Lee at al. (2009) showed changes in metabolite profiling during Caco-2 cell differentiation.

125

Results showed that taurine level was significant in cells fed with maltose and glucose (Figure 4.8). Taurine level was significantly higher at 72 hours compared to cells fed other carbohydrates (Figure 4.9). Myo-inositol level was significantly higher in cells fed with maltose compared to cells fed with other carbohydrates (Figure 4.10). In cells fed with maltose at 72 hour, myo-inositol was significantly higher compare to cells fed with other carbohydrates (Figure 4.11). The DMEM used for culturing Caco-2 cells does not contain taurine. Taurine is made from cysteine (Beetsch et al., 1998) and behaves as an osmolyte to regulate extracellular osmolarity (Satsu et al., 1999). It has been shown that in an acute hypotonic stimulus, cells release taurine and myo-inositol to respond the volume increase and to avoid swelling-induced cell lysis (Ullrich et al., 2006). These results suggest that maltooligosaccharides induced biosynthesizing of taurine and glycerophosphorylcholine which are seen in more differentiated cells (it is important to note that the osmolality of DMEM with different carbohydrates are same).

Phosphorylcholine level of cells fed with maltose, was significantly higher compared cells fed with other carbohydrates (Figure 4.12 & 4.13). Results showed that glycerophosphocholine level in cells fed with maltose, maltotriose, maltose+low glucose, and sucrose were significantly higher than cells fed with glucose and low glucose (Figure

4.14 & Figure 4.16). Phosphorycholine was another metabolite that significantly increased only in Caco-2 cells fed with maltooligosaccharides. This metabolite is involved in phospholipids sphingomyelin structure that can be found abundantly in lipid raft membrane (Yeagle, 2004). In Chapter 3, maltose was shown to induce a higher molecular SI which are more lipid raft associated. This data may further supports that higher lipid raft structure was caused by maltooligosaccharides.

126

As starch and its derived products (e.g., maltodextrins) are the principal dietary supply of glucose, investigation of their physiological effects on intestinal enterocyte cells is important for this aspect of human health. This is the first study to use maltooligosaccharides as the energy source of cells for studying tight junction barrier, cell permeability, and cell differentiation. This study revealed that maltooligosaccharides have the ability to promote tight junction formation and cell differentiation compared to glucose and sucrose. This event might be related to sensing maltose, perhaps by the sweet taste receptor, to result in better formation of tight junctions and barrier function.

However, more investigation is required to study potential signaling pathways involved in improving cell permeability in presence of maltose.

127

Table 4.1 Metabolite assignments and chemical shifts of identified peaks. Assigned number Metabolite Chemical shift (ppm) 1 Pantothenate 0.88, 0.93 2 Leucine 0.95 3 Isoleucine 0.95, 1.00 4 Valine 0.98, 1.03 5 3-hydroxy butyrate 1.21 6 Lactate 1.32, 4.10 7 Alanine 1.47 8 Arginine 1.69 9 Acetate acid 1.91 10 Proline 1.99, 2.06 11 Glutamate 2.04, 2.11, 2.34 12 Glutamine 2.12, 2.45 13 Succinate 2.39 14 Beta-Alanine 2.54, 3.16 15 Creatine 3.03, 3.92 16 Glycerophosphocholine 3.20, 4.32 17 Phosphorylcholine 3.21 18 Taurine 3.26, 3.42 19 Myo-Inositol 3.53 20 Glycine 3.56 21 NAD 6.02, 6.09,6.13, 8.82, 9.32, 9.12 22 Fumarate 6.51 23 Tyrosine 6.89, 7.18 24 Histidine 7.10 25 Tryptophane 7.53, 7.32 26 Phenylalanine 7.32, 7.34, 7.42 27 UDP derivate 7.93 28 Formate 8.41 29 AMP derivate 8.59, 8.16

128

120

A A 100

B 80 BC High Glucose C Low Glucose 60 Maltose

TEER Maltotriose D Low Glucose + Maltose 40 Sucrose (% of differentiated cell in 0 hour) cell (% of differentiated

0 Type of carbohydrate

Figure 4.1 Intestinal epithelial tight junction barrier function in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose over 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Different letters denote significant differences at α=0.05.

129

140

120 *

100 * * * * 80 * * * * * *

TEER 60 * * * * * 40 * (% of differentiated cell in 0 hour) cell (% of differentiated

0 12 24 48 72 Time (h)

High Glucose Low Glucose Maltose Maltotriose Low Glucose + Maltose Sucrose

Figure 4.2 Intestinal epithelial tight junction barrier function in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Asterisks denote significant differences at α=0.05 between the carbohydrate treatments at that time point using high glucose as the reference.

130

160 A

140

120 B B B B 100 High glucose Low glucose Permeability Permeability 80 Maltose C Maltotriose Maltose+low glucose 60 Sucrose Paracellular

(% of differentiated cell in 0 hour) cell (% of differentiated 40

0 Type of carbohydrate

Figure 4.3 Epithelial paracellular permeability in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose over 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Different letters denote significant differences at α=0.05.

131

180

160 * * * *

140 *

120

100 Permeability Permeability

80 * * * 60 * Paracellular (% of differentiated cell in 0 hour) cell (% of differentiated 40

0 12 24 48 72 Time (h)

High glucose Low glucose Maltose Maltotriose Maltose+low glucose Sucrose

Figure 4.4 Epithelial paracellular permeability in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Asterisks denote significant differences at α=0.05 between the carbohydrate treatments at that time point using high glucose as the reference.

10.0 9.59.0 8.58.0 7.57.0 6.56.0 5.55.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 Chemical Shift (ppm)

Figure 4.5 Representative 1D 1H NMR noesygppr1d spectrum of metabolites extracted from Caco-2 cells fed with glucose. 132

2,3

4 1 13 4 14 11,12 10,11

12 3

2.75 2.70 2.65 2.602.55 2.50 2.45 2.40 2.35 2.302.25 2.20 2.15 2.10 2.05 2.00 1.95 1

5 9 7 16

14 1.25 1.20 1.15 1.10 1.05 1.00 0.95 0.90 0.85 6

20 18

18 15 DSS 19 10 16 8

4.6 4.4 4.2 4.03.8 3.63.4 3.23.0 2.82.6 2.42.2 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0 Chemical Shift (ppm)

Figure 4.6 Aliphatic region of a representative 1D 1H NMR noesygppr1d spectrum of metabolites extracted from Caco-2 cells fed with glucose. Identified metabolites have been labeled with numbers, according to Table 4.1. 133

28 27 21

29 23 21 23 26 26 21 26 21

21 21 24 25 25 22

9.4 9.2 9.08.8 8.68.4 8.28.0 7.87.6 7.47.2 7.0 6.8 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 Chemical Shift (ppm)

Figure 4.7 Aromatic region of a representative 1D 1H NMR noesygppr1d spectrum of metabolites extracted from Caco-2 cells fed with glucose. Identified metabolites have been labeled with numbers, according to Table 4.1 134

135

1.2

A 1 A

AB 0.8 taurine High glucose BC BC Low glucose 0.6 C Maltose Maltotriose Maltose+low glucose 0.4 Sucrose Relative populations Relative of

0.2

0 Type of carbohydrate

Figure 4.8 Relative population of taurine in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose over 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco-2 cells. Different letters denote significant differences at α=0.05.

136

1.6 *

1.4

1.2

taurine 1 *

0.8 * * * * * * * 0.6 * * * Relative populations Relative of 0.4

0.2

0 12 24 36 48 72 Time (h)

High glucose Low glucose Maltose Maltotriose Maltose+low glucose Sucrose

Figure 4.9 Relative population of taurine in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours. Results represent the mean ± SEM of at least three wells of treated Caco-2 cells. Asterisks denote significant differences at α=0.05 between the carbohydrate treatments at that time point using high glucose as the reference.

137

2.5

A 2

AB AB -inositol -inositol B B B High glucose

myo 1.5 Low glucose Maltose Maltotriose 1 Maltose+low glucose Sucrose Relative populations Relative of 0.5

0 Type of carbohydrate

Figure 4.10 Relative population of myo-inositole in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose over 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Different letters denote significant difference at α=0.05.

138

3.5

2.5 -inositol -inositol myo 2 * * * * * * * * 1.5 * * *

1 Relative populations Relative of

0.5

0 12 24 36 48 72 Time (h) High glucose Low glucose Maltose Maltotriose Maltose+low glucose Sucrose

Figure 4.11 Relative population of myo-inositol in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Asterisks denote significant differences at α=0.05 between the carbohydrate treatments at that time point using high glucose as the reference.

139

2.5 B B B B

2 C High glucose

phosphorylcholine Low glucose 1.5 Maltose Maltotriose Maltose+low glucose 1 Sucrose

Relative populations Relative of 0.5

0 Type of carbohydrate

Figure 4.12 Relative population of phosphorylcholine in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose over 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Different letters denote significant difference at α=0.05.

140

3 * * * * * * 2.5 * * * * * * * * * * * * *

2 * * phosphorylcholine 1.5

1 Relative populations Relative of 0.5

0 12 24 36 48 72 Time (h) High glucose Low glucose Maltose Maltotriose Maltose+low glucose Sucrose

Figure 4.13 Relative population of phosphorylcholine in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Asterisks denote significant differences at α=0.05 between the carbohydrate treatments at that time point using high glucose as the reference.

141

A AB AB 2.5 BC C

2 D High glucose Low glucose glycerophosphocholine

1.5 Maltose Maltotriose Maltose+low glucose 1 Sucrose

0.5 Relative populations Relative of

0 Type of carbohydrate

Figure 4.14 Relative population of glycerophosphocholine in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose over 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Different letters denote significant difference at α=0.05.

142

3 * * * * * * * * * * * * * * * * * 2.5 * * * * *

2 * *

glycerophosphocholine 1.5

0.5 Relative populations Relative of

0 12 24 36 48 72 Time (h)

High glucose Low glucose Maltose Maltotriose Maltose+low glucose Sucrose

Figure 4.15 Relative population of glycerophosphocholine in Caco-2 cells in presence of different types of carbohydrates. Cells were fed carbohydrate-free media containing 25 mM glucose (high glucose), 5.5 mM glucose (low glucose), 12.5 mM maltose, 8.33 mM maltotriose, 10 mM maltose and 5.5 mM glucose (low glucose+maltose), or 12.5 mM sucrose for 12, 24, 48, and 72 hours. The data was normalized with time zero (after cells differentiated). Results represent the mean ± SEM of at least three wells of treated Caco- 2 cells. Asterisks denote significant differences at α=0.05 between the carbohydrate treatments at that time point using high glucose as the reference.

143

4.6 References

Anderssen, E., Dyrstad, K., Westad, F., & Martens, H. (2006). Reducing over-optimism in variable selection by cross-model validation. Chemometrics and Intelligent Laboratory Systems, 84(1), 69-74.

Barker, M., & Rayens, W. (2003). Partial least squares for discrimination. Journal of Chemometrics, 17(3), 166-173.

Beckonert, O., Keun, H. C., Ebbels, T. M., Bundy, J., Holmes, E., Lindon, J. C., & Nicholson, J. K. (2007). Metabolic profiling, metabolomic and metabonomic procedures for NMR spectroscopy of urine, plasma, serum and tissue extracts. Nature Protocols, 2(11), 2692-2703.

Beetsch, J. W., & Olson, J. E. (1998). Taurine synthesis and cysteine metabolism in cultured rat astrocytes: effects of hyperosmotic exposure. American Journal of Physiology - Cell Physiology, 274, C866–874.

Bolten, C. J., & Wittmann, C. (2008). Appropriate sampling for intracellular amino acid analysis in five phylogenetically different yeasts. Biotechnology Letters, 30(11), 1993- 2000.

Bond Jr, J. H., & Levitt, M. D. (1976). Fate of soluble carbohydrate in the colon of rats and man. Journal of Clinical Investigation, 57(5), 1158.

Cani, P. D., Neyrinck, A. M., Maton, N., & Delzenne, N. M. (2005). Oligofructose Promotes Satiety in Rats Fed a High‐Fat Diet: Involvement of Glucagon‐Like Peptide‐1. Obesity Research, 13(6), 1000-1007.

Castrillo, J. I., Hayes, A., Mohammed, S., Gaskell, S. J., & Oliver, S. G. (2003). An optimized protocol for metabolome analysis in yeast using direct infusion electrospray mass spectrometry. Phytochemistry, 62(6), 929-937.

Cherbut, C. (2003). Motor effects of short-chain fatty acids and lactate in the gastrointestinal tract. Proceedings of the Nutrition Society, 62(01), 95-99.

Craig, A., Cloarec, O., Holmes, E., Nicholson, J. K., & Lindon, J. C. (2006). Scaling and normalization effects in NMR spectroscopic metabonomic data sets. Analytical Chemistry, 78(7), 2262-2267.

144

Davis, R. A., Charlton, A. J., Godward, J., Jones, S. A., Harrison, M., & Wilson, J. C. (2007). Adaptive binning: An improved binning method for metabolomics data using the undecimated wavelet transform. Chemometrics and Intelligent Laboratory Systems, 85(1), 144-154.

De Meyer, T., Sinnaeve, D., Van Gasse, B., Tsiporkova, E., Rietzschel, E. R., De Buyzere, M. L., & Van Criekinge, W. (2008). NMR-based characterization of metabolic alterations in hypertension using an adaptive, intelligent binning algorithm. Analytical Chemistry, 80(10), 3783-3790.

Delaglio, F., Grzesiek, S., Vuister, G. W., Zhu, G., Pfeifer, J., & Bax, A. D. (1995). NMRPipe: a multidimensional spectral processing system based on UNIX pipes. Journal of Biomolecular NMR, 6(3), 277-293.

Delmée, E., Cani, P. D., Gual, G., Knauf, C., Burcelin, R., Maton, N., & Delzenne, N. M. (2006). Relation between colonic proglucagon expression and metabolic response to oligofructose in high fat diet-fed mice. Life Sciences, 79(10), 1007-1013.

Delzenne, N. M., Cani, P. D., Daubioul, C., & Neyrinck, A. M. (2005). Impact of inulin and oligofructose on gastrointestinal peptides. British Journal of Nutrition, 93(S1), S157- S161.

Dettmer, K., Nürnberger, N., Kaspar, H., Gruber, M. A., Almstetter, M. F., & Oefner, P. J. (2011). Metabolite extraction from adherently growing mammalian cells for metabolomics studies: optimization of harvesting and extraction protocols. Analytical and Bioanalytical Chemistry, 399(3), 1127-1139.

Dieterle, F., Ross, A., Schlotterbeck, G., & Senn, H. (2006). Probabilistic quotient normalization as robust method to account for dilution of complex biological mixtures. Application in 1H NMR metabonomics. Analytical Chemistry, 78(13), 4281-4290.

Hashimoto, K., Matsunaga, N., & Shimizu, M. (1994). Effect of vegetable extracts on the transepithelial permeability of the human intestinal Caco-2 cell monolayer. Bioscience, Biotechnology, and Biochemistry, 58(7), 1345-1346.

Jenkins, D. J., & Kendall, C. W. (2002). Glycemic index: overview of implications in health and disease. The American Journal of Clinical Nutrition. 76(1): 266S- 273S.

Jensen-Jarolim, E., Gajdzik, L., Haberl, I., Kraft, D., Scheiner, O., & Graf, J. (1998). Hot spices influence permeability of human intestinal epithelial monolayers. The Journal of Nutrition, 128(3), 577-581.

145

Faijes, M., Mars, A. E., & Smid, E. J. (2007). Comparison of quenching and extraction methodologies for metabolome analysis of Lactobacillus plantarum. Microbial Cell Factories, 6(1), 27.

Farquhar, M. G., & Palade, G. E. (1963). Junctional complexes in various epithelia. The Journal of Cell Biology, 17(2), 375-412.

Fasano, A. (2011). Zonulin and its regulation of intestinal barrier function: the biological door to inflammation, autoimmunity, and cancer. Physiological reviews, 91(1), 151-175. Johnson, K. (2013). Leaky Gut Syndrome. Encyclopedia of Autism Spectrum Disorders, 1706-1712.

Fiehn, O. (2002). Metabolomics–the link between genotypes and phenotypes. Plant Molecular Biology, 48(1-2), 155-171.

Florian, C. L., Preece, N. E., Bhakoo, K. K., Williams, S. R., & Noble, M. D. (1995). Cell type-specific fingerprinting of meningioma and meningeal cells by proton nuclear magnetic resonance spectroscopy. Cancer Research, 55(2), 420-427.

Gates, S. C., & Sweeley, C. C. (1978). Quantitative metabolic profiling based on gas chromatography. Clinical Chemistry, 24(10), 1663-1673.

Gee, J. M., & Johnson, I. T. (2005). Dietary lactitol fermentation increases circulating peptide YY and glucagon-like peptide-1 in rats and humans. Nutrition, 21(10), 1036- 1043.

Gottschalk, M., Ivanova, G., Collins, D. M., Eustace, A., O'connor, R., & Brougham, D. F. (2008). Metabolomic studies of human lung carcinoma cell lines using in vitro1H NMR of whole cells and cellular extracts. NMR in Biomedicine, 21(8), 809-819.

Gu, H., Pan, Z., Xi, B., Asiago, V., Musselman, B., & Raftery, D. (2011). Principal component directed partial least squares analysis for combining nuclear magnetic resonance and mass spectrometry data in metabolomics: Application to the detection of breast cancer. Analytica Chimica Acta, 686(1), 57-63.

Guan, W., Zhou, M., Hampton, C. Y., Benigno, B. B., Walker, L. D., Gray, A., & Fernández, F. M. (2009). Ovarian cancer detection from metabolomic liquid chromatography/mass spectrometry data by support vector machines. BMC Bioinformatics, 10(1), 259.

146

Horning, E. C., & Horning, M-G. (1971). Metabolic profiles: gas-phase methods for analysis of metabolites. Clinical Chemistry 17.8, 802-809.

Khoo, S. H. G., & Al-Rubeai, M. (2009). Metabolic characterization of a hyper- productive state in an antibody producing NS0 myeloma cell line. Metabolic Engineering, 11(3), 199-211.

Khoo, S. H. G., & Al-Rubeai, M. (2007). Metabolomics as a complementary tool in cell culture. Biotechnology and Applied Biochemistry, 47(2), 71-84.

Koning, W. D., & Dam, K. V. (1992). A method for the determination of changes of glycolytic metabolites in yeast on a subsecond time scale using extraction at neutral pH. Analytical Biochemistry, 204(1), 118-123.

Lee, I. J., Hom, K., Bai, G., & Shapiro, M. (2009). NMR metabolomic analysis of caco-2 cell differentiation. Journal of Proteome Research, 8(8), 4104-4108.

Le Belle, J. E., Harris, N. G., Williams, S. R., & Bhakoo, K. K. (2002). A comparison of cell and tissue extraction techniques using high resolution 1H NMR spectroscopy. NMR in Biomedicine, 15(1), 37-44.

Lu, S., Gough, A. W., Bobrowski, W. F., & Stewart, B. H. (1996). Transport properties are not altered across Caco-2 cells with heightened TEER despite underlying physiological and ultrastructural changes. Journal of Pharmaceutical Sciences, 85(3), 270-273.

Madara, J. L., & Pappenheimer, J. R. (1987). Structural basis for physiological regulation of paracellular pathways in intestinal epithelia. The Journal of Membrane Biology, 100(1), 149-164.

Matter, K., Aijaz, S., Tsapara, A., & Balda, M. S. (2005). Mammalian tight junctions in the regulation of epithelial differentiation and proliferation. Current Opinion in Cell Biology, 17(5), 453-458.

Mitchell, S., Holmes, E., & Carmichael, P. (2002). Metabonomics and medicine: the Biochemical Oracle. Biologist, 49(5), 217-221.

147

Nicholson, J. K., Lindon, J. C., & Holmes, E. (1999). 'Metabonomics': understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica, 29(11), 1181- 1189.

Oliver, S. G., Winson, M. K., Kell, D. B., & Baganz, F. (1998). Systematic functional analysis of the yeast genome. Trends in Biotechnology, 16(9), 373-378.

Pappenheimer, J. R. (1987). Physiological regulation of transepithelial impedance in the intestinal mucosa of rats and hamsters. The Journal of Membrane Biology, 100(1), 137- 148.

Patterson, A. D., Li, H., Eichler, G. S., Krausz, K. W., Weinstein, J. N., Fornace, A. J., & Idle, J. R. (2008). UPLC-ESI-TOFMS-based metabolomics and gene expression dynamics inspector self-organizing metabolomic maps as tools for understanding the cellular response to ionizing radiation. Analytical Chemistry, 80(3), 665-674.

Ritter, J. B., Genzel, Y., & Reichl, U. (2008). Simultaneous extraction of several metabolites of energy metabolism and related substances in mammalian cells: optimization using experimental design. Analytical Biochemistry, 373(2), 349-369.

Sakata, T. (1987). Stimulatory effect of short-chain fatty acids on epithelial cell proliferation in the rat intestine: a possible explanation for trophic effects of fermentable fibre, gut microbes and luminal trophic factors. British Journal of Nutrition, 58(01), 95- 103.

Satsu, H., Miyamoto, Y., & Shimizu, M. (1999). Hypertonicity stimulates taurine uptake and transporter gene expression in Caco-2 cells. Biochimica et Biophysica Acta, 1419, 89–96.

Sellick, C. A., Hansen, R., Maqsood, A. R., Dunn, W. B., Stephens, G. M., Goodacre, R., & Dickson, A. J. (2008). Effective quenching processes for physiologically valid metabolite profiling of suspension cultured mammalian cells. Analytical Chemistry, 81(1), 174-183.

Schnabel, M., Marlovits, S., Eckhoff, G., Fichtel, I., Gotzen, L., Vecsei, V., & Schlegel, J. (2002). Dedifferentiation-associated changes in morphology and gene expression in primary human articular chondrocytes in cell culture. Osteoarthritis and Cartilage, 10(1), 62-70.

148

Schumann, A., Nutten, S., Donnicola, D., Comelli, E. M., Mansourian, R., Cherbut, C. & Garcia-Rodenas, C. (2005). Neonatal antibiotic treatment alters gastrointestinal tract developmental gene expression and intestinal barrier transcriptome. Physiological Genomics, 23(2), 235-245.

Sousa S.A.A., Magalhães A., & Ferreira M.M.C. (2013). Optimized bucketing for NMR spectra: Three case studies. Chemometrics and Intelligent Laboratory Systems, 122:93– 102.

Stretch, C., Eastman, T., Mandal, R., Eisner, R., Wishart, D. S., Mourtzakis, M., & Baracos, V. E. (2012). Prediction of skeletal muscle and fat mass in patients with advanced cancer using a metabolomic approach. The Journal of Nutrition, 142(1), 14-21.

Tate, A. R., Crabb, S., Griffiths, J. R., Howells, S. L., Mazucco, R. A., Rodrigues, L. M., & Watson, D. (1995). Lipid metabolite peaks in pattern recognition analysis of tumour in vivo MR spectra. Anticancer Research, 16(3B), 1575-1579.

Tate, A. R., Griffiths, J. R., Martínez-Pérez, I., Moreno, A., Barba, I., Cabañas, M. E., ... & Arús, C. (1998). Towards a method for automated classification of 1H MRS spectra from brain tumours. NMR in Biomedicine, 11(45), 177-191.

Teng, Q., Huang, W., Collette, T. W., Ekman, D. R., & Tan, C. (2009). A direct cell quenching method for cell-culture based metabolomics. Metabolomics, 5(2), 199-208.

Tweeddale, H., Notley-McRobb, L., & Ferenci, T. (1998). Effect of slow growth on metabolism of Escherichia coli, as revealed by global metabolite pool (“metabolome”) analysis. Journal of Bacteriology, 180(19), 5109-5116.

Ullrich, N., Caplanusi, A., Brone, B., Hermans, D., Lariviere, E., Nilius, B., Van Driessche, W., & Eggermont, J. (2006). Stimulation by caveolin-1 of the hypotonicity- induced release of taurine and ATP at basolateral, but not apical, membrane of Caco-2 cells. American Journal of Physiology - Cell Physiology, 290, C1287–1296.

Vachon, P. H., & Beaulieu, J. F. (1992). Transient mosaic patterns of morphological and functional differentiation in the Caco-2 cell line. Gastroenterology, 103(2), 414-423.

Van Vliet, E., Morath, S., Eskes, C., Linge, J., Rappsilber, J., Honegger, P., Hartung, T., & Coecke, S. (2008). A novel in vitro metabolomics approach for neurotoxicity testing, proof of principle for methyl mercury chloride and caffeine. Neurotoxicology, 29(1), 1- 12.

149

Veselkov, K. A., Lindon, J. C., Ebbels, T. M., Crockford, D., Volynkin, V. V., Holmes, E., & Nicholson, J. K. (2008). Recursive segment-wise peak alignment of biological 1H NMR spectra for improved metabolic biomarker recovery. Analytical Chemistry, 81(1), 56-66.

Viant, M. R., Bundy, J. G., Pincetich, C. A., de Ropp, J. S., & Tjeerdema, R. S. (2005). NMR-derived developmental metabolic trajectories: an approach for visualizing the toxic actions of trichloroethylene during embryogenesis. Metabolomics, 1(2), 149-158.

Watts, T., Berti, I., Sapone, A., Gerarduzzi, T., Not, T., Zielke, R., & Fasano, A. (2005). Role of the intestinal tight junction modulator zonulin in the pathogenesis of type I diabetes in BB diabetic-prone rats. Proceedings of the National Academy of Sciences USA, 102(8), 2916-2921.

Westerhuis, J. A., Hoefsloot, H. C., Smit, S., Vis, D. J., Smilde, A. K., van Velzen, E. J., & Van Dorsten, F. A. (2008). Assessment of PLSDA cross validation. Metabolomics, 4(1), 81-89.

Wold, S., & Sjöström, M. (1987). Comments on a recent evaluation of the SIMCA method. Journal of Chemometrics, 1(4), 243-245.

Wold, S., Sjöström, M., & Eriksson, L. (2001). PLS-regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratory Systems, 58(2), 109-130.

Yeagle, P. L. (Ed.). (2010). The structure of biological membranes. CRC press.

150

CHAPTER 5. OVERALL CONCLUSIONS AND FUTURE RECOMMENDATIONS

The findings of this thesis supports the overall hypothesis that maltooligosaccharides (generated by α-amylase degradation of dietary starch) cause the maturation, trafficking, and activation of HMW SI to the apical membrane of intestinal enterocytes. It additionally provided further evidence of maltose sensing at the enterocyte surface that signaled cell differentiation events and an increase in monolayer tight junctions as measured by cell permeability tests. Maltooligosaccharide sensing appeared to occur via the gut sweet taste receptor to trigger the processing and trafficking of SI.

The expression of sweet taste receptor subunits (T1R2/T1R3) was increased in presence of maltose. The increase of T1R2 expression was 30-fold for maltose compared to 18- fold for glucose. Also the expression of T1R3 increased 9-fold for maltose compared to

6-fold for glucose. It seems sweet taste receptor subunits sense the existence of maltooligosaccharides via the gastrointestinal track and signal the cell to express a higher molecular weight species of SI that is mobilized to the enterocyte apical surface and therein activated.

This increase in molecular weight in SI is due to a post-translational events. The deglycosylation endo F and endo H enzyme treatments showed that most of the new glycosylations on SI are the N-glycosylation type, with some possible O-glycosylation.

SI glycosylation occurs at its co-translation and post-translation stages and is critical to

151

the maturation of the SI enzyme active form. As shown in Chapter 3, the new N- and O-linked glycans promoted different events related to maturation and trafficking of

SI, as shown in a series of experiments described as follows.

First, the change in glycosylation of SI induced by maltose altered the distribution of SI in cell compartments (intercellular vs. apical membrane). The result from this experiment showed significant increase in the amount of SI on apical side of membrane.

Second, this study showed that the extra O-linked glycans on SI induced by maltose enhanced association of SI with lipid rafts. SI is more associated with the lipid raft region on the enterocyte membrane (Le Bivic et al., 1990), which has significance to starch digestion as SI has higher activity when it is found in lipid raft regions (Alfalah et al., 1999). This sorting of SI to lipid raft happens via O-linked glycans (Jacob et al.,

2000).

Third, the pulse-chase study clearly revealed that the mature form of SI (fully glycosylated and active form) was induced by maltose within one hour. This shows that maltose not only induces the formation of higher N- and O-glycosylated SI, but also the mature form of SI trafficks faster to the apical membrane for digestion of maltooligosaccharides in small intestine. This could be of practical importance to the body related to the ileal brake mechanism, as ileal received maltooligosaccharides would signal the final maturation and activation of SI within a short time period where it could digest well the starchy food retained in the stomach to be releases over longer postprandial period.

152

Fourth, the enzyme activity of high molecular SI revealed that maltase and sucrase activity was similar to SI induce by glucose. This is an important finding considering the high amount of N- and O-glycosylation did not have an effect on enzyme activity of HMW SI.

Finally, we study the effect of maltooligosaccharides on two important characters of Caco-2 cell line; intestinal permeability and cell differentiation. While the main function of small intestine is digestion and absorption of nutrients, it also functions as a barrier preventing harmful substances from entering the body. While high glucose concentration induces tight junctions, maltose promoted lower permeability compared to other carbohydrates tested.

Cell differentiation is another important character of small intestinal enterocytes.

Cell differentiation is the process by which a less specialized cell becomes a more specialized cell type. Monitoring alteration of metabolites in presence of maltooligosaccharides showed that maltose induced a significant increase in taurine level.

In addition, glycerophosphocholine and phosphorycholine levels significantly increased in presence of maltooligosaccharides. Considering these metabolites are known to be biomarkers for cell differentiation, it appears maltooligosaccharides improve this important cell function. Enhancement of cell differentiation rate with maltose may be then related to the observed lower monolayer permeability.

Maltooligosaccharides, as important digestion products of starch by α-amylase, symbolize the presence of glucose-composed starch, which is the main glycemic carbohydrate on which humans evolved. Chronic rapid digestion of glycemic carbohydrates and absorption of glucose can be detrimental to health, particularly for

153 diabetics and pre-diabetics, due to glycemic spikes that stress the glucose homeostasis regulatory system. This study presented evidence that the sweet taste receptor senses maltooligosaccharides and triggers intracellular pathways to express HMW SI. We showed that HMW SI can traffic faster to the apical membrane and digest starch presence in small intestine. An approach might be envisioned to be developed to manipulate the block sweet taste receptor so as to modulate postprandial glycemic response for improved health.

To apply this approach (control of glucogenesis) with the concept of controlled glucose delivery to the body, further investigations are required related to pathway analysis of the sweet taste receptor and formation of HMW SI. Further study is also recommended to evaluate the expression of gut hormones in presence of maltooligosaccharides, as this may be the mechanism of carbohydrates triggering the ileal brake and gut-brain access for appetite and food intake control. Since Caco-2 cells only express SI, in vivo study is needed to evaluate the effect of maltooligosaccharides also on

MGAM.

154

1.5 References

VITA

155

VITA

Mohammad Chegeni was born in Iran. He received the degree of Bachelor of

Science in Biology from Ferdowsi University, Mashhad, Iran, in 2001. He entered the

Graduate School in the Department of Health Science at Ball State University, Muncie,

IN, in 2007, under the supervision of Dr. James McKenzie, and received a Master's of

Science degree in May 2009. Mohammad started his Ph.D. studies in the Department of

Food Science at Purdue University, West Lafayette, IN, in 2010. While at Purdue

University he joined Professor Bruce Hamaker’s research group where he studied maltose sensing and sucrase-isomaltase maturation and trafficking.