<%BANNER%>

Factors Impacting Fructose Bioavailability and Its Adverse Metabolic Effects.

Permanent Link: http://ufdc.ufl.edu/UFE0041402/00001

Material Information

Title: Factors Impacting Fructose Bioavailability and Its Adverse Metabolic Effects.
Physical Description: 1 online resource (306 p.)
Language: english
Creator: Le, Myphuong
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adverse, caco, corn, drink, effects, expression, fructose, gene, glucose, hfcs, high, khk, metabolic, permeability, pharmacodynamic, pharmacokinetic, polymorphisms, si, slc2a2, slc2a5, slc5a1, soft, sucrose, syrup
Pharmaceutics -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Fructose consumption in the United States has spiked since the past four decades. There is growing evidence that excessive consumption of fructose may play a pivotal role in the current epidemic of a number of health disorders. The objective of these studies was to determine factors that may impact fructose bioavailability and its adverse metabolic effects. From the genetic association study, several potentially interesting polymorphisms were associated with various metabolic phenotypes, including triglycerides, serum uric acid, home diastolic and systolic blood pressure, glucose, high-density lipoprotein (HDL), and body mass index. Importantly, SLC2A2 (solute carrier family 2, member 2) rs8192675 was significantly associated with a decrease in HDL levels in European Americans in the PEAR study and was replicated in the GERA study. From the clinical study, we detected higher fructose concentrations in the systemic circulation from high fructose corn syrup (HFCS) versus sucrose. There were treatment differences in triglycerides, systolic blood pressure, and serum uric acid levels. However, we were unable to detect correlations between higher fructose AUC (area under the curve of plasma concentrations versus time) or Cmax (maximum observed concentration) with higher fructose-induced metabolic effects. In addition, we did not detect any strong correlations between higher chronic fructose intake and increased metabolic responses. We did detect almost a four-fold difference in the inter-individual variability in fructose AUC from HFCS. From the in vitro studies, we detected that sugar composition, sugar concentration, and duration of sugar exposure impacted fructose transport and expression of genes involved in fructose absorption and metabolism. We did not detect a significant impact of glucose on fructose permeability. However, after exposing cells to high glucose dose for 72 hr, glucose was shown to have strong negative regulatory impact on the expressions of SI (sucrase-isomaltase), SLC2A2, SLC2A5 (solute carrier family 2, member 5), SLC5A1 (solute carrier family 5, member 1), and KHK (ketokexokinase). The studies conducted and described herein showed that various factors may influence an individual s exposure and response to fructose. Further studies are needed to elucidate the long-term effects of these factors on the inter-individual variability towards the development of fructose-induced adverse metabolic effects and the increased risk for developing various health problems, such as cardiovascular disease and metabolic syndrome.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Myphuong Le.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Johnson, Julie A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041402:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041402/00001

Material Information

Title: Factors Impacting Fructose Bioavailability and Its Adverse Metabolic Effects.
Physical Description: 1 online resource (306 p.)
Language: english
Creator: Le, Myphuong
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2010

Subjects

Subjects / Keywords: adverse, caco, corn, drink, effects, expression, fructose, gene, glucose, hfcs, high, khk, metabolic, permeability, pharmacodynamic, pharmacokinetic, polymorphisms, si, slc2a2, slc2a5, slc5a1, soft, sucrose, syrup
Pharmaceutics -- Dissertations, Academic -- UF
Genre: Pharmaceutical Sciences thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Fructose consumption in the United States has spiked since the past four decades. There is growing evidence that excessive consumption of fructose may play a pivotal role in the current epidemic of a number of health disorders. The objective of these studies was to determine factors that may impact fructose bioavailability and its adverse metabolic effects. From the genetic association study, several potentially interesting polymorphisms were associated with various metabolic phenotypes, including triglycerides, serum uric acid, home diastolic and systolic blood pressure, glucose, high-density lipoprotein (HDL), and body mass index. Importantly, SLC2A2 (solute carrier family 2, member 2) rs8192675 was significantly associated with a decrease in HDL levels in European Americans in the PEAR study and was replicated in the GERA study. From the clinical study, we detected higher fructose concentrations in the systemic circulation from high fructose corn syrup (HFCS) versus sucrose. There were treatment differences in triglycerides, systolic blood pressure, and serum uric acid levels. However, we were unable to detect correlations between higher fructose AUC (area under the curve of plasma concentrations versus time) or Cmax (maximum observed concentration) with higher fructose-induced metabolic effects. In addition, we did not detect any strong correlations between higher chronic fructose intake and increased metabolic responses. We did detect almost a four-fold difference in the inter-individual variability in fructose AUC from HFCS. From the in vitro studies, we detected that sugar composition, sugar concentration, and duration of sugar exposure impacted fructose transport and expression of genes involved in fructose absorption and metabolism. We did not detect a significant impact of glucose on fructose permeability. However, after exposing cells to high glucose dose for 72 hr, glucose was shown to have strong negative regulatory impact on the expressions of SI (sucrase-isomaltase), SLC2A2, SLC2A5 (solute carrier family 2, member 5), SLC5A1 (solute carrier family 5, member 1), and KHK (ketokexokinase). The studies conducted and described herein showed that various factors may influence an individual s exposure and response to fructose. Further studies are needed to elucidate the long-term effects of these factors on the inter-individual variability towards the development of fructose-induced adverse metabolic effects and the increased risk for developing various health problems, such as cardiovascular disease and metabolic syndrome.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Myphuong Le.
Thesis: Thesis (Ph.D.)--University of Florida, 2010.
Local: Adviser: Johnson, Julie A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-04-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2010
System ID: UFE0041402:00001


This item has the following downloads:


Full Text

PAGE 1

1 FACTORS IMPACTING FRUCTOSE BIOAVAILABILITY AND ITS ADVERSE METABOLIC EFFECTS By MYPHUONG THI LE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR T HE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2010

PAGE 2

2 2010 MyPhuong Thi Le

PAGE 3

3 To my family: my father, Thanh Quang Le, my mother, Bay Vo Le, my sisters, BachCuc, MyHanh, MyDung, and MyLien Thi Le, and my brothers, Tien, Phuoc and Tri Quang Le

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my graduate supervisory committee, Drs. Julie Johnson, Richard Johnson, Reginald Frye, Mark Segal, and Sihong Song, for t heir invaluable wisdom, guidance, and encouragement while working on my research project. I would also like to thank Dr. Ed Mougey Dr. Elizabeth Dudenhausen, and Dr. A lan Shuldiner for their collaborative efforts. I am also grateful for the support and advice from my colleagues and all faculty and staff members in the department of Pharmaceutics and the department of Pharmacotherapy and Translational Research In particular, Dr Taimour Langaee, Dr. Yan Gong, Dr. Jing Cheng, Dr. Jaekyu Shin, Dr. Michael Pac a nowski, Dr. Anzeela Schentrup, Dr Maximillian Lobmeyer, Dr. Elvin Price, Dr. Hong Li, Dr. Kim McFann, Dr. Mohamed Mohamed, Dr. Mohamed Shafiu, Dr. Jason Karnes, Ms. Cheryl Galloway, Mr. Hrishikesh Navare, Mr. Ben Burkley, Ms. Lynda Stauffer and Dr. Marcus Campbell for their assistance. I would like to acknowledge the GCRC and its staff for all their efforts and help with the clinical study. I would like to especially thank my family and friends for their love, support, encouragement, and laughter: my parents, Thanh and Bay ; my sisters BachCuc, MyHanh, MyDung, MyLien; my brothers Tien, Phuoc and Tri ; the little ones Trang, Thao, Tai, Loc, Khiem, Kieu, Kelsey, Nhuy, and Ky; my friends Carisa Miranda, Matt Maza, and Leah Villegas. I would also like t o thank Mmi for her dedication, effort, and sense of humor in helping me achieve this goal.

PAGE 5

5 TABLE OF CONTENTS ACKNOWLEDGMENTS .................................................................................................. 4 page LIST OF TABLES .......................................................................................................... 10 LIST OF FIGURES ........................................................................................................ 13 ABSTRACT ................................................................................................................... 18 CHAPTER 1 BACKGROUND AND SIGNIFICANCE ................................................................... 20 Introduction ............................................................................................................. 20 Dietary Sources and Consumption of Fructose ...................................................... 20 Structure of Fructose .............................................................................................. 21 Absorption of Fructose ............................................................................................ 21 Metabolism of Fructose ........................................................................................... 22 Toxicities of Fructose & Potential Role in Disease Risk .......................................... 23 Metabolic Disorders and Genetic Polymorphisms of Fructose ................................ 24 Fructose Bioavailability and Metabolic Effects Research Plans .............................. 25 2 GENOTYPE ASSOCIATION STUDY: IMPACT OF GENETIC POLYMORPHISMS OF SLC2A2, SLC2A5 SI AND KHK ON METABOLIC PHENOTYPES IN PEAR, HAPI, AND GERA ......................................................... 35 Introduction ............................................................................................................. 35 Materials and Methods ............................................................................................ 36 Study Populations ............................................................................................ 36 PEAR ......................................................................................................... 36 HAPI Heart ................................................................................................. 36 GERA I and II ............................................................................................. 36 SNP Selection .................................................................................................. 37 Putatively functional SNPs ......................................................................... 37 IBC genotyping chip ................................................................................... 41 Custom OPA genotyping chip .................................................................... 42 DNA Isolation and Quantification ...................................................................... 43 Genotyping ....................................................................................................... 44 Data Analysis .......................................................................................................... 45 IBCchip Analysis ............................................................................................. 45 OPA Ana lysis ................................................................................................... 45 SNPTrait Associations ..................................................................................... 46 Results .................................................................................................................... 47 Genotype and Data Quality Control .................................................................. 47 Discovery Cohorts: Primary Phenotypes .......................................................... 48

PAGE 6

6 Uric acid ..................................................................................................... 48 Discovery Cohorts: Secondary Phenotypes ..................................................... 48 Triglycerides ............................................................................................... 48 HOMA ........................................................................................................ 49 HDBP ......................................................................................................... 49 HSBP ......................................................................................................... 49 Glucose ...................................................................................................... 50 Insulin ........................................................................................................ 50 HDL ............................................................................................................ 50 LDL ............................................................................................................ 50 BMI ............................................................................................................ 50 MS ............................................................................................................. 50 Summary of Associations from Discovery Cohorts ........................................... 50 Replication ........................................................................................................ 51 Conclusion .............................................................................................................. 52 3 CLINICAL STUDY: PHARMACOKINETICS AND PHARMACODYNAMICS OF FRUCTOSE FOLLOWING SOFT DRINK CONSUMPTION: SUCROSE VERSUS HIGH FRUCTOSE CORN SYRUP .......................................................... 87 Introduction ............................................................................................................. 87 Materials and Methods ............................................................................................ 89 Study Design .................................................................................................... 89 Exclusion Criteria ............................................................................................. 89 Finger stick blood glucose ......................................................................... 90 Pregnancy test ........................................................................................... 90 Participants ....................................................................................................... 90 Sugar Load from Soft Drinks ............................................................................ 91 Sugar content ............................................................................................. 91 Food Intake Data .............................................................................................. 91 Threeday food record ............................................................................... 92 Food frequency questionnaire .................................................................... 92 Physical Activity Data ....................................................................................... 93 Study Protocol .................................................................................................. 93 Treatment ................................................................................................... 93 Metabolic response phenotypes ................................................................ 94 Sample collections ..................................................................................... 94 Analytical Methods ........................................................................................... 95 BMI, hip and waist circumferences ............................................................ 95 Blood pressure and heart rate .................................................................... 96 Glucose and lactate ................................................................................... 96 Fructose ..................................................................................................... 96 Triglycerides, uric acid, creatinine, insulin, and FEUA ............................... 98 Data Analysis .......................................................................................................... 99 Pharmacokinetic Parameters ........................................................................... 99 Statistical Methods ......................................................................................... 100 Results .................................................................................................................. 101

PAGE 7

7 Conclusion ............................................................................................................ 105 4 IMPACT OF SUGAR EXPOSURE ON THE TRANSPORT OF FRUCTOSE IN CACO 2 CELLS .................................................................................................... 133 Introduction ........................................................................................................... 133 Materials and Methods .......................................................................................... 134 Cel l Line and Cell Culture ............................................................................... 134 Subculturing ................................................................................................... 135 Transwell Plates ............................................................................................. 135 Integrity of Caco 2 Monolayers ....................................................................... 135 TEER measurements ............................................................................... 135 Mannitol transport .................................................................................... 136 Sugars ............................................................................................................ 136 Transport Experiments ................................................................................... 136 Experimental design ................................................................................ 137 Sampling .................................................................................................. 137 Measurements ......................................................................................... 137 Data Analysis ........................................................................................................ 138 P ermeability .................................................................................................... 139 Results .................................................................................................................. 139 Evaluation of Monolayer Integrity ................................................................... 139 Effect of Sugar Composition and Concentration on Permeability ................... 140 Permeability of fructose ........................................................................... 140 Permeability of glucose ............................................................................ 141 Conclusion ............................................................................................................ 141 5 IMPACT OF SUGAR EXPOSURE ON THE EXPRESSION OF KHK, SI SLC2A2 SLC2A5 AND SLC5A1 IN CACO 2 CELLS .......................................... 152 Introduction ........................................................................................................... 152 Materials and Methods .......................................................................................... 153 Expression Experiments ................................................................................. 153 Experimental design ................................................................................ 153 Sampling .................................................................................................. 154 RNA ................................................................................................................ 154 cDNA ....................................................................................................... 155 RT PCR ................................................................................................... 155 Protein ............................................................................................................ 155 Quantification ........................................................................................... 155 Antibodies ................................................................................................ 156 Western Blot ............................................................................................ 157 Data A nalysis ........................................................................................................ 158 In Vitro Studies RT PCR ................................................................................. 158 Clinical Study RT PCR ................................................................................... 159 Results .................................................................................................................. 160 In Vitro Sugar Exposure Experiments ............................................................ 160

PAGE 8

8 Evaluation of monolayer integrity ............................................................. 160 Effects of sugar composition, sugar concentration, and duration of sugar exposure on gene expression ..................................................... 160 Conclusion ............................................................................................................ 163 6 SUMMARY AND CONCLUSION .......................................................................... 178 APPENDIX A GENOTYPE ASSOCIATION STUDY DATA ......................................................... 185 Sequencing and Genotyping Data ........................................................................ 185 HAPI Heart Data ................................................................................................... 192 B CLINICAL STUDY DOCUMENTS ........................................................................ 193 Volunteer Health Information Questionnaire ......................................................... 193 Sugar Analyses ..................................................................................................... 196 April 05, 2008 ................................................................................................. 196 May 12, 2009 .................................................................................................. 198 ThreeDay Food Record ....................................................................................... 200 Serving Size Booklet ............................................................................................. 202 Food Frequency Questionnaire ............................................................................ 208 International Physical Activity Questionnaire ........................................................ 214 Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire ........................................................................................ 218 QC Data ................................................................................................................ 226 C CLINICAL STUDY SECONDARY ANALYSES .................................................. 227 Correlation of AUC and Cmax Fructose with Changes in Metabolic Phenotypes 227 Correlation between Fructose Intake and Pretreatment (0 min) Levels of Metabolic Phenotypes ....................................................................................... 240 Impact of Fructose Intake on Changes in Metabolic Phenotypes ......................... 250 D CELL CULTURE MEDIA ....................................................................................... 260 20% FBS Growth Media for Cell Growth in Plates ................................................ 260 High Glucose Growth Media for Cell Growth in Transwell .................................... 260 Low Glucose Growth Media for Cell Growth in Transwell ..................................... 260 Subculturing Process ............................................................................................ 261 HBSS (Without Glucose) ...................................................................................... 261 Hank's Full Strength Composition .................................................................. 261 Hanks Stock Solutions ................................................................................... 262 Stock #1 ................................................................................................... 262 Stock #2 ................................................................................................... 262 Stock #3 ................................................................................................... 262 Stock #4 ................................................................................................... 262

PAGE 9

9 Stock #5 ................................................................................................... 262 Hank's Premix ( Combine the solutions in following order) .............................. 262 Hank's Full Strength (mix prior to use) ........................................................... 262 E In Vitro Transport DATA ........................................................................................ 263 Mannitol Transport and TEER Values ................................................................... 263 Fructose and Glucose Permeability Data .............................................................. 265 F WESTERN BLOT MEDIA AND PROTOCOLS ..................................................... 269 4X Sample Dilution Buffer ..................................................................................... 269 Running Buffer ...................................................................................................... 269 Transfer Buffer ...................................................................................................... 269 10X TBS (Tris buffered saline) ............................................................................. 269 Washing Buffer ..................................................................................................... 269 Blocking Reagent .................................................................................................. 270 Staining Protocol ................................................................................................... 270 Fast Green Stain (0.1%) ................................................................................. 270 Fast Green DeStain ...................................................................................... 270 Staining/De staining of the Blot ...................................................................... 270 G GENE EXPRESSION DATA ................................................................................. 271 In V itro RT PCR Results ....................................................................................... 271 Clinical RT PCR Results ....................................................................................... 284 LIST OF REFERENCES ............................................................................................. 291 BIOGRAPHICAL SKETCH .......................................................................................... 306

PAGE 10

10 LIST OF TABLES Table page 1 1 Sugar contents of fruits and vegetables (g / 100 g). ........................................... 26 1 2 Sugar contents of commercial sweeteners (g / 100 g). ....................................... 26 1 3 Summary of genes. ............................................................................................ 27 2 1 Su mmary of dbSNPs and pfSNPs for SLC2A2 SLC2A5 KHK, SI, and ALDOB. .............................................................................................................. 55 2 2 Primer design of amplicons for unvalidated pfSNPs of KHK. ............................. 56 2 3 Primer design of amplicons for unvalidated pfSNPs of SLC2A5 ........................ 58 2 4 Sequencing results of pfSNPs of KHK. ............................................................... 60 2 5 Seque ncing results of pfSNPs of SLC2A5 ......................................................... 61 2 6 Validated candidate pfSNPs for KHK, SI SLC2A2 and SLC2A5 ...................... 62 2 7 Quality control of PEAR genotyping data. .......................................................... 63 2 8 PEAR: Baseline characteristics. ......................................................................... 64 2 9 SNPs associated with baseline serum uric acid in PEAR. .................................. 65 2 10 SNPs associated with baseline triglycerides in PEAR. ....................................... 66 2 11 SNPs associated with baseline HOMA in PEAR. ............................................... 68 2 12 SNPs associated with baseline HDBP in PEAR. ................................................ 69 2 13 SLC2A5 SNPs associated with baseline HSBP in PEAR. .................................. 70 2 14 SNPs associated with baseline glucose in PEAR. .............................................. 71 2 15 SNPs associated with baseline insulin in PEAR. ................................................ 72 2 16 SNPs associated with baseline HDL in PEAR. ................................................... 73 2 17 SNPs associated with baseline LDL in PEAR. .................................................... 74 2 18 SNPs asso ciated with baseline BMI in PEAR. .................................................... 75 2 19 SNPs associated with baseline MS in PEAR. ..................................................... 76

PAGE 11

11 2 20 Summary of SNPs significantly associ ated after adjustments in PEAR. ............. 77 2 21 GERA I and II: Replication results. ..................................................................... 78 3 1 Study variables and blood and urine collectio n time points. ............................. 109 3 2 Baseline demographics. ................................................................................... 112 3 3 Metabolic phenotypes of study visits 1 and 2 measured prior to treatment. ..... 113 3 4 Effect of HFCS versus sucrose on the response of various metabolic phenotypes. ...................................................................................................... 117 3 5 Effect of HFCS versus sucrose on fructose concentrations over time. ............. 120 3 6 Effect of HFCS versus sucrose on glucose concentrations over time. ............. 122 3 7 Effect of HFCS versus sucrose on SUA concentrations over time. .................. 123 3 8 Effect of HFCS versus sucrose on Tg concentrations over time. ...................... 124 3 9 Effect of HFCS versus sucrose on SBP levels over time. ................................. 125 3 10 Effect of HFCS versus sucrose on insulin concentrations over time. ................ 126 3 11 Effect of HFCS versus sucrose on lactate concentrations over time. ............... 127 3 12 Effect of HFCS versus sucrose on DBP levels over time. ................................ 128 3 13 Effect of HFCS versus sucrose on HR over time. ............................................. 129 3 14 Changes in metabolic phenotypes from soft drinks. ......................................... 132 4 1 Sugar treatments of Caco 2 cells in transport experiments. ............................. 144 5 1 Sugar treatments of Caco 2 cells in expression experiments. .......................... 167 5 2 Comparison of gene expression between 1 hr vs 72 hr after exposure to 300 mM of sugar treatments. ................................................................................... 168 5 3 Comparison of gene expression between 5 mM vs 300 mM after exposur e to sugar treatments for 72 hr. ............................................................................... 169 A 1 List of Coriell genomic DNA samples used to determine allelic frequencies of pfSNPs. ............................................................................................................ 185 A 2 Primer design of amplicons for pfSNPs of ALDOB. .......................................... 186 A 3 Sequencing results of pfSNPs of ALDOB. ........................................................ 188

PAGE 12

12 A 4 Genotype results of PEAR samples using IBC and OPA genotyping chips. ..... 189 A 5 Associations with metabolic phenotypes in HAPI Heart. .................................. 192 B 1 Precisio n (R.S.D%) and accuracy (R.E.%) for plasma fructose (eight replicates per day for 3 days). .......................................................................... 226 C 1 Correlation of AUC fructose with changes in metabolic phenotypes. ............... 236 C 2 Correlation of Cmax fructose with changes in metabolic phenotypes. .............. 238 E 1 % of total mannitol transported during 60 min and TEER values. ..................... 263 G 1 Gene expression data for KHK, SI SLC2A2 SLC2A5 and SLC5A1 .............. 272

PAGE 13

13 LIST OF FIGURES Figure page 1 1 Chemical structures of fructose, glucose, and sucrose. ..................................... 28 1 2 Production of high fructose corn syrup from corn. .............................................. 29 1 3 U.S. per capita consumption of sugars and sweeteners. .................................... 30 1 4 Digestion and absorption of carbohydrates. ....................................................... 31 1 5 Metabolism of fructose. ...................................................................................... 32 1 6 Production of uric acid from the breakdown of ATP. ........................................... 33 1 7 U.S. per capita consumption of sweeteners and preval ence of overweight and obesity. ........................................................................................................ 34 2 1 Genotype calling from sequence alignment. ....................................................... 79 2 2 R2 plot of KHK candidate SNPs generated from HapMap data of YRI population. .......................................................................................................... 80 2 3 R2 plot of KHK candidate SNPs generated from HapMap data of CEU population. .......................................................................................................... 81 2 4 R2 plot of SLC2A2 candidate SNPs generated from HapMap data of YRI population. .......................................................................................................... 82 2 5 R2 plot of SLC2A2 candidate SNPs generated from HapMap data of CEU population. .......................................................................................................... 83 2 6 R2 plot of SLC2A5 candidate SNPs generated from HapMap data of YRI population. .......................................................................................................... 84 2 7 R2 plot of SLC2A5 candidate SNPs generated from HapMap data of C EU population. .......................................................................................................... 85 2 8 Effects of rs8192675 on HDL levels in the PEAR and GERA I and II study populations.. ....................................................................................................... 86 3 1 Study design. .................................................................................................... 110 3 2 Study enrollment. .............................................................................................. 111 3 3 Comparison of fructose AUC and Cmax between HFCS and sucrose. ............ 114 3 4 Comparison of glucose AUC and Cmax between HFCS and sucrose. ............ 115

PAGE 14

14 3 5 Comparison of Cmax between HFCS and sucrose. ......................................... 116 3 6 Comparison of fructose concentrations over time between HFCS and sucrose ............................................................................................................. 119 3 7 Comparison of glucose concentrations over time between HFCS and sucrose. ............................................................................................................ 121 3 8 Comparison of SUA concentrations over time between HFCS and sucrose. ... 123 3 9 Comparison of Tg concentrations over time between HFCS and sucrose. ...... 124 3 10 Comparison of SBP levels over time between HFCS and sucrose. .................. 125 3 11 Comparison of insulin concentrations over time between HFCS and sucrose. 126 3 12 Comparison of lactate concentrations over time between HFCS and sucrose. 127 3 13 Comparison of DBP levels over time between HFCS and sucrose ................. 128 3 14 Comparison of HR levels over time between HFCS and sucrose. .................... 129 3 15 Comparison of FEUA over time between HFCS and sucrose. ......................... 130 3 16 Correlation between fructose intake from 3Day Food Log and Food Frequency Questionnaire. ................................................................................ 131 4 1 Caco 2 monolayer on Transwell insert. .......................................................... 145 4 2 Ln apparent permeability of fructose vs fructose concentration. ....................... 146 4 3 Apparent permeation rate of fructose vs fructose concentration. ...................... 147 4 4 Ln apparent permeability of fructose vs glucose concentration. ....................... 148 4 5 Ln apparent permeability of glucose vs glucose concentration. ........................ 149 4 6 Apparent permeation rate of glucose vs glucose concentration. ...................... 150 4 7 Ln apparent permeability of glucose vs fructose concentration. ....................... 151 5 1 Relative quantity (RQ) of KHK expression. ....................................................... 170 5 2 Relative quantity (RQ) of SI expression. ........................................................... 171 5 3 Relative quantity (RQ) of SLC2A2 expression. ................................................. 172 5 4 Relative quantity (RQ) of SLC2A5 expression. ................................................. 1 73

PAGE 15

15 5 5 Relative quantity (RQ) of SLC5A1 expression. ................................................. 174 5 6 Comparison of HFCS vs Sucrose on the relative quantity of KHK and SLC2A5 after 6 hr exposure. ............................................................................ 175 5 7 Correlation of pretreatment expression levels of KHK with A) pretreatment concentrations of fructose, B) Cmax fructose, and C) AUC fructose. ............... 176 5 8 Correlation of pretreatment expression levels of SLC2A5 with A) pretreatment concentrations of fructose, B) Cmax fruc tose, and C) AUC fructose. ............................................................................................................ 177 C 1 Correlation of fructose levels with changes in glucose. .................................... 227 C 2 Correlation of fructose level s with changes in insulin. ...................................... 228 C 3 Correlation of fructose levels with changes in lactate. ...................................... 229 C 4 Correlation of fructose level s with changes in Tg. ............................................ 230 C 5 Correlation of fructose levels with changes in SBP. ......................................... 231 C 6 Correlation of fructose levels with ch anges in DBP. ......................................... 232 C 7 Correlation of fructose levels with changes in HR. ........................................... 233 C 8 Correlation of fructose levels with changes in SUA. ......................................... 234 C 9 Correlation of fructose levels with changes in FEUA. ....................................... 235 C 10 Correlation between fructose intake and pretreatment fructose levels. ............ 240 C 11 Correlation between fructose intake and pretreatment glucose levels. ............. 241 C 12 Correlation between fruc tose intake and pretreatment insulin levels. ............... 242 C 13 Correlation between fructose intake and pretreatment lactate levels. .............. 243 C 14 Correlation between fructose intake and pretreatment Tg levels ...................... 244 C 15 Correlation between fructose intake and pretreatment SBP levels. .................. 245 C 16 Correlation between fructose intake and pretreatment DBP levels. .................. 246 C 17 Correlation between fructose intake and pretreatment HR levels. .................... 247 C 18 Correlation between fructose intake and pretreatment SUA levels. .................. 248 C 19 Correlation between fructose intake and pretreatment FEUA levels. ............... 249

PAGE 16

16 C 20 Correlation of fructose intake with changes in fructose. ................................... 250 C 21 Correlation of fructose intake with changes in glucose. .................................... 251 C 22 Correlation of fructose intake with changes in insulin. ...................................... 252 C 23 Correlation of fructose intake with changes in lactate. ...................................... 253 C 24 Correlation of fructose intake with changes in Tg. ............................................ 254 C 25 Correlation of fructose intake with changes in SBP. ......................................... 255 C 26 Correlation of fructose intake with changes in DBP. ......................................... 256 C 27 Correlation of fructose intake with changes in HR. ........................................... 257 C 28 Correlation of fructose intake with changes in SUA. ......................................... 258 E 1 Permeability of fructose with varying fructose concentrations. ......................... 265 E 2 Permeability of fructose with varying glucose concentrations. .......................... 266 E 3 Permeability of glucose with varying glucose concentrations. .......................... 267 E 4 Permeability of glucose with varying fructose concentrations. .......................... 268 G 1 Relative quantity (RQ) of KHK expression. ....................................................... 279 G 2 Relative quantity (RQ) of SI expression. ........................................................... 280 G 3 Relative quantity (RQ) of SLC2A2 expression. ................................................. 281 G 4 Relative quantity (RQ) of SLC2A5 expression. ................................................. 282 G 5 Relative quantity (RQ) of SLC5A1 expression. ................................................. 283 G 6 Relative quantity (RQ) of KHK expression after consumption of soft drinks containing either HFCS or sucrose. .................................................................. 284 G 7 Relative quantity (RQ) of KHK expression by study visit and treatment. .......... 285 G 8 Relative quantity (RQ) of KHK expression after consumption of soft drinks containing either HFCS or sucrose. .................................................................. 286 G 9 Relative quantity (RQ) of SLC2A5 expression after consumption of soft drinks containing either HFCS or sucrose. ....................................................... 287 G 10 Relative quantity (RQ) of SLC2A5 expression by study visit and trea tment. .... 288

PAGE 17

17 G 11 Relative quantity (RQ) of SLC2A5 expression after consumption of soft drinks containing either HFCS or sucrose. ....................................................... 289 G 12 Correlation of BMI with pretreatment expression levels of A) KHK and B) SLC2A5 ........................................................................................................... 290

PAGE 18

18 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy FACTORS IMPACTING FRUCTOSE BIOAVAILABILITY AND ITS ADVERSE METABOLIC EFFECTS By MyPhuong Thi Le May 2010 Chair: Julie Johnson Major: Pharmaceutical Sciences Fructose consumption in the United States has spiked since the past four decades. There is growing evidence that excessive consumption of fructose may play a pivotal role in the current epidemic of a number of health disorders. The objective of these studies was to determine factors that may impact fructose bioavailability and its adverse metabolic effects. From the genetic association study, several potentially interesting polymorphisms were associated with various metabolic phenotypes, including tr iglycerides, serum uric acid, home diastol ic and systolic blood pressure, glucose, highdensity lipoprotein (HDL), and body mass index. Importantly, SLC2A2 (solute carrier family 2, member 2) rs8192675 was significantly associated with a decrease in HDL levels in European Americans in the PEAR s tudy and was replicated in the GERA study. From the clinical study, we detected higher fructose concentrations in the systemic circulation from high fructose corn syrup (HFCS) versus sucrose. There were treatment differences in triglycerides, systolic bl ood pressure, and serum uric acid levels. However, we were unable to detect correlations between higher fructose AUC (area under the curve of plasma concentrations versus time) or Cmax (maximum observed concentration) with higher fructose induced metaboli c effects. In addition, we

PAGE 19

19 did not detect any strong correlations between higher chronic fructose intake and increased metabolic responses. We did detect almost a four fold difference in the inter individual variability in fructose AUC from HFCS. From t he in vitro studies, we detected that sugar composition, sugar concentration, and duration of sugar exposure impacted fructose transport and expression of genes involved in fructose absorption and metabolism. W e did not detect a significant i mpact of glucose on fr uctose permeability. However, after exposing cells to high glucose dose for 72 hr, glucose was shown to have strong negative regulatory impact on the expressions of SI (sucrase isomaltase), SLC2A2 SLC2A5 (solute carrier family 2, member 5), SLC5A1 (solut e carrier family 5, member 1), and KHK (ketokexokinase). The studies conducted and described herein showed that various factors may influence an individuals exposure and response to fructose. Further studies are needed to elucidate the longterm effect s of these factors on the inter individual variability towards the development of fructose induced adverse metabolic effects and the increased risk for developing various health problems, such as cardiovascular disease and metabolic syndrome.

PAGE 20

20 CHAPTER 1 BACKGROUND AND SIGNIFI CANCE Introduction Of great concern is the sharp rise in consumption of fructose in the United States. Over the past four decades, the total intake of fructose has increased approximately 30% Over the same period, the prevalence of s everal disorders, including obesity, hypertension, metabolic syndrome, and diabetes has reached epidemic proportions. Numerous studies in animals and humans provide growing evidence that suggest excessive fructose consumption to be partly responsible for the rise in these health problems due to its ability to cause a variety of harmful metabolic effects. Dietary Sources and Consumption of Fructose Fructose exists naturally in many foods, such as fruits and vegetables (Table 11) .1 Small amounts of fructose can also be converted from sorbitol by aldose reductase. Sorbitol is also found in fruits and vegetables and as a sweetener in diet foods.2 However, the majority of dietary fructose comes from two sweeteners sucrose and high fructose corn syrup (HFCS), which are commonly used in manufactured foods and beverages (Table 12) .1, 3, 4 Sucrose, commonly known as table sugar, is manufactured from sugar cane and sugar beet. As a disaccharide, sucrose releases equal amounts of glucose and fructose into the small intestine after being hydrolyzed by sucrase whic h is encoded by the gene SI (Figure 11 ).5 On the other hand, HFCS which is produced through enzymatic processes f rom corn, provides a direct source of free fructose (Figure 12 ) .6, 7 HFCS was introduced to the United States in 1967. Since then, HFCS has encroached into the market, competing with sucrose for the top spot as the most

PAGE 21

21 consumed sweetener. The annual per capita intake of HFCS has dramatically increased from 0.03 lbs to 58.2 lbs in 2006 (Figure 1 3 ), while sucrose has decreased from 98.5 lbs to 62.3 lbs.810 Nevertheless both are heavily used in numerous food items such as carbonated beverages and other sweetened drinks, baked goods, candies, and dairy products .11, 12 As a result, to tal fructose consumption has significantly inc reased by 2535% over the past several decades. The average daily int ake by adults is approximately 75100 g of fructose, and is potentially still rising.2, 8, 10, 12 Structure of Fructose Fructose (C6H1206) is a n isomer of glucose. However, unlike glucose which has an aldehyde group attached to C1, fructose has a keto group attached to C2 (Figure 11 A). As a ketohexose, fructose can form a six member ring. Nevertheless, fructose predominantly exists as a five member ring (furanose), resulting in four dif ferent stereoisomers (Figure 11 B).13 Importantly, this slight difference in the functional group of fructose profoundly impacts its absorption and metabolism. Absorption of Fructose The small intestine is selectively permeable to different sugars, resulting in varying absorption capacities. Although there are species differences, galactose is generally absorbed the fastest, followed by glucose, and then by fructose.1416 Overall, the absorption rate of fructose is about half of that of glucose.5, 17 In humans, i ngested fructose is absorbed by facilitated diffusion and is transported across the apical membrane of enterocytes by glucose transporter 2 ( GLUT2) and glucose transporter 5 ( GLUT5; Figure 15). These transporters are encoded by the genes SLC2A2 and SLC2A5 respectively (Table 13). GLUT5 is considered to be the

PAGE 22

22 primary fructose transporter due to its higher infinity for fructose. However, it is less efficient than other sugar transporters due to its lower capacity and slower transport rate for fructose, and thus, is probably responsible for the slower absorption rate of fructose. Once fructose is in the ent erocytes, it has to compete with glucose and galactose for the GLUT2 transporters located on the basolateral membrane in order to cross into the systemic circulation (Figure 14 ). Interestingly ingestion of fructose together with either glucose or galact ose, have been shown to increase the intestinal absorption of fructose, primarily through the increased utilization of GLUT2.2 Metabolism of Fructose The live r, kidney, and small intestine are primarily responsible for the rapid metabolism of 5070% of absorbed fructose. Metabolism of the remaining percentage of fructose remains unclear but may be utilized by other tissues such as muscle and adipose tissues.2, 12, 18 Fructose is primary metabolized by a specific pathway containing three enzymes : ketohexokinase or fructokinase ( encoded by the gene KHK), aldolase B ( ALDOB) and triokinase (Figure 15 Table 13) First, fructose is converted into fructose1 phosp hate by ketohexokinase Next, this molecule is cleaved by aldolase B into dihydroxyacetone phosphate and glyceraldehyde. Triokinase converts glyceraldehyde into glyceraldehyde3 phosphate. This metabol ite and d ihydroxyacetone phosphate are further metabolized in the glycolytic gluconeogenic pathway.2, 12 Meanwhile, within the adipose tissues, fructose is broken down by a secondary metabolic pathway. Fructose is initial ly converted by a hexokinase into fructose 6 phosphate, which is also shunted into the glycolytic gluconeogenic pathway (Figure 15 ). U ltimately, the breakdown of fructose results in the production of numerous potentially harmful molecules, such as lactat e, lipids, glucose, fatty acids, and

PAGE 23

23 fructose 1 phosphates .2, 5, 17, 18 In addition, from the rapid breakdown of fructose, ATP levels are depleted, which results in the production of uric acid (Figure 16 ).2, 5, 19 Importantly, elevated levels of serum uric acid have recently been speculated to play a ca usal in the pathogenesis of cardiorenal diseases.2023 Toxicities of Fructose & Potential Role in Disease Risk Intolerance to fructose was first reported by Chambers and Pratt in 1956.24 After the ingestion of fructose or sucrose, a 24 year old woman suffered from nausea, vomiting, and abdominal pain. The effects became more severe with greater amounts of fructose. Chambers and Pratt called this inborn error of carbohydrate metabolism an idiosyncrasy to fructose and speculated that the symptoms were due to the accumulation of a toxic intermediate arising specifically from the breakdown of fructose. For individuals who suffer from this disease, continued exposure to fructose can lead to liver and kidney failure and eventually death.2 Since the discovery of heredit ary fructose intolerance additional fructose toxicities have been documented in numerous animal and human studies.2 Excessive fructose ingestion and the IV administration of large doses of fructose have been shown to cause liver injury25, 26, lactic acidosis27, hypertriglyceridemia and lipogenesis28, insulin resistance29, high blood pressure30, suppression of leptin31, and weight gain32, 33. In addition, high doses of IV administered fructose have been shown to reduce up to 75% of hepatic ATP levels, resulting in fructose induced hyperuricemia.25, 26, 34 Recent studies indicate that elevated serum uric acid levels may be an independent causal factor for the predisposition of hypertension, metabolic syndrome, and cardiorenal diseases.20 23 Thus, fructose induced hyperuricemia may be the key m echanism driving the development of these toxicities .35

PAGE 24

24 There is growing concern that the current excessive consumption of fructose may pose a great health risk.31, 33, 354 2 It has been hypothesized that fructose may be part ially responsible for the 20% rise in obesity in adults and the 10% increase in the prevalence of overweight in children and adolescents f rom 1971 to 2006 (Figure 1 7 ) In particular, some have speculated that HFCS may be the culprit since the sharp rise in obesity occurred after the introduction and rise in HFCS consumption while sucrose consumption was decreasing.43 Overall, these studies suggest that fructose may play an important role in the epidemic increases in several health disorders, including obesity10, 44, gout45, hypertension46 48, metabolic syndrome49, and diabetes.12, 36, 5052 Metabolic Disorders and Genetic Polymorphisms of Fructose There are various abnormalities in the absorption and metabolism of fructose It is estimated that one out of three adults and two out of three children are malabsorbers of fructose, suffering from various gastrointestinal adverse effects, such as bloating and diarrhea.2, 12 SLC2A5 which encodes for the main intestinal fructose transporter GLUT5, may be responsible for this malabsorption. Meanwhile, genetic mutations in S LC2A2 KHK, and SI can cause known, but rare, inherited disorders in humans. Mutations in SLC2A2 can lead to the impaired utilization of glucose and galactose, resulting in glycogen accumulation in the liver and kidney and proximal renal tubular dysfuncti on.53, 54 Essential fructosuria is a benign, asymptomatic metabolic disorder caused by the inactivity of ketohexokinase resulting in the ineffective breakdown of fructose.2 Defects in SI can cause congenital sucrase isomaltase deficiency, which results from abnormalities in the breakdown of sucrose.55 These rare genetic defects highlight the role of these genes and their encoded proteins on the regulation o f fructose levels Therefore, common polymorphisms in these genes can potentially influence

PAGE 25

2 5 inter individual variability in fructose bioavailability and the development of fructose induced adverse metabolic effects. Fructose Bioavailability and Metabolic Effects Research Plans Various factors can influence an individuals response to fructose. We hypothesize that three main factors are ( 1) polymorphisms in genes involved in fructose absorption and metabolism ; ( 2) the amount of chronic fructose consumed by an individual ; and ( 3) the composition of fructosecontaining sweeteners. We hypothesize t hese factors impact the inter individual variability towards fructose, resulting in varying degrees of susceptibility for developing fructose induced adverse metabolic effects, such as increased serum uric acid and triglyceride levels. Currently, there are limited studies investigating the effects of these factors on the variable response to fructose by individuals. In order to advance our understanding of the pot ential role of fructose as an important causal factor in the pathogenesis of several common health disorders, further characterization of the impact of these three factors on fructose utilization is required. For this study, the goals were to conduct (1) a genotype association study to elucidate the effects of polymorphisms in genes involved in the absorption ( SLC2A2 SLC2A5 and SI ) and metabolism ( KHK) of fructose on metabolic phenotypes ; (2) an in vitro study to investigate the effects of sugar formulat ion, sugar concentration, and duration of sugar exposure on fructose transport and gene expression; and ( 3) a clinical study to compare the pharmacokinetic and pharmacodynamic effects of sucrose v ersus high fructose corn syrup.

PAGE 26

26 Table 11. Sugar contents of fruits and vegetables (g / 100 g). Food Fructose Glucose Sucrose Total Total Fructose as a % of Total Sugars Carbohydrates Carbohydrates Fruits Apple 5.90 2.43 2.07 10.39 13.81 42.72 Orange 2.25 1.97 4.28 8.50 12.54 17.94 Grape 8.13 7.20 0.15 15.48 18.10 44.92 Banana 4.85 4.98 2.39 12.23 22.84 21.23 Nectarine 1.37 1.57 4.87 7.89 10.55 12.99 Pear 6.23 2.76 0.78 9.80 15.46 40.30 Vegetables Carrot 0.55 0.59 3.59 4.74 9.58 5.74 Corn 0.48 0.50 2.06 3.22 19.02 2.52 Red Pepper 2.26 1.94 0.00 4.20 6.03 37.48 Onion 1.29 1.97 0.99 4.24 9.34 13.81 Sweet Potato 0.70 0.96 2.52 4.18 20.12 3.48 Lettuce 0.43 0.36 0.00 0.78 2.79 15.41 Data obtained from the United States Department of Agriculture Nutrient Database for Standard Reference, Release 21.1 Table 1 2 Sugar contents of commercial s weeteners (g / 100 g). Sweetener Fructose Glucose Sucrose Total Total Fructose as a % of Total Sugars Carbohydrates Carbohydrates Brown Sugar 1.1 1.4 94.6 97.0 98.1 1.1 Corn Syrup 25/42C 0.0 7.0 0.0 100.0 100.0 0.0 Corn Syrup 63/44C 0.0 36.0 0.0 100.0 100.0 0.0 Crystalline FructoseA 99.9 0.1 0.0 100.0 100.0 99.9 Granulat ed Sugar 0.0 0.0 99.9 100.0 100.0 0.0 HFCS 42C 42.0 52.0 0.0 100.0 100.0 42.0 HFCS 55C 55.0 41.0 0.0 100.0 100.0 55.0 HFCS 90A 90.0 8.5 0.0 100.0 100.0 90.0 Honey 40.9 35.8 0.9 82.1 82.4 49.7 Maltodextrin CR18A 0.0 1.6 0.0 100.0 100.0 0.0 Maltodextri n CR24A 0.0 7.0 0.0 100.0 100.0 0.0 Maple Syrup 0.9 2.4 56.3 59.5 67.1 1.3 Molasses 12.8 11.9 29.4 55.5 74.7 17.1 Data obtained from: A = Archer Daniels Midland Company C = Cargill Incorporated, and U nited States Department of Agriculture Nutrient Data base for Standard Reference, Release 21. 1, 3, 4

PAGE 27

27 Table 13. Summary of genes. Gene Chromosome Gene Size Name Alias Protein size Location (bp) (aa) ALDOB 9q21.3 q22.2 14,448 aldolase B, fructosebisphosphate 364 KHK 2p23.3 14,009 ketohexokinase, fructokinase 299 SLC5A1 22q12.3 67,387 solute carrier family 5 (sodium/glucose cotransporter), member 1 SGLT1 664 SI 3q25.2 q26.2 99,597 sucrase isomaltase, alphaglucosidase 1,827 SLC2A2 3q26.1 q26.2 30,632 sol ute carrier family 2 (facilitated glucose transporter), member 2 GLUT2 524 SLC2A5 1p36.2 32,664 solute carrier family 2 (facilitated glucose/fructose transporter), member 5 GLUT5 501 Data obtained from National Center for Biotechnology Information ( http: //www.ncbi.nlm.nih.gov/gene) and University of California, Santa Cruz Genome Bioinformatics ( http://genome.ucsc.edu/ ), NCBI Build 36.1 .56, 57

PAGE 28

28 Figure 11 Chemical structures of fructose, glucose, and sucrose. A) Sucrose is digested by sucrase to produce glucose and fructose. B) Fructose stereoisomers.

PAGE 29

29 Figure 12 Production of high fructose corn syrup from corn. Through multiple manufacturing processes and enzymatic reactions, corn is milled into starch which is then hydrolyzed into glucose. Next, glucose is converted into fructose by glucose isomerase. Three common commercial grades of HFCS can be produced with fructose contents of 4 2%, 55%, and 90%.

PAGE 30

30 Figure 13 U.S. per capita consumption of sugars and sweeteners. Data obtained from USDA/Economic Research Service. Last updated Feb. 15, 2007.8

PAGE 31

31 Figure 14. Digestion and absorption of carbohydrates.

PAGE 32

32 Figure 15 Metabolism of fructose.

PAGE 33

33 Figure 16 Production of uric acid from the breakdown of ATP.

PAGE 34

34 Figure 17 U.S. per capita c onsumption of sweeteners and prevalence of overweight and obesity. Data obtained from the Center for Disease Control and Prevention.58, 59

PAGE 35

35 CHAPTER 2 GENOTYPE ASSOCIATION STUDY: IMPACT OF GENETIC POLYMO RPHISMS OF S LC2A2 SLC2A5 SI AND KHK ON METABOLIC PHENOTYPES IN PEAR, HAPI, AND GERA Introduction Five candidate genes were initially selected for the genotype association study (Tables 21). The genes were selected based on their ability to regulate fr uctose concentrations in the body. SI (sucrase isomaltase), SLC2A2 (solute carrier family 2, member 2), SLC2A5 (solute carrier family 2, member 5) were chosen because of their involvement in the absorption/transport of fructose. Polymorphisms in these genes could impact the absorption of dietary fructose and greatly affect its bioavailability. Meanwhile, KHK (ketohexokinase) and ALDOB (aldolase B) were chosen because of their involvement in the metabolism of fructose.2, 55 Polymorphisms in these genes could impact the level of exposure to various metabolic byproducts cleaved from fructose (Figure 16). Importantly, polymorphisms in these five genes could impart variability in individual risk for developing advers e metabolic phenotypes, such as increased serum uric acid and triglyceride concentrations, which ultimately could lead to the development of cardiovascular and renal diseases. A genetic association study was conducted to elucidate the impact of single nuc leotide polymorphisms (SNPs) of ALDOB, KHK, SI SLC2A2 and SLC2A5 on the increased susceptibility of developing adverse metabolic phenotypes. Data from four study populations were utilized: Pharmacogenomic Evaluation and Antihypertensive Responses (PEAR) Hereditary and Phenotype Intervention (HAPI) Heart, and Genetic Epidemiology of Responses to Antihypertensives (GERA) I and II studies.

PAGE 36

36 Materials and Methods Study Populations For the purposes of this association study, we only focused on baseline fasting laboratory data collected from the four study populations. PEAR PEAR is a pharmacogenomic study investigating the impact of polymorphisms in a number of candidate genes on antihypertensive and adverse metabolic responses to a blocker (atenolol) and a thiazide diuretic (hydrochlorothiazide).60 The study plans to recruit 800 subjects, who are mild to moderate essential hypertensive, male or female, of any race or ethnicity, and between the ages of 17 and 65. A total of 418 subjects were genotyped for this project. Only data from African Americans and European Americans were used for the association analyses, thus eliminating 14 subjects belonging to other racial groups. HAPI Heart The HAPI Heart study recruited 868 relatively healthy Amish adults. The study investigated the genetic determinants affecting acute cardiovascular response to four different interve ntions.61 GERA I and II The GERA study investigated whether polymorphisms in genes predicted inter individual responses to blood pressure response to antihypertensive drugs. Men and women with essential hypertens ion between the ages of 30 and 59 were recruited. In an earlier study (GERA I), 289 African Americans and 295 European Americans were recruit ed and were given a thiazide di uretic (hydrochlorothiazide).62 In a later study (GERA II), an independent sample of 203 African Americans and 236 European

PAGE 37

37 Americans were recruited and were treated with an angiotensin II receptor blocker (candesartan).63 SNP Selection To best capture the genetic variability of fructose absorption and metabolism, all SNPs listed in NCBI dbSNP Build 1 28 ( http://www.ncbi.nlm.nih.gov/SNP/ ) for KHK SI SLC2A2 SLC2A5 and ALDOB were initially considered.64, 65 At the start of the study (April 2, 2007), t here were a total of 54 SNPs that have been identified in KHK, 283 SNPs in SI 128 SNPs in SLC2A2 151 SNPs in SLC2A5 and 82 SNPs in ALDOB (Tables 21). Thus, there were initially a total of 698 dbSNPs for the five genes. Putatively f unctional SNPs To help prioritize the SNP selection, the potential functionalities of the dbSNPs were predicted using in silico software. The SNPs, including 3 kb upstream and downstream of the genes, were analyzed for the following functional properties using P upaS uite 2.0. 066 ( http://pupasuite.bioinfo.cipf.es/ ) and fastSNP67 ( http://fastsnp.ibms.sinica.edu.tw/pages/input_CandidateG eneSearch.jsp ) : nonsynonymous SNPs (polymorphisms that cause changes in the amino acid sequence), transcription factor binding site, exonic splicing enhancer or silencer, intron/exon boundaries, and triplex forming sequences. From the analysis, there wer e 30 putatively functional SNPs (pfSNPs) for KHK 10 pfSNPs for SI 24 pfSNPs for SLC2A2 30 pfSNPs for SLC2A5 and 22 pfSNPs for ALDOB. In addition, two nonsynonymous SNPs of KHK that were not listed in dbSNP were also added as pfSNPs. These two SNPs we re discovered in three siblings with essential fructosuria. One is a null mutation ( Gly40Arg) that potentially affects the proteins tertiary structure

PAGE 38

38 and the other is a leaky mutation (Ala43Thr) that destabilizes only KHKC at physiological temperatures .2, 68, 69 The majority of the pfSNPs had not been validated. The validation of SNPs was primarily conducted by the International HapMap Project (HapMap, http://hapmap.ncbi.nlm.nih.gov/index.html.en).70 Its genomewide approach (Phase I) was to genotype one SNP every 5 kb across most of the human genome. This was done in four different populations: Yoruba in Ibadan, Nigeria (YRI), Japanese in Tokyo, Japan (JPT), Han Chinese in Beijing, China (CHB), and CEPH which consists of Utah residents with ancestry from northern and western Europe (CEU) .71, 72 Although over an additional 3.1 million SNPs were genotyped in Phase II, many of the genetic polymorphisms have still not been fully characterized.73 Based upon Hap map data from Phase I and II, 10 of the 32 pfSNPs of KHK, 9 of the 10 pfSNPs of SI 6 of the 24 pfSNPs of SLC2A2 8 of the 30 pfSNPs of SLC2A5 and 9 of the 22 pfSNPs of ALDOB have been validated (Tables 21). Thus, of the total 118 pfSNPs, only 42 have been validated by HapMap. To increase coverage across the genes and to decide on whether to include them in the genetic association study, the unvalidated potential pfSNPs of KHK, SLC2A5 and ALDOB were sequenced. As the project progressed and fructose ab sorption/transport was emphasized, ALDOB was eliminated as a candidate gene for the genotype association study. Data for ALDOB are presented in Appendix A: Sequencing and Genotyping Data. Because the addition of SLC2A2 and SI as candidate genes occurred near the end of the sequencing process and because many of the SNPs listed on dbSNP were most likely monomorphic, it was determined to postpone the sequencing of the unvalidated pfSNPs of these two genes until the

PAGE 39

39 outcome of the sequencing results for KHK, ALDOB and SLC2A5 could be considered. If the sequencing results showed a high detection rate of valid common SNPs, then the unvalidated pfSNP of SL2A2 and SI would be sequenced. Sequencing. Sin ce the frequencies of variant alleles can differ widely amon gst races, a total of 48 samples were sequenced using the genomic DNA of 24 individuals of European descent and 24 individuals of African descent (Table A 1) All samples were obtained from the Coriell Institute for Medical Research Human Variation Collections of the NIGMS Repository (Coriell Institute for Medical Research, Camden, NJ). Short amplicons spanning the regions of the various unvalidated pfSNPs of KHK, SLC2A5 and ALDOB were generated by PCR (polymerase chain reaction) Primers, starting about 100 bp from the location of the candidate SNP, were designed using Vector NTI Advance 10.3.0 (Invitrogen Corporation, Carlsbad, CA) and were blasted against NCBIs GenBank for specificity. Some of the amplicons were able to capture more than one SNP. Thus, the 22 unvalidated pfSNPs of KHK were captured by 9 PCR products (Table 22), the 22 pfSNPs of SLC2A5 were captured by 16 PCR products (Table 23), and the 13 pfSNPs of ALDOB were captured by 10 PCR products (Table A 2). Sequencing was performed by the Molecular Services Core Laboratory at the University of Floridas Interdisciplinary Center for Biotechnology Research (UFICBR, Gainesville, FL). The amplicons were sequenced by conventional capillary based sequencing using BigDye chemistry and were run on the ABI3100 platform (Applied

PAGE 40

40 Biosystems Inc., Foster City, CA). Most amplicons were sequenced with both forward and reverse primers. Gene sequences of KHK, SLC2A5 and ALDOB were downloaded from NCBIs RefSeq.74 The sequencing results were then aligned against these reference sequences using ContigExpress, a component of Vector NTI. The genotypes were called by visual analysis and confirmed by three analyzers (Figure 2 1). The sequencing results of the SN Ps were analyzed for MAF and Hardy Weinburg equilibrium (HWE). Of the 22 potential pfSNPs of KHK, 18 SNPs were monomorphic in both the African and European samples (Table 24). One SNP was common (rs7602823) and one SNP (rs7573066) had a MAF below 5%. SNPs surrounded by long runs of homopolymers were difficult to sequence. Thus for KHK, two SNPs failed sequencing. For SLC2A5 4 out of the 22 potential pfSNPs were monomorphic, while five SNPs were out of HWE which was defined as having a pvalue below 0.05 in both races (Table 25). In addition, five SNPs failed to be sequenced. However, two SNPs (GT_9043651 and CT_9011003) not reported on dbSNP were discovered and added to the selection process. 8 of the SLC2A5 pfSNPS were validated with a MAF 5%. For ALDOB, 9 pfSNPs were monomorphic and 2 SNPs failed to be sequenced (Table A 3). Two of the pfSNPs were validated with a MAF 5%. Since ALDOB was eliminated as a candidate gene, these pfSNPs were not selected for the association study. Overall, of the 57 unvalidated pfSNPs sequenced, we were able to validate 11 of those SNPs. Importantly, all of the nonsynonymous SNPs were eliminated because they were monomorphic in our tested samples.

PAGE 41

41 For KHK and SLC2A5 the sequencing results were combined with previously validated pfSNPS with a MAF 5%. Thus, there were a total of 6 potentially interesting pfSNPs for KHK and 14 pfSNPs for SLC2A5 Because of the low rate of detection of valid common SNPs (19%), the unvalidated pfSNPs of SI and SLC2A2 were not sequenced. Only previously validated pfSNPs with a MAF two genes, resulting in 9 pfSNPs for SI and 10 pfSNPs for SLC2A2 Ultimately, there were a total of 39 common pfSNPS (Table 26). Some of these candidate pfSNPs were used to supplement the predesigned cardiov ascular genotyping chip that was used to genotype the PEAR population. IBC genotyping chip After this dissertation project was initiated, there was a decision to utilize the HumanCVD Genotyping BeadChip (Illumina, San Diego, CA) to genotype the PEAR populaton.75, 76 The HumanCVD Genotyping BeadChip was designed by the Institute of Tr anslational Medicine and Therape u tics (ITMAT, Philadelphia, PA), the Broad Institute (Cambrdige, MA), and Candidategene Association Re source (CARE) consortium ( http://bmic.upenn.edu/cvdsnp/ ) along with other investigators, including those from our laboratory. T he ITMAT/Broad/CARE (IBC) chip contained over 49,000 SNPs from approximately 2,100 genes that were directly or indirectly associated with cardiovascular, metabolic, and inflammatory syndromes. Most of the genes were prioritized based upon the strength of their relationship to these syndromes. Group 1 contained genes and regions that wer e most likely to have functional significance. As a result, the criteria used to select tagSNPs for these loci were MAF 2 0.8. Group 2 consisted of potentially important loci and the criteria used for the tagSNP selection was MAF with an r2

PAGE 42

42 with a lower likelihood of functional significance. For these genes, only nonsynonymous SNPs and known functional polymorphisms of MAF In addition to other potentially important SNPs, about 1500 ancestry informative markers used to distinguish African versus European descent were included on the IBC chip. For our association study, SLC2A2 and SLC2A5 were in priority Group 2. T here were 13 SLC2A2 SNPs and 11 SL C2A5 SNPs on the IBC chip. There were also 5 SNPs of KHK. However, the IBC chip had no SI SNPs Custom OPA genotyping chip In order to increase the characterization of KHK, SLC2A2 and SLC2A5 the IBC chip was supplemented with additional SNPs from a cu stom chip we designed. These SNPs were generated using a cosmopolitan tagSNP approach, which minimizes tagSNPs amongst racial populations. SeattleSNPs Genome Variation Server ( http://gvs.gs.washington.edu/G VS/ Seattle, WA) was used to generate the list of tagSNPs. Since the study populations consisted primarily of African and European heritage, only YRI and CEU populations of the HapMap datasets were selected and analyzed by the MultiPopTagSelect algorit hm.77 3 kb upstream and downstream of each gene were includ ed in the tagSNP selection process. The minor allele frequency (MAF) cutoff was 5% since we were only interested in common SNPs. The r2 threshold, which is a measure of linkage disequilibrium (LD), was set at 0.8. Stratified by African and European popu lations, r2 plots of KHK, SLC2A2 and SLC2A5 are shown in Figures 2 2 to 27. 12 multiple population tagSNPs were generated for KHK, 18 for SLC2A2 and 21 for SLC2A5

PAGE 43

43 To minimize overlap, the SNPs on the IBC chip were crossreferenced against these tagSN Ps. The additional tagSNPs were added to a custom SNP array (oligo pool all, OPA) T he OPA chip was designed using Illuminas Assay Design Tool (ADT). The ADT algorithm tested each SNP for its ability to be successfully genotyped. SNPs with a design ra nk value of 0 have a low success rate and were excluded. Meanwhile, only SNPs with moderate (design rank = 0.5) and high (design rank = 1) success rate were included into the OPA chip. The OPA chip was designed with 96 SNPs which included 16 additional cosmopolitan tagSNPs for SLC2A2 and SLC2A5 along with 6 pfSNPs of SLC2A2, 8 pfSNPs of SLC2A5 and 1 pfSNP of KHK. Thus OPA chip contained 1 KHK SNP, 11 SLC2A2 SNPs, and 19 SLC2A5 SNPs. T he remaining SNPs on the chip were for another project In total the IBC and OPA chip contained 60 SNPs for these three genes Because of later changes to the SNPs on the IBC chip, two SNPs on the two chips overlapped ( KHK rs2304681 and SLC2A2 rs11925298). Because SI was not covered on the IBC chip, and was lower priority than the other three genes, we did not add SNPs for this gene to the custom array, but relied on existing GWAS data in HAPI where there was good coverage for this gene. DNA Isolation and Q uantification Genomic DNAs of subjects from PEAR were isol ated from whole blood using the FlexiGene DNA kit and were processed according to the manufacturers protocol at the Center for Pharmacogenomics (Qiagen Inc., Valencia, CA). DNA concentrations were measured using Quant iT PicoGreen dsDNA reagents (Invit rogen Corporation, Carlsbad, CA).

PAGE 44

44 For absolute quantification, Lambda DNA standard was used. The standard was serially diluted to the following concentrations: 75, 50, 25, 12.5, 6.25, 3.12, 1.56, and 0 ng/ L. DNA samples were diluted 1:20 (5 l DNA + 9 5 l water) and 2 l of the solution were used for the measurements. Samples and standards were measured in triplicates and were placed in 384well plates and assayed by the Synergy HT Multi Mode Microplate Reader (BioTek, Winooski, VT). DNA was normaliz ed to 50 ng/ L. Genotyping PEAR was genotyped using the IBC chip and the OPA chip. Because these chips contained SNPs for KHK, SLC2A2 and SLC2A5 PEAR was used as the discovery cohort for associations for these three genes. Meanwhile, HAPI Heart study w as the discovery cohort for associations with SI Associations found in the HAPI Heart population are presented in Appendix A: HAPI Heart Data. Interesting SNPs for any of the genes were then replicated in the GERA populations. PEAR was genotyped at the Center for Pharmacogenomics of the University of Florida and UFICBR (Gainesville, FL). Illuminas Infinium II, a whole genome beadchip genotyping technology, was used to assay the IBC chips .78, 79 The PEAR DNA samples were processed according to the manufacturers protocol. The intensities of the fluorescence were detected by the Illumina BeadArray Reader. The image files were then analyzed by Illuminas BeadStudio Genotyping Analysis Module 3.3.7. The PEAR DNA samples were assayed with the custom OPA chips using Illuminas Veracode GoldenGate chemistry and were performed according to manufacturers protocol.80 The bead plates were then scanned by Iluminas BeadXpress Station. The r esulting files were analyzed by BeadStudio Genotyping Analysis Module 3.3.7.

PAGE 45

45 Genotyping for HAPI Heart and GERA have already been conducted by their institutions. HAPI Heart utilized the Affymetrix 500K chip which contained 2 KHK SNPs, 223 SI SNPs, 10 SLC 2A2 SNPs, and 8 SLC2A5 SNPs ( http://www.affymetrix.com/analysis/index.affx ). GERA I was genotyped using the Affymetrix 100K SNP array while GERA II was genotyped with the Affymetrix 6.0 SNP array Although these chips had limited coverage of the candidate genes, the chips were able to provide genotyping data for the interesting SNPs from the discovery cohorts, thus no additional genotyping was needed. Data Analysis IBCchip Analysis Samples were excluded if they were contaminated, which was determined by the Genome Viewer component of BeadStudio, had sex gender estimate mismatches, or had call rates < 0.90. SNPs, which were clustered using a Gen Call Threshold of 0.15, were removed if they had poor clustering scores (GenTrainScore < 0.3), were monomorphic, or had call rates <0.95. For the 29 SNPs for this association study, the SNPs were stratified by race and HWE was calculated. SNPs with a pvalue of <0.05 in both races were also eliminated. O PA Analysis Samples were excluded if they had call rates <0.90. SNPs were clustered with a Gen Call Threshold of 0.25. SNPs were excluded if they were had low clustering scores (GenTrainScore < 0.03), were monomorphic, or had a call rate < 0.95. From the HWE analysis, SNPs were excluded if the chi square test p values were < 0.05 in both African American and European American populations.

PAGE 46

46 SNPTrait Associations For PEAR, SNPtrait association analyses were performed for each gene using SAS 9.2 and JMP Ge nomics 4.0 (SAS Institute Inc., Cary, NC). Because of the differences in LD structures and frequencies of variant alleles, the data were stratified by race (European American and African American). SNPs with a MAF < 5% in both races were excluded from analysis due to lack of power. The primary adverse metabolic phenotype was fasting serum uric acid (SUA). Secondary phenotypes were fasting glucose, insulin, triglycerides (Tg), highdensity lipoprotein (HDL), low density lipoprotein (LDL), home diastolic blood pressure (HDBP), home systolic blood pressure (HSBP), body mass index (BMI), homeostatic model assessment (HOMA), and metabolic syndrome (MS).81 Because of sex differences, the interaction between gender and the SNP were also analyzed for SUA and HDL.8284 If the interaction was significant, the data were also stratified by gender. All data were from the baseline study period in PEAR. For each adverse metabolic phen otype, unadjusted analysis of covariance (ANOVA) was performed. The three possible genotypes were treated as categories (nominal): homozygote common allele (AA), heterozygote (AB), and homozygous minor allele (BB). From this first pass analysis, SNPs wit h a pvalue 0.05 were then reanalyzed, adjusting for age, BMI, and/or gender. Based upon the means, an additive, dominant (AA vs AB + BB) or recessive (AA + AB vs BB) model was also performed for some SNPs and only if there were three genotype groups. In these lin ear regression models, the data were treated as ordinal values and the genotypes were coded as: AA = 0, AB = 1, or BB = 2. Although multiple comparisons were conducted, Bonferroni adjustment was considered too stringent. Therefore to reduce the possibilit y of type I errors, while also

PAGE 47

47 minimizing risk for Type II errors, only SNPs with an adjusted pvalue 0.005 were considered significant in the initial discovery process. For SNPs with a MAF of 0.05, there was a greater than 91% power to detect an effect size of 1.0; and for MAF of 0.20, there was a great than 99% power to detect effect size of 0.8 ( = 0.005, n = 200 by race).85, 86 Some of the SNPs meeting statistical criteria were excluded from replication if the SNP impacted less than 5% of the targeted population, or if the SNP had opposite effects in either race or gender populations, or if the SNPs effect showed the greatest impact for heterozygotes, since this is unlikely to represent a true genetic effect. The remaining SNPs were then replicated in GERA I and II and those with a pvalue were considered significant. For HAPI Heart and GERA I and II, the analyses were conducted by the studies central data coordinating centers. HAPI Heart data were adjusted for age, age2, and gender. GERA data were adjusted for age, BMI, and gender. Results Genotype and Data Quality Control Table 27 lists the number of samples and SNPs excluded during quality control. Of the 418 DNA samples from PEAR, one sample failed to be genotyped due to low yield and quality. For the IBC chip, one sample w as excluded due to contamination and a low call rate (< 90%) and another sample was excluded because of sex gender match. For the OPA chip, three samples failed due to a low call rate (<90%). There were complete genotype data for 414 subjects. However, subjects who were deemed not fasting at the study visits were excluded. After quality control, 407 PEAR subjects remained and their characteristics are shown in Table 28. For the association

PAGE 48

48 analyses, only African Americans and European Americans were utilized, resulting in 393 PEAR subjects. Of the 58 SNPs genotyped, two SNPs had a call rate lower than 95% on the IBC chip. The OPA chip also had two SNPs with a call rate < 95%. However due to the overlap in SNPs between the two chips, only two of these SNPs were excluded from analysis (Table A 4). Four additional SNPs were excluded because they were monomorphic and four more were excluded since the SNPs had a MAF < than 5% in both races. 48 SNPs, consisting of 18 SLC2A2 SNPs, 28 SLC2A5 SNPs, and 2 KHK SNPs, were analyzed. Discovery Cohorts: Primary Phenotypes Uric acid Of the five SI SNPs, one SLC2A2 SNP, and two SLC2A5 SNPs initially associated with uric acid, only SLC2A5 rs5438 remained significant after adjusting for covariates (Tables 29, A 5). B ecause of the significant interaction between the SNP and sex (pvalue = 0.0011), the SNP was stratified by gender. Thus, the association was seen only in European American males (pvalue = 0.0007). 21 or 17.1% of white men carried one copy of the minor allele, which was associated with approximately 1.1 mg/dL higher serum uric acid concentrations. There were no homozygote minor allele carriers. Discovery Cohorts: Secondary Phenotypes Triglycerides Listed in Tables 210 and A 5 are five SLC2A2 SNPs and eleven SLC2A5 SNPs that have an unadjusted pvalue 0.05 with triglycerides. After adjusting for age, BMI, and gender, four SNPs ( SLC2A2 : rs11924032, rs5398, and rs8192675; SLC2A5 : rs12086036) were significantly associated with triglycerides with pvalue

PAGE 49

49 The associations for all three SLC2A2 SNPs were observed only in European Americans. All three minor alleles appeared to have an additive effect on triglyceride levels. With each copy, triglyceride increased by about 30 mg/dL. In the additive model, rs5396 was also found to be significant (pvalue = 0.0015. Again, each copy of the minor allele was associated with an increase of about 30 mg/dL in triglycerides. For SLC2A5 rs12086036, the SNPs association was in African Americans. Within the r ecessive model (pvalue = 0.0011), two copies of the minor allele (GG) appeared to increase triglyceride levels by about 95 mg/dL. HOMA Only two SNPs had an initial association with HOMA (Table 211). After adjusting for age, BMI, and gender, none of the SNPs were below the significance threshold. HDBP Two SLC2A2 SNPs, three SLC2A5 SNPs, and one KHK SNP were initially associated with home diastolic blood pressure (Table 212). The results remained significant for SLC2A5 rs12145292 (recessive pvalue = 0.0015) after adjusting for covariates. The SNP affected African Americans. DBP was higher by about 7 mmHg in individuals who were homozygous for the minor allele. HSBP Table 213 lists the four SNPS that had an unadjusted pvalue 0.05 with home systolic blood pressure. Once adjusted, KHK rs7588333 (pvalue = 0.0006) was significantly associated with HSBP in European Americans. However, the SNP was rare. Only 2 out of 231 European Americans carried one copy of the minor allele.

PAGE 50

50 Glucose Two SLC2A5 SNPs were initially associated with fasting glucose but did not meet the adjusted significance threshold (Table 214). Insulin One SNP of SLC2A5 was initially associated with insulin (Table 215). However, the SNP was not significant after adjusting for age, BMI, or gender in either the additive or recessive models. HDL After adjusting for covariates, two SLC2A2 SNPs were determined to be significant in European Americans (Table 216). Individuals whom were carriers of one copy of t he minor allele of either rs5398 (pvalue = 0.0028) or rs8192675 (pvalue = 0.0034) showed a decrease in HDL levels by about 4 mg/dL. LDL One SLC2A5 SNP was initially associated with LDL but did not meet the adjusted significance threshold in either the additive or recessive models (Table 217). BMI After adjusting for age and gender, SLC2A5 rs1612895 (additive pvalue 0.0049) was significantly associated with a decrease in BMI in African Americans (Table 218). MS After adjusting for age and gender, two initial SNPs failed to meet the adjusted significance threshold for an association with MS (Table 219). Summary of Associations from Discovery Cohorts Following adjusted analyses, 9 SNPs of KHK, SLC2A2 and SLC2A5 were associated with various metabolic phenotypes (Table 220).

PAGE 51

51 KHK. One KHK SNP was associated with HSBP (rs7588333). SLC2A2 Four SLC2A2 SNPS were associated with Tg (rs11924032, rs5396, rs5398, rs8192675) and/or HDL (rs5398, rs8192675). SLC2A5 SLC2A5 SNPs were associated with SUA (rs5438), Tg (rs12086036), HDBP (rs12145292), and BMI (rs1612895). Replication For the replication, we focused on the SLC2A2 SNPs since there were multiple associations with Tg and HDL. In addition, two SNPs (rs5398 and rs8192675) were both associated with Tg and HDL. The other SNPs were excluded based on if the SNP impacted less than 5% of the targeted population, if the SNP had opposite effects in either race or gender populations, or if there was a lack of genotyping data for the SNP in GERA. Furthermore, SL C2A5 rs5438, which was associated with our primary phenotype, was not replicated because there was a lack of uric acid data in GERA. Of the SLC2A2 SNPs, GERA had data for only rs8192675. However, the SNP tagged the other 3 SLC2A2 SNPs: rs5398 (r2 = 1), rs 11924032 (r2 = 1), rs5396 (r2 = 0.916). Rs8192675 was significantly associated with HDL in GERA I (pvalue = 0.0664), GERA II (p value = 0.0692), and the combined study populations (pvalue = 0.0126; Table 221, Figure 28). Unlike PEAR where one copy of the minor allele appeared to decrease HDL levels by 4 mg/dL, there seems to be little effect in GERA. Only when two copies of the minor allele are present is there possibly a decrease of about 8 mg/dL in HDL. Nevertheless, the net change is similar bet ween both study groups. For Tg, rs8192675 was replicated only in the GERA I population (pvalue = 0.0890). However, the Tg level for heterozygotes was lower than the other two genotypes.

PAGE 52

52 Conclusion From the discovery cohorts, a number of potentially inte resting associations were detected between genetic polymorphisms of KHK, SI SLC2A2 and SLC2A5 with various metabolic phenotypes, including serum uric acid, Tg, home diastolic and systolic blood pressure, glucose, HDL, and BMI. Importantly, SLC2A2 rs8192675, which tagged three other SNPs (rs5398, rs11924032, rs5396), was associated with both Tg and HDL levels in the Caucasian subjects in PEAR, was also significantly associated with a lower HDL levels in the GERA populations. SLC2A2 rs8192675 is located in intron 5. As for the other three tagged SNPs, rs5398 causes a synonymous polymorphism and is located in exon 11, rs11924032 is located in intron 1, and rs5396 is located in the 5 region of the gene and was a pfSNP with potential promoter regulatory function. Although the functional consequences of these polymorphisms are unknown, mutations of SLC2A2 which is highly expressed in the liver and the cell islets, causes Fanconi Bickel syndrome. This disease is associated with hyperglycemia, hypoinsulinemia, and hypertriglyceridemia.54, 8790 Polymorphisms of SLC2A2 have also been linked with increased risk of developing type 2 diabetes.87, 90 Rs5393 and rs5400 were pfSNPs in this study, but both failed to be significantly associated with glucose or insulin levels. Importantly, recent studies have implicated insulinr esistance and hypertriglyceridemia with the lowering of HDL levels, potentially through the increased metabolism of apoA I, an essential component of HDL particles.89, 9193 Thus, while the potential functional mec hanism of our findings is not clear, the literature suggests such phenotypic associations could be consistent with impaired function of the GLUT2 transporter.

PAGE 53

53 Due to the limited studies conducted on these four genes, there was a lack of complete sequenci ng data. As a result, the association study relied on dbSNP for an initial list of SNPs. However, the majority of these SNPs have not been validated and from our sequencing efforts many of these SNPs are probably rare. Only 11 of the 57 unvalidated pfSN Ps that were sequenced had a MAF 5%. Thus, the lack of complete coverage of these genes was an important limitation of the study because important functional SNPs may have not been reported or characterized. At the start of the study, there were 54 KHK SNPs, 283 SI SNPs, 128 SLC2A2 SNPs, and 151 SLC2A5 SNPs reported on dbSNP. Currently, the database shows that the number of SNPs for each gene has more than doubled. There are now 127 KHK SNPs, 637 SI SNPs 256 SLC2A2 SNPs and 337 SLC2A5 SNPs listed (db SNP BUILD 130).94 These new SNPs may either be rare or could have been already tagged by SNPs in this study. Nevertheless, it does show that the genetic variability of these genes may be greater than what was initially covered at the start of the study. Due to the lack of coverage, potentially functional SNPs may have been omitted. For future studies, better coverage of the genetic variability across these four genes may be needed. Importantly, this limitation can be addressed with the future availability of data from the 1000 Genomes Project.95 Due to the reduced capacity on the O PAchip, we were only able to supplement the IBC chip with mainly SLC2A2 and SLC2A5 SNPs. Thus, KHK coverage was limited. Furthermore, there were no SI SNPs genotyped in the PEAR study. We, instead, relied on HAPI Heart for our analysis of SI associations. Although the association was not strong, there was potentially an interesting finding between SI

PAGE 54

54 SNPs and serum uric acid levels. Since sucrase plays an essential role in the breakdown of sucrose, polymorphisms of SI could potentially increase the release of free fructose, thus allowing more to be absorbed into the body. As the only sugar found to induce uric acid levels, higher fructose bioavailability could potentially increase serum uric acid concentrations. In conclusion, the study detected some potentially interesting associations between polymorphisms of KHK, SI SLC2A2 and SLC2A5 with various aspects of the metabolic syndrome in the study population of PEAR The data suggest that these genes, especially SLC2A2 may have an important role in increasing an individuals risk for developing adverse metabolic phenotypes, such as decreased HDL levels. More intensive studies are needed in larger populations to better characterize the impact of polymorphisms of KHK SI SLC2A2 and SLC2A5 on the dev elopment of adverse metabolic effects and increased disease risks, especially since recent studies have linked SNPs of SLC2A2 to the development of type 2 diabetes. Along with better coverage of these genes, the interesting SNPs from this study can be uti lized as candidate SNPS in future studies.

PAGE 55

55 Table 21. Summary of dbSNPs and pfSNPs for SLC2A2 SLC2A5 KHK, SI and ALDOB. Gene # of dbSNPs # Previously Validated in dbSNPs # pfSNPs # Validated pfSNPs # pfSNPs included for New Validation # Newly Valida ted SNPs* KHK 54 9 32 10 2 2 2 SI 283 113 10 9 SLC2A2 128 42 24 6 SLC2A5 151 19 30 8 22 8 ALDOB 82 19 22 9 13 2 Monomorphic dbSNPs and SNPs out of HWE were excluded.

PAGE 56

56 Table 22. Primer design of amplicons for unvalidated pfSNPs of KHK. Amplicon Size (bp) Primers (5' SNP ID rs# Region Chromosomal Position KHK_PCR1 401 FP: CAGGTATGTGCACTGAGACCCTTGCT 1 rs7424556 5' near gene 27161101 RP: ACATCTGCTGAGGCTCTCCTGTTGTG KHK_PCR2 906 FP: GACGTGTCTACGTCGGCTGAAGAGAC 2 rs7602823 5'near gene 27162910 RP: GATGGACTCACAGCTGATGCAAACCT KHK_PCR3 532 FP: GGGCCTAACAACTTTTAGTTTTAAAATACGG 3 rs11369867 intron 1 27166824 RP: ACCCAGCTGGTCAAAGATTTTCAAGCT KHK_PCR4 950 FP: TAATGATTTCTCATTTGGGCTCCAGTAACA 4 Gly40Arg exon 2 27169099 R P: TTCTGCTGGACACAACCCATTTCTAATC 5 Ala43Thr exon 2 27169108 KHK_PCR5 595 FP: AACTCCCAGAACAGGACTCTTCTTCTCTTAGAG 6 rs1063550 intron 3 27171231 RP: CACAGCCAAGCTGCTGATTAACGATTTAA 7 rs1063551 intron 3 27171249 8 rs1063552 intron 3 27171251 9 rs1063553 intron 3 27171256 1 0 rs1063554 intron 3 27171276 1 1 rs1063555 intron 3 27171280 KHK_PCR6 631 FP: TGAGAAGTCCTGAGTTCTAGCTCCATCAT 1 2 rs2075862 intron 4 27173782 RP: ATTCTACTCCTTGCTCACTGCTGAGGC 1 3 rs41288797 exon 5 27174019 KHK_PCR8 982 FP: ATGGAGACTACCATTGCGGCTGCATC 1 4 rs3769141 exon 8/3' UTR 27176302 RP: CATCAGGCATCATGCTGGGCATTTTAT 1 5 rs7573066 exon 8/3' UTR 27176702 1 6 rs6733022 exon 8/3' UTR 27177007 FP forward primer; RP reverse primer.

PAGE 57

57 Table 22. Continued. Amplicon Size (bp) Primers (5' SNP ID rs# Region Chromosomal Position KHK_PCR9 678 FP: GTGTTCTTAGACTCCAGTGTTTCCCCTG 1 7 rs1057393 3' near gene 27177793 RP: GTAAAAGGAGGGCTCAGAGCTTGCAC 1 8 rs11893427 3' near gene 27177844 19 rs1057391 3' near gen e 27177858 2 0 rs1057389 3' near gene 27177875 2 1 rs11889832 3' near gene 27177909 KHK_PCR10 430 FP: CCAGCATACCCATTGCACAGTATTTATAAA 2 2 rs3072688 3' near gene 27179761 RP: TTTTGTTATCAAGACCTTGAGGTGGGC FP forward primer; RP reverse pri mer.

PAGE 58

58 Table 23. Primer design of amplicons for unvalidated pfSNPs of SLC2A5 Amplicon Size (bp) Primers (5' SNP ID rs# Region Chromosomal Position SLC2A5_PCR1 1027 FP: ACAGAGTTCCCTCTGCAACACCAGGAG 1 rs12739286 5' near gene 9052719 RP: CCCTAGGTCATTCTGAGCTCCTTAAAATGG SLC2A5_PCR2 765 FP: GGGTATAGTGAGACCCCGTCTCTTCAA 2 rs2478868 intron 1 9049803 RP: ACTTAGGTGTTTGGGGTCCAGGGAGGT 3 rs2478869 intron 1 9049746 SLC2A5_PCR3 520 FP: GTTTCACAAACAGGTAGTGGGAGGGGC 4 rs12751666 intron 1 9048599 RP: ACCCATTTGTAACACCTACTACAGACCAGG SLC2A5_PCR4 401 FP: CTAACTGGACTCCAGGAAGCACAGTACC 5 rs1751680 intron 1 9044989 RP: CTACAGAGCACATGGAGGGAACAGATGG SLC2A5_PCR5 480 FP: TGCTCTAAGTGTTGGGGATATAGCGG 6 rs12086036 intron 1 9043612 RP: GCCATCACACTCAGATAGAAGGGATTTT SLC2A5_PCR6 609 FP: CAAACACTGTCAATGTCTGATCTTCT TTTCTG 7 rs12732059 intron 3 9039965 RP: ATTCATGGAAGACTTCCCCTTGACGT 8 rs12732058 intron 3 9039961 9 rs11121310 intron 3 9039958 SLC2A5_PCR7 417 FP: ATCTGCAAGGCAAGCGCGGG 10 rs12075583 exon 7 9022622 RP: GGCCGCCAAGAAAGGTAACGAGG 11 rs12564090 exon 7 9022493 SLC2A5_PCR8 208 FP: TGGATGAACGGGAAGATCAAGCC 12 rs11804131 exon 11 9020646 RP: TACCTGGTGGGTCTGCATGGCCA SLC2A5_PCR9 1026 FP: GTTTCTGACATCACACCTTCCTAGACAGG 13 rs762605 intron 11 9020438 RP: TACCTGGTGGGTCTGCATGGCCA FP forward primer; RP reverse primer.

PAGE 59

59 Table 23. Continued. Amplicon Size (bp) Primers (5' SNP ID rs# Region Chromosomal Position SLC2A5_PCR10 620 FP: GTTTCTGACATCACACCTTCCTAGACAGG 14 rs1063137 exon 12/3'UTR 9019838 RP: GCCAAGACGTTCATAGAGAT CAACCAG SLC2A5_PCR11 715 FP: TCTACTACCACATGGATCATTCGCGC 15 rs12068539 3' near gene 9017816 RP: CATAGATGGGTGGCAGAGGCTGGT SLC2A5_PCR12 874 FP: AGCCAGTGGCCTTCCGACACA 16 rs707452 3' near gene 9017118 RP: AGGTGTGGAGGCATTGAGGGATG 17 rs12025713 3' near gene 9017107 SLC2A5_PCR13 559 FP: TTTCATCCTCTCAGTTGGCAAACCTGG 18 rs12130301 3' near gene 9016483 RP: AGGGACATAAGTTTGGGGAAACCATAT 19 rs12119987 3' near gene 9016477 SLC2A5_PCR14 500 FP: TCAGTCCTCTCTGCTGGCTTTCCTGTG 20 rs11121303 3' near gene 9016021 RP: CCACGAGACAGCAAGGAAATCAGATCC SLC2A5_PCR15 832 FP: CCTCCTCCATGGAGCAGAAGGTTAGG 21 rs10779708 3' near gene 9015330 RP: ACCTCAGGGAGTGCTGTCAGGGGTAGT SLC2A5_PCR16 407 FP: AACAGGTGGGTAAATGACAAG ACTGAGAG 22 rs12120126 3' near gene 9011010 RP: GTGAACCAAGGCATATGAGACCACATG FP forward primer; RP reverse primer.

PAGE 60

60 Table 24. Sequencing results of pfSNPs of KHK. SNP ID rs# African Americans European Americans Test for HWE Test for HWE MA MAF X 2 p value MA MAF X 2 p value 2 rs7602823 A 0.40 1.119 0.290039 A 0.33 1.882 0.170154 MAF <5% 15 rs7573066 T 1.00 C 0.04 0.048 0.827435 Monomorphic SNPS 1 rs7424556 G 1.00 G 1.00 4 Gly40Arg G 1.00 G 1.00 5 Ala43Thr G 1.00 G 1.00 6 rs1063550 G 1.00 G 1.00 7 rs1063551 G 1.00 G 1.00 8 rs1063552 T 1.00 T 1.00 9 rs1063553 C 1.00 C 1.00 10 rs1063554 A 1.00 A 1.00 11 rs1063555 C 1.00 C 1.00 12 rs2075862 C 1.00 C 1.00 13 rs41288797 G 1.00 G 1.00 14 rs3769141 G 1.00 G 1.00 16 rs6733022 G 1.00 G 1.00 17 rs1057393 G 1.00 G 1.00 18 rs11893427 G 1.00 G 1.00 19 rs1057391 C 1.00 C 1. 00 20 rs1057389 C 1.00 C 1.00 21 rs11889832 T 1.00 T 1.00 Failed Sequencing 3 rs11369867 22 rs3072688 HWE Hardy Weinberg Equilibrium ; MA minor allele; MAF minor allele frequency.

PAGE 61

61 Table 2 5. Sequencing results of pfSNPs of SLC2A5 SNP ID rs# African Americans European Americans Test for HWE Test for HWE MA MAF X 2 p value MA MAF X 2 p value 2 rs2478868 C 0.31 1.630 0.201707 C 0.35 0.813 0.367211 3 rs24 78869 C 0.48 0.160 0.688834 C 0.42 0.960 0.327187 5 rs1751680 A 0.29 1.059 0.303387 A 0.35 0.813 0.367211 6 rs12086036 G 0.21 0.003 0.958873 G 0.33 0.179 0.672399 14 rs1063137 G 0.50 0.667 0.414216 G 0.29 0.002 0.967162 15 rs12068539 A 0.46 0. 445 0.504869 A 0.29 0.002 0.967162 21 rs10779708 A 0.45 1.766 0.183864 A 0.29 0.002 0.967162 22 rs12120126 T 1.00 G 0.29 1.059 0.303387 23 GT_9043651 G 0.08 0.198 0.656058 T 1.00 24 CT_9011003 C 0.17 0.480 0.488422 T 1.00 Monomorphic SNPS 10 rs12075583 G 1.00 G 1.00 11 rs12564090 A 1.00 A 1.00 12 rs11804131 G 1.00 G 1.00 13 rs762605 G 1.00 G 1.00 Out of HWE 16 rs707452 T 0.32 13.75 0.000209 T 0.14 8.29 0.003976 17 rs12025713 C 0.14 22.00 0.000003 C 0.09 22.00 0.000003 18 rs12130301 T 0.15 20.00 0.000008 T 0.05 21.00 0.000005 19 rs12119987 A 0.15 20.00 0.000008 A 0.05 21.00 0.000005 20 rs11121303 C 0.40 10.00 0.001565 C 0.15 13.00 0.000311 Failed Sequencing 1 rs12739286 4 rs12751666 7 rs12732059 8 rs12732058 9 rs11121310 HWE Hardy Weinberg Equilibrium ; MA minor allele; MAF minor allele frequency. potentially a new SNP discovered.

PAGE 62

62 Table 26. Validated candidate pfSNPs for KHK, SI SLC2A2 and SLC2A5 Sequenced Data # Gene rs# Population MA MAF Population MA MAF 1 KHK rs7602823 AF A 0.40 EU A 0.33 2 rs2119026 YRI A 0.18 CEU G 0.40 3 r s12714092 YRI A 0.08 CEU A 0.38 4 rs2304681 YRI A 0.28 CEU A 0.40 5 rs2384571 YRI G 0.33 CEU G 0.40 6 rs2384572 YRI T 0.22 CEU G 0.41 7 SLC2A5 rs770041 YRI A 0.09 CEU A 0.34 8 rs3820034 YRI A 0.07 CEU A 0.26 9 rs5438 YRI T 0.19 CEU A 0.04 10 rs2478868 AF C 0.31 EU C 0.35 11 rs2478869 AF C 0.48 EU C 0.42 12 rs1751680 AF A 0.29 EU A 0.35 13 rs12086036 AF G 0.21 EU G 0.33 14 rs1063137 AF G 0.50 EU G 0.29 15 rs5840 YRI T 0.50 CEU T 0.36 16 rs12068539 AF A 0.46 EU A 0.29 17 rs10779708 AF A 0.45 EU A 0.29 18 rs12120126 AF T 1.00 EU G 0.29 19 GT_9043651 AF G 0.08 EU T 1.00 20 CT_9011003 AF C 0.17 EU T 1.00 21 SLC2A2 rs5393 CEPH C 0.33 22 rs5395 HYP2 C 0.05 23 rs5396 CEU G 0.27 24 rs11720145 YRI A 0.43 CEU A 0.13 25 rs7643425 YRI G 0.40 26 rs11925298 YRI A 0.07 CEU A 0.02 27 rs5400 YRI T 0.49 CEU T 0.14 28 rs5401 HYP3 C 0.05 29 rs5404 YRI A 0.32 CEU A 0.12 30 rs5408 HYP1 C 0.05 31 SI rs764158 6 YRI G 0.15 CEU G 0.35 32 rs9290264 YRI C 0.33 CEU C 0.33 33 rs13086543 YRI A 0.08 CEU G 1.00 34 rs9283633 YRI C 0.44 CEU T 0.33 35 rs9290257 YRI A 0.41 CEU G 0.38 36 rs9838509 YRI A 0.43 CEU C 0.32 37 rs4855271 YRI C 0.03 CEU C 0.07 38 rs6799858 YRI A 0.16 CEU G 1.00 39 rs9917722 YRI C 0.16 CEU G 1.00 MA minor allele; MAF minor allele frequency. Populations used for genotyping: AF genomic DNA of African descent used for sequencing; CEPH Centre d'Etude du Polymorphism e Human; CEU: CEPH Utah residents with ancestry from northern and western Europe (HapMap); EU genomic DNA of European descent used for sequencing; HYP1 Harare, Zimbabwe; HYP2 Tecumseh, Michigan and CEPH; HYP3 Harare, Zimbabwe, T ecumseh, Michigan, and CEPH; YRI Yoruba in Ibadan, Nigeria (HapMap).

PAGE 63

63 Table 27. Quality control of PEAR genotyping data. Variable IBC Chip OPA chip Samples genotyped 418 418 Samples excluded low DNA yield and quality 1 1 call rate < 90%/contamination 1 0 call rate < 90% 0 3 sex gender mismatch 1 0 Project SNPs 29 31 Project SNPs excluded call rate < 95% 2 2 MAF < 5% 4 0 m onomorphic 1 3 out of HWE 0 0 SNPs included i n analyses 4 8 HWE H ardy Weinberg Equilibrium ; MAF minor allele frequency; IBC ITMAT/Broad/CARE genotyping chip; OPA oligo pool all custom SNP genotyping chip ; PEAR Pharmacogenomic Evaluation and Antihypertensive Responses .

PAGE 64

64 Tabl e 28 PEAR: Baseline ch aracteristics. Subjects (n = 407) Age 50.1 8.9 Female 229 (56.3) Race White, European American 234 (57.5) Black, African American 159 (39.1) Asian 5 (1.2) Other/Multiracial 9 (2.2) Primary Phenotypes Tg (mg/d L) 127 .3 95.3 UA (mg/dL) 5.6 1.4 HOMA 2.2 2.1 Secondary Phenotypes HDBP (mmHg) 93.5 6.5 HSBP (mmHg) 145.9 10.9 Glucose (mg/dL) 91.4 10.0 Insulin (mg/dL) 9.5 8.3 HDL (mg/dL) 49.5 14.2 LDL (mg/dL) 12 1.7 30.9 BMI (kg/m 2 ) 30.9 5.7 MS 43.5% BMI body mass index; HDBP home diastolic blood press u re; HDL highdensity lipoprotein; HOMA homeostatic model assessment; HSBP home systolic blood pressure; LDL low density lipoprotein; MS metabolic s yndrome; Tg triglycerides; UA uric acid. Data are given as mean standard deviation or n (%).

PAGE 65

65 Table 29. SNPs associated with baseline serum uric acid in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (% ) Mean n (%) Mean P value (mg/dL) (mg/dL) (mg/dL) SLC2A2 rs10513684 PEAR S UA EA ANOVA 0.0471 A 210 (90.1) 5.6 1.3 23 (9.9) 6.2 1.9 ANOVA 0.0159 B 5.6 0.1 6.1 0.2 SLC2A5 rs12145292 PEAR S UA AA ANOVA 0.0458 A 77 (49.4) 5.4 1.3 67 (42.9) 5.2 1.6 12 (7.7) 6.3 1.7 ANOVA 0.9068 B 5.8 0.1 5.7 0.2 5.9 0.4 ADD 0.8950 B REC 0.8088 B SLC2A5 rs5438 PEAR S UA EA: ANOVA 0.0002 A 102 (82.9) 6.1 1.2 21 (17.1) 7.2 1.4 Male ANOVA 0.0007 C 6.1 0.1 7.1 0.3 A unadjusted; B adjusted for age, BMI, and gender; C adjusted for age and BMI; PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African American; EA European American. SUA serum uric acid. ADD addit ive; ANOVA analysis of covariance; REC recessive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard de viation. Adjusted data are given as mean standard error. Shaded SNPs have an adjusted pvalue

PAGE 66

66 Table 210. SNPs associated with baseline triglycerides in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mg/dL) (mg/dL) (mg/dL) SLC2A2 r s10513684 PEAR Tg EA ANOVA 0.0407 A 210 (90.1) 142.0 90.2 23 (9.9) 186.1 149.9 ANOVA 0.0342 B 140.6 6.6 185.1 19.8 SLC2A2 rs11924032 PEAR Tg EA ANOVA 0.0095 A 124 (53.2) 131.2 69.1 97 (41.6) 158.0 109.9 12 (5.2) 209.2 190.5 AN OVA 0.0038 B 128.6 8.5 157.8 9.6 211.3 27.1 ADD 0.0011 B SLC2A2 rs5396 PEAR Tg EA ANOVA 0.0160 A 116 (49.8) 130.1 69.2 100 (42.9) 157.4 109.4 17 (7.3) 192.6 162.3 ANOVA 0.0066 B 126.8 8.8 158.3 9.5 190.7 22.8 ADD 0.0015 B SLC2A2 rs5398 PEAR Tg EA ANOVA 0.0070 A 111 (48.3) 128.6 69.6 104 (45.2) 160.5 107.3 15 (6.5) 197.8 172.2 ANOVA 0.0046 B 126.8 9.0 159.4 9.3 197.0 24.3 ADD 0.0010 B SLC2A2 rs8192675 PEAR Tg EA ANOVA 0.0 051 A 114 (48.9) 127.6 69.0 104 (44.6) 160.5 107.3 15 (6.4) 197.8 172.2 ANOVA 0.0034 B 126.0 8.8 159.4 9.3 197.1 24.2 ADD 0.0007 B SLC2A5 rs12086036 PEAR Tg AA ANOVA 0.0004 A 75 (51.0) 88.7 46.7 62 (42.2) 94.2 69 10 (6. 8) 202.3 249.5 ANOVA 0.0043 B 97.4 10.4 104.1 11.5 194.5 26.5 ADD 0.0180 B REC 0.0011 B A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Respons es AA African American; EA European American. Tg triglycerides. ADD additive; ANOVA analysis of covariance; DOM dominant; REC recessive. HCA homoz ygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error. Shaded SNPs have an adjusted p value

PAGE 67

67 Table 210. Continued. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mg/dL) (mg/dL) (mg/dL) SLC2A5 rs12145292 PEAR Tg AA ANOVA 0.0037 A 77 (49.4) 90.1 47.2 67 (42.9) 92.0 67.0 12 (7.7) 176.3 233.6 ANOVA 0.0388 B 98.0 10.1 101.5 11.0 166.7 24.2 ADD 0.0722 B REC 0.0110 B SLC2A5 rs1751680 PEAR Tg EA ANOVA 0.0241 A 93 (39.9) 128.3 58.5 109 (46.8) 153.1 109.6 31 (13.3) 180.5 136.2 ANOVA 0.0234 B 128.0 9.9 150.5 9.1 180.1 17.0 ADD 0.0063 B SLC2A5 rs2478868 PEAR Tg EA ANOVA 0.0456 A 92 ( 39.5) 127.5 58.3 112 (48.1) 157.3 117.0 29 (12.4) 167.5 112.1 ANOVA 0.0451 B 127.1 9.9 154.2 9.0 169.9 17.6 ADD 0.0140 B DOM 0.0181 B SLC2A5 rs6680123 PEAR Tg AA ANOVA 0.0228 A 122 (78.2) 89.3 49.9 34 (21.8) 127.1 156.4 ANOVA 0.0294 B 98.9 8.3 134.9 14.5 SLC2A5 rs6694527 PEAR Tg AA ANOVA 0.0104 A 127 (80.8) 89.0 49.3 30 (19.2) 133.5 165.6 ANOVA 0.0088 B 98.6 8.0 143.3 15.4 SLC2A5 rs770041 PEAR Tg EA ANOVA 0.0014 A 95 (40.8 ) 131.0 6 2.0 116 (49.8) 152.2 114.5 22 (9.4 ) 182.1 1 22.0 ANOVA 0.0 337 B 129.1 9.8 150.4 8.9 184.7 20.2 ADD 0.0103 B A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensi ve Responses AA African American; EA European American. Tg triglycerides. ADD additive; ANOVA analysis of covariance; DOM dominant; REC recessive. HCA homoz ygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error. Shaded SNPs have an adjusted p value

PAGE 68

68 Table 211. SNPs associated with baseline HOMA in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Mod el n (%) n (%) n (%) P value Mean Mean Mean SLC2A2 rs11925298 PEAR HOMA AA ANOVA 0.0303 A 124 (79.5) 2.2 2.0 31 (19.9) 1.7 1.3 1 (0.6) 6.6 ANOVA 0.0175 B 2.3 0.2 1.7 0.3 6.3 1.8 ADD 0.4089 B SLC2A5 rs25 05972 PEAR HOMA AA ANOVA 0.0474 A 79 (50.6) 2.1 1.8 60 (38.5) 2.4 2.2 17 (10.9) 1.1 0.7 ANOVA 0.1502 B 2.1 0.2 2.5 0.2 1.5 0.4 ADD 0.6854 B REC 0.1054 B A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African American. HOMA homeostatic model assessment. ADD additive; ANOVA analysis of covariance. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error.

PAGE 69

69 Table 212. SNPs associated with baseline HDBP in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mmHg) (mmHg) (mmHg) SLC2A2 rs16855638 PEAR HDBP EA ANOVA 0.0162 A 232 (99.6) 92.9 5.8 1 (0.4) 107 ANOVA 0.0330 B 92.9 0.4 105.3 5.8 SLC2A2 rs5393 PEAR HDBP EA ANOVA 0.0374 A 174 (74.7) 93.5 6.1 56 (24 .0) 91.3 5.0 3 (1.3) 95.3 1.5 ANOVA 0.0564 B 93.4 0.4 91.4 0.8 95.7 3.3 ADD 0.0924 B SLC2A5 rs12086036 PEAR HDBP AA ANOVA 0.0457 A 75 (51.7) 94.3 6.4 60 (41.4) 93.9 7.6 10 (6.9) 99.9 8.6 ANOVA 0.1048 B 94.6 0.9 94.6 1.0 99.8 2.3 ADD 0.1946 B REC 0.0333 B SLC2A5 rs12145292 PEAR HDBP AA ANOVA 0.0008 A 77 (50.0) 94.7 6.5 65 (42.2) 93.2 7.4 12 (7.8) 101.8 9.0 ANOVA 0.0047 B 94.9 0.9 93.9 0.9 101.6 2.1 ADD 0.1669 B REC 0.0015 B SLC2A5 rs7656 17 PEAR HDBP AA ANOVA 0.00 87 A 33 (21.4) 95.7 6.3 71 (46.1) 92.7 7.2 50 (32.5) 96.7 7.7 ANOVA 0.0 164 B 96.3 1.3 93.4 0.9 9 7 .0 1.0 ADD 0. 4116 B KHK rs7588333 P EAR HDBP EA ANOVA 0.0300 A 231 (99.1) 92.9 5.8 2 (0.9) 101.9 7.2 ANOVA 0.0320 B 92.9 0.4 101.7 4.1 A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African American ; EA European American. HDBP home diastolic blood pressure. ADD additive; ANOVA analysis of covariance; DOM dominant; REC rec essive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error. Shaded SNPs have an adjusted pvalue

PAGE 70

70 Table 213. SLC2A5 SNPs associated with baseline HSBP in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mmHg) (mmHg) (mmHg) SLC2A5 rs1705295 PEAR HSBP AA ANOVA 0.0 434 A 150 (97.4 ) 146.5 1 1.7 4 (2.6 ) 134.2 17.5 ANOVA 0.0 217 B 147.5 1. 0 133.8 5.8 SLC2A5 rs17842190 PEAR HSBP EA ANOVA 0. 0 306 A 225 (96.6 ) 145.6 9.9 8 (3.4 ) 153.6 15.1 ANOVA 0. 0290 B 145.7 0. 7 153.4 3.5 SLC2A5 rs6680123 PEAR HSBP AA ANOVA 0.0219 A 121 (78.6) 145.0 11.5 33 (21.4) 150.4 13.1 ANOVA 0.0709 B 146.1 1.2 150.3 2.1 KHK rs7588333 PEAR HSBP EA ANOVA 0.0015 A 231 (99.1) 145.7 10.0 2 (0.9) 168.6 4.8 ANOVA 0.0006 B 145.7 0.6 169.9 6.9 A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African Ameri can; EA European American. HSBP home systolic blood pressure. ADD additive; ANOVA analysis of covariance. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted dat a are given as mean standard error. Shaded SNPs have an adjusted pvalue

PAGE 71

71 Table 214. SNPs associated with baseline glucose in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mg/dL) (mg/dL) (mg/dL) SLC2A5 rs12080175 PEAR Glucose EA ANOVA 0.0456 A 125 (53.6) 92.0 10.7 92 (39.5) 91.5 8.9 16 (6.9) 98.2 10.0 ANOVA 0.0301 B 91.7 0.9 91.4 1.0 98.2 2.4 ADD 0.1225 B REC 0.00 82 B SLC2A5 rs765617 PEAR Glucose EA ANOVA 0.0355 A 104 (44.6) 91.3 9.7 105 (45.1) 93.9 9.9 24 (10.3) 88.7 11.5 ANOVA 0.0108 B 91.0 0.9 93.8 0.9 87.8 2.0 ADD 0.9835 B REC 0.0728 A A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses EA European American. ADD additive; ANOVA analysis of covariance; REC recessive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error.

PAGE 72

72 Table 215. SNPs associated with baseline insulin in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (% ) Mean n (%) Mean n (%) Mean P value (IU/mL) (IU/mL) (IU/mL) SLC2A5 rs2505972 PEAR Insulin AA ANOVA 0.0351 A 79 (50.6) 8.9 6.8 60 (38.5) 10.6 9.2 17 (10.9) 5.3 3.1 ANOVA 0.1186 B 9.2 0.9 10.8 0.9 6.8 1.8 ADD 0 .7660 B REC 0.0250 A 139 (89.1) 9.7 8.0 17 (10.9) 5.3 3.1 REC 0.1014 B 9.9 0.7 6.9 1.8 A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA Af rican American; EA European American. ADD additive; ANOVA analysis of covariance; REC recessive. HCA homozygote common allele; HET heterozy gote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error.

PAGE 73

73 Table 216. SNPs associated with baseline HDL in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model P value n (%) Mean n (%) Mean n (%) Mean (ANOVA) (mg/dL) (mg/dL) (mg/dL) SLC2A2 rs 5396 PEAR HDL AA ANOVA 0.0081 A 66 (42.3) 47.5 12.6 73 (46.8) 54.4 14.9 17 (10.9) 56.2 17.6 ANOVA 0.0389 B 47.1 1.7 52.2 1.7 54.8 3.3 ADD 0.0123 B DOM 0.0141 B SLC2A2 rs5398 PEAR HDL EA ANOVA 0.0197 A 111 (48 .3) 50.6 14.2 104 (45.2) 46.8 13.3 15 (6.5) 41.5 12.2 ANOVA 0.0028 B 51.2 1.2 47.0 1.2 41.5 3.1 ADD 0.0006 B SLC2A2 rs8192675 PEAR HDL AA ANOVA 0.0331 A 76 (51.7) 49.1 14.9 62 (42.2) 54.5 13.8 9 (6.1) 59.3 17.3 ANOVA 0.0791 B 48.3 1.7 52.4 1.9 57.5 4.7 ADD 0.0245 B SLC2A2 rs8192675 PEAR HDL EA ANOVA 0.0228 A 114 (48.9) 50.4 14.1 104 (44.6) 46.8 13.3 15 (6.4) 41.5 12.2 ANOVA 0.0034 B 51.0 1.1 47.0 1.2 41.5 3.1 ADD 0.0008 B SLC2A5 rs12086036 PEAR HDL AA ANOVA 0.0182 A 75 (51.0) 51.7 13.6 62 (42.2) 54.3 16.0 10 (6.8) 40.2 11.1 ANOVA 0.0361 B 51.2 1.7 52.8 1.9 40.0 4.4 ADD 0.3064 B REC 0.0126 B SLC2A5 rs12145292 PEAR HDL AA ANOVA 0.0236 A 77 (49.4) 51.2 13.5 67 (42.9) 54.0 15.8 12 (7.7) 41.7 11.3 ANOVA 0.0888 B 50.8 1.7 52.0 1.8 42.1 4.0 ADD 0.3467 B REC 0.0326 B A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African American; EA European American. HDL highdensity lipoprotein. ADD additive; ANOVA analysis of covariance; DOM dominant; REC recessive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error. Shaded SNPs have an adjusted pvalue

PAGE 74

74 Table 217. SNPs associated with baseline LDL in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mg/dL) (mg/dL) (mg/dL) SLC2A5 rs12736085 PEAR LDL EA ANOVA 0.0476 A 126 (54.1) 117.5 28.5 86 (36.9) 121.7 29.5 2 1 (9.0) 133.9 25.4 ANOVA 0.0396 B 117.3 2.6 122.0 3.1 134.2 6.3 ADD 0.0159 B REC 0.0239 B A unadjusted; B adjusted for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses EA European American. LDL low density lipoprotein. ADD additive; ANOVA analysis of covariance; REC recessive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjust ed data are given as mean standard error.

PAGE 75

75 Table 218. SNPs associated with baseline BMI in PEAR. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (kg/m 2 ) (kg/m 2 ) (kg/m 2 ) SLC2A2 rs11711437 PEAR BMI EA ANOVA 0.0217 A 171 (73.7) 31.0 5.6 59 (25.4) 30.2 5.1 2 (0.9) 41.0 3.2 ANOVA 0.0305 D 31.0 0.4 30.2 0.7 40.5 3.9 ADD 0.9455 D REC 0.0140 D SLC2A5 rs1612895 PEAR BMI AA ANOVA 0. 0176 A 79 (50.3) 32.3 5.9 64 (40.8) 30.2 6.3 14 (8.9) 28.1 1.8 ANOVA 0.0192 D 31.9 0.7 29.9 0.8 27.8 1.5 ADD 0.0049 D A unadjusted; D adjusted for age and gender. PEAR Pharmacogenomic Evaluation and Antihy pertensive Responses AA African American; EA European American. BMI body mass index. ADD additive; ANOVA analysis of covariance; REC recessive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Unadjusted data are given as mean standard deviation. Adjusted data are given as mean standard error.

PAGE 76

76 Table 219. SNPs associated with baseline MS in PEAR. MS = 1 MS = 0 Gene Marker Study Trait Race Model P value HCA HET HMA HCA HET HMA SLC2A5 rs12736085 PEAR MS AA X 2 0.0368 A 18 (36.0) 29 (58.0) 3 (6.0) 47 (52.2) 32 (35.6) 11 (12.2) X 2 0.0279 B ADD 0.3022 B SLC2A5 rs6694527 PEAR MS AA X 2 0.0303 A 38 (71.7) 15 (28.3) 83 (86.5) 13 (13.5) X 2 0.0111 B A unadjusted; B adjust ed for age, BMI, and gender. PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African American; MS metabolic syndrome. ADD additive; X2 chi square HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Data are given as n (%).

PAGE 77

77 Table 220. Summary of SNPs significantly associated after adjustments in PEAR. # Gene RS# Study Trait Race Model P value Replicated /Excluded 1 KHK rs7588333 PEAR HSBP EA ANOVA 0.0006 B Excluded 2 SLC2A2 rs11924032 PEAR Tg EA ADD 0.0011 B Tagged 3 SLC2A2 rs5396 PEAR Tg EA ADD 0.0015 B Tagged 4 SLC2A2 rs5398 PEAR HDL EA ADD 0.0006 B Tagged 5 SLC2A2 rs5398 PEAR Tg EA ADD 0.0010 B Tagged 6 SLC2A2 rs8192675 PEAR HDL EA ADD 0.0008 B R eplicated in GERA 7 SLC2A2 rs8192675 PEAR Tg EA ADD 0.0007 B Not replicated in GERA 8 SLC2A5 rs12086036 PEAR Tg AA REC 0.0011 B Excluded 9 SLC2A5 rs12145292 PEAR HDBP AA REC 0.0015 B Excluded 10 SLC2A5 rs1612895 PEAR BMI AA ADD 0.0049 D Excluded 11 SLC2A5 rs5438 PEAR S UA EA: Male ANOVA 0.0007 C Excluded B adjusted for age, BMI, and gender; C adjusted for age and BMI; D adjusted for age and gender. GERA Genetic Epidemiology of Responses to Antihypertensives ; PEAR Pharmacogenomic Evaluation and Antihypertensive Responses AA African American; EA European American. BMI body mass index; HDBP home diastolic blood pressure; HDL highdensity lipoprotein; HSBP home systolic blood pressure; SUA serum uric acid; Tg triglycerides. ADD additive; ANOVA analysis of covariance; DOM dominant; REC recessive. Shaded SNPs were replicated.

PAGE 78

78 Table 221. GERA I and II: Replication results. Genotype HCA HET HMA Gene Marker Study Trait Race n (%) Mean n (%) Mean n (%) Mean Model P value (mg/dL) (mg/dL) (mg/dL) SLC2A2 rs8192675 GERA I HDL EA ANOVA 0.0664 B 98 (52.4) 44.7 75 (40.1) 43.8 14 (7.5) 37.7 GERA II HDL EA ANOVA 0.0692 B 65 (52.0) 52.2 53 (42.4) 54.9 7 (5.6) 42.3 GERA I + II HDL EA ANOVA 0.0126 B 163 (52.2) 47.4 128 (41.0) 48 21 (6.7) 39.4 GERA I Tg EA ANOVA 0.089 0 B 98 (52.1) 193.5 76 (40.4) 166.2 14 (7.4) 213.2 GERA II Tg EA ANOVA 0.9896 B 98 (50.0) 165.5 87 (44.4) 164.3 11 (5.6) 160.8 GERA I + II Tg EA ANOVA 0.3170 B 196 (51.0) 179.8 163 (42.4) 166.1 25 (6.5) 191.4 B adjusted for age, BMI, and gend er. GERA Genetic Epidemiology of Responses to Antihypertensives EA European American. HDL highdensity lipoprotein; Tg triglycerides. ANOVA analysis of covariance. HCA homozygote common allele; HET heterozygote; HMA ho mozygote minor allele.

PAGE 79

79 Figur e 21. Genotype calling from sequence alignment. Sequence data was aligned against reference sequence of SLC2A5 PCR fragment 5. For SNP rs5840, samples NA18516 and NA17138 had C/T genotype. NA07029 and NA19238 had C/C genotype. NA05377 and NA19917 had T/T genotype.

PAGE 80

80 Figure 22. R2 plot of KHK candidate SNPs generated from HapMap data of YRI population. The numbers in the diamondshaped boxes represent the r2 values between the two corresponding SNPs. Black diamond boxes represent r2 = 1. Plot was generated by HaploView 4.1 ( http://www.broadinstitute.org/mpg/haploview ).96

PAGE 81

81 Figure 23. R2 plot of KHK candidate SNPs generated from HapMap data of CEU population. The numbers in the diamondshaped boxes represent the r2 values between the two correspondi ng SNPs. Black diamond boxes represent r2 = 1. Plot was generated by HaploView 4.1 ( http://www.broadinstitute.org/mpg/haploview ).96

PAGE 82

82 Figure 24. R2 plot of SLC2A2 candidate SNPs generated from HapMap data of YRI population. The numbers in the diamondsh aped boxes represent the r2 values between the two corresponding SNPs. Black diamond boxes represent r2 = 1. Plot was generated by HaploView 4.1 ( http://www.broadinstitute.org/mpg/haploview ).96

PAGE 83

83 Figure 25. R2 plot of SLC2A2 candidate SNPs generated from HapMap data of CEU population. The numbers in the diamondshaped boxes represent the r2 values between the two corresponding SNPs. Black diamond boxes represent r2 = 1. Plot was generated by HaploView 4.1 ( http://www.broadinstitute.org/mpg/haploview ).96

PAGE 84

84 Figure 2 6. R2 plot of SLC2A5 candidate SNPs generated from HapMap data of YRI population. The numbers in the diamondshaped boxes represent the r2 values between the two corresponding SNPs. Black diamond boxes represent r2 = 1. Plot was generated by HaploV iew 4.1 ( http://www.broadinstitute.org/mpg/haploview ).96

PAGE 85

85 Figure 27. R2 plot of SLC2A5 candidate SNPs generated from HapMap data of CEU population. The numbers in the diamondshaped boxes represent the r2 values between the two corresponding SNPs. Black di amond boxes represent r2 = 1. Plot was generated by HaploView 4.1 ( http://www.broadinstitute.org/mpg/haploview ).96

PAGE 86

86 Figure 28. Effects of rs8192675 on HDL levels in the PEAR and GERA I and II study populations. ANOVA was adjusted for age, BMI, and gender.

PAGE 87

87 CHAPTER 3 CLINICAL STUDY: PHARMACOKINETICS AND PHA RMACODYNAMICS OF FRUCTOSE FOLLOWING S OFT DRINK CONSUMPTIO N: SUCROSE VERSUS HI GH FRUCTOSE CORN SYRUP Introduction United States is in the midst of a cardiorenal disease crisis. During the past four decades a n increasing percentage of the population is suffering from various health disorders, including hypertension, obesity, metabolic syndrome, diabetes, and kidney disease. Currently, onethird of the population has hypertension, 7% has diabetes, about 20 million has kidney disease, and onethird of adults and onesixth of children are obese.52, 59, 83, 97, 98 Importantly, a similar rise in the consumption of fructose has also occurred during this same time period and recent studies have implicated fructose to be at the root of these epidemics.2, 8, 10, 12 Fructose has been shown to cause a variety of harmful metabolic effects, such as l actic acidosis, lipogenesis h ypertrig lyceridemia and increased weight gain (Chapter 1: Toxicities of Fructose & Potential Role in Disease Risk).2, 5, 10, 12, 31, 33 In addition to these adverse effects, fructose is the only natural sugar capable of r aising uric acid concentrations which have been shown to in crease by 14 mg/dL after a fructose meal .13 With recent studies linking elevated serum uric acid levels with increased risk for hypertension and metabolic syndrome2023, fructose induced hyperuricemia may be a key mechanism driving the cardiorenal epidemic s.52 The two most commonly used sweeteners in the Western diet are high fructose corn syrup (HFCS) and sucrose. However, the sharp increase in fructose consumption is primarily due to the increase use of high fructose corn syrup (HFCS) in the Western diet. From 1967 to 2006, the annual per capital intake of HFCS increased from 0.03 lbs

PAGE 88

88 to 58.2 lbs (Figure 13).810 As a result, Bray et al has hypothesized that there is a causal link between the increased consumption of HFCS and the epidemic in obesity.10 Meanwhile, others have suggested that the excessive consumption of fructosecontaining sweeteners, regardless of its composition, may play an important role in the pathogenesis of a number of health disorders, such as high blood pressure and metabolic syndrome.31, 33, 3542 Studies that directly compare the metabolic effects of sucrose and HFCS are lacking. Thus, the purpose of the clinical study was to compare the impact of sucrose versus HFCS on fructose bioavailability and acute metabolic changes by measuring response phenotypes, such as serum uric acid, lactate, and triglyceride levels. Soft drinks made wit h either cane sugar (sucrose) or HFCS were given to healthy volunteers to consume. Although the compositions of the two sweeteners are very similar, the HFCS grade used in soft drinks contains 5% more fructose than sucrose. Thus, soft drinks sweetened wi th HFCS, which contains a higher amount of fructose, would provide more fructose into the systemic circulation than soft drinks sweetened with sucrose. Furthermore, HFCS provides an immediate source of free fructose and glucose, while sucrose must first be broken down by sucrase. The expression and function of sucrase can be influenced by negative factors, such as genetic polymorphisms and negative regulatory inhibition by glucose. Thus, the amount of fructose available for absorption is reduced.99104 Therefore, due to the potential inefficiency of sucrase, the relative fructose bioavailability from HFCS may also be higher. Ultimately, we hypothesized that higher fructose systemic concentrations, either through the higher

PAGE 89

89 fructose dose or also from increased fructose bioavailability from HFCS, would lead to increased fructose induced adverse metabolic effects. In addition, w e hypothesize d that repeated exposure to excessive fructose ingestion leads to progressiv e adverse effect s. Therefore, individuals with higher chronic intake of fructose may potentially have a greater risk for developing higher levels of adverse metabolic phenotypes, such as increased levels of serum uric acid. Materials and Methods Study D esign The clinical study was a prospective, randomized, single blinded, cross over study. Adult participants aged 18 years or older, of either gender, and of any ethnicity were recruited. A participants eligibility was determined at a screening visit. Qualified participants were then randomized in blocks of four, using Proc Plan in SAS 9.1.3, to receive either sucrose or HFCS sweetened soft drinks at the first study visit, then crossed over to the other formulation (Figure 31). The two study visits were separated by a minimum of 2 days and were conducted by the General Clinical Research Center (GCRC) at the University of Florida, Gainesville, FL. Exclusion Criteria Volunteers were excluded from participating in the study if they met any of the follow ing criteria: 1. history of hypoglycemia 2. history of gout 3. history of liver or kidney disease 4. history of diabetes mellitus or fasting glucose glucose

PAGE 90

90 5. taking any medication (except for oral contraceptives) 6. consume more than 7 alcoholic drinks per week 7. pregnant or breast feeding 8. donated blood within 8 weeks prior to the screening visit. The presence of exclusion criteria were ascertained through a Volunteer Health Information Questionnaire (Appendix B: Volunteer Health Infor mation Questionnaire) and limited laboratory analysis. Finger stick b lood g lucose Finger stick blood tests were used to determine glucose levels. OneTouch UltraSoft Lancets were used to prick the fingertips of participants. Blood was captured on the O neTouch Ultra Test Strips and glucose was measured by OneTouch Ultra 2 Blood Glucose Meter (LifeScan, Inc., Milpitas, CA). Pregnancy test Urine pregnancy tests were conducted on women of childbearing potential. Leader Early Result Pregnancy Tests were used (Cardinal Health, Dublin, OH). Participants Sixty nine relatively healthy men and women were screened at the University of Florida in Gainesville, FL from April 15, 2008 to April 2, 2009. Participants were recruited from the general population through advertisements, Craigslist, or from Genetics Screening Database (IRB#4402005). Informed consent was obtained from all participants (IRB#6822007). Forty eight subjects, from ages 18 to 52, participated in the study. However, one subject was withdrawn after the first study visit due to reasons unrelated to the study. An additional seven were withdrawn due to developing asymptomatic hypoglycemia

PAGE 91

91 from the sugar load. Hypoglycemia was defined as blood glucose confirmed by two separate measurements. Overall, 40 individuals completed both study visits. The study visits wer e conducted from May 16, 2008 to May 4, 2009. Sugar Load from Soft Drinks Dr Pepper sweetened with high fructose corn syrup was purchased (Lot# NOV 24 08 12:33 to 12:58RS02218X, Dr Pepper/Seven Up, Inc., Plano, TX)). Cane sugar (sucrose) sweetened Dr P epper was purchased from the Dr Pepper Bottling Company (Lot# 800807:11 TBC, http://www.dublindrpepper.com/ Dublin, TX). The total carbohydrate load from 24 oz is about 80 g. Sugar content To confirm the sugar content of the soft drinks, three cans of each type of Dr Pepper were analyzed before (April 05, 2008) and after the end of study (May 12, 2009) for their sugar profile (Appendix B: Sugar Analyses). The analyses were conducted by Silliker, Inc. (Illinoi s Laboratory, Chicago Heights, IL). From the analyses, the average fructose dose was 34.6 g from sucrose and 39.2 g or 13% more fructose from the soft drinks containing HFCS. The average glucose dose from sucrose was 34.8 g and from HFCS was 28.8 g. Food Intake Data Short term food intake data were collected through a threeday food log. Longterm diet intake was obtained through a food frequency questionnaire (FFQ). Food logs and FFQs were collected during a study visit. However, a couple of patients who were withdrawn from the study, failed to turn in their food intake data.

PAGE 92

92 Threeday food record Once a participant was deemed to qualify for the study, they were given a threeday food log (GCRC, University of Florida, Gainesville, FL; Appendix B: Thr eeDay Food Record) and a servingsize booklet (Fred Hutchinson Research Cancer Center, Seattle, WA; Appendix B: Serving Size Booklet ) at the end of the screening visit. They were asked to record food intake for three consecutive days, which included at l east one weekend day. The food log was analyzed by dieticians at the GCRC. To reflect the marketplace throughout the study, dietary intake data were collected and analyzed using Nutrition Data System for Research (NDSR) software versions 2007 and 2008 developed by the Nutrition Coordinating Center (NCC), University of Minnesota, Minneapolis, MN.105 F inal calculations were completed using NDSR version 2008. The NDSR time related database updates analytic data while maintaining nutrient profiles true to the version used for data collection. Food frequency questionnaire Nutrient data were collected using a food frequency questionnaire developed by the Nutrition Assessment Shared Resource (NASR) of Fred Hutchinson Cancer Research Center (Appendix B: Food Frequency Questionnaire). The general form of the FFQ was self administered during the first study visit. The FFQ asked participants to report the frequency of consumption and portion size of foods and beverages over a time period of 3 months. To minimize errors during processi ng of the FFQs, all forms were checked for completeness and errors according to the recommended quality assurance guidelines (http://www.fhcrc.org/science/shared_resources/nutrition/ffq/qa.html). Energy drinks were marked as regular soda drinks. Gatorade was marked as a regular soda drink.

PAGE 93

93 Nutrient calculations were performed using the NDSR software version 2008, developed by the NCC, University of Minnesota, Minneapolis, MN. Physical Activity Data To assess a participants physical activity level, the long format of the International Physical Activity Questionnaire (IPAQ, version May 2001) was used ( http://www.ipaq.ki.se/ipaq.htm l, Stockholm, Sweden; Appendix B: International Physical Activity Questionnaire ).106 The questionnaire was self administered and asked participants to record their physical activity for the past 7 days before the first study visit. The physical questionnaires were analyzed based upon the Guidelines for Data Processing and Analysis of t he International Physical Activity Questionnaire (Stockholm, Sweden; Appendix B: Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire ).107 Study Protocol Treatment Prior to the study visits, participants abstained from consuming alcohol for at least three days. They also fasted and did not exercise overnight for a minimum of 8 hours before the oral sugar challenges. On the study days, participants reported to the GCRC in the morning. Their height and weight were measured. They were then assigned to a hospital room and allowed to rest for 15 min before measurement of BP and HR. Intravenous catheters were inserted into participants by GCRC nurses. The participants were randomly challenged with cold, 24 oz of carbonated soft drinks sweetened with either sucrose or high fructose corn syr up, which were poured into 4 cups (~6 oz/cup). Participants were given approximately 4 minutes to ingest the sugar load, 1 minute/cup.

PAGE 94

94 Participants were allowed to drink only water during the 6 hr study period. At the end, participants were fed. Metabol ic response phenotypes Data collected at various study visits are shown in Table 31. Height, weight, hip and waist circumferences were collected. The following are the response phenotypes collected: fructose, glucose, insulin, lactate, Tg, BP, HR, serum and urine uric acid, and serum and urine creatinine. Sample collections Table 31 lists the blood and urine collections during the two 6hr study visits. Aliquots of serum and urine samples were stored in O ring cryovials and stored at 80C until analys is. Blood. Whole blood was collected from the participants at baseline and at the following time points: 15 min, 30 min, 60 min, 90 min, 2 hr, 3 hr, 4.5 hr, and 6 hr (Table 3 1). Approximately 185 mL of blood were drawn from each participant per study v isit, overall a total of about 370 mL were collected. Tubes were immediately centrifuged and separated by GCRC technicians. DNA. DNA was isolated from whole blood collected in BD Vacutainer purple top tubes. After centrifuging the tubes at 3000 rpm at 4 C for 10 min, buffy coats were isolated and stored at 80C in 2 mL cryogenic tubes. Blood for DNA were only collected at the baseline of the first study visit. RNA. RNA was isolated from ~6 mL of whole blood. For both study visits, blood were collect ed at pretreatment and at the end of the 6 hr study period in PAXgene Blood RNA tubes (PreAnalytiX, Qiagen Inc, Valencia, CA). The tubes were stored at 20C. RNA was extracted according to manufacturers protocol for PAXgene Blood

PAGE 95

95 RNA Kit (Qiagen). RNA was quantified on the NanoDrop ND1000 Spectrophotometer (Thermo Fisher Scientific Inc., Waltham, MA). 5 l of RNA was diluted in 15 l of 10 mM Tris HCl pH 7.5. 2 l of the mixture was used for the measurements. All samples were measured in duplicat es. Protein. Whole blood (~5 mL) collected at baseline and at the end of the 6 hr study visits in BD Vacutainer purple top tubes containing EDTA were used to extract proteins. The tubes were kept on ice and were immediately processed. Whole blood was diluted with 6 mL of 1X D P BS (Dulbeccos Phosphate Buffered Saline, Cellgro, Mediatech, Inc., Herndon, VA or SigmaAldrich, St. Louis, MO) and was spun down at 800 x g at 22C for 20 min in ACCUSPIN System Histopaque1077 tubes ( Sigma Aldrich). The mononuclear cell layer was isolated and transferred to a 15 mL conical tube and ~9 mL of 1X DPBS was added to the tube, which was then spun down at 800 x g at 22C for 20 min. The supernatant was decanted and ~13 mL of 1X DPBS was added to the mononuclear cells and mixed. The tube was then spun down at 800 x g at 22C for 10 min. Purified mononuclear cells were resuspended in 1.5 to 2.5 mL of 1X DPBS and transferred into four 2 mL cryogenic tubes and stored at 80C until analysis. Results for RNA and protein experiments are contained in Chapter 5. Urine. Two urine samples were collected. One was collected at baseline. The second sample consisted of 6hr urine collections after the consumption of t he soft drinks. The volumes of the urine samples were measured by GCRC technicians. Analytical Methods BMI, hip and waist circumferences The average height and weight of each participant at the two study visits were used to determine BMI.

PAGE 96

96 Blood pressure and heart rate Pretreatment BP and HR were measured at a resting state in triplicates, separated by 5 minutes. Afterwards, single readings of BP and HR were recorded at every collection time points. The BP device used for the measurements was Microlife Model #3AC1 AP ( Minneapolis MN), which has been approved by the British Society of Hypertension.108 Glucose and lactate Whole blood collected in BD Vacutainer gray top tubes containing sodium fluoride and potassium oxalate (BD, Franklin Lakes, NH) were kept on ice and immediately assayed for glucose and lactate by GCRC technicians. Concentrations at all sampling time points were determined by the oxidation principle implemented on the YSI 2300 STAT Plus (YSI Inc, Yellow Springs, OH). Fructose Plasma from whole blood collected in BD Vacutainer green top tubes containing sodium heparin were used to quantify fructose. An LC/MS/MS assay was developed which utilized Alltech Prevail Carbohydrate ES HPLC Column ( 4.6 x 250 mm, Grace Davison, Columbia, MD). A stock solution of internal standard containing 1.1 mM fructose D6 was made in ddH2O (D Fructose, U 13C6, 99%, catalogue #CLM 15530.25, Cambridge Isotope Laboratories, Inc., Andover, MA). Before analyzing the samples on the LC/MS/MS, the plasma samples were protein precipitated. 10 l of internal standard (100 M fructose D6) and 300 l cold acetonitrile were added to 100 l of plasma samples. The mixture was mixed by vortexing at maximum speed for 2 min. The samples were then centrifuged for 10 min at max speed ~14,000 rpm. 150 l of supernatant was added to autosampler vials and ran on the LC/MS/MS.

PAGE 97

97 Standard curve concentrations were 2, 5, 25, 100, 500, 1000, and 1500 M. The QC standard concentrations were 10, 400, and 1250 M. The standards were protein precipitated as the pl asma samples. The assay was linear over the concentration range of 2 to 1500 M The intraand inter day precision (coefficient of variation) and accuracy (relativ e error) values were within 11.4% at low (10 M ), 11.3% at medium (400 M ) and 6.5% at hig h (1250 M ) fructose concentrations (Appendix B: QC Data). T he precision and accuracy at the lower limit of quantitation (2 M ) were 14.1% and 6.1%, respectively. The LC/MS/MS system consisted of a Surveyor HPLC autosampler, Surveyor MS quaternary pump and a TSQ Quantum Discovery triple quadrupole mass spectrometer (Thermo Scientific, San Jose, CA). The TSQ Quantum mass spectrometer was equipped with an electrospray (ESI) ion source and operated in the negative mode. The ESI source parameters were tuned f or maximum abundance of [M H]ions of fructose at the LC flow rate of 1.0 ml/min (25% mobile phase A:75% mobile phase B). Mobile phase A was 1 mM ammonium chloride, pH 9.0. Mobile phase B consisted of 75% ACN and 25% acetone. For quantification, the TS Q Quantum was operated in single reaction monitoring mode (SRM). The acquisition parameters were: spray voltage 3kV, source CID 10V, and heated capillary temperature at 325C. Nitrogen was used as the sheath and auxiliary gas and set to 40 and 15 (arbitr ary units), respectively. The argon collision gas pressure was set to 1.5 mTorr. The collision energy was 15eV for fructose and fructoseD6 (internal standard). The SRM monitoring followed transitions of the [M H]precursor to selected product ions wit h the following values: m/z 215.1 89.0 for fructose and 221.1 95.0 for fructose D6. The peak width (FWHM) was set to m/z 0.7

PAGE 98

98 at both Q1 and Q3. The scan time was 500 ms for each transition. SRM data were acquired and processed using XCalibur software version 1.4, service release 1 (Thermo Scientific, San Jose, CA, USA). Triglycerides, uric acid, creatinine, insulin, and FEUA Whole blood collected in BD Vacutainer SST tubes were used to assay for triglycerides, uric acid, creatinine, and insulin. Triglycerides, serum and urine uric acid, and serum and urine creatinine were measured by the Division of Renal Diseases and Hypertension, University of Colorado Denver. Serum and urine samples were first thawed to 37C in a water bath for 10 minutes followed by vortexing for 15 seconds to ensure all constituents were in solution. Samples were then centrifuged at room temperature at 1000 g for 10 minutes. Sample supernatant were loaded into well cups and analyzed using an ACE clinical analyzer (Alpha W asserman, West Caldwell, NJ). For triglycerides, an enzymatic glycerol phosphate oxidase assay was used with a reportable range of 6 1000 mg/dL. Serum and urine creatinine were determined by the alkaline picrate (Jaffe Reaction) without Lloyds reagent and with a reportable range of 0.2 25 mg/dL. Serum and urine uric acid were analyzed by the uricase method with a reportable range of 0.9 16 mg/dL. An automatic internal dilution was performed by the instrument using system diluent (0.5 mL per liter Triton X100) to allow for analysis of concentrated samples. A system calibration standard and Level 1 and 2 range standards (Alpha Wasserman) were used to calibrate and validate all analyses as recommended by the manufacturer and to conform to CLIA and CAP requirements. For insulin, an ELISA immunoassay was used with a reportable range of 0.798200 IU/mL (ALPCO Diagnostics, Salem, NH). Samples collected at baseline, 30, 60, and 120 min were measured in duplicates. Using Biotek Synergy HT, t he optical

PAGE 99

99 density was measured by spectrophotometer at 450 nm with a reference wavelength of 650 nm. The calculation method used was linear log and the standard curve was plotted using a log/log scale. A zero standard was used as a blank with its average value subtracted from each sample. Fractional excretion of uric acid (FEUA) was calculated from the following equation: where UUA is urine uric acid, SCreat is serum creatinine, SUA is serum uric acid, and UCreat is urine creatinine. All measurement s are in mg/dL.109 Data Analysis All analyses were conducted using SAS 9.2. Statistical significance was defined as p < 0.05. All graphs were generated using SigmaPlot 11.0 (Systat Software, Inc., San Jose, CA). Pharmacokinetic Parameters WinNonlin Professional Edition Version 2.1 (Pharsight Corporation, Mountain View, CA) was used to calculate the following pharmacokinetic parameters: area under the curve (AUC) of plasma concentration versus time, maximum observed concentration (Cmax), half life ( HL), mean residence time (MRT), and time of Cmax (Tmax). Noncompartmental analysis was conducted using linear/log trapezoidal as the calculation method. Relative fructose bioavailability between sucrose and HFCS were calculated using the following equat ions (DS = sucrose dose, DH = HFCS dose, AUCS = area under the

PAGE 100

100 curve from sucrose, AUCH = area under the curve from HFCS ClS = clearance from sucrose, and ClH= clearance from HFCS ): Equation 1: Equation 2: Equation 3: Equation 4: Paired t test was used to compare relative bioavailability. Statistical Methods To compare the effects between HFCS and sucrose treatments, linear mixed effect models were used to analyze the crossover study.110 Fixed effects were study visits and treatment. Random effect was subjects within sequence, which was a variable that accounted for the order participants received the treatments. In addition, metabolic phenotype values measured at pretreatment (0 min) during study visits 1 and 2 were covariates. Pharmacokinetic parameter s of fructose, glucose, insulin, lactate, triglycerides (Tg), systolic blood pressure (SBP), diastolic blood pressure (DBP), HR, and SUA were compared. In addition, changes in FEUA were examined. AUC/D and Cmax/D of fructose and glucose, which were normalized by the average doses of the respective sugars from each treatment, were also compared. To analyze the longitudinal data, repeated measures mixed effect models for a crossover design was used to compare HFCS and sucrose treatments. For these linear m ixed effect models, the random effect was subjects within sequence. A repeated

PAGE 101

101 statement for the covariance structure was used to model for the repeated time and treatment measures. Preplanned contrasts were utilized to compare the treatments at each ti me point. Values measured at 0 min were covariates. Based on the ShapiroWilk test, fructose, glucose, insulin, lactate, and triglyceride were not normally distributed. Analyses for these variables were based on log10 transformed data. Results were reported by back transforming the least square means and 95% confidence interval (CI). Fructose and glucose concentrations were also normalized by their respective doses from each treatment. Concentrations at all time points were normalized except at 0 min; therefore, normalized data were not adjusted for 0 min during analysis. Furthermore, t test was used to compare the maximum changes in the various metabolic phenotypes after drinking the soft drinks, regardless of sweetener. Simple linear models were use d to examine if higher fructose levels predicted higher changes in glucose, insulin, lactate, Tg, SBP, DBP, HR, SUA, and FEUA. Both AUC and Cmax fructose were used as variables to represent the fructose levels in an individual. The AUC of the response phenotypes were examined along with changes in Cmax which equaled the values measured at Tmax minus the values at 0 min. FEUA equaled the values measured at 360 min minu s the values at 0 min. In addition, the impacts of fructose intake on pretreatment concentrations and peak changes in the metabolic phenotypes were examined. The data are shown in Appendix C: Clinical Study Secondary Analyses. Results Table 32 shows the baseline demographics of the 40 subjects who completed both treatments of the crossover study (Table 32). Subjects were randomized to two

PAGE 102

102 different sequences. Subjects randomized to the first sequence received HFCS at study visit 1 and sucrose at study visit 2. Subjects randomized to the second sequence received sucrose at study visit 1 and HFCS at study visit 2. Table 33 lists the values of the metabolic phenotypes measured at 0 min at the two study visits for both of the sequences. The sequence and visit effects were assessed in the linear mixed effect models and were insignific ant factors; the p values were > 0.05 for all of the metabolic phenotypes. From the linear mixed effect model analyses, AUC of fructose was greater from the HFCS treatment than from the sucrose treatment (pvalue <0.0001, Figure 33, Table 34). However, dose adjusted fructose AUC (AUC/D) was not significant (pvalue = 0.1076). This was also seen with Cmax. Fructose Cmax was significantly different between the two treatments (pvalue = 0.0043), but dose normalized fructose Cmax (Cmax/D) was no longer significant (pvalue = 0.8039). Although there was a slight increase, relative fructose bioavailability (FH/FS = 1.08, pvalue 0.0728) between HFCS and sucrose was not significantly different (Table 34). This pattern was the opposite in glucose. AUC (pvalue = 0.6492) and Cmax (pvalue = 0.1778) of glucose between the two treatments were not significant (Figure 34, Table 34). Meanwhile, dose normalized glucose AUC (pvalue <0.0001) and Cmax (pvalue < 0.0001) were significant. In addition, glucose MRT was higher with HFCS (172.5 min) than with sucrose (170.4 min; Table 34). Tg Cmax (HFCS 101.7 mg/dL vs sucrose 108.4 mg/dL) and SBP Cmax (HFCS 133.5 mmHg vs sucrose 130.2 mmHg) were significantly different between the two treatments (Figure 35, Table 34). There

PAGE 103

103 were no treatment differences for insulin, lactate, DBP, HR, SUA, and FEAU (Table 34). From the longitudinal data analyses of the repeated measures, there was a significant treatment effect of HFCS versus sucrose on fructose (pvalue = 0.0033; Figure 36; Table 35) and normalized glucose (pvalue <0.0001; Figure 37; Table 36). Fructose levels tracked higher in the HFCS treatment at 30, 60, and 90 min but became insignificant when the concentrations were normalized against the fructose doses (Figure 36; Table 35). Glucose concentrations were higher during the sucrose treatment at 120 min. However, when the concentrations were normalized against the glucose doses, the differences between the treatments were magnified, resulting in higher c oncentrations from the HFCS treatment at all time points (Figure 37; Table 36). There was a significantly higher effect from HFCS than from sucrose on SUA (pvalue = 0.0030; Figure 38; Table 37). SUA levels tracked higher from HFCS treatment at 60, 90, 120, 180, and then again at 360 min. Although there was not an overall effect of treatment, triglyceride levels were higher in the sucrose treatment at 30 min (Figure 39; Table 38). Meanwhile, the HFCS treatment showed higher increases in SBP at 3 0 and 180 min (Figure 310; Table 39). There were no contrast differences at all time points between HFCS versus sucrose in insulin, lactate, DBP, HR, and FEUA (Figures 311 to 315; Tables 310 to 313 and 34). There were no correlations between AUC or Cmax of fructose with changes in metabolic phenotypes from the HFCS treatment (Figures C1 to C9, Tables C1 to C2). On the other hand after adjusting for covariates, AUC fructose was correlated with

PAGE 104

104 changes in AUC lactate (pvalue = 0.0448) and AUC HR (pvalue = 0.0169) from the sucrose treatment (Table C 1). In addition, Cmax fructose was correlated with changes in Cmax insulin (pvalue = 0.0292), Cmax HR (p value = 0.0093), and Cmax SUA (p value = 0.0172) from the sucrose treatment (Table C 2). From the food data collected, there was variability in the amount of fructose ingested among the participants. From the 3 day food log, the daily intake of fructose was estimated to range from 9 g to 107 g. From the FFQ, the daily intake of fructose ranged from 14 g to 177 g. Fructose intake from the 3 day food log and from the food frequency questionnaire was correlated (R = 0.58073, pvalue <0.0001; Figure 316). Importantly, regardless of treatment, the consumption of 24 oz of soft drinks significantly induced responses in all of the metabolic phenotypes (Table 314). Fructose increased by an average of 335.9 M, gl ucose increased by 41.0 mg/dL, insulin by 51.5 IU/mL, lactate by 1.35 mg/dL, Tg by 18.0 mg/dL, SBP by 12.8 mmHg, DBP by 9.2 mmHg, HR by 8.8 bpm, SUA by 0.47 mg/dL, and FEUA by 2.02%. All pvalues were below 0.0001. Fructose intake, generally, was not str ongly correlated with pretreatment concentrations of the various metabolic phenotypes. Although fructose intake from the food log were correlated with pretreatment concentrations of glucose (R = 0.33339, pvalue = 0.0381) and lactate (R = 0.33262, pvalue = 0.0386) before the HFCS treatment, the correlations were not strong (Figures C10 to C19). In addition, correlations between fructose intake and maximal changes in the metabolic phenotypes were limited (Figures C20 to C29). Fructose intake from the FFQ were associated with Cmax SUA (R = 0.41321, pvalue = 0.0080), while fructose intake from the food log

PAGE 105

105 was correlated with FEUA (R = 0.44012, pvalue = 0.0045). These were only with responses from the sucrose treatment Conclusion We had hypothesized that the formulation of HFCS, which has higher fructose content than sucrose, would result in greater fructose exposure in the body. The results of the study confirmed our hypothesis. The overall AUC was 20% greater in the study participants after they were exposed to HFCS, w hich contained 39.2 g of fructose, than when they were exposed to sucrose, which contained 34.6 g of fructose. Thus, food and drinks sweetened with HFCS could potentially increase an individuals exposure to fructose. We had also hypothesized that due to the potential inhibition of sucrase either at the protein or expression level could result in decrease breakdown of sucrose. Thus, less fructose would be available for transport. Although the relative bioavailability between HFCS and sucrose was higher, it was not significant. Our ability to detect a potential difference in relative bioavailability was probably hindered by the use of soft drinks as the mode of treatment. It was determined from the sugar analyses conducted on the soft drinks that sucrose was broken down. At the start of the study, about 60% of the sucrose had already been hydrolyzed. At the end of the study when the soft drinks were reanalyzed, all of the sucrose had been broken down. As a result, the potential important role of sucras e was marginalized. However, the reason we chose to use soft drinks versus a sugar solution we created, was to increase the external validity of our data. Soft drinks represent a major source of fructose and glucose in the diet.111, 112 Additionally, we detected a significant difference in dose normalized glucose AUC and Cmax. This was surprising since the glucose dose from sucrose (34.8 g) was about

PAGE 106

106 17% higher than from HFCS (28.8 g). This enhanced bioavailabi lity of glucose needs to be further elucidated. We had expected that HFCS, which provides greater levels of fructose, would result in greater acute fructose induced metabolic responses. However, the overall treatment effects on acute metabolic changes were minimal. In contrast to our hypothesis, we detected a greater increase in Tg levels from sucrose than from HFCS. This may be due to the greater demand fructose metabolism has on the ATP pool in the body. Therefore, both glucose and fructose are potentially initially shunted towards replenishing the ATP levels. Since sucrose exposure results in reduced fructose levels, the excess glucose could be shunted towards de novo synthesis of triglycerides.13, 113, 114 Furthermore, there was a significant increase in SBP from HFCS compared to sucrose. Although the acute increase may not be clinically important, about 3 mmHg, the effect between HFCS versus sucrose may be magnified with longterm exposure. Importantly, w e detected a treatment difference in serum uric acid levels, which were greater from the HFCS treatment than from the sucrose treatment. Although the difference was small about 0.2 mg/dL, the results does show that higher SUA levels can potentially develo p from higher levels of fructose in the body. Longterm fructose exposure may lead to greater impact on uric acid levels, thus potentially supporting the hypothesis that fructose can induce hyperuricemia, which has been speculated to be a key mechanism in the pathogenesis of various cardiorenal diseases.2023, 52 We had also hypothesized that greater levels of fructose in the body would predict greater changes in metabolic phenotypes. By using both AUC fructose and Cmax fructose to represent an individuals fructose level, we did not find any strong

PAGE 107

107 correlations. Thus, greater levels of fructose AUC or Cmax were not correlated with increased acute metabolic responses. Speculating that repeated exposure to exc essive fructose ingestion leads to progressive adverse effect s, we used two different methods to estimate chronic fructose consumption. We determined that fructose concentrations measured prior to the treatments was not a good surrogate marker for chronic fructose intake levels. In addition, fructose intake was determined not to be a good predictor of changes in the various metabolic phenotypes. Importantly, we determined that there was a large inter individual variability towards transport and/or metaboli sm of fructose. There was almost a 4fold difference in fructose AUC, ranging from 26560 to 90537 min* M following ingestion of the HFCS formulated Dr. Pepper. Once in the body, fructose was rapidly metabolized with an average half life of about 35 min with inter individual variability that ranged from 24 min to 51 min. Overall the study shows that the consumption of 24 oz of a soft drink can cause significant acute changes in all of the metabolic phenotypes measured. This emphasizes the growing concern over the health hazard of sweetened soft drinks. Recent studies have associated soft drinks with increased risk for a number of health disorders, such as metabolic syndrome41, gout45, and pancreatic disease115. There are several limitations that may have impacted our inability to support our hypotheses. First, as mentioned, the sucrose in the soft drink was already hydrolyzed, which could have minimized our ability to detect treatment effects. For future studies, a more controlled environment can be obtained by having the sugar mixtures made

PAGE 108

108 immediately prior to the study visits. Second, the study population consisted of young and healthy individuals. Their responses may have been less dramatic t han older in dividuals who are metabolically at risk, such as those with abdominal obesity or those with metabolic syndrome. Third, the inter individual variability was large which could have prevented our ability to correlate fructose intake to increased metabolic changes. For instance, although an individual may consume large amounts of fructose, he/she may be protected against its effects by not efficiently absorbing or metabolism the sugar as well as another person. In conclusion, we were able to deter mine that the consumption of HFCS resulted in higher fructose AUC than from sucrose. However, once normalized by dose, the fructose relative bioavailability was similar bet ween the two sugar formulations. There were treatment effects seen with glucose, Tg, SBP, and SUA. Although the treatment effects on acute metabolic phenotypes were small, the effects may increase with continued longterm exposure to these sweeteners. In addition, we were unable to correlate higher fructose levels in the body and higher chronic fructose intake with greater increases in adverse metabolic phenotypes but this could potentially be due to the large inter individual variability detected. The longterm effects of higher fructose AUC on fructose induced metabolic responses, s uch as serum uric acid, are unknown. Further studies are needed to evaluate the impact of variable fructose absorption/metabolism on longterm metabolic responses and disease risks.

PAGE 109

109 Table 31. Study variables and blood and urine collection time point s. Variables Collection Volume Per Time Point/ Blood Tube Screening Visit Study Visits 1 and 2 Times of Collections (min) Baseline 15 30 60 90 120 180 270 360 sex x age x ethnicity x race x waist circumference x hip circumference x height x weight x BP and HR x x B x x x x x x x x urinary pregnancy test x finger stick test: blood glucose x 3 day food log x FFQ x IPAQ x Urine Measurements urinary fructose x x random urine uric acid x x random urine creatinine x x Blood Measurements glucose 4 m L /Gray x x x x x x x x x lactate x x x x x x x x x fructose 6 m L /Green x x x x x x x x x Protein 10 m L /Purple x x RNA 6 m L/PAXgene DNA 1 5 m L /Purple x serum uric acid 8 m L /SST x x x x x x x x x triglyceride x x x x x x x x x serum creatinine x x ACollected only at study visit 1; BMeasured three times at baseline only; FFQ = food frequency questionnaire; IPAQ = international physical activity questionnaire.

PAGE 110

110 Figure 31. Study design.

PAGE 111

111 Figure 32. Study enrollment.

PAGE 112

112 Table 32. Bas eline demographics Subjects (n = 40) Age 27.1 8.6 Female 24 (60.0) Race White, European American 23 (57.5) Black, African American 4 (10.0) Asian 7 (17.5) Other/Multiracial 6 (15.0) BMI 25.9 4.9 Physical Activity: IPA Q (MET minutes/week) 4893.9 4692.7 Fructose: FFQ (g/day) 59.1 38.9 R Fructose: 3DAY (g/day) 43.7 24.3 R 3DAY 3 day food log; BMI body mass index; FFQ food frequency questionnaire; IPAQ international physical activity questionnaire; MET metabolic equivalent of task. Data give as either mean standard deviation or n (%). R fructose intake from FFQ and 3Day were correlated: R = 0.58073, pvalue < 0.0001.

PAGE 113

113 Table 33. Metabolic phenotypes of study visits 1 and 2 measured prior to treatment. V ariable Treatment HFCS Sucrose Sucrose HFCS Visit 1 Visit 2 Visit 1 Visit 2 Fructose ( M) 5.37 3.94 4.39 1.37 5.40 5.00 4.64 2.00 Glucose (mg/dL) 79.67 5.27 81.39 6.51 82.09 4.09 80.33 6.69 Insulin (IU/mL) 10.38 15.70 8.96 13.13 9.21 9.78 9 .62 10.74 Lactate (mg/dL) 0.61 0.18 0.65 0.18 0.76 0.27 0.71 0.29 Tg (mg/dL) 92.22 46.54 79.89 45.68 81.86 32.16 94.57 44.01 SBP (mmHg) 118.13 9.48 119.83 9.50 118.70 10.06 119.14 10.24 DBP (mmHg) 74.69 6.18 74.39 8.49 75.33 6.77 74.90 6.82 HR (beats/min) 66.46 7.30 66.44 7.41 65.92 9.23 66.48 9.31 SUA (mg/dL) 4.91 0.89 4.84 0.79 4.93 1.04 4.97 1.04 FEUA (%) 5.47 1.45 5.10 2.18 5.52 2.34 5.03 2.13 DBP diastolic blood pressure; FEUA f ractional excretion of uric acid; HR heart rate, SBP systolic blood pressure; SUA serum uric acid; Tg triglycerides. Sequence and visit effects were not significant. Data give n as mean standard deviation.

PAGE 114

114 Figure 33. Comparison of fructose AUC and Cmax between HFCS and s ucrose. Values are least squar e means standard errors. A) Fructose AUC. B) AUC data were normalized by the fructose dose of each treatment resulting in AUC/D. C) Fructose Cmax D) Cmax data were normal i zed by the fructos e dose of each treatment resulting in Cmax/D.

PAGE 115

115 Figure 34 Comparison of glucose AUC and Cmax between HFCS and s ucrose. Values are least squar e means standard errors. A) Glucose AUC. B) AUC data were normalized by the glucose dose of each treatment resulting in AUC/D. C) Glucose Cmax D) Cmax data were normal i zed by the glucose dose of each treatment resulting in Cmax/D.

PAGE 116

116 A. B. Figure 35 Comparison of Cmax between HFCS and s ucrose. Values are least square means standard errors. A) Tg Cmax B) SBP Cmax.

PAGE 117

117 Table 34. Effect of HFCS versus sucrose on the response of various metabolic phenotypes. Treatment HFCS Sucrose Variable Phenotype P value Mean 95% CI Mean 95% CI AUC (min*M) Fructose <.0001* 38791.0 1623.8 35533.0 42049.0 32327.0 1614.1 29087.0 35567.0 AUC/D ((min*M)/g) Fructose 0.1076 989.5 43.3 902.6 1076.3 934.4 43.0 848.0 1020.8 Cmax ( M) Fructose 0.0043* 363.4 17.6 328.1 398.8 317.0 17.5 281.9 352.2 Cmax/D (M/g) Fructose 0.8039 9.3 0.5 8.3 10.2 9.2 0.5 8.2 10.1 Half life (min) Fructose 0.2079 35.4 2.2 31.0 39.8 39.3 2.2 34.9 43.7 MRT (min) Fructose 0.3063 87.8 2 83.8 91.8 89.6 2 85.6 93.5 Tmax (min) Fructose 0.6172 57.4 4.2 49.1 65.8 59.7 4.1 51.4 68.0 Relative bioavailabilityR Fructose 0.0728 1.08 0.24 AUC (min*mg/dL) Glucose 0.6492 29911.0 322.5 29266.0 30557.0 30053.0 319.8 29413.0 30694.0 AUC/D ((min*mg/dL)/g) Glucose <.0001* 1038.2 9.9 1018.4 1058.0 863.1 9.8 843.5 882.7 Cmax (mg/dL) Glucose 0.1778 120.3 2.6 115.2 125.4 123.5 2.5 118.4 128.6 Cmax/D ((mg/dL)/g) Glucose <.0001* 4.2 0.1 4.0 4.3 3.5 0.1 3.4 3.7 MRT (min) Glucose 0.0292 172.5 0.9 170.8 174.3 170.4 0.9 168.6 172.1 Tmax (min) Glucose 0.9842 30.1 2.7 24.8 35.4 30.2 2.6 24.9 35.4 AUC (min*UI/mL) Insulin 0. 6846 3717.2 304.4 3106.7 4 327.8 3824.3 302.2 3217.8 4 430.7 Cmax (UI/mL) Insulin 0.5905 58.1 5.8 4 6.4 69.8 61.2 5.8 49.6 72.8 Tmax (min) Insulin 0.1441 35.1 2.0 31.1 39.2 39.0 2 .0 35.0 43.0 AUC (min*mg/dL) Lactate 0.8907 355.4 11.1 333.1 377.8 354.2 11.0 332.1 376.4 Cmax (mg/dL) Lactate 0.4401 2.0 0.1 1.9 2.2 2.1 0.1 1.9 2.2 Tmax (min) Lactate 0. 2627 57.6 3.2 51.4 63.9 52.9 3.1 46.7 59.1 AUC area under the curve; AUC/D AUC divided by the respective sugar dose of the treatment; Cmax maximum observed concentration; Cmax/D Cmax divided by the respective sugar dose of the treatment; MRT mean residence time; Tmax = time of Cmax. DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate, SBP systolic blood pressure; SUA serum uric acid; Tg triglycerides. CI confidence interval. Data given as least square mean stand ard error. Linear mixed effect models were used to analyze the parameters. R FH/FS paired t test was used. pvalue < 0.05.

PAGE 118

118 Table 34. Continued. Treatment HFCS Sucrose Variable Phenotype P value Mean 95% CI Mean 95% CI AUC (min*mg/dL) Tg 0.1949 30661.0 798.2 29069.0 32253.0 31964.0 790.3 30387.0 33540.0 Cmax (mg/dL) Tg 0.0108* 101.7 2.4 97.0 106.5 108.4 2.4 103.7 113.1 Tmax (min) Tg 0.4176 174.4 23.5 127.6 221.3 150.1 23.2 103.7 196.4 AUC (min*mmHg) SBP 0.4802 43832.0 309.0 43214.0 44449.0 43588.0 306.1 42977.0 44200.0 Cmax (mmHg) SBP 0.0047* 133.5 1.0 131.4 135.5 130.2 1.0 128.1 132.2 Tmax (min) SBP 0.5134 118.1 18.9 80.5 155.7 100.6 18.7 63.4 137.8 AUC (min*mmH g) DBP 0.7425 27460.0 263.0 26935.0 27985.0 27363.0 260.4 26842.0 27883.0 Cmax (mmHg) DBP 0.9470 84.1 0.8 82.4 85.7 84.0 0.8 82.4 85.6 Tmax (min) DBP 0.9229 106.4 17.9 70.7 142.0 103.9 17.7 68.6 139.2 AUC (min*bpm) HR 0.7745 23856.0 263.2 23328.0 24384.0 23791.0 261.5 23266.0 24316.0 Cmax (bpm) HR 0.7488 75.3 0.9 73.5 77.1 75.0 0.9 73.2 76.8 Tmax (min) HR 0.1441 143.8 18.7 106.5 181.0 104.7 18.5 67.9 141.5 AUC (min*mg/dL) S UA 0.0827 1848.6 15.1 1818.5 1878.7 1811.3 14.9 1781.5 1841.0 Cmax (mg/dL) SUA 0.4947 5.4 0.1 5.3 5.5 5.4 0.1 5.2 5.5 Tmax (min) SUA 0.9777 83.2 13.0 57.2 109.1 83.7 12.9 58.0 109.4 FEUA (%) FEUA 0.2671 7.6 0.4 6.8 8.3 7.0 0.4 6.2 7.7 AUC area under the curve; AUC/D AUC divided by the respective sugar dose of the treatment; Cmax maximum observed concentration; Cmax/D Cmax divided by the respective sugar dose of the treatment; MRT mean residence time; Tmax = t ime of Cmax. DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate, SBP systolic blood pressure; SUA serum uric acid; Tg triglycerides. CI confidence interval. Data given as least square mean standard error. Linear mixed effect models were used to analyze the parameters. R FH/FS paired t test was used. pvalue < 0.05.

PAGE 119

119 Figure 36. Comparison of fructose concentrati ons over time between HFCS and sucrose. Values are least square means standard errors. p val ue <0.05. A) Fructose concentrations. B) Concentrations were normalized by the fructose dose of each treatment.

PAGE 120

120 Table 3 5. Effect of HFCS versus s ucrose on fructose concentrations over time. Treatment HFCS Sucrose Time P value Mean Fructose (M) 95% CI Mean Fructose (M) 95% CI Treatment 0.0033* 15 min 0.1682 161.1 1.1 141.1 184.0 143.3 1.1 120.9 169.7 30 min 0.0012* 291.5 1.1 255.3 332.8 220.7 1.1 186.0 261.8 60 min 0.0137* 301.8 1.1 264.5 344.2 244.8 1.1 206.6 290.1 90 min 0.0008* 258.8 1.1 226.9 295.2 194.2 1.1 163.7 230.4 120 min 0.1121 159.4 1.1 139.3 182.3 139.1 1.1 117.4 164.8 180 min 0.2829 38.3 1.1 33.6 43.7 35 1.1 29.5 41.4 270 min 0.8829 8.7 1.1 7.6 9. 9 8.5 1.1 7.2 10.1 360 min 0.6531 5.1 1.1 4.5 5.9 5.3 1.1 4.5 6.3 Normalized by doseN: Treatment 0.7686 15 min 0.9437 4.1 1.1 3.6 4.7 4.1 1.1 3.5 4.9 30 min 0.0692 7.4 1.1 6.5 8.5 6.4 1.1 5.4 7.6 60 min 0.3087 7.7 1.1 6.8 8.8 7.1 1.1 6 .0 8.4 90 min 0.0537 6.6 1.1 5.8 7.5 5.6 1.1 4.7 6.7 120 min 0.8763 4.1 1.1 3.6 4.7 4 .0 1.1 3.4 4.8 180 min 0.7022 1 .0 1.1 0.9 1.1 1 .0 1.1 0.9 1.2 270 min 0.1898 0.2 1. 1 0.2 0.3 0.2 1.1 0.2 0.3 360 min 0.0564 0.1 1.1 0.1 0.1 0.2 1.1 0.1 0.2 Data given as least square mean standard error N Concentrations were normalized by the respective sugar dose of the treatment. During analysis, normalized data were not adjusted for values measured prior to the treatments. Repeated measures mixed effect models were used. p value < 0.05.

PAGE 121

121 Figure 37. Comparison of glucose concentrati ons over time between HFCS and sucrose. Values are least square m eans standard errors. p value <0.05. A) Glucose concentrations. B) Concentrations were normalized by the glucose dose of each treatment.

PAGE 122

122 Table 3 6. Effect of HFCS versus s ucrose on glucose concentrations over time. Treatment HFCS Sucrose Time P value Mean Glucose (mg/dL) 95% CI Mean Glucose (mg/dL) 95% CI Treatment 0.3155 15 min 0.4720 107.8 1.0 103.4 112.3 109.8 1.0 104.7 115.1 30 min 0.3368 113.1 1.0 108.5 117.9 115.9 1.0 110.5 121.6 60 min 0.0844 89. 1 1.0 85.6 92.9 93.1 1.0 88.8 97.7 90 min 0.3844 81.3 1.0 78.1 84.7 83.1 1.0 79.2 87.2 120 min 0.0282* 75.8 1.0 72.6 79.0 80.1 1.0 76.4 84.0 180 min 0.3538 75.3 1.0 72.3 78.5 73.6 1.0 70.2 77.2 270 min 0.4459 79. 4 1.0 76.3 82.8 77.9 1.0 74.3 81.7 360 min 0.5445 79.5 1.0 76.3 82.8 78.3 1.0 74.6 82.1 Normalized by dose N : Treatment <.0001* 15 min <.0001* 3.7 1.0 3.6 3.9 3.2 1.0 3 .0 3.3 30 min <.0001* 3.9 1.0 3 .8 4.1 3.3 1.0 3.2 3.5 60 min <.0001* 3.1 1.0 3 .0 3.2 2.7 1.0 2.6 2.8 90 min <.0001* 2.8 1.0 2.7 2.9 2.4 1.0 2.3 2.5 120 min <.0001* 2.6 1.0 2.5 2.7 2.3 1.0 2.2 2.4 180 min <.0001* 2.6 1.0 2.5 2.7 2.1 1.0 2 0 2.2 270 min <.0001* 2.7 1.0 2.6 2.9 2.2 1.0 2.1 2.4 360 min <.0001* 2.8 1.0 2.6 2.9 2.3 1.0 2.2 2.4 Data given as least square mean standard error N Concentrations were normalized by the respective sugar dose of the treatment During analysis, normalized data were not adjusted for values measured prior to the treatments. Repeated measures mixed effect models were used. p value < 0.05.

PAGE 123

123 Figure 38. Comparison of SUA concentrati ons over time between HFCS and s ucrose Values are least square means standard errors. p value <0.05. Table 3 7. Effect of HFCS versus s ucrose on SUA concentrations over time. Treatment HFCS Sucrose Time P value Mean SUA (mg/dL) 95% CI Mean SUA (mg/dL) 95% CI Treatment 0. 0030 15 min 0.7450 5.01 0.05 4.91 5.11 4.99 0.05 4.90 5.09 30 min 0.1897 5.21 0.05 5.11 5.30 5.13 0.05 5.03 5.23 60 min 0.0072* 5.26 0.05 5.17 5.36 5.11 0.05 5.01 5.20 90 min 0.0343 5.21 0.05 5.11 5.30 5.09 0.05 4.99 5.18 120 min 0.0030* 5.22 0.05 5.12 5.31 5.04 0.05 4.95 5.14 180 min 0.0299* 5.16 0.05 5.06 5.25 5.03 0.05 4.93 5.12 270 min 0.3547 5.04 0.05 4.95 5.14 4.99 0.05 4.89 5.08 360 min 0.0264 5.07 0.05 4.98 5 .17 4.94 0.05 4.85 5.04 SUA serum uric acid. CI confidence interval. Data given as least square mean standard error Repeated measures mixed effect models were used. p value < 0.05.

PAGE 124

124 Figure 39. Comparison of Tg concentrati ons over time between HFCS and s ucrose. Values are least square means standard errors. p value <0.05. Table 3 8. Effect of HFCS versus s ucrose on Tg concentrations over time. Treatment HFCS Sucrose Time P value Mean Tg (mg/dL) 95% CI Mean Tg (mg/dL) 9 5% CI Treatment 0.2696 15 min 0.3450 83.3 1.0 78.5 88.3 86.4 1.0 81.3 91.9 30 min 0.0137* 82.8 1.0 78.1 87.8 91.3 1.0 85.8 97.1 60 min 0.3992 78.0 1.0 73.6 82.8 80.6 1.0 75.8 85.7 90 min 0.9695 71.4 1.0 67.4 75.8 71.5 1.0 67.3 76.1 120 min 0.5991 70.7 1.0 66.6 75.0 69.2 1.0 65.1 73.6 180 min 0.7809 76.5 1.0 72.1 81.1 77.3 1.0 72.7 82.2 270 min 0.1732 81.7 1.0 77.1 86.7 86.2 1.0 81.1 91.6 360 min 0.1531 80.2 1.0 75.6 85.0 84.8 1.0 79.7 90.1 Tg triglycerides. CI confidence interval. Data given as least square mean standard error Repeated measures mixed effect models were used. p value < 0.05.

PAGE 125

125 Figure 310. Comparison of SBP lev els over time between HFC S and s ucrose. Values are least square means standard errors. pvalue <0.05. Table 3 9. Effect of HFCS versus s ucrose on SBP levels over time. Treatment HFCS Sucrose Time P value Mean SBP (mmHg) 95% CI Mean SBP (mmHg) 95% CI Treatment 0.2996 15 min 0.2758 124.8 1.4 122.0 127.5 123.0 1.2 120.6 125.5 30 min 0.0403* 125.1 1.4 122.4 127.8 121.8 1.3 119.4 124.3 60 min 0.7536 124.1 1.4 121.4 126.8 123.6 1.2 121.2 126.1 90 min 0.3069 121.4 1.4 118.7 124.1 123.1 1.3 120.6 125.5 120 min 0.1007 122.7 1.4 119.9 125.4 120.1 1.3 117.6 122.5 180 min 0.0190 117.6 1.4 114.9 120.3 121.3 1.2 118.9 123.8 270 min 0.2472 121.3 1.4 118.6 124.0 119.5 1.2 117.1 121.9 360 min 0 .4348 123.2 1.4 120.5 125.9 122.0 1.2 119.6 124.4 SBP systolic blood pressure. C I confidence interval. Data given as least square mean standard error Repeated measures mixed effect models were used. p value < 0.05.

PAGE 126

126 Figure 311. Comparison of insulin concentrations over time between HFCS and sucrose. Values are least square means standard errors. p value <0.05. Tabl e 310. Effect of HFCS versus s ucrose on insulin concentrations over time. Treatment HFCS Sucrose Time P value Mean Insulin (I U /mL) 95% CI Mean Insulin ( I U /mL) 95% CI Treatment 0.7572 30 min 0.0861 49.8 1.1 41.2 60.3 41.8 1.1 34 51.3 60 min 0.8795 29.7 1.1 24.6 35.9 30.0 1.1 24.4 36.8 120 min 0.3302 10.4 1.1 8.6 12.6 11.6 1.1 9.4 14.2 C I confidence interval. Data given as least square mean standard error Repeated measures mixed effect models were used.

PAGE 127

127 Figure 312. Comparison of lactate concentrati ons over time between HFCS and sucrose. Values are least square means standard errors. p value <0.05. Tabl e 311. Effect of HFCS versus s ucrose on lactate concentrations over time. Treatment HFCS Sucrose Time P value Mean Lactate (mg/dL) 95% CI Mean Lactate (mg/dL) 95% CI Treatment 0.7195 15 min 0.6315 1.01 1.04 0.93 1.1 1.04 1.05 0.95 1.13 30 min 0.2366 1.63 1.05 1.50 1.78 1.54 1.05 1.41 1.69 60 min 0.5032 1.84 1.04 1.69 2.00 1.90 1.05 1.74 2.08 90 min 0.3030 1.62 1.04 1.49 1.77 1.54 1.05 1.41 1.69 120 min 0.7702 1.15 1.05 1.06 1.26 1.17 1.05 1.07 1.28 180 min 0.4440 0.68 1.04 0.63 0.74 0.66 1.05 0.60 0.72 270 min 0.6084 0.58 1.04 0.54 0.64 0.60 1.05 0.55 0.65 360 min 0.5755 0.60 1.04 0.55 0.65 0.58 1. 05 0.53 0.63 C I confidence interval. Data given as least square mean standard error Repeated measures mixed effect models were used.

PAGE 128

128 Figure 313. Comparison of DBP lev els over time between HFCS and s ucrose. Values are least square means standard errors. p value <0.05. Tabl e 312. Effect of HFCS versus s ucrose on DBP levels over time. Treatment HFCS Sucrose Time P value Mean DBP (mmHg) 95% CI Mean DBP (mmHg) 95% CI Treatment 0.9388 15 min 0.4263 79 .0 1 .0 77 .0 81 .0 78 .0 1.1 75.9 80.1 30 min 0.6585 78.6 1.0 76.5 80.6 78.0 1.1 75.9 80.1 60 min 0.1416 76.5 1.0 74.5 78.5 78.4 1.1 76.3 80.5 90 min 0.2605 76.4 1.0 74.3 78.4 77.8 1.1 75.7 79.9 120 min 0.0667 77.6 1 .0 75.6 79.7 75.2 1.1 73.1 77.4 180 min 0.1271 74.0 1 .0 72.0 76.0 75.9 1.1 73.8 78.0 270 min 0.2145 75.7 1.0 73.7 77.7 74.1 1.1 72.0 76.2 360 min 0.9766 77.2 1.0 75.2 79.2 77.2 1.1 75.1 79.3 DBP diastolic blood pressure. C I confi dence interval. Data given as least square mean standard error Repeated measures mixed effect models were used.

PAGE 129

129 Figure 314. Comparison of HR lev els over time between HFCS and s ucrose. Values are least square means standard errors. pvalue < 0.05. Table 313. Effect of H FCS versus s ucrose on HR over time. Treatment HFCS Sucrose Time P value Mean HR (bpm) 95% CI Mean HR (bpm) 95% CI Treatment 0.7840 15 min 0.2027 63 .0 1.1 60.8 65.1 64.5 1.1 62.4 66.6 30 min 0.7204 65.3 1.1 63.1 67.4 64.8 1.1 62.7 67.0 60 min 0.1092 67.5 1.1 65.4 69.7 69.5 1.1 67.4 71.6 90 min 0.8105 70.5 1.1 68.4 72.6 70.2 1.1 68.1 72.3 120 min 0.8921 68.6 1.1 66.5 70.8 68.8 1.1 66.6 70.9 180 min 0. 8659 65.2 1.1 63.1 67.3 65.0 1.1 62.9 67.1 270 min 0.9775 64.0 1.1 61.9 66.2 64.1 1.1 62.0 66.2 360 min 0.2487 67.8 1.1 65.7 70.0 66.4 1.1 64.3 68.5 HR heart rate. CI confidence interval. Data given as least square mean standard error R epeated measures mixed effect models were used.

PAGE 130

130 Figure 315. Comparison of FEUA over time between HFCS and s ucrose. Values are least square means standard errors. p value <0.05.

PAGE 131

131 Figure 316. Correlation between fructose i ntake from 3Day Food Log and Food Frequency Questionnaire.

PAGE 132

132 Table 314. Changes in metabolic phenotypes from soft drinks. Phenotype Mean Min Max P Value (M) 335.9 110.6 171.6 747.8 <0.0001* Cmax Glucose (mg/dL) 41.0 15.6 10 80 <0.0001* Cmax Insulin (IU/mL) 51.5 36.5 10.7 204.3 <0.0001* Cmax Lactate (mg/ dL) 1.35 0.50 0.3 2.5 <0.0001* Cmax Tg (mg/dL) 18.0 14.8 0 5 8 <0.0001* Cmax SBP (mmHg) 12.8 6.8 0 34 <0.0001* Cmax DBP (mmHg) 9.2 5.3 0 23,3 <0.0001* Cmax HR (bpm) 8.8 6.0 0 31 <0.0001* Cmax SUA (mg/dL) 0.47 0.36 0 1.6 <0.0001* FEUA (%) 2.02 2.70 6.6 11.31 <0.0001* = differenc e between maximum concentration and pretreatment concentration. FEUA = difference between FEUA at 360 min and 0 min. DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate; SBP systolic blood pressure; SUA serum uric acid; T g triglycerides.

PAGE 133

133 CHAPTER 4 IMPACT OF SUGAR EXPO SURE ON THE TRANSPORT OF FRUCTOSE IN CACO 2 CELLS Introduction Several studies have suggested that fructose bioavailability in humans can be influenced by various factors such as the sugar composition and the total amount of sugar ingested.116 From these studies, it has been shown that the ability to absorb fructose varies widely in adults and children. Using a fructose breathH2 test to semi quantitatively measure for malabsorption, the number of individuals affected increased as the amount of fructose ingested increas ed. About 2/3 of children given 1 to 2 g/kg (fructose/body weight) up to 50 g of fructose were malabsorbers. Meanwhile, about 3050% of adults given 50 g of fructose were estimated to be malabsorbers. Interestingly, if a mixture of fructose and glucose were given, the number of individuals with a positive fructose breathH2 test (> 20 ppm) decreased. Furthermore, individuals given sucrose also improved their intestinal absorption of fructose.117121 A mechanism for this improvement has been proposed and involves the utilization of sodium dependent glucose transporter 1 (SGLT1), which allows glucose to induce the translocation of glucose transporter 2 (GLUT2) to the apical membrane of enterocytes. As a glucose and fructose transporter, GLUT2 increases the uptake of both fructose and glucose.122124 A positive fructose breathH2 test implies that fructose is not being transported efficiently. However, there is a lack of cor relation among individuals who have a positive test and those who suffer from malabsorption symptoms, such as abdominal pain, diarrhea, nausea, and vomiting. Thus, a positive test may not truly represent the malabsorption of fructose. Therefore, the purpose of the study was to characterize the

PAGE 134

134 impact of sugar composition and concentration on fructose and glucose transport, specifically, to investigate 1) the impact of glucose on fructose permeability and 2) the impact of fructose on glucose permeability. Experiments were conducted in Caco 2 cell line. Although it is derived from human colon carcinoma, Caco 2 cells spontaneously develop morphological and functional characteristics of enterocytes, and thus, are commonly used as an in vitro model of the sm all intestine.125, 126 In addition, these cells have been found to express proteins involved in su gar digestion and absorption: sucrase GLUT2, glucose transporter 5 ( GLUT5 ) and SGLT1.127, 128 The physiological levels of sugars in the lumen of the intestine can vary greatly depending on individual diet. T he digestion of dietary sugars can generate very high concentrations of free glucose, approximately 200300 mM in the mic roenvironment surrounding cell junctions and transporters .129 Accordingly, Caco 2 monolayer s were exposed to test media containing various concentrations of fructose and glucose up to a final concentration of 400 mM.15, 16 Materials and Methods Cell Line and Cell Culture Caco 2 cell line was obtained from Dr. Sihong Song at t he University of Florida, Gainesville, FL. The cells were grown in 100 mm or 150 mm cell culture plates (Corning Inc., Corning, NY) in Dulbeccos modified Eagles minimum essential medium (DMEM) containing 25 mM glucose. The medium was supplemented with 20% heat inactivated fetal bovine serum (FB S), 1% nonessential amino acids, 100 U/mL penicillin, and 100 g/mL streptomycin (Appendix D: 20% FBS Growth Media for Cell Growth in Plates ) Cells were grown at 37C under a 5% CO29 5 % air atmosphere.124, 130 Growth media was replaced every 23 days.

PAGE 135

135 Subculturing Cells were subcultured according to the Subculturing Process, Appendix D. Single cell suspensions were obtained with 0.05% trypsin in 0.53 mM EDTA in HBSS with out sodium bicarbonate, calcium, and magnesium. The cells were passaged every 3 5 days when confluency was about 7080%. Transwell Plates Cells between passages 42 to 50 were used in the experiments. By hemacytometer, a cell count was obtained (Fisher Sc ientific Fair Lawn, NJ). For experiments, Caco 2 cells were seeded onto 6Well Transwell culture plates at a density of about 3 X 105 cells per permeable insert (Corning Inc, Corning, NY) .131 The Transwell inse rt was made of polyester membranes and was 24 mm in diameter, with a surface area of 4.67 cm2 and a pore size of 0.4 m. Medium (2 mL in the insert and 3.0 mL in the well) was initially changed five days after seeding and then every 23 days.132 For the first 10 days, cells were grown in high glucose growth medium (Appendix D: High Glucose Growth Media for Cell Growth in Transwell). Afterwards, the culture medium was switched to a low glucose growth medium (Appendix D: Low Glucose Gr owth Media for Cell Growth in Transwell). Integrity of Caco2 Monolayers TEER measurements The transepi thelial electrical resistance (TEER) measurements are frequently used to assess the integrity of monolayers in transport studies.116, 126, 133 TEER values are inversely related to the capacity of the monolayer to support paracellular transport. Caco 2 cell monolayer integrity was measured using an epithelial voltohmmeter used was the EVOM with the STX2 electrode (World Precision Instruments, Inc., Sarasota,

PAGE 136

136 FL). Resistance values were corrected for background (106.33 ) and then divided by two for the effect of using the STX2 probe for 6well inserts. TEER values were determined by multiplying the corrected electrical resistance by the surface area of the insert. After 20 to 22 days of growth, only confluent cell monolayers with TEER values above 600 2 were utilized for experiments .126, 133135 Mannitol transport The paracellular transport of mannitol is another method used to determine the integrity of cell monolayer. Mannitol is nonionizable and impermeable to the cell membrane. Thus, mannitol permeability is low, usually < 0.5% per hour.136138 Sugars D (+) glucose (G5146), D ( ) fructose (F3510), and D mannitol (M9546) were pur chased from Sigma Aldrich Inc, St. Louis, MO. 1 M solutions were filtered and stored at 20C. Transport Experiments The impact of sugar concentration and sugar composition on fructose and glucose transport was investigated. On the day of the experiments the Caco 2 monolayers (Figure 41) were washed three times with HBSS medium. The medium contained no glucose and was used as the transport medium: 137.0 mM NaCl, 5.4 mM KCl, 0.25 mM Na2HPO4, 0.44 mM KH2PO4, 1.3 mM CaCl2, 1.0 mM MgSO4, and 4.2 mM NaHCO3 (Appendix D: HBSS (Without Glucose)). The Transwell plates were then incubated at 37C. After incubating for a minimum of 15 min, TEER values were measured twice for each well. Furthermore, new basolateral plates were filled with 3 mL of transport medium and incubated at 37C. The inserts were then transferred to these plates before conducting the transport studies.

PAGE 137

137 Experimental design The apical surfaces of the Caco 2 monolayers were exposed to the sugar treatments listed in Table 41. For each sugar treatment, sugar solutions were diluted in the transport medium and D mannitol was added to maintain osmolarity (400 mM). The monolayers were then incubated at 37C for 1 hr. To test the integrity of the monolayers after the experiment, the TEER values were measured. Sampling 300 l of transport medium in the basolateral chamber were sampled at the following time points: 0, 2, 5, 10, 30, and 60 min. After each sampling, 300 l of HBSS without glucose was replaced in the basolateral chamber. Measurement s An LC/MS/MS assay was develope d to measure for fructose, glucose, and mannitol using LC/MS/MS. The MS instrument settings and pump settings were the same as reported in Chapter 3: Fructose. In addition to fructose, masses for glucose and mannitol were added and the SRM monitoring followed transitions of the [M H]precursor to selected product ions with the following values: m/z 215.1 89.0 for fructose, 221.1 95.0 for fructose D6, m/z 179.0 179.0 for glucose, 185.0 185.0 for glucose D6, and m/ z 181.0 181.0 for mannitol. A stock solution of internal standards containing 0.6 mM fructose D6 and 6.0 mM glucose D6 were made with HBSS transport medium (Cambridge Isotope Laboratories, Inc., Andover, MA). Transport samples were processed before run ning them on the system. 20 l of internal standards (100 M of fructose D6, 1000 M glucose D6) and 100 l of acetonitrile were vortexed with 100 l transport sample at maximum speed for 2 min. The samples were then centrifuged for 10 min at 4C at 14,000 rpm. 150 l of

PAGE 138

138 supernatant was transferred into autosampler vials and the samples were run on the LC/MS/MS. Standard curve concentrations were 5, 25, 100, 500, 1000, and 1500 M. The QC standard concentrations were 5 0, 400, and 1250 M. The standar ds were processed as the transport samples. For fructose, the assay was linear over the concentration range of 5 to 1500 M The intraand inter day precision (coefficient of variation) and accuracy (relative error) values were within 4.2% and 7.3%, resp ectively, at low (50 M ), medium (400 M ) and high (1250 M ) fructose concentrations. For glucose, assay was linear over the concentration range of 25 to 1500 M The intraand inter day precision and accuracy values were within 7.3% and 5.2%, respective ly, at low (50 M ), medium (400 M ) and high (1250 M ) glucose concentrations. For mannitol, the assay was linear over the concentration range of 5 to 1500 M The intraand inter day precision (coefficient of variation) and accuracy (relative error) val ues were within 10.2% and 13.4%, respectively, at low (50 M ), medium (400 M ) and high (1250 M ) mannitol concentrations. There was not enough data to report the variability at the lower limit of quantitation. Data Analysis Data points were derived from the average of three replicates. Replicates were excluded if the integrity of the monolayers was compromised as determined by the mannitol transport was > 0.5% per hour or the TEER values were below 600 2.135, 138 Analysis of variance (ANOVA) was used to assess the effects of sugar composition and sugar concentration on the permeability of fructose and glucose. Significance was defined as pvalue 0.05. SAS 9.2 was used for all analyses (SAS Institute Inc., Cary,

PAGE 139

139 NC). All graphs were generated using SigmaPlot 11.0 (Systat Software, Inc., San Jose, C A). Permeability The apparent permeability (Papp, cm/s) of fructose and glucose was calculated from the amount of sugar transported per time using the following equation: where dQ/dt is the transport rate or flux ( mol/sec) of the sugar into the basolateral chamber, A is the surface area of the insert (4.67 cm2), and C0 is the initial concentration of the sugar in the donor chamber ( M). The donor concentration for each sampling interval was corrected for the amount removed by transport.137139 The apparent permeation rate (PRapp, nmol/s/cm2) was calculated using the following equation: where J (nmol/s) is the flux of fructose or glucose and A is the surface area of the membrane (cm2).140 Results Evalua tion of Monolayer Integrity In this study, initial TEER values were typically between 666 to 1424 2 and dropped to between 599 to 1132 2 during the 1 hr transport period. Only one well dropped below 600 2 and was consequently excluded from analysis (Table E 1). Mannitol permeability is a more sensitive measurement of cell integrity, thus, five

PAGE 140

140 additional monolayers were excluded based on their high transport of mannitol (> 0.5%/hr) (Table E 1).136, 137 Effect of Sugar Composition and Concentration on Permeability Permeability of fructose over time and the permeability of glucose over ti me are shown in Appendix E: Fructose and Glucose Permeability Data. Due to analytical problems, primarily due to deterioration of sensitivity, the glucose assay was not sensitive enough to measure for all of the glucose treatments. Thus, glucose data was only obtained for the following sugar treatments: 100 mM fructose plus 25, 50, or 75 mM glucose and 100 mM glucose plus 0 mM, 25 mM, 50 mM or 75 mM fructose (treatment #s 24 and 811 in Table 4.1). Permeability of f ructose The apparent permeability of f ructose significantly decreased as the fructose concentration increased from 25 to 300 mM (Figure 42). There was an initial decrease in permeability from 25 to 50 mM, which then stabilized between 50 to 200 mM. At 300 mM, there was a large reduction in apparent permeability of fructose (p < 0.0001, Figure 4 2). The overall decrease in Papp indicates that fructose transporter(s) may be either saturated and/or down regulated. Figure 43 illustrates the corresponding apparent permeation rate of fructose. Between 25 to 200 mM, the PR linearly increased (p <0.0001). Before saturation could occur, the PR decreased at 300 mM, suggesting that the expression of fructose transporter(s) was downregulated. Figure 44 illustrates the effects of glucose on apparent permeability of fructose. Although fructose Papp initially decreased with 25 mM glucose, the apparent permeability, in general, increases as glucose concentration increases. The largest fructose Papp occurred at 200 mM glucose. However, the effect was not significant (p-

PAGE 141

141 value = 0.1278); and thus, the data suggests that glucose does not have an effect on fructose permeability. Permeability of glucose For the limited glucose data collected, the apparent permeability of glucose significantly decreased between 25 mM to 50 mM glucose (pvalue = 0.0271, Figure 45). The observed increase in permeability of glucose at 75 mM was not significant. The corresponding apparent permeation rate shows the potential saturation of a high affinity glucose transporter between 25 to 50 mM (Figure 46). Between 50 to 75 mM glucose, there was an increase in the PR of glucose (p = 0.0408), suggesting that another transporter with a lower affinity becomes involved in glucose transport. The plot of glucose Papp vs fructose co ncentrations is shown in Figure 47. Although there was an initial decrease in Papp from 25 to 50 mM fructose and the direction reversed from 50 to 75 mM, there was no effect of fructose on glucose permeability. Conclusion The in vitro Caco 2 experiment s show that the permeability of fructose varies with sugar composition and concentration. Specifically, we showed that at a constant glucose concentration, fructose permeability decreased with increasing fructose concentrations. Probably due to the limit ed number of data points and limited sampling, saturation was not detected in the study. Importantly, when considering the apparent permeability data along with the permeation rate data, putative fructose transporter(s) may be downregulated at high fruct ose concentrations. From our gene expression study (Chapter 5), SLC2A2 was downregulated after being exposed to 300 mM of glucose, fructose, or fructose:glucose. However, this decrease was initially detected

PAGE 142

142 only after 24 hr of exposure, suggesting that the acute decrease in fructose permeability is most likely due to the inhibition and/or downregulation of protein levels. In addition, if SLC2A2 is the transporter of interest, the decrease in permeabil ity could be due to the translocation of GLUT2 from the apical membranes into the enterocytes.141 When fruc tose concentration was held constant and glucose concentrations varied, we saw a general increase in fructose Papp with increasing glucose concentrations. However, our results were not statistically significant, suggesting that glucose does not influence the permeability of fructose. Further studies are needed to elucidate the impact of glucose on fructose permeability, especially if there are differential effects at lower constant concentrations of fructose. Even though the glucose data were limited, these experiments showed that glucose permeability varies with sugar composition and concentration. The permeability was initially high at the lowest concentration and then decreased as the concentration was titrated up to 50 mM. When coupled with the perme ation rate data, the decrease in glucose permeability may be due to the saturation of a high affinity transporter, most likely SLGT1.142, 143 Between 50 to 75 mM glucose, the permeability and permeation rate increased, indicative of the involvement of another, low affinity glucose transporter, most likely GLUT2.144 In addition to concentration, there may also be a temporal effect since the increase in glucose permeability was seen only after the 30 min time interval (Figure E 3). This may substantiate the recent findings that suggest GLUT2 is translocated from the cytoplasm to the apical membrane of the enterocytes.122, 145 When glucose concentration was held constant and fructose varied, we did not observe an effect of fructose on glucose permeability. Because of our inability to

PAGE 143

143 measure glucose at higher sugar treatments, we were unable to determine the i mpact of fructose on glucose permeability at higher concentrations. Thus, further studies are needed to elucidate the impact of fructose on glucose permeability at higher concentrations and if the effect varies at lower concentrations of glucose. In addi tion, further studies are needed, with greater sampling, to confirm our observations on f ructose and glucose permeability and permeation rates.

PAGE 144

144 Table 4 1. Sugar treatments of Caco 2 cells in transport experiments # Sugar Composition Fructose Glucose D Mannitol Final Concentration (mM) (mM) (mM) (mM) 1 Fructose only 100 300 400 2 + 25% Glucose 100 25 275 400 3 + 50% Glucose 100 50 250 400 4 + 75% Glucose 100 75 225 400 5 + 100% Glucose 100 100 200 400 6 + 200% Glucose 100 200 100 400 7 + 300% Glucose 100 300 400 8 Glucose only 100 300 400 9 + 25% Fructose 25 100 275 400 10 + 50% Fructose 50 100 250 400 11 + 75% Fructose 75 100 225 400 12 + 200% Fructose 2 00 100 100 400 13 + 300% Fructose 3 00 100 400 14 Mannitol 400 400

PAGE 145

145 Figure 41. Caco 2 monolayer on Transwell insert. The magnification was 160X (Axiovert 200 microscope, Carl Zeiss MicroImaging GmbH, Gottingen, Germany).

PAGE 146

146 Figure 42. Ln apparent permeability of fructose vs fructose concentration. Ca co 2 cells treated with 100 mM glucose and increasing concentrations of fructose. All comparisons were against 25 mM fructose. p 0.05; ** p 0.0001.

PAGE 147

147 Figure 43. Apparent permeation rate of fructose vs fructose concentration. Caco 2 cells treated with 100 mM glucose and increasing concentrations of fructose. All comparisons were against 25 mM fructose. p 0.05; ** p 0.0001.

PAGE 148

148 Figure 44. Ln apparent permeability of fructose vs glucose concentration. Caco 2 cells treated with 100 mM fructose and increasing concentrations of glucose. All comparisons were against 0 mM glucose. There was no significant effect of increasing glucose concentration on fructose permeability.

PAGE 149

149 Figure 45. Ln apparent permeability of glucose vs glucose concentration. Caco 2 cells treated with 100 mM fructose and increasing concentrations of glucose. All c omparisons were against 25 mM glucose. p 0.05.

PAGE 150

150 Figure 46. Apparent permeation rate of glucose vs glucose concentration. Caco 2 cells treated with 100 mM fructose and increasing concentrations of glucose. All comparisons were against 25 mM glucose. p 0.05.

PAGE 151

151 Figure 47. Ln apparent permeability of glucose vs fructose concentration. Caco 2 cells treated with 100 mM glucose and increasing concentrations of fructose. All comparisons were against 0 mM fructose. There was no significant effect of increasing fructose concentration on glucose permeability.

PAGE 152

152 CHAPTER 5 IMPACT OF SUGAR EXPO SURE ON THE EXPRESSI ON OF KHK, SI SLC2A2 SLC2A5 AND SLC5A 1 IN CACO 2 CELLS Introduction In addition to sugar composition and the total amounts of sugars consumed, another important factor that can greatly influence fructose bioavailability in humans is the expression of various proteins involved in carbohydrate hydrolysis, transport, and metabolism.116 Several studies have shown that h igh sugar diets can increase the expression of intestinal proteins, such as GLUT2, GLUT5, and sucrase which are involved in sugar absorption.104, 120, 123, 146 Thus as a continuation of the in vitro experiments conducted in Chapter 4, the purpose of this study was to investigate the effects of sugar concentration and sugar composition on the expression of genes involved in sugar absorption ( SI SLC2A2 SLC2A5 and SLC5A1 ) and fructose metabolism ( KHK). Furthermore, the expression of various proteins involved in sugar transport can also be a ffected by short term and longterm exposure to diffe rent sugar compositions thus impacting sugar uptake. Accordingly, Caco 2 cells were exposed up to 3 days with medium containing 5, 25, 100, and 300 mM of the various sugar compositions .143, 147 We hypothesized that the greatest change in gene expression would occur after 3 days exposure to the highest sugar concentration. In addition, RNA and protein samples collected from the fructose clinical study (Chapter 3) were also analyzed for the effects of HFCS and sucr ose on gene expression. We hypothesized that pretreatment expression levels would be correlated with fructose concentrations within the body.

PAGE 153

153 Materials and Methods Refer to the Materials and Methods section of Chapter 4 for information on cell line and c ell culture, subculturing, transwell plates, TEER measurements, and sugars used for these experiments. Expression Experiments The impact of sugar concentration, sugar composition, and duration of sugar exposure on the gene expression of KHK, SI SLC2A2 SL C2A5 and SLC5A1 was investigated. On the day of the experiments, the integrity of Caco 2 monolayers was tested by measuring for TEER values. The apical chamber was rinsed with DMEM containing no glucose ( GIBCO Life Technologies Gaithersburg, MD, USA). Meanwhile, the basolateral chamber was washed with low glucose DMEM (Sigma Aldrich Inc, St. Louis, MO). Afterwards, the Transwell plates were then incubated at 37C. Meanwhile, new basolateral plates were filled with 3 mL of low glucose DMEM (Sigma Aldrich Inc, St. Louis, MO, which represents physiological conditions (5.55 mM or 100 mg/dL glucose), and incubated at 37C for at least 15 min. Before conducting the expression studies, inserts from the Transwell plates were transferred to the basolateral plates. Experimental design To test the integrity of the cell monolayers, TEER values of each well were measured prior to performing the expression experiments. DMEM containing no glucose was used to allow for the manipulation of the sugar concentration and composition. There were four concentrations 5, 25, 100, and 300 mM and three sugar compositions fructose, glucose, and 50:50 fructose:glucose. Mannitol was used to control for osmolarity and to obtain a final concentration of 400 mM for each test

PAGE 154

154 media. Table 51 lists the sugar treatments that were exposed to the apical surfaces of Caco 2 cell monolayers. All treatments were performed in triplicates. There were a total of four sets of plates exposed to the same sugar treatments. The plates were then incubated at 37C. The cells of one set of plates were collected after 1 hr for RNA isolation, the 2nd set of plates were collected after 6 hr, the 3rd set of plates were collected after 24 hr, and the fourth set of plates were collected after 72 hr. For the 72 hr incubation, media was changed daily. To determine if the cell integrity was maintained, TEER values were conducted at 24 hr, 48 hr, and 72 h. Sampling Cells were collected and immediately isolated for RNA after each incubation period: 1, 6, 24, and 72 h. In addition, untreated cells were collected before washing and after washing of the monolayers before start of the expression studies. RNA According to the manufacturers protocol for the RNeasy Mini Kit (Qiagen), 500 l of Buffer RLT was used to lyse the cells directly off the Transwell membrane. A Qiashredder spin column was used to homogenize the lysate. Total RNA was then isolated and purified using the RNeasy spin column. RNA was quantified on the NanoDrop ND1000 Spectrophot ometer (Thermo Fisher Scientific Inc., Waltham, MA). 5 l of RNA was diluted in 15 l of 10 mM Tris HCl pH 7.5. 2 l of the mixture was used for the measurements. All samples were measured in duplicates.

PAGE 155

155 cDNA 500 ng of RNA was used to generate cDNA using the High Capacity cDNA Reverse Transcription Kit (#4368813, Applied Biosystems Inc, Foster City, CA). Using DEPCtreated water (Fisher BioReagents), 50 l reactions were setup. RT PCR TaqMan gene expression assays were used to quantitate mRNA levels. Using TaqMan Universal PCR Master Mix (Applied Biosyst ems Inc.), 2 l of cDNA was used. 20 l reactions in 96well plates were prepared for singleplex reactions. For all samples, duplicate reactions were performed. Probes for KHK (#Hs00240827_m1), SI (#Hs00356112_m1), SLC2A2 (#Hs01096908_m1), SLC2A5 (#Hs00161720_m1), and SLC5A1 (#Hs00165793_m1) were obtained from Applied Biosystems. In addition, the endogenous control was actin ( ACTB, #Hs99999903_m1). The RTPCR reactions were analyzed on ABI 7300 Real Time PCR System (Applied Biosystems). For the fruct ose clinical study samples, only a subset of the total samples were analyzed (11 out of 40 subjects). Because the RNA samples were obtained from whole blood, only KHK, SLC2A2 and SLC2A5 and ACTB probes were used. Protein Protein samples were obtained a t baseline and at the end of both study visits from participants in the fructose clinical study (Chapter 3: Protein). Quantification Protein concentrations were measured in triplicates using the bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL) The diluent used was 1X DPBS. Absorbance was measured at 562 nm on the Biotek Synergy HT (BioTek Instruments, Inc, Winooski, VT). The average measurements of the blank standard were subtracted

PAGE 156

156 from all samples. A standard curve was generated using a nonlinear regression, four parameter (quadratic) algorithm: Curve formula: Antibodies Mouse monoclonal antibody for KHK (#H00003795M01 ~3033 kDa)68, 148 was purchased from A bnova Corporation, Walnut, CA. Mouse monoclonal antibodies were also obtained for GLUT2 (#MAB1414, ~5362 kDa)149, 150 and GLUT5 (#MAB1349, ~4558 kDa)150152 from R&D Systems, Inc., Minneapolis, M N. The antibodies were reconstituted with 1X PBS. The antibodies were further diluted using 2% blocking reagent (Appendix F: Blocking Reagent). For all three antibodies, a final concentration of 5 g/mL was used during the Western blotting. For the endogenous control, polyclonal rabbit antibody for Anti Actin (~42 kDa, #A2066, Sigma) was used and was a gift from Dr. Elizabeth Dudenhausen at the University of Florida, Gainesville, FL.46, 147 The antibody, whic h was at a concentration of 0.2 0.4 g/mL, was diluted 1:3,000 for use in the Western blotting. Horseradish peroxidase (HRP) conjugated rabbit anti mouse IgG was obtained as the secondary antibody for KHK, GLUT2, and GLUT5 (#SG019, Applied Biological Mat erials Inc., Richmond, British Columbia). However, there was extreme cross reactivity. Instead, goat anti mouse IgG HRP conjugated secondary antibody was used. In addition, goat anti rabbit IgG HRP conjugated secondary antibody was used to detect for An ti Actin. These were gifts from Dr. Elizabeth Dudenhausen, which were obtained from KPL (Kirkegaard & Perry Laboratories, Inc., Gaithersburg, MD). For

PAGE 157

157 Western blotting, the secondary antibodies, which were at a stock concentration of 1 mg/mL, were diluted to 2.5 to 5.0 pg/mL Western Blot 24 l of protein samples were mixed with 8 l of 4X Sample Dilution Buffer (Appendix F: 4X Sample Dilution Buffer). After heating the mixture at 95C for 5 min using an Eppendorf Thermomixer (Hauppauge, NY), the tubes were immediately put on ice. 30 l were loaded onto an 18well Criterion T ris HCl Gel (4 15% polyacrylamide, Bio Rad Laboratories, Hercules, CA). BioRad Precision Plus Protein (#1610373) standards were used for the ladder. For gels probing for fruc tokinase, Hep G2 cells, derived from a human liver carcinoma, were used as the positive control.148 The Hep G2 cells were obtained from Dr. Elizabeth Dudenhausen. Caco 2 cells were used as the positive control for probing of GLUT2 and GLUT5 expression. Using 1X Tris/Glycine/SDS buffer (Appendix F: Running Buffer), the gels were run at 200 V for 75 min under ice. Running in 1X Tris/Glycine/Methanol buffer (Appendix F: Transfer Buffer) at 50 V for 1 h under ice, proteins were transferred onto Amersham Hybond ECL nitrocellulose membranes (0.20 m pore size, GE Healthcare BioSciences Corp., Piscataway, NJ). The membranes were rinsed with 1X Tris buffered saline Tween (TBS T, Appendix F: Washing Buffer) and then stained with fast green for visualization (Appendix F: Staining Protocol). Following the instructions for the Amersham ECL Advanced Western Blotting Detection Kit, the membranes were blocked for 1 h at room temperature in 2% blocking reagent. The membranes wer e incubated in diluted primary antibody overnight at 4C on an orbital shaker. After gently washing the membranes, they were incubated in diluted secondary antibodies for 1 h at room temperature on an orbital shaker. After

PAGE 158

158 washing with rapid shaking for 4 x 10 min with 1X TBS T, the membranes were incubated with ECL Advance detection reagent for 5 min at room temperature. Using chemiluminesc ence, the bands were imaged by the Molecular Imager ChemiDoc XRS System (Bio Rad). Data Analysis The foldexpression changes in the genes were calculated using the comparative CT method, which utilizes the following equation: where CT refers to the cyc le number when amplification crosses the threshold.153 The outcome variable used was CT for all RT PCR analyses.154, 155 All analyses were conducted using SAS 9.2 (SAS Institute Inc., Cary, NC). All graphs were generated using SigmaPlot 11.0 (Systat Software, Inc., San Jose, CA). P values were adjusted for multiple comparisons utilizing. Due to multiple comparisons, pvalues were adjusted using a stepdown Bonferroni method, which is less conservative than traditional Bonferroni method and controls for Type I error.156 In Vitro Studies RTPCR To normali ze for total RNA, ACTB was the endogenous control and was used as the active reference. For the in vitro samples, the threshold settings were the following: KHK 0. 398527, SI 0.373658, SLC2A2 0.206691, SLC2A5 0.368589, SLC5A1 0.231042, and ACTB 0. 264073 Relative quantification (RQ) was used to show changes in gene expression in the treated samples relative to the samples treated with 5 mM of glucose, which was used as the reference sample or calibrator for each incubation time set.

PAGE 159

159 The method used to analyze the impact of sugar composition, sugar concentration, and duration of sugar exposure on gene expression was a simple regression model. To compare gene expression between different incubation times, the covariates used were treatment, time and the interaction between treatment and time. To compare gene expression between different sugar concentrations, the covariates used were treatment, concentration, and the interaction between treatment and concentration. Clinical Study RT PCR For the clinical samples, the threshold settings were the following: KHK 0.311936, SLC2A2 0.0820251, SLC2A5 0.258501, and ACTB 0.262955. ACTB was used to normalize for total RNA. The expression levels of the end of each study visit were calibrated agains t the levels before the ingestion of the soft drinks for each subject. The data of one subject was excluded from analysis due to the lack of fasting. To compare the effects between HFCS and sucrose treatments on gene expression, repeated measures mixed effect models for a crossover design were used. The random effect was subjects within sequence or the order participants received the treatments. The CT values were adjusted for baseline CT values of each study visit. In addition, simple linear models were used to examine if higher pretreatment expression levels correlated with 1) pretreatment concentrations of fructose concentrations, 2) Cmax fructose, and 3) AUC fructose.

PAGE 160

160 Results In Vitro Sugar Exposure Experiments Evaluation of monolayer integrity TEER values of every well were measured before performing the expression studies. The initial TEER measurements were 783 79 cm2. Three monolayers had TEER values < 600 2 and were excluded. The TEER values continued to increase after exposure to the sugar treatments.138 TEER values were 1035 151 cm2 at 24 hr, 1230 206 cm2 at 48 hr, and 1358 143 cm2 at 72 hr, indicating that monolayer integrity was maintained for the duration of the experiments.136 Effects of sugar composition, sugar concentration, and duration of sugar exposure on gene expression RT PCR results from the in vitro expression experiments are shown in Appendix G:In Vitro RT PCR Results. The expression patterns varied depending on the gene, sugar composition, sugar concentration, and the duration of sugar exposure. Because of this lack of overall pattern, we narrowed our analysis to examine the effects of the sugar treatments where the data showed the greatest changes in gene expression (Figures G1 to G5). 300 mM glucose consistently decreased the expression of all five genes over time (Table 52, Figures 51 to 55). From 1 to 72 hr, the relative quantity of KHK significantly decreased from 1.37 0.27 to 0.65 0.16 ( p value = 5.63 x 108). SI decreased from 0.91 0.24 to 0.17 0.09 ( p value = 5.65 x 1011). The expression of SLC2A2 reduced from 1.06 0.18 to 0.11 0.06 ( p value = 1.42 x 1013). S LC2A5 was reduced from 1.26 0.29 to 0.67 0.25 ( p value = 1.63 x 104). S LC5A1 was also reduced from 1.31 0.26 to 0.43 0.12 ( p value = 3.20 x 1011).

PAGE 161

161 Over time, the effect of 300 mM fructose on the expression of the five genes were variable (Table 52, Figures 51 to 55). The relative expression of KHK, SLC2A5 and SLC5A1 were unchanged across the various time points at the high fructose concentration. SI slightly decreased from 1 hr to 72 hr but the effect was not significant. The expression of SLC2A2 was significantly decreased from 1.01 0.28 to 0.54 0.03 ( p value = 2.88 x 103). Interestingly, the effect of 150 mM fructose and 150 mM glucose on t he pattern of expression for KHK, SI and SLC2A2 was similar to the effect from 300 mM glucose, while the pattern of expression for SLC2A5 and SLC5A1 was unaffected by this exposure, similar to what was observed for these genes with 300 mM fructose (Table 5 2, Figures 51 to 55). From 1 hr to 72 hr, the relative quantity of KHK significantly reduced from 1.20 0.01 to 0.74 0.09 ( p value = 6.67 x 105). T he expression of SI reduced from 0.93 0.97 to 0.39 0.10 ( p value = 1.27 x 105) and SLC2A2 reduced from 1.32 0.14 to 0.29 0.09 ( p value = 1.51 x 109). There was little change from 1 hr to 72 hr for SLC2A5 and SLC5A1 Comparing the effects of 5 mM vs 300 mM after 72 hr of sugar exposure, the higher glucose dose resulted in significant decreases in expression of KHK, SI SLC2A2 SLC2A5 and SLCA1 (Table 53, Figures 51 to 55). Prolonged exposure led to KHK expression that was reduced by about 35% ( p value = 9.34 x 105), SI was reduced by about 80%, ( p value = 1.62 x 1013), SLC2A2 was decreas ed by about 90% ( p value = 3.54 x 101 7), SLC2A5 was decreased by about 30% ( p value = 2.34 x 103), and SLC5A1 was reduced by about 60% ( p value = 2.07 x 109).

PAGE 162

162 The effect of fructose concentration was more varied (Table 53, Figures 51 to 55). For KHK, there was no difference in the level of expression between 5 mM and 300 mM after 72 hr of incubation. For SI and SLC2A2 the relative quantities of the genes were decreased at the higher dose. SI was reduced by about 35% ( p value = 2.35 x 102) while S LC2A2 was reduced by about 35% ( p value = 4.29 x 103). Although the expression of SLC2A5 increased by about 30% ( p value = 1.09 x 101) and SLC5A1 ( p value = 1.92 x 101).increased by about 20% at the higher fructose dose, the adjusted pvalues were not significant. The sugar composition containing 150 mM fructose and 150 mM glucose was significantly associated with a decrease in expression of KHK, SI and SLC2A2 compared to 5 mM of fructose:glucose. KHK was reduced by about 40% ( p value = 9.34 x 105), SI was reduced by about 65% ( p value = 5.63 x 108), and SLC2A2 was reduced by 1fold ( p value = 2.83 x 1011). Meanwhile, there were no differences in expression between 5 mM and 300 mM fructose:glucose at 72 hr for SLC2A5 and SLC5A1 Clinical Sugar Exposure Experiments Over all, the responses to the treatments were variable (Appendix G: Clinical RTPCR Results). Although there was a slight increase in gene expression of KHK and SLC2A5 from HFCS compared to sucrose, the differences 6 hours after the consumption of the soft drinks were not significant (Figure 56). Interestingly, the data showed that the pretreatment expression levels, especially for KHK, were variable amongst the study participants. Higher pretreatment expression levels of KHK, which a re indicated by lower CT values, were significantly correlated with lower pretreatment

PAGE 163

163 or fasting fructose concentrations (R = 0.50054, pvalue = 0.0246, Figure 57). There was a slight trend with Cmax fructose but not with AUC fructose levels. There were no significant correlations with SLC2A5 pretreatment expression levels. Western Blots Western blots were unsuccessful due to the inefficiency or lack of sensitivity of the antibodies. Conclusion For the in vitro experiments, we had exposed the cell monolayers to four different sugar concentrations (5, 25, 100, and 300 mM) for 1, 24, and 72 hr. Typically physiological concentrations during fasting conditions are approximately 5 mM. However it has been estimated that after a feeding, the glucose concent rations in the intestinal lumen can vary from 50 to 500 mM in rats and 48 to 300 mM in humans.157159 More importantly, it has been estimated that the digestion of dietary sugars can generate high concentrations, approximately 200 to 300 mM, of free glucose in the microenvironment surrounding the membrane transporters.129 Thus, we had hypothesized that the gene expressions of genes involved in fructose absorption/transport and metabolism would be affected the greatest after being exposed at the highest sugar dose for 72 h. The results of the in vitro ex periments showed that the gene expressions of KHK, SI SLC2A2 SLC2A5 and SLC5A1 were affected by sugar composition, sugar concentration, and duration of sugar exposure; however, the overall patterns of expression were complex. Nevertheless, high sugar dose (300 mM), particularly of glucose or fructose:glucose, with longer exposure time (72 hr) affected the gene

PAGE 164

164 expression levels for most genes. However, the long exposure time may be a potential limitation of the study since the physiological relevance i s unclear. From the changes in expression, the data suggest glucose plays an important role in its own absorption. After long exposure to 300 mM glucose, the expression of SLC5A1 (a sodium glucose cotransporter), SLC2A2 (facilitative glucose and fructose transporter), and SI (an enzyme that digests sucrose) were all downregulated, thus, which might be a mechanism potentially preventing the transport of excessive glucose into the cell or body. Importantly, these experiments uniquely demonstrate that high glucose concentrations can potentially interfere in the transport and metabolism of fructose by reducing the expression of both SLC2A5 and KHK, respectively. Previous studies have shown that low concentrations of glucose can upregulate SLC2A5 expression.160, 161 Although we did not analyze the effects in our study, we did detect higher expression of SLC2A5 from 25 mM and 100 mM glucose. However, our data suggests that at higher concentrations, glucose instead tak es on a negative regulatory role. Further studies are needed to elucidate the regulatory role of glucose on the expression of these five genes. In contrast to glucose, fructose does not appear to be as strong a regulator of gene expression for these proteins. Although fructose seems to slightly decrease SI and SLC2A2 expression, fructose, in general, does not appear to downregulate its own transport or metabolism. Both SLC2A5 and KHK expressions were maintained over the 72 hr exposure time even at high doses of fructose. The combination of fructose and glucose seems to take the characteristics of either glucose only or fructose only effects on expression depending on the gene.

PAGE 165

165 Although there is only half the amount of glucose (150 mM), the expression levels of KHK, SI and SLC2A2 were all down regulated similar to the glucose only treatment. Since KHK was not affected by 300 mM fructose, this suggests that the glucose component in the sugar mixture is potentially driving the negative effects. In partic ular, lower concentration of glucose (150 mM) may be enough to induce the downregulation in the three genes. Meanwhile, the expression of SLC2A5 and SLC5A1 were maintained similar to the effects of fructose only treatment. This suggests either the fruct ose component of the sugar mixture is preventing the downregulation by glucose or that the amount of glucose is not enough to reduce the expression of the two genes. Importantly, the results show that the mixture of fructose:glucose can differentially ef fect the expression of the genes compared to glucose only or fructose only. By consuming a diet consisting of fructose:glucose versus glucose only, the overall reduction in the transport of glucose and fructose can potentially be circumvented since SLC5A1 and SLC2A5 expressions were maintained and the reduction in SLC2A2 was not as strong as from glucose only treatment. For the clinical study, the small sample size probably prevented our ability to detect differences in gene expression between HFCS and suc rose. Nevertheless, of potential interest was the variability in the expression of KHK and SLC2A5 amongst the study participants. KHK expression varied from about 1 to 5 fold while SLC2A5 varied from about 1 to 3 fold. From our limited numbers, we were able to detect potential correlations between the KHK expression and fasting fructose levels. Further studies are needed to investigate the potential impact of higher expression of these genes on greater exposure to fructose and its adverse metabolic eff ects.

PAGE 166

166 In conclusion, sugar composition, sugar concentrations, and duration of sugar exposure have varying impacts on the gene expression of KHK, SI SLC2A2 SLC2A5 and SLC5A1. Further studies are needed to explore the longterm impact of high concentrati ons of the sugars on gene expression. Importantly, the role of high dose of glucose on the negative regulation of fructose transport and metabolism needs to be further elucidated.

PAGE 167

167 Table 5 1. Sugar treatments of Caco 2 cells in expression experiments # Sugar Composition Fructose Glucose D Mannitol Final Concentration (mM) (mM) (mM) (mM) 1 Fructose 5 395 400 2 Fructose 25 375 400 3 Fructose 100 300 400 4 Fructose 300 100 400 5 Glucose 5 395 400 6 Glucose 25 375 400 7 Glucose 100 300 400 8 Glucose 300 100 400 9 50:50 Fructose : Glucose 2.5 2.5 395 400 10 50:50 Fructose : Glucose 12.5 12.5 375 400 11 50:50 Fructose : Glucose 50 50 300 400 12 50:50 Fructose : Glucose 150 150 100 400 13 Mannitol 400 400

PAGE 168

168 T able 52. Comparison of gene expression between 1 hr vs 72 hr after exposure to 300 mM of sugar treatments. Gene Treatment RQ1hr SD RQ 72hr SD Change Raw P value Adjusted P value KHK 300 mM Glucose 1.37 0.27 0.65 0.16 2.56 x 10-09 5.63 x 10-08 300 mM Fructose 1.09 0.20 0.93 0.05 7.55 x 10-02 6.80 x 10-01 150 mM Fructose:150 mM Glucose 1.20 0.01 0.74 0.09 3.51 x 10-06 6.67 x 10-05 SI 300 mM Glucose 0.91 0.24 0.17 0.09 2.26 x 10-12 5. 65 x 10-11 300 mM Fructose 0.74 0.11 0.59 0.04 1.14 x 10-01 9.12 x 10-01 150 mM Fructose:150 mM Glucose 0.93 0.07 0.39 0.10 6.34 x 10-07 1.27 x 10-05 SLC2A2 300 mM Glucose 1.06 0.18 0.11 0.06 4.74 x 10-15 1.42 x 10-13 300 mM Fructose 1.01 0.28 0.54 0.03 2.06 x 10-04 2.88 x 10-03 150 mM Fructose:150 mM Glucose 1.32 0.14 0.29 0.09 6.28 x 10-11 1.51 x 10-09 SLC2A5 300 mM Glucose 1.26 0.29 0.67 0.25 1.02 x 10-05 1.63 x 10-04 300 mM Fruct ose 1.15 0.27 1.29 0.03 2.96 x 10-01 1 150 mM Fructose:150 mM Glucose 1.35 0.14 1.54 0.31 3.22 x 10-01 1 SLC5A1 300 mM Glucose 1.31 0.26 0.43 0.12 1.23 x 10-12 3.20 x 10-11 300 mM Fructose 1.28 0.14 1.23 0.07 7.27 x 10-01 1 150 mM Fructose:150 mM Glucose 1.29 0.06 1.14 0.19 1.23 x 10-01 9.12 x 10-01 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation. Change in RQ: = no change, / = < 25%, = 25 < 50%,

PAGE 169

169 Table 53. Comparison of gene expression between 5 mM vs 300 mM after exposure to sugar treatments for 72 hr. Gene Treatment RQ5mM SD RQ300mM SD Change Raw P value Adjuste d P value KHK Glucose 1.00 0.03 0.65 0.16 5.19 x 10-06 9.34 x 10-05 Fructose 0.90 0.03 0.93 0.05 7.05 x 10-01 1 Fructose:Glucose 1.15 0.14 0.74 0.09 5.19 x 10-06 9.34 x 10-05 SI Glucose 1.00 0.04 0.17 0.09 5.75 x 10-15 1.62 x 10-13 Fructose 0.88 0.04 0.59 0.04 1.96 x 10-03 2.35 x 10-02 Fructose:Glucose 1.05 0.08 0.39 0.10 2.68 x 10-09 5.63 x 10-08 SLC2A2 Glucose 1.00 0.08 0.11 0.06 3.54 x 10-17 3.54 x 10-17 Fructose 0.88 0.08 0.54 0.03 3.30 x 10-04 4.29 x 10-03 Fructose:Glucose 1.30 0.22 0.29 0.09 1.05 x 10-12 2.83 x 10-11 SLC2A5 Glucose 1.00 0.01 0.67 0.25 1.56 x 10-04 2.34 x 10-03 Fructose 0.98 0.07 1.29 0.03 9.92 x 10-03 1.09 x 10-01 Fructose:Glucose 1.50 0.24 1.54 0.31 8.25 x 10-01 1 SLC5A1 Glucose 1.00 0.09 0.43 0.12 9.01 x 10-11 2.07 x 10-09 Fructose 1.01 0.13 1.23 0.07 1.92 x 10-02 1.92 x 10-01 Fructose:Glucose 1.19 0.15 1.14 0.19 5.64 x 10-01 1 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation. Change in RQ: = no change, / = < 25%, = 25 < 50%,

PAGE 170

170 Figure 51. Relative quantity (RQ) of KHK expression. A) Comparison between 1 hr vs 72 hr. B) Comparison between 5 mM vs 300 mM. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. P value: = 0.05, ** = 0.001, *** = 0.0001.

PAGE 171

171 Figure 5 2. Relative quantity (RQ) of SI expression. A) Comparison between 1 hr vs 72 hr. B) Comparison between 5 mM vs 300 mM. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. P value: = 0.05, ** = 0.001, *** = 0.0001.

PAGE 172

172 Figure 53. Relative quantity (RQ) of SLC2A2 expression. A) Comparison between 1 hr vs 72 hr. B) Comparison between 5 mM vs 300 mM. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. P value: = 0.05, ** = 0.001, *** = 0.0001.

PAGE 173

173 Figur e 54. Relative quantity (RQ) of SLC2A5 expression. A) Comparison between 1 hr vs 72 hr. B) Comparison between 5 mM vs 300 mM. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. P value: = 0.05, ** = 0.001, *** = 0.0001.

PAGE 174

174 F igure 55. Relative quantity (RQ) of SLC5A1 expression. A) Comparison between 1 hr vs 72 hr. B) Comparison between 5 mM vs 300 mM. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. P value: = 0.05, ** = 0.001, *** = 0.0001.

PAGE 175

175 Figure 56. Comparison of HFCS vs Sucrose on the relative quantity of KHK and SLC2A5 after 6 hr exposure.

PAGE 176

176 Figure 57. Correlation of pretreatment expression levels of KHK with A) pretreatment concentrations of fructose, B) Cmax fructose, and C) AU C fructose.

PAGE 177

177 Figure 58. Correlation of pretreatment expression levels of SLC2A5 with A) pretreatment concentrations of fructose, B) Cmax fructose, and C) AUC fructose.

PAGE 178

178 CHAPTER 6 SUMMARY AND CONCLUSION The purpose of this project was to investigate fact ors that impact fructose bioavailability and its metabolic effects. The objectives were to 1) determine the impacts of genetic polymorphisms of genes involved in fructose absorption ( SI SLC2A2 SLC2A5 ) and metabolism ( KHK) on metabolic phenotypes (Chapter 2), 2) compare the effects of sucrose versus high fructose corn syrup on the pharmacokinetics and pharmacodynamics of fructose (Chapter 3), and 3) characterize the effects of varying sugar composition, sugar concentration, and duration of sugar exposure on fructose permeability (Chapter 4) and gene expression (Chapter 5). From the genetic association study, several significant associations were detected in the study population of PEAR between genetic polymorphi sms of SLC2A2 SLC2A5 and KHK with various m etabolic phenotypes, such as HDL, serum uric acid, triglycerides, and home diastolic and systolic blood pressure. Four SLC2A2 SNPs (rs8192675, rs5398, rs11924032, rs5396) were strongly associated with Tg and HDL in the Caucasian population in the PEAR study. Importantly, rs8192675, which tagged the three other SNPs, was tested for replication in the GERA study. However, an association with only HDL was found. Homozygote carriers were associated with about 8 mg/dL lower HDL level. The potential function of these polymorphisms still needs to be elucidated. The strongest candidate genes were SLC2A2 and SLC2A5 ; both had multiple associations with various aspects of the metabolic syndrome. However, this may be due to greater coverage of these two genes on the chips used to genotype the PEAR study.

PAGE 179

179 From the clinical study, we had hypothesized that higher systemic concentrations of fructose, either from the higher fructose dose or from increased fructose b ioavailability from HFCS versus sucrose, would lead to increased fructose induced adverse metabolic responses. We found that the fructose AUC was 20% greater from the consumption of 24 oz of a soft drink sweetened with high fructose corn syrup than sweetened with sucrose. Although fructose AUC was higher from HFCS versus sucrose, it was not significant. The overall treatment effects on acute metabolic changes were minimal. Nevertheless, we detected a greater increase in serum uric acid from HFCS than from sucrose. Although the difference was small, about 0. 2 mg/dL, the effect could potentially magnify with continued longterm fructose exposure, which could support the hypothesis that fructose can induce hyperuricemia, a key mechanism in the pathogenesis of various cardiorenal diseases.2023, 52 However, the longterm effect of higher fructose systemic concentrations on increased uric acid levels is unknown and further studies are needed to evaluate the impact of variable fructose AUC on long term metabolic responses and disease risks. We had also hypothesized that repeated exposure to excessive fructose consumption would cause increased adverse effects. Using a food log and a food frequency questionnaire to estimate chronic fructose intake, we were unable to det ect any strong correlations between higher fructose intake and increased development of metabolic responses. Interestingly, we discovered a large inter individual variability amongst the study participants towards the fructose. There was almost a four fol d difference in fructose AUC from the HFCS sweetened soft drink In addition, we did detect that once

PAGE 180

180 normalized for the dose, glucose AUC was higher from the HFCS treatment, indicating that glucose bioavailability was probably higher from HFCS than sucrose. The mechanism for the enhanced glucose bioavailability remains unclear and further studies are needed to elucidate the method of action. Nevertheless, the results suggest that f ood and drinks sweetened with HFCS could potentially increase an individu als exposure to fructose. More importantly, the consumption of soft drinks was strongly associated with acute changes in all of the metabolic phenotypes that were measured. From the in vitro Caco 2 transport study, we were able to determine that the perm eability of fructose and glucose could be affected by varying sugar compositions and concentrations. Specifically, fructose permeability decreased with increasing fructose concentrations. Before saturation was detectable, putative fructose transporter(s) were potentially down regulated at high fructose concentrations (300 mM). Meanwhile, glucose permeability initially decreased as glucose concentrations increased from 25 to 50 mM, then increased as glucose concentrations increased from 50 to 75 mM. When considering the permeation rate data, the early decrease was potentially due to the saturation of a high affinity glucose transporter while the increase in permeability was probably due to the involvement of a low affinity glucose transporter. Importantl y, we were unable to detect an effect of increasing glucose concentrations on fructose permeability and also did not detect an effect of fructose on glucose permeability. From the in vitro Caco 2 gene expression study, we determined that sugar composition, sugar concentration, and the duration of sugar exposure impacted the gene expression of KHK, SI SLC2A2 SLC2A5 and SLC5A1 After exposing the cells

PAGE 181

181 to 72 hr of 300 mM glucose, all five genes were downregulated, suggesting that glucose could have an im portant role in self regulating its excessive absorption. On the other hand, fructose appeared to not be a strong regulator of gene expression. In general, fructose did not self regulate its transport or metabolism since both SLC2A5 and KHK expression levels were maintained. The impact of equimolar mixtures of fructose and glucose on gene expression varied. The impact on KHK, SI and SLC2A2 were most similar to glucose only effects, causing significant decreases in gene expression. The effects on SLC2A5 and SLC5A1 were similar to fructose only effects, showing no changes in gene expression levels. By preventing the downregulation of these genes, both fructose and glucose absorption are sustained, potentially circumventing the bodys ability to negativ ely control excessive sugar absorption. Importantly, we showed that fructose could reduce the expression of SLC2A2 The result support s our previous findings from the transport study that showed a decrease in fructose apparent permeability and permeation rate with increasing fructose concentration. However, the decrease in gene expression of SLC2A2 was detectable only after 24 hr of exposure. This delay suggests that proteins are first inhibited, downregulated, and/or translocated and only after prolonged exposure to glucose are the transcript expressions affected. The exact mechanism of this negative regulatory process remains to be elucidated. The expression experiments also support previous studies showing the negative regulatory effect of glucose on SI expression.104, 162, 163 In addition, SI is inhibited by high concentrations of fructose. Thus, the activity and expression of sucrase may be an important factor in the absorption of fructose. Because of the limitation of the clinical

PAGE 182

182 study where sucrose was degraded in the soft drinks, we were unable to truly see a negative effect, which may have also reduced our ability to detect treatment differences between high fructose corn syrup and sucrose. From the clinical study samples, we detected wide variability in the pretreatment expressions of KHK and SLC2A5 Amongst 10 study participants, KHK expression varied from 1to 5 fold while SLC2A5 varied from 1 to 3 fold. Even with such a small sample size, we were able to find a significant correlation between pretreatment levels of KHK and fructose. Further studies are needed to explore the impact of the variability in the expression of these genes on greater fructose exposure and its adverse metabolic effec ts. In conclusion, various factors were detected that could influence an individuals response to fructose. We determined that genetic polymorphisms of genes involved in fructose absorption and metabolism can potentially impact the development of adverse metabolic effects, such as decreased levels of HDL. Interesting SNPs from this study could serve as candidate SNPs in future studies investigating the role of genetic variations of KHK, SI SLC2A2 and SLC2A5 SNPs on various metabolic phenotypes and incr eased health risks. We were unable to find strong correlations between higher levels of chronic fructose intake with added risk towards developing higher adverse metabolic responses. The lack of correlations may have been affected by the wide inter indiv idual variability in fructose AUC. In addition, we determined that the composition of sugar sweeteners, the sugar concentration, and the duration of sugar exposure can greatly affect fructose permeability and also the expression of genes involved in its absorption and metabolism. Further studies are needed to elucidate the

PAGE 183

183 longterm effects of these factors on the inter individual variability towards the development of various metabolic phenotypes and potentially increase a persons risk for various healt h problems, such as cardiovascular disease and metabolic syndrome. Specifically, continued replication of the significant SNPs detected in the discovery cohorts could be undertaken, especially SLC2A5 rs5438 that was associated with serum uric acid, our primary phenotype. Importantly, recent genomewide association studies have associated polymorphisms in another fructose transporter, SLC2A9 with increased uric acid levels and the risk for gout.164, 165 Interest ingly, SLC2A9 was found to have a dual ability; in addition to fructose, the protein can also transport uric acid. Therefore, it may be of interest to span the list of candidate genes to include other fructose transporters, SLC2A7166, 167 and SLC2A11168, 169. Although we had initially included it as a candidate gene, ALDOB was removed due to our interest in fructose absorption. ALDOB may be an important gene in manipulating the effects from fructose exposure. Mutations in these genes have been shown to cause hereditary fructose intolerance.170, 171 Some symptoms are abdominal pain, vomiting, hypoglycemia, and liver failure. Death may also occur if fructose exposure is left unchecked. However, this does indicate that polymorphisms in ALDOB may play a role in the development of adverse metabolic effects. In addition, ALDOB has been linked to hepatocellular carcinoma and type 2 diabetes.172, 173 Thus, ALDOB and other fructose transporters should be considered important genes in future genetic association studies that probe the effects of fructose. Furthermore, with the potential inhibition of sucrase, invitro experiments can be performed to characterize both the transport and expression differences between HFCS

PAGE 184

184 and sucrose. However, to determine the potential physiological impacts, another clinical study may be conducted to examine the longterm effects of interindividual variability on fructose systemic concentrations and the development of fructose induced adverse effects. Importantly, the study can help us determine if fructose AUC can be used as a surrogate marker for the prediction of more severe metabolic imbalances. Our gene expression study showed that glucose may play a role in fructose transport and metabolism. The exposure levels may not be physiological but it does point towards the investigation of the impact of high blood glucose on modulating fructose levels and its effects. It has been shown that diabetics have higher fasting fructose concentrations compared to healthy individuals.50 In addition, individuals with worsening glucos e tolerance showed a greater glycemic response to fructose.174 This could be due to the greater conversion of fructose to glucose in the diabetic state due to a decrease in fructose metabolism. Thus, other tissues and organs can be exposed to higher circulating fructose, which could c ause either global effects or tissue/organspecific effects due to the localization of higher fructose concentrations. Animal studies may need to be performed to elucidate the impact of higher blood glucose on the transport and/or metabolism of fructose and its adverse metabolic effects.

PAGE 185

185 APPENDIX A GENOTYPE ASSOCIATION STUDY DATA Sequencing and Genoty p ing Data Table A 1. List of Coriell genomic DNA samples used to determine allelic frequencies of pfSNPs. Race # Coriell ID Race # Coriell ID African 1 NA17136 European 1 NA05377 2 NA17137 2 NA05380 3 NA17138 3 NA05392 4 NA17144 4 NA05398 5 NA17145 5 NA05408 6 NA17147 6 NA05422 7 NA18516 7 NA06991 8 NA18517 8 NA07019 9 NA19035 9 NA07029 10 NA19036 10 NA07048 11 NA19206 11 NA07348 12 NA19207 12 NA10846 13 NA19238 13 NA10847 14 NA19239 14 NA10851 15 NA19309 15 NA10854 16 NA19310 16 NA10855 17 NA19444 17 NA10856 18 NA19445 18 NA10857 19 NA19700 19 NA10859 20 NA19701 20 NA10865 21 NA19916 21 NA1233 5 22 NA19917 22 NA12336 23 NA20340 23 NA12375 24 NA20341 24 NA12376

PAGE 186

186 Table A 2. Primer design of amplicons for pfSNPs of ALDOB. Amplicon Size (bp) Primers (5' SNP ID rs# Region Chromosomal Position ALDOB_PCR1 540 FP: GGCAAACCAATCTTCCCTTCTATAAGCT 1 rs17551011 5' near gene 103237999 RP: CCAAGTTCATGTGTTTTGTAAGTGTTACAGC ALDOB_PCR2 237 FP: GGCAAACCAATCTTCCCTTCTATAAGCT 2 rs484567 exon 1 103237819 RP: GAAAGGGGAGGGCAAATATTGGAAAA ALDOB_PCR3 578 FP: CAGAAAGTTCATTTGCTTCTCTATGTACAAGACA 3 rs508191 intron 1 103236833 RP: CAGCTTAAGATAGTAGGAATCAAAAGGTTGTGT ALDOB_PCR4 508 FP: ACTCAAATGCCATCTATGAGCACCCTG 4 rs2854707 intron 1 103235436 RP: GCTTGAGATCTGAGAGAGCAGAGGCTT ALDOB_PCR5 324 FP: CTCGTGGAAAAGGATCACACCCC 5 rs41281039 exon 3 103232046 RP: CCTAACTAGCCACCTGAGAGCAACCA ALDOB_PCR6 878 FP: TGGGCAAGTTATGTTACTCTTCGTAGTTCA 6 rs551640 intron 3 103231246 RP: CATTCACACCTCACTTCTGCTTTGTCC 7 rs11460994 intron 3 103231242 ALDOB_PCR7 598 FP: ATTATTTGGATTGGAACTCGGTCAGAGC 8 rs1800546 exon 5 103229677 RP: GGATCAGGTACAAAGGTACAAAGAAGCCT 9 rs17852652 exon 5 103229638 ALDOB_PCR8 1518 FP: TGAATAGAAGCAGATTTCTGGACAAACAGA 10 rs35885472 exon 6 103228739 RP: AAAGGCAGTAGCTAGGTTCTGAGGCAG ALDOB_PCR9 378 FP: TGAATAGAAGCAGATTTCTGGACAAACAGA 11 rs1803086 exon 7 103227709 RP: AACAAATAAGAATTCCTTCATCCTGCCTC 12 rs41281035 exon 7 103227607 FP = forward primer; RP = reverse primer

PAGE 187

187 Table A 2. Continued. Amplicon Size (bp) Primers (5' SNP ID rs# Region Chromosomal Position ALDOB_PCR10 375 FP: TGTTCTTTGGATGAGGAGCCGATATT 13 rs1803088 exon 9 103223938 RP: TATGTTCACACGGGTTCTTCTGGGG FP = forward primer; RP = reverse primer

PAGE 188

188 Table A 3. Sequencing results of pfSNPs of ALDOB. SNP ID rs# African Americans European Americans Test for HWE Test for HWE MA MAF X 2 p value MA MAF X 2 p value 4 rs2854707 A 0.48 0.444 0.505055 A 0.24 0.130 0.717917 6 rs551640 T 0.42 0.353 0.552248 C 0.36 1.998 0.157504 Monomorphic SNPS 2 rs484567 G 1.00 G 1.00 5 rs41281039 T 1.00 T 1.00 7 rs11460994 T 1.00 T 1.00 8 rs1800546 C 1.00 C 1.00 9 rs17852652 C 1.00 C 1.00 10 rs35885472 A 1.00 A 1.00 11 rs1803086 C 1.00 C 1.00 12 rs41281035 C 1.00 C 1.00 13 rs1803088 G 1.00 G 1.00 Failed Sequencing 1 rs17551011 3 rs508 191 HWE = Hardy Weinberg Equilibrium ; MA = minor allele; MAF = minor allele frequency.

PAGE 189

189 Table A 4. Genotype results of PEAR samples using IBC and OPA genotyping chips. # Gene rs# CP Genotype Chip Variant African Americans European Americans Test for HWE Test for HWE MA MAF X 2 pvalue MA MAF X 2 pvalue 1 SLC2A2 rs10513684 172206904 IBC C/T T 0.03 0.206 0.650306 T 0.05 0.628 0.428066 2 rs10513685 172207084 IBC A/G A 0.39 1.543 0.2141 98 A 0.13 2.795 0.094562 3 rs10513688 172209912 OPA A/G A 0.30 0.277 0.598407 A 0.13 5.094 0.024014 4 rs11711206 172215639 OPA A/G G 0.29 0.234 0.628644 G 0.11 3.519 0.060675 5 rs11711437 172209057 IBC C/G C 0.48 3.009 0.082812 G 0.14 1.754 0. 185424 6 rs11924032 172217793 IBC A/G A 0.40 1.358 0.243946 A 0.26 1.598 0.206203 7 rs11924648 172200690 OPA A/G G 0.43 4.737 0.029515 G 0.14 3.719 0.053797 8 rs11925298 P 172219412 IBC, OPA A/G A 0.11 0.472 0.492108 A 0.02 0.090 0.764221 9 rs 12487486 172200825 OPA C/T T 0.39 2.323 0.127488 T 0.14 3.334 0.067865 10 rs12488694 172227055 IBC C/T T 0.05 0.449 0.502609 T 0.02 0.071 0.789757 11 rs16855638 172201058 IBC G/T G 0.10 4.781 0.028776 G 0.00 0.001 0.973813 12 rs3774046 17221969 7 OPA C/T C 0.09 0.597 0.439879 C 0.16 0.127 0.721878 13 rs5393 P 172227614 OPA A/C C 0.49 0.326 0.567752 C 0.13 0.408 0.523048 14 rs5396 P 172227509 IBC A/G A 0.34 0.265 0.606700 G 0.29 0.525 0.468666 15 rs5398 172198524 OPA C/T C 0.39 0.000 0. 986901 T 0.29 2.082 0.149067 16 rs5400 P 172214994 IBC C/T T 0.46 6.581 0.010307 T 0.14 3.334 0.067865 17 rs8192675 172207577 OPA A/G A 0.27 0.526 0.468382 G 0.29 1.861 0.172501 18 rs9828378 172226766 IBC C/G G 0.11 2.360 0.124463 G 0.15 0.0 21 0.885825 CP chromosomal position; IBC ITMAT/Broad/CARE genotyping chip; HWE H ardy Weinberg Equilibrium ; MA minor allele; MAF minor allele frequency; OPA oligo pool all custom SNP genotyping chip. P putatively functional SNP.

PAGE 190

190 Table A 4. Continued. # Gene rs# CP Genotype Chip Variant African Americans European Americans Test for HWE Test for HWE MA MAF X 2 pvalue MA MAF X 2 pvalue 19 SLC2A5 rs1060998 9023066 OPA C/T T 0.16 0.583 0.445194 T 0.27 0.63 9 0.424084 20 rs1063137 P 9019838 OPA C/T C 0.47 1.232 0.267038 T 0.36 0.317 0.573429 21 rs12025713 9017107 OPA C/T C 0.49 0.797 0.371856 C 0.36 1.273 0.259132 22 rs12068539 P 9017816 OPA A/G A 0.49 0.797 0.371856 A 0.36 0.904 0.341624 23 rs120 80175 9039330 OPA C/T C 0.12 0.405 0.524302 C 0.27 0.028 0.867360 24 rs12086036 P 9043612 OPA A/G G 0.28 0.277 0.598524 G 0.28 0.210 0.646890 25 rs12117043 9055107 OPA A/G A 0.15 0.225 0.635513 A 0.32 0.934 0.333818 26 rs12119987 9016477 OPA A/T A 0.49 0.957 0.327965 A 0.35 1.178 0.277816 27 rs12145292 9041183 IBC A/C A 0.29 0.183 0.668727 A 0.26 0.282 0.595509 28 rs12736085 9053237 OPA A/G G 0.32 0.105 0.746362 G 0.27 1.265 0.260680 29 rs1612895 9045144 IBC A/G A 0.29 0.015 0.903998 G 0.44 0.498 0.480198 30 rs1705295 9044969 IBC C/T T 0.01 0.026 0.871974 T 0.11 1.449 0.228708 31 rs1751680 P 9044989 OPA C/T T 0.26 0.876 0.349211 T 0.37 0.011 0.915896 32 rs17842190 9041010 OPA C/T T 0.05 0.419 0.517638 T 0.02 0.071 0.789757 33 rs2478868 P 9049803 OPA G/T C 0.31 0.535 0.464411 C 0.36 0.322 0.570128 34 rs2505972 9039486 IBC A/C A 0.30 1.059 0.303398 C 0.38 1.394 0.237717 35 rs3004245 9034213 OPA A/C A 0.26 0.106 0.744450 C 0.45 1.108 0.292540 37 rs3737661 9030436 IBC A/C A 0.10 0.294 0.587692 A 0.07 0.541 0.461864 38 rs3765962 9050045 IBC A/T T 0.14 0.352 0.553082 T 0.29 0.549 0.458605 39 rs3820034 P 9052454 OPA C/T T 0.03 0.145 0.703845 T 0.20 1.792 0.180672 40 rs4908526 9025544 OPA C/T T 0.48 0.308 0. 579190 T 0.40 0.010 0.921419 41 rs4908809 9038287 OPA C/T C 0.43 0.130 0.718832 C 0.36 0.254 0.614404 42 rs5438 P 9052207 OPA A/G A 0.13 0.104 0.746956 A 0.07 1.195 0.274316 43 rs6680123 9040335 IBC C/T C 0.11 2.297 0.129644 C 0.01 0.040 0.8421 84 44 rs6694527 9041615 IBC A/G A 0.09 1.738 0.187333 G 1.00 45 rs7521322 9036678 IBC A/G G 0.49 0.621 0.430693 G 0.39 0.287 0.592396 46 rs765617 9044369 IBC C/T T 0.44 0.411 0.521696 T 0.33 0.110 0.740031 47 rs770041 P 9052532 IBC A/G A 0.16 0.389 0.532782 A 0.34 2.524 0.112108 CP chromosomal position; IBC ITMAT/Broad/CARE genotyping chip; HWE H ardy Weinberg Equilibrium ; MA minor allele; MAF minor allele frequency; OPA oligo pool all custom SNP genotyping chip. P putatively functional SNP.

PAGE 191

191 Table A 4. Continued. # Gene rs# CP Genotype Chip Variant African Americans European Americans Test for HWE Test for HWE MA MAF X 2 p value MA MAF X 2 p value 48 KHK rs1131375 27176889 IBC C/T T 0.15 0.74 4 0.388535 T 0.36 0.015 0.901449 50 rs7588333 27165775 IBC C/G C 0.09 1.379 0.240258 C 0 0.004 0.947541 Call Rate < 95% EXCLUDED SNPS 49 KHK rs2304681 P 27168756 IBC, OPA A/G A 0.31 0.397 0.528792 A 0.3 6 0.073 0.78721 36 SLC2A5 rs3004249 9043984 OPA A/G A 0.45 0.562 0.453498 G 0.39 5.659 0.017368 MAF < 5% 51 KHK rs36045790 27161663 IBC C/T C 1 T 0.03 0.258 0.611676 52 KHK rs7561117 27165936 IBC A/G A 0.04 0.246 0.61 977 G 1 53 SLC2A2 rs5397 172198529 IBC C/G C 1 G 0 0.001 0.973813 54 SLC2A2 rs7610064 172198030 IBC A/T T 1 A 0.01 0.01 0.919082 Monomorphic SNPS 55 SLC2A2 rs2229608 172198839 IBC C/T T 1 T 1 56 SLC2A2 rs 5395 P 172227514 OPA C/T T 1 T 1 57 SLC2A2 rs5408 P 172199641 OPA C/T T 1 T 1 58 SLC2A5 CT_9011003 9011003 OPA C/T T 1 T 1 CP chromosomal position; IBC ITMAT/Broad/CARE genotyping chip; HWE H ardy Weinberg Equilibrium ; MA minor alle le; MAF minor allele frequency; OPA oligo pool all custom SNP genotyping chip. P putatively functional SNP.

PAGE 192

192 HAPI Heart Data Table A 5. Associations with metabolic phenotypes in HAPI Heart. Genotype HCA HET HMA Gene Marker Study Trait Race Model n (%) Mean n (%) Mean n (%) Mean P value (mg/dL) (mg/dL) (mg/dL) SI rs12186083 HAPI UA EA ADD 0.0328 E 454 (53.7) 4.08 0.0 335 (39.6) 4.22 0.0 56 (6.6) 4.32 0.1 SI rs12496714 HAPI UA EA ADD 0.0349 E 462 (54.3) 4.09 0.0 332 (39.0) 4.22 0.0 57 (6.7) 4.32 0.1 SI rs9714197 HAPI UA EA ADD 0.0343 E 453 (53.5) 4.09 0.0 336 (39.7) 4.22 0.0 57 (6.7) 4.32 0.1 SI rs9812124 HAPI UA EA ADD 0.0340 E 462 (54.2) 4.09 0.0 334 (39.2) 4.22 0.0 57 (6.7) 4.32 0.1 SI rs 9856642 HAPI UA EA ADD 0.0285 E 457 (53.8) 4.09 0.0 335 (39.5) 4.22 0.0 57 (6.7) 4.32 0.1 SLC2A5 rs10864381 HAPI Tg EA ADD 0.0316 E 410 (50.4) 72.4 1.1 330 (40.6) 65.9 1.2 73 (9.0) 62.1 2.1 SLC2A5 rs1612895 HAPI Tg EA ADD 0.0313 E 386 (46.0) 72.1 1.1 363 (43.3) 65.6 1.1 90 (19.7) 63.8 2.0 SLC2A5 rs4908526 HAPI Tg EA ADD 0.0175 E 400 (46.7) 72.2 1.1 365 (42.6) 65.6 1.1 91 (10.6) 63.4 2.0 SLC2A5 rs6667506 HAPI Tg EA ADD 0.0383 E 429 (50.1) 71.5 1.1 341 (39.8) 65.9 1.1 86 ( 10.0) 63.1 2.0 E adjusted for age, age2, and gender. HAPI Heredit ary and Phenotype Intervention Heart EA European American. Tg triglycerides; UA uric acid. ADD additive. HCA homozygote common allele; HET heterozygote; HMA homozygote minor allele. Adjusted data are given as mean standard error. single nucleotide polymorphism with an adjusted p value

PAGE 193

193 APPENDIX B CLINICAL STUDY DOCUM ENTS Volunteer Health Information Questionnaire

PAGE 194

194

PAGE 195

195

PAGE 196

196 Sugar Analyses April 05, 2008

PAGE 197

197

PAGE 198

198 May 12, 2009

PAGE 199

199

PAGE 200

200 ThreeDay Food Record

PAGE 201

201

PAGE 202

202 Serving Size Booklet

PAGE 203

203

PAGE 204

204

PAGE 205

205

PAGE 206

206

PAGE 207

207

PAGE 208

208 Food Frequency Questionnaire

PAGE 209

209

PAGE 210

210

PAGE 211

211

PAGE 212

212

PAGE 213

213

PAGE 214

214 International Physical Activity Questionnaire

PAGE 215

215

PAGE 216

216

PAGE 217

217

PAGE 218

218 Guidelines for Data Processing and Analysis of the International Physical Act ivity Questionnaire

PAGE 219

219

PAGE 220

220

PAGE 221

221

PAGE 222

222

PAGE 223

223

PAGE 224

224

PAGE 225

225

PAGE 226

226 QC Data Table B 1. Precision (R.S.D%) and accuracy (R.E.%) for plasma fructose (eight replicates per day for 3 days). R.S.D. (%) a R.E. (%) Nominal Measured (mean) Intra day Inter day 10 9.54 11.4 11.4 4.6 400 399.48 10.9 11.3 0.1 1250 1328.75 6.4 6.5 6.3 a Estimated using oneway ANOVA.175

PAGE 227

227 A P PENDIX C CLINICAL STUDY SECONDARY ANALYSES Correlation of AUC and Cmax Fructose with Changes in Metabolic Phenotypes Figure C 1. Correlation of fructose levels with changes in glucose. A) AUC fructose with AUC glucose from HFCS. B) AUC fructose with AUC glucose from sucrose. C) Cmax fructose with Cmax glucose from HFCS. D) Cmax fructose with Cmax glucose from sucrose.

PAGE 228

228 Figure C 2. Correlation of fructose levels with changes in insulin. A) AUC fructose with AUC insulin from HFCS. B) AUC fructose with AUC insulin from sucrose. C) Cmax fructos e with Cmax insulin from HFCS. D) Cmax fructose with Cmax insulin from sucrose.

PAGE 229

229 Figure C 3. Correlation of fructose levels with changes in lactate. A) AUC fructose with AUC lactate from HFCS. B) AUC fructose with AUC lactate from sucrose. C) Cmax fructose with Cmax lactate from HFCS. D) Cmax fructose with Cmax lactate from sucrose.

PAGE 230

230 Figure C 4. Correlation of fructose levels with changes in Tg. A) AUC fructose with AUC Tg from HFCS. B) AUC fructose with AUC Tg from sucrose. C) Cmax fructose with Cmax Tg from HFCS. D) Cmax fructose with Cmax Tg from sucrose.

PAGE 231

231 Figure C 5. Correlation of fructose levels with changes in SBP. A) AUC fructose with AUC SBP from HFCS. B) AUC fructose with AUC SBP from sucrose. C) Cmax fructose with Cmax SBP from HFCS. D) Cmax fructose with Cmax SBP from sucrose.

PAGE 232

232 Figure C 6. Correlation of fructose levels with changes in DBP. A) AUC fructose with AUC DBP from HFCS. B) AUC fructose with AUC DBP from sucrose. C) Cmax fructose with Cmax DBP from HFCS. D) Cmax fructose with Cmax DBP from sucrose.

PAGE 233

233 Figure C 7. Correlation of fructose levels with changes in HR. A) AUC fructose with AUC HR from HFCS. B) AUC fructose with AUC HR from sucrose. C) Cmax fructose with Cmax HR from HFCS. D) Cmax fructose with Cmax HR from sucrose.

PAGE 234

234 Figure C 8. Correlation of fructose levels with changes in SUA. A) AUC fructose with AUC SUA from HFCS. B) AUC fructose with AUC SUA from sucrose. C) Cmax fructose with Cmax SUA from HFCS. D) Cmax fructose with Cmax SUA from sucrose.

PAGE 235

235 Figure C 9. Correlation of fructose levels with changes in FEUA. A) AUC fructose with AUC FEUA from HFCS. B) AUC fructose with AUC FEUA from sucrose. C) Cmax fructose with FEUA from HFCS. D) Cmax fructose with FEUA from sucrose.

PAGE 236

236 Table C1. Correlation of AUC fructose with changes in metabolic phenotypes. Variables Treatment Adjustments R P Value AUC Glucose HFCS Unadjusted 0.16257 0.3228 AUC Glucose HFCS Adjusted1 0.36123 0.5609 AUC Glucose Sucrose Unadjusted 0.05183 0.7508 A UC Glucose Sucrose Adjusted1 0.17407 0.8768 AUC Insulin HFCS Unadjusted 0.06943 0.6787 AUC Insulin HFCS Adjusted1 0.5774 5 0.2319 AUC Insulin Sucrose Unadjusted 0.00263 0.9873 AUC Insulin Sucrose Adjusted1 0.3413 4 0.7068 AUC Lactate HFCS Unadjusted 0 .15415 0.3488 AUC Lactate HFCS Adjusted1 0.37174 0.2411 AUC Lactate Sucrose Unadjusted 0.41446 0.0078* AUC Lactate Sucrose Adjusted1 0.3249 6 0.0448* AUC Tg HFCS Unadjusted 0.07455 0.6520 AUC Tg HFCS Adjusted1 0.90865 0.3403 AUC Tg Sucrose Unadjuste d 0.12339 0.4481 AUC Tg Sucrose Adjusted1 0.9105 3 0.7562 AUC SBP HFCS Unadjusted 0.09208 0.5772 AUC SBP HFCS Adjusted1 0.6266 6 0.6793 AUC SBP Sucrose Unadjusted 0.14219 0.3815 AUC SBP Sucrose Adjusted1 0.61881 0.7927 AUC DBP HFCS Unadjusted 0.02134 0.8974 AUC DBP HFCS Adjusted1 0.5729 8 0.9145 AUC DBP Sucrose Unadjusted 0.28945 0.0701 AUC DBP Sucrose Adjusted1 0.68046 0.7019 AUC HR HFCS Unadjusted 0.03111 0.8509 AUC HR HFCS Adjusted1 0.67293 0.9870 AUC HR Sucrose Unadjusted 0.22389 0.1649 AUC HR Sucrose Adjusted1 0.57697 0.0169* AUC SUA HFCS Unadjusted 0.31388 0.0517 AUC SUA HFCS Adjusted1 0.95206 0.3334 AUC SUA Sucrose Unadjusted 0.07989 0.6241 AUC SUA Sucrose Adjusted1 0.9266 4 0.3688 1 adjusted for pretreatment values fructose intake from 3 day food log, physical activity and sequence of study visit and treatment AUC area under the curve; DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate; SBP systolic blood pressure; SUA serum uric acid; Tg triglyc erides. p value < 0.05.

PAGE 237

237 Table C 1. Continued. Variables Treatment Adjustments R P Value FEUA HFCS Unadjusted 0.04133 0.8027 FEUA HFCS Adjusted1 0.1833 9 0.7410 FEUA Sucrose Unadjusted 0.24498 0.1276 Sucrose Adjusted1 0.2437 2 0.1388 1 adjusted for pretreatment values fructose intake from 3 day food log, physical activity and sequence of study visit and treatment AUC area under the curve; DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate; SBP systolic blood pressure; SUA ser um uric acid; Tg triglycerides. FEUA = difference between FEUA at 360 m in and 0 min. p value < 0.05.

PAGE 238

238 Table C 2. Correlation of Cmax fructose with changes in metabolic phenotypes. Variables Treatment Adjustments R P Value Cmax Glucose HFCS Unadjusted 0.07643 0.6437 Cmax Glucose HFCS Adjusted1 0.01619 0.6806 Cmax G lucose Sucrose Unadjusted 0.25268 0.1157 Sucrose Adjusted1 0.12251 0.1891 Cmax Insulin HFCS Unadjusted 0.18142 0.2757 Cmax Insulin HFCS Adjusted1 0.27572 0.1942 Cmax Insulin Sucrose Unadjusted 0.2991 0 0.0644 Cmax Insulin Sucrose Adju sted1 0.23618 0.0292* Cmax Lactate HFCS Unadjusted 0.15877 0.3343 Cmax Lactate HFCS Adjusted1 0.08662 0.2916 Cmax Lactate Sucrose Unadjusted 0.15203 0.3490 Cmax Lactate Sucrose Adjusted1 0.04575 0.3471 Cmax Tg HFCS Unadjusted 0.19634 0.2309 Cm ax Tg HFCS Adjusted1 0.08238 0.1987 Cmax Tg Sucrose Unadjusted 0.07386 0.6506 Cmax Tg Sucrose Adjusted1 0.05024 0.5707 Cmax SBP HFCS Unadjusted 0.06182 0.7085 HFCS Adjusted1 0.14757 0.8732 Cmax SBP Sucrose Unadjusted 0.05871 0.7190 Cmax SBP Sucrose Adjusted1 0.02063 0.7059 Cmax DBP HFCS Unadjusted 0.10525 0.5237 Cmax DBP HFCS Adjusted1 0.13934 0.6472 Cmax DBP Sucrose Unadjusted 0.07443 0.6481 Cmax DBP Sucrose Adjusted1 0.02341 0.6518 Cmax HR HFCS Unadjusted 0.01589 0.9235 Cmax HR HFCS Adjusted1 0.03460 0.8974 Sucrose Unadjusted 0.39894 0.0108* Cmax HR Sucrose Adjusted1 0.18472 0.0093* HFCS Unadjusted 0.02108 0.8987 Cmax SUA HFCS Adjusted1 0.08469 0.8101 Cmax SUA Sucrose Unadjusted 0.37752 0 .0163* Sucrose Adjusted1 0.20857 0.0172* 1 adjusted for values at pretreatment fructose intake from 3day food log, physical activity and sequence of study visit and treatment DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate; SBP systolic blood pressure; SUA ser um uric acid; Tg triglycerides. = difference between maximum concentration and pretreatment concentration. FEUA = difference between FEUA at 360 min and 0 min. p value < 0.05.

PAGE 239

239 Table C 2. Continued. Variables Treatment Adjustments R P Value FEUA HFCS Unadjusted 0.11589 0.4823 FEUA HFCS Adjusted1 0.19416 0.4567 FEUA Sucrose Unadjusted 0.02815 0.8631 FEUA Sucrose Adjusted1 0.20396 0.5153 1 adjusted for values at pretreatment fru ctose intake from 3day food log, physical activity and sequence of study visit and treatment DBP diastolic blood pressure; FEUA fractional excretion of uric acid; HR heart rate; SBP systolic blood pressure; SUA serum uric acid; Tg triglycerides = difference between maximum concentration and pretreatment concentration. FEUA = difference between FEUA at 360 min and 0 min. p value < 0.05.

PAGE 240

240 Correlation between Fructose Intake and Pretreatment (0 min) Levels of Metabolic Phenotypes Figure C 10 Correlation between fructose intake and pretreatment fructose levels A) Fructose intake from FFQ with fructose levels before HFCS treatment. B) Fructose intake from FFQ with fructose levels before sucrose treatment. C) Fructose intake from 3 day food log with fructose levels before HFCS treatment. D) Fructose intake from 3day food log with fructose levels before sucrose treatment.

PAGE 241

241 Figure C 11 Correlation between fructose intake and pretreatment glucose levels A) Fructose intake from FFQ with glucose levels before HFCS treatment. B) Fructose intake from FFQ with glucose levels before sucrose treatment. C) Fructose intake from 3 day food log with glucose levels before HFCS treatment. D) Fructose intake from 3day food log with glucose level s before sucrose treatment.

PAGE 242

242 Figure C 12 Correlation between fructose intake and pretreatment insulin levels A) Fructose intake from FFQ with insulin levels before HFCS treatment. B) Fructose intake from FFQ with insulin levels before sucrose treat ment. C) Fructose intake from 3day food log with insulin levels before HFCS treatment. D) Fructose intake from 3day food log with insulin levels before sucrose treatment.

PAGE 243

243 Figure C 13 Correlation between fructose intake and pretreatment lactate le vels A) Fructose intake from FFQ with lactate levels before HFCS treatment. B) Fructose intake from FFQ with lactate levels before sucrose treatment. C) Fructose intake from 3 day food log with lactate levels before HFCS treatment. D) Fructose intake from 3 day food log with lactate levels before sucrose treatment.

PAGE 244

244 Figure C 14 Correlation between fructose intake and pretreatment Tg levels A) Fructose intake from FFQ with Tg levels before HFCS treatment. B) Fructose intake from FFQ with Tg leve ls before sucrose treatment. C) Fructose intake from 3day food log with Tg levels before HFCS treatment. D) Fructose intake from 3day food log with Tg levels before sucrose treatment.

PAGE 245

245 Figure C 15 Correlation between fructose intake and pretreatme nt SBP levels A) Fructose intake from FFQ with SBP levels before HFCS treatment. B) Fructose intake from FFQ with SBP levels before sucrose treatment. C) Fructose intake from 3 day food log with SBP levels before HFCS treatment. D) Fructose intake fro m 3 day food log with SBP levels before sucrose treatment.

PAGE 246

246 Figure C 16 Correlation between fructose intake and pretreatment DBP levels A) Fructose intake from FFQ with DBP levels before HFCS treatment. B) Fructose intake from FFQ with DBP levels b efore sucrose treatment. C) Fructose intake from 3 day food log with DBP levels before HFCS treatment. D) Fructose intake from 3day food log with DBP levels before sucrose treatment.

PAGE 247

247 Figure C 17 Correlation between fructose intake and pretreatment HR levels A) Fructose intake from FFQ with HR levels before HFCS treatment. B) Fructose intake from FFQ with HR levels before sucrose treatment. C) Fructose intake from 3 day food log with HR levels before HFCS treatment. D) Fructose intake from 3day food log with HR levels before sucrose treatment.

PAGE 248

248 Figure C 18 Correlation between fructose intake and pretreatment SUA levels A) Fructose intake from FFQ with SUA levels before HFCS treatment. B) Fructose intake from FFQ with SUA levels before s ucrose treatment. C) Fructose intake from 3 day food log with SUA levels before HFCS treatment. D) Fructose intake from 3day food log with SUA levels before sucrose treatment.

PAGE 249

249 Figure C 19 Correlation between fructose intake and pretreatment FEUA l evels A) Fructose intake from FFQ with FEUA levels before HFCS treatment. B) Fructose intake from FFQ with FEUA levels before sucrose treatment. C) Fructose intake from 3day food log with FEUA levels before HFCS treatment. D) Fructose intake from 3d ay food log with FEUA levels before sucrose treatment.

PAGE 250

250 Impact of Fructose Intake on Changes in Metabolic Phenotypes Figure C 20. Correlation of fructose intake with changes in fructose. A) Fructose intake from FFQ with Cmax fructose from HFCS. B) Fructose intake from FFQ with Cmax fructose from sucrose. C) Fructose intake from 3 day food log with Cmax fructose from HFCS. D) Fructose intake from 3 day food log with Cmax fructose from sucrose.

PAGE 251

251 Figure C 21. Corr elation of fructose intake with changes in glucose. A) Fructose intake from FFQ with Cmax glucose from HFCS. B) Fructose intake from FFQ with Cmax glucose from sucrose. C) Fructose intake from 3day food log with Cmax glucose from HFCS. D) Fructose intake from 3day food log with Cmax glucose from sucrose.

PAGE 252

252 Figure C 22. Correlation of fructose intake with changes in insulin. A) Fructose intake from FFQ with Cmax insulin from HFCS. B) Fructose intake from FFQ with Cmax insulin from sucrose. C) Fructose intake from 3day food log with Cmax insulin from HFCS. D) Fructose intake from 3 day food log with Cmax insulin from sucrose.

PAGE 253

253 Figure C 23. Correlation of fructose intake with changes in lactate. A) Fructose intake from FFQ with Cmax lactate from HFCS. B) Fructose intake from FFQ with Cmax lactate from sucrose. C) Fructose intake from 3 day food log with Cmax lactate from HFCS. D) Fructose intake from 3 day food log with Cmax lactate from sucrose.

PAGE 254

254 Figure C 24. Correlation of fructose intake with changes in Tg. A) Fructose intake from FFQ with Cmax Tg from HFCS. B) Fructose intake from FFQ with Cmax Tg from sucrose. C) Fructose intake from 3day food log with Cmax Tg from HFCS. D) Fructose intake from 3day food log wit h Cmax Tg from sucrose.

PAGE 255

255 Figure C 25. Correlation of fructose intake with changes in SBP. A) Fructose intake from FFQ with Cmax SBP from HFCS. B) Fructose intake from FFQ with Cmax SBP from sucrose. C) Fructose intake from 3day food log with Cm ax SBP from HFCS. D) Fructose intake from 3day food log with Cmax SBP from sucrose.

PAGE 256

256 Figure C 26. Correlation of fructose intake with changes in DBP. A) Fructose intake from FFQ with Cmax DBP from HFCS. B) Fructose intake from FFQ with Cmax DBP from sucrose. C) Fructose intake from 3day food log with Cmax DBP from HFCS. D) Fructose intake from 3day food log with Cmax DBP from sucrose.

PAGE 257

257 Figure C 27. Correlation of fructose intake with changes in HR. A) Fructose intake from FFQ with Cma x HR from HFCS. B) Fructose intake from FFQ with Cmax HR from sucrose. C) Fructose intake from 3day food log with Cmax HR from HFCS. D) Fructose intake from 3 day food log with Cmax HR from sucrose.

PAGE 258

258 Figure C 28. Correlation of fructose intake w ith changes in SUA. A) Fructose intake from FFQ with Cmax SUA from HFCS. B) Fructose intake from FFQ with Cmax SUA from sucrose. C) Fructose intake from 3day food log with Cmax SUA from HFCS. D) Fructose intake from 3day food log with Cmax SUA fr om sucrose.

PAGE 259

259 Figure C 29. Correlation of fructose intake with changes in FEUA. A) Fructose intake from FFQ with FEUA from HFCS. B) Fructose intake from FFQ with FEUA from sucrose. C) Fructose intake from 3day food log with FEUA from HFCS. D) Fr uctose intake from 3day food log with FEUA from sucrose.

PAGE 260

260 APPENDIX D CELL CULTURE MEDIA 20% FBS Growth Media for Cell Growth in Plates 1. 500 mL DMEM with 4.5 g/L glucose, Lglutamine, and sodium bicarbonate (#D5796, SigmaAldrich Inc, St. Louis, MO). 2. 100 mL heat inactivated FBS (#12306C, SigmaAldrich Inc, St. Louis, MO). 3. 6.0 mL 100X Penicillin ( 10,000 U/mL ) Streptomycin (10 ,000 g/mL) Solution (#30002CI, Cellgro, Mediatech, Inc, Manassas, VA). 4. 6.0 mL 100X Nonessential Amino Acid Solution (#M7145, Sigma Aldrich Inc, St. Louis, MO). 5. Store at 4C. High Glucose Growth Media for Cell Growth in Transwell 1. 500 mL DMEM with 4.5 g/L glucose, Lglutamine, and sodium bicarbonate (#D5796, SigmaAldrich Inc, St. Louis, MO). 2. 50 mL heat inactivated FBS (#12306C, Sigm a Aldrich Inc, St. Louis, MO). 3. 5.5 mL 100X Penicillin ( 10,000 U/mL ) Streptomycin (10 ,000 g/mL) Solution (#30002CI, Cellgro, Mediatech, Inc, Manassas, VA). 4. 5.5 mL 100X Nonessential Amino Acid Solution (#M7145, SigmaAldrich Inc, St. Louis, MO). 5. Store at 4C. Low Glucose Growth Media for Cell Growth in Transwell 1. 500 mL DMEM with 1.0 g/L glucose, Lglutamine, and sodium bicarbonate (#D6046, SigmaAldrich Inc, St. Louis, MO). 2. 50 mL heat inactivated FBS (#12306C, SigmaAldrich Inc, St. Louis, MO). 3. 5.5 mL 100X Penicillin ( 10,000 U/mL ) Streptomycin (10 ,000 g/mL) Solution (#30002CI, Cellgro, Mediatech, Inc, Manassas, VA). 4. 5.5 mL 100X Nonessential Amino Acid Solution (#M7145, SigmaAldrich Inc, St. Louis, MO). 5. Store at 4C

PAGE 261

261 Subculturing Process 1. Aspirate old growth media from cell culture plate. 2. Wash cells with 3 mL 1X PBS (phosphatebuffered saline) to 100 mm plate or 6 mL to 150 mm plate (#21040CV, Cellgro, Mediatech, Inc, Manassas, VA). 3. Aspirate out PBS. 4. Dissociate cells from plate with 3 mL Trypsin to 100 mm plate or 6 mL to 150 mm plate (#25052CI, Cellgro, Mediatech, Inc, Manassas, VA). 5. Incubate at 37 C in a 5% CO2 95% air atmosphere until single cell suspensions develop (~5 min). If needed, mix with pipette to break up cell clumps. 6. Add 1:1 v/v growt h media containing FBS to nullify action of trypsin: 3 mL to 100 mm plate or 6 mL to 150 mm plate. 7. Transfer to conical tube. 8. Centrifuge at 1000 rpm for 5 min at RT (~25C). 9. Aspirate media. 10. Resuspend cell pellet with growth media. 11. Transfer part of cell susp ension into new cell culture plate: 3 x 106 cells per 100 mm plate, 5 X 106 cells per 150 mm plate, or 3 X 105 cells per 6well plate insert (#3450, Corning Inc, Corning, NY). 12. Add 10 mL growth media to 100 mm plate or 20 mL to 150 mm plate. 13. Grow cells by incubating at 37 C in a 5 % CO2 9 5 % air atmosphere. HBSS (Without Glucose) Hank's Full Strength Composition 0.137 M NaCl 5.4 mM KCl 0.25 mM Na2HPO4 0.44 mM KH2PO4 1.3 mM CaCl2 1.0 mM MgSO4 4.2 mM NaHCO3

PAGE 262

262 Hanks Stock Solutions All stocks were filtered and st ored at 4oC. Stock #1 1. Dissolve the following in 90ml of distilled H2O o g NaCl (#S5886, SigmaAldrich Inc, St. Louis, MO) o 0.4 g KCl (#P5405, SigmaAldrich Inc, St. Louis, MO) 2. qs to 100 ml with ddH2O Stock #2 1. Dissolve the following in 90ml of distilled H2O o 0.358 g anhydrous Na2HPO4 (#S3741, Fisher Scientific Fair Lawn, NJ) o 0.60 g KH2PO4 (#P5655, SigmaAldrich Inc, St. Louis, MO) 2. qs to 100 ml with ddH2O Stock #3 1. Add 0.72 g CaCl2 (#C5670, Sigma Aldrich Inc, St. Louis, MO) 2. qs to50 ml with ddH2O Stock #4 1. Add 1.23 g MgSO4x7H2O (#M63 500, Fisher Scientific Fair Lawn, NJ) 2. q s to 50 ml with ddH2O Stock #5 1. Add 0.35 g NaHCO3 (#S 5761, SigmaAldrich Inc, St. Louis, MO) 2. qs to10 ml with ddH2O Hank's Premix ( Combine the solutions in following order) 1. 10.0 ml Stock #1 2. 1.0 ml Stock #2 3. 1.0 ml Stock #3 4. 8 6.0 ml distilled H2O 5. 1. 0 ml Stock #4 Hank's Full Strength (mix prior to use) 1. 9.9 ml Hank's Premix 2. 0.1 ml Stock #5

PAGE 263

263 A PPENDIX E IN VITRO TRANSPORT DATA Mannitol Transport and TEER Values Table E 1. % of total mannitol trans port ed during 60 min and TEER values Treatment Well # % Mannitol Transport TEER AW TEER 1 H Excluded F100:G0 1 0.254 N/A N/A 2 0.117 N/A N/A 3 0.125 N/A N/A 4 0.138 792 753 5 0.151 787 739 6 0.065 827 784 F100:G25 1 0.306 N/A N/A 2 0.140 N/A N/A 3 0.162 N/A N/A F100:G50 1 0.134 1027 827 2 0.098 1180 757 3 0.117 1157 735 F100:G75 1 0.855 1257 818 Yes 2 0.167 1091 689 3 2.523 1252 822 Yes F100:G100 1 0.229 1420 1024 2 0.274 1331 901 3 0.351 1152 978 F100:G200 1 2.324 1218 860 Yes 2 N/A 1290 862 3 0.764 1314 916 Yes F100:G300 1 N/A 1334 849 2 N/A 1286 834 3 N/A 1363 822 F fructose; G glucose; HFCS high fr uctose corn syrup; M mannitol. TEER for trans epithelial electrical resistance; AW TEER measured after washing of monolayers; 1H TEER measured after 1 H transport experiments. F or the treatments, concentrations shown are in mM. Monolayers were excluded if % mannitol transport was > 0.5%. ** Monol ayers were excluded if TEER values < 600 2.

PAGE 264

264 Table E 1. Cont. Treatment Well # % Mannitol Transport TEER AW TEER 1 H Excluded F0:G100 1 0.089 N/A N/A 2 0.078 N/A N/A 3 0.208 N/A N/A 4 0.011 855 773 5 0.014 835 818 6 0.040 909 822 F25:G100 1 0.087 N/A N/A 2 0.127 N/A N/A 3 0.096 N/A N/A F50:G100 1 0.116 1328 748 2 0.107 1166 680 3 0.121 1323 718 F75:G100 1 0.205 1289 708 2 0.165 1093 744 3 0.147 1304 751 F200:G100 1 0.393 1392 874 2 0.225 1412 946 3 0.607 1424 1000 Yes F300:G100 1 N/A 1404 869 2 N/A 1358 881 3 N/A 666 599 Yes ** M400 1 0.132 1352 969 2 0.351 1210 1132 3 0.113 1405 950 F fructose; G glucose; HFCS high fructose corn syrup; M manni tol.. TEER for transepithelial electrical resistance; AW TEER measured after washing of monolayers; 1H TEER measured after 1 H transport experiments. For the treatments, concentrations shown are in mM. Monolayers were excluded if % mannitol transport was > 0.5%. ** Monolayers were excluded if TEER values < 600 2.

PAGE 265

265 Fructose and Glucose Permeability Data Figure E 1. Permeability of fructose with varying fructose concentrations. Caco 2 cells treated with 100 mM glucose and increasing concentrations of fructose (F, mM): 25 F; 50 F; 75 F; 100 F; 200 F; 300 F.

PAGE 266

266 Figure E 2. Permeability of fructose with varying glucose concentrations. Caco 2 cells treated with 100 mM fructose and increasing concentrations of glucose (G, mM): 0 G; 25 G; 50 G; 75 G; 100 G; 200 G; 300 G.

PAGE 267

267 Figure E 3. Permeability of glucose with varying glucose concentrations. Caco 2 cells treated with 100 mM fructose and increasing concentrations of glucose (G, mM): 25 G; 50 G; 75 G.

PAGE 268

268 Figure E 4. Permeability of glucose with varying fructose concentr ations. Caco 2 cells treated with 100 mM glucose and increasing concentrations of fructose (F, mM): 0 F; 25 F; 50 F; 75 F.

PAGE 269

269 A PPENDIX F WESTERN BLOT MEDIA AND PROTOCOLS 4X Sample Dilution Buffer 4 mL glycerol (5.04 g) 0.8 g SDS 2.5 mL 1M Tris HCL 80 l bromophenol blue slurry (5 mg/ml in water; vortex well before using) qs with ddH2O to 8 mL Mercaptoethanol (BME) to 20% (i.e. 0.2 mL BME to 0.8 ml 4X SDB) before using (#194705, MP Biomedicals, LLC, Solon, OH) Running Buffer 1. Make stock solution of 10X Tris/Glycine/SDS 250 mM Tris 30.29 g 1920 mM glycine 144.13 g 0.1% SDS (w/v) 10 g qs with ddH2O to 1 L 2. Dilute with ddH2O to 1X before use Transfer Buffer 1. Make stock solution of 10X Tris/Glycine 250 mM Tris 30.29 g 1920 mM glycine 144.13 g qs with ddH2O to 1 L 2. Prior to us, add 20% Methanol and dilute with ddH2O to 1X 10X TBS (Trisbuffered saline) 30 mM Tris 36.5 g 200 mM NaCl 116.9 g qs with ddH2O to 1 L Adjust the pH to 7.5 with concentrated HCl Washing Buffer 1X TBST (0.1% w/v) o 10X TBS 100 mL o ddH2O 900 mL o Tween20 (#9005645, Acros Organics, Morris Plains, NJ) 1 mL

PAGE 270

270 Blocking Reagent 1X TBST 100 mL ECL Advance Blocking Agent 2 g (Amersham ECL Advance Western Blotting Detection Kit, GE Healthcare BioSciences Corp., Pi scataway, NJ). Staining Protocol Fast Green Stain (0.1%) 0.1 g fast green 50% methanol 40% water 10% acetic acid Solution is hazardous! Fast Green DeStain 50% Methanol 10% glacial acetic acid 40% water Staining/Destaining of the Blot Blot was stained by immersion in Fast Green Stain until the protein bands were visible (~ 5 minutes). The Fast Green Stain was saved and reused. Blot was immersed in Fast Green DeStain (~ 30 ml) and wash on rotator for several minutes. The process was repeated until discrete protein bands were visible, but no background staining was apparent (~ 15 min). Blot was rinsed with 1X TBS T, wrapped in SaranWrap and photographed under white light. Wrapped in SaranWrap, Blot was stored at 4C.

PAGE 271

271 A PPENDIX G GENE EXPRESSIO N DATA In Vitro RT PCR Results For the following graphs, t he sugar treatments were glucose (G), fructose (F), and fructose:glucose (FG). The Mannitol (M) was the osmolarity control. The sugar concentrations were 5, 25, 100, and 300 mM. Mannitol was added to achieve a final concentration of 400 mM. The cells exposed to the sugars and incubated for 1 H, 6 H, 24 H, and 72 H. 5 mM glucose treatment (RQ = 1) was used as the ca librator for each time set. RQ values > 1 represent an increase in gene expression. RQ values < 1 represent a decrease in gene expression.

PAGE 272

272 Table G 1. Gene expression data for KHK, SI SLC2A2 SLC2A5 and SLC5A1 Gene Sugar Incubation Time (H) Concentration (mM) RQ KHK Fructose 1 5 0.98 0.11 KHK Fructose:Glucose 1 5 1.51 0.08 KHK Glucose 1 5 1 .00 0.07 KHK Fructose 1 25 1.18 0.25 KHK Fructose:Glucose 1 25 1.35 0.25 KHK Glucose 1 25 1.36 0.1 7 KHK Fructose 1 100 1.25 0.23 KHK Fructose:Glucose 1 100 1.31 0.35 KHK Glucose 1 100 1.13 0.13 KHK Fructose 1 300 1.0 9 0.20 KHK Fructose:Glucose 1 300 1.2 0 0.01 KHK Glucose 1 300 1.37 0.27 KHK M annitol 1 400 1 .10 0.28 KHK Fructose 6 5 0.73 0.11 KHK Fructose:Glucose 6 5 1.07 0.06 KHK Glucose 6 5 1 .00 0.12 KHK Fructose 6 25 1.17 0.19 KHK Fructose:Gl ucose 6 25 1.15 0.12 KHK Glucose 6 25 1.29 0.07 KHK Fructose 6 100 1.32 0.08 KHK Fructose:Glucose 6 100 1.46 0.09 KHK Glucose 6 100 1.23 0.18 KHK Fructose 6 300 1.32 0.1 0 KHK Fructose:Glucose 6 300 1.04 0.10 KHK Glucose 6 300 1.18 0. 07 KHK M annitol 6 400 1.1 0 0.11 KHK Fructose 24 5 0.91 0.04 KHK Fructose:Glucose 24 5 0.96 0.07 KHK Glucose 24 5 1 .00 0.05 KHK Fructose 24 25 0.89 0.09 KHK Fructose:Glucose 24 25 1.16 0.17 KHK Glucose 24 25 1.09 0.16 KHK Fructose 24 1 00 0.94 0.05 KHK Fructose:Glucose 24 100 1.13 0.08 KHK Glucose 24 100 1.14 0.05 KHK Fructose 24 300 1.02 0.06 KHK Fructose:Glucose 24 300 1.03 0.10 KHK Glucose 24 300 1.21 0.07 KHK M annitol 24 400 1.03 0.06 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation.

PAGE 273

273 Table G 1. Continued. Gene Sugar Incubation Time (H) Concentration (mM) RQ KHK Fructose 72 5 0.9 0 0.03 KHK Fructose:Glucose 72 5 1.15 0.14 KH K Glucose 72 5 1 .00 0.03 KHK Fructose 72 25 1.23 0.05 KHK Fructose:Glucose 72 25 1.19 0.09 KHK Glucose 72 25 1.36 0.06 KHK Fructose 72 100 1.32 0.15 KHK Fructose:Glucose 72 100 1.3 0 0.12 KHK Glucose 72 100 1.08 0.24 KHK Fructose 72 300 0.93 0.05 KHK Fructose:Glucose 72 300 0.74 0.09 KHK Glucose 72 300 0.65 0.16 KHK Mannitol 72 400 1.15 0.14 SI Fructose 1 5 1.03 0.0 3 SI Fructose:Glucose 1 5 0.97 0.16 SI Glucose 1 5 1.01 0.13 SI Fructose 1 25 1.08 0.22 SI Fructose: Glucose 1 25 0.92 0.04 SI Glucose 1 25 1.2 2 0.19 SI Fructose 1 100 0.85 0.10 SI Fructose:Glucose 1 100 0.8 0 0.05 SI Glucose 1 100 0.88 0.07 SI Fructose 1 300 0.74 0.11 SI Fructose:Glucose 1 300 0.93 0.07 SI Glucose 1 300 0.91 0.24 S I Mannitol 1 400 1.12 0.15 SI Fructose 6 5 0.91 0.08 SI Fructose:Glucose 6 5 0.88 0.08 SI Glucose 6 5 1 .00 0.09 SI Fructose 6 25 1.08 0.15 SI Fructose:Glucose 6 25 1.02 0.17 SI Glucose 6 25 1.27 0.02 SI Fructose 6 100 1.02 0.03 SI Fructose:Glucose 6 100 0.98 0.01 SI Glucose 6 100 1.16 0.03 SI Fructose 6 300 0.85 0.12 SI Fructose:Glucose 6 300 0.81 0.09 SI Glucose 6 300 0.8 0 0.02 SI Mannitol 6 400 1.11 0.02 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation.

PAGE 274

274 Table G 1. Continued. Gene Sugar Incubation Time (H) Concentration (mM) RQ SI Fructose 24 5 0.99 0.17 SI Fructose:Glucose 24 5 0.99 0.12 SI Glucose 24 5 1.01 0.2 0 SI Fruct ose 24 25 0.92 0.12 SI Fructose:Glucose 24 25 0.88 0.17 SI Glucose 24 25 1 .00 0.17 SI Fructose 24 100 0.82 0.03 SI Fructose:Glucose 24 100 0.86 0.09 SI Glucose 24 100 1.07 0.06 SI Fructose 24 300 0.61 0.03 SI Fructose:Glucose 24 300 0. 56 0.07 SI Glucose 24 300 0.49 0.08 SI Mannitol 24 400 0.72 0.05 SI Fructose 72 5 0.88 0.04 SI Fructose:Glucose 72 5 1.05 0.08 SI Glucose 72 5 1 .00 0.04 SI Fructose 72 25 1.15 0.13 SI Fructose:Glucose 72 25 1.1 0 0.2 0 SI Glucose 72 2 5 1.39 0.06 SI Fructose 72 100 1.27 0.13 SI Fructose:Glucose 72 100 1.08 0.09 SI Glucose 72 100 1.16 0.09 SI Fructose 72 300 0.59 0.04 SI Fructose:Glucose 72 300 0.39 0.10 SI Glucose 72 300 0.17 0.09 SI Mannitol 72 400 1.03 0.08 SLC 2A2 Fructose 1 5 0.99 0.08 SLC2A2 Fructose:Glucose 1 5 1.52 0.17 SLC2A2 Glucose 1 5 1 .00 0.06 SLC2A2 Fructose 1 25 1.08 0.07 SLC2A2 Fructose:Glucose 1 25 1.38 0.24 SLC2A2 Glucose 1 25 1.17 0.22 SLC2A2 Fructose 1 100 1.04 0.18 SLC2A2 Fr uctose:Glucose 1 100 1.29 0.23 SLC2A2 Glucose 1 100 1.04 0.1 0 SLC2A2 Fructose 1 300 1.01 0.28 SLC2A2 Fructose:Glucose 1 300 1.32 0.14 SLC2A2 Glucose 1 300 1.06 0.18 SLC2A2 Mannitol 1 400 1.31 0.28 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation.

PAGE 275

275 Table G 1. Continued. Gene Sugar Incubation Time (H) Concentration (mM) RQ SLC2A2 Fructose 6 5 0.79 0.06 SLC2A2 Fructose:Glucose 6 5 1.19 0.12 SLC2A2 Gl ucose 6 5 1.01 0.19 SLC2A2 Fructose 6 25 1.34 0.10 SLC2A2 Fructose:Glucose 6 25 1.27 0.02 SLC2A2 Glucose 6 25 1.49 0.24 SLC2A2 Fructose 6 100 1.55 0.1 0 SLC2A2 Fructose:Glucose 6 100 1.53 0.06 SLC2A2 Glucose 6 100 1.43 0.23 SLC2A2 Fruct ose 6 300 1.26 0.19 SLC2A2 Fructose:Glucose 6 300 1.11 0.16 SLC2A2 Glucose 6 300 0.78 0.09 SLC2A2 Mannitol 6 400 1.19 0.19 SLC2A2 Fructose 24 5 1.11 0.13 SLC2A2 Fructose:Glucose 24 5 1.78 0.15 SLC2A2 Glucose 24 5 1.01 0.21 SLC2A2 Fruct ose 24 25 1.02 0.20 SLC2A2 Fructose:Glucose 24 25 1.77 0.32 SLC2A2 Glucose 24 25 1 .00 0.13 SLC2A2 Fructose 24 100 0.92 0.01 SLC2A2 Fructose:Glucose 24 100 1.6 0 0.44 SLC2A2 Glucose 24 100 1.09 0.11 SLC2A2 Fructose 24 300 0.68 0.06 SLC2A 2 Fructose:Glucose 24 300 0.46 0.05 SLC2A2 Glucose 24 300 0.43 0.05 SLC2A2 Mannitol 24 400 1.01 0.16 SLC2A2 Fructose 72 5 0.88 0.08 SLC2A2 Fructose:Glucose 72 5 1.3 0 0.22 SLC2A2 Glucose 72 5 1 .00 0.08 SLC2A2 Fructose 72 25 1.13 0.03 SL C2A2 Fructose:Glucose 72 25 1.18 0.09 SLC2A2 Glucose 72 25 1.35 0.11 SLC2A2 Fructose 72 100 1.3 0 0.17 SLC2A2 Fructose:Glucose 72 100 1.16 0.13 SLC2A2 Glucose 72 100 1.06 0.03 SLC2A2 Fructose 72 300 0.54 0.03 SLC2A2 Fructose:Glucose 72 300 0.29 0.09 SLC2A2 Glucose 72 300 0.11 0.06 SLC2A2 Mannitol 72 400 1.21 0.12 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation.

PAGE 276

276 Table G 1. Continued. Gene Sugar Incubation Time (H) Concentration (mM) RQ SLC2A5 Fructose 1 5 1.01 0.0 3 SLC2A5 Fructose:Glucose 1 5 1.73 0.12 SLC2A5 Glucose 1 5 1 .00 0.04 SLC2A5 Fructose 1 25 1.35 0.33 SLC2A5 Fructose:Glucose 1 25 1.67 0.31 SLC2A5 Glucose 1 25 1.5 3 0.31 SLC2A5 Fructose 1 100 1.84 0.24 SLC2A5 Fructose:Glucose 1 100 1.48 0.36 SLC2A5 Glucose 1 100 1.57 0.44 SLC2A5 Fructose 1 300 1.15 0.27 SLC2A5 Fructose:Glucose 1 300 1.35 0.14 SLC2A5 Glucose 1 300 1.26 0.29 SLC2A5 Mannitol 1 400 1 .00 0 .06 SLC2A5 Fructose 6 5 0.79 0.06 SLC2A5 Fructose:Glucose 6 5 1.05 0.03 SLC2A5 Glucose 6 5 1.05 0.37 SLC2A5 Fructose 6 25 1.32 0.04 SLC2A5 Fructose:Glucose 6 25 1.03 0.09 SLC2A5 Glucose 6 25 1.18 0.56 SLC2A5 Fructose 6 100 1.4 0 0.22 S LC2A5 Fructose:Glucose 6 100 1.69 0.31 SLC2A5 Glucose 6 100 1.15 0.18 SLC2A5 Fructose 6 300 1.4 0 0.25 SLC2A5 Fructose:Glucose 6 300 1.31 0.10 SLC2A5 Glucose 6 300 1.18 0.20 SLC2A5 Mannitol 6 400 0.96 0.11 SLC2A5 Fructose 24 5 1.03 0.07 SLC2A5 Fructose:Glucose 24 5 0.8 0 0.21 SLC2A5 Glucose 24 5 1.01 0.14 SLC2A5 Fructose 24 25 0.96 0.11 SLC2A5 Fructose:Glucose 24 25 0.94 0.2 0 SLC2A5 Glucose 24 25 0.95 0.13 SLC2A5 Fructose 24 100 0.92 0.05 SLC2A5 Fructose:Glucose 24 100 1 .19 0.07 SLC2A5 Glucose 24 100 1.32 0.23 SLC2A5 Fructose 24 300 0.97 0.06 SLC2A5 Fructose:Glucose 24 300 0.97 0.17 SLC2A5 Glucose 24 300 0.76 0.04 SLC2A5 Mannitol 24 400 0.84 0.05 RQ relative quantity. 5 mM glucose (RQ = 1) was used as t he calibrator for each time set. Data shown as mean standard deviation.

PAGE 277

277 Table G 1. Continued. Gene Sugar Incubation Time (H) Concentration (mM) RQ SLC2A5 Fructose 72 5 0.98 0.07 SLC2A5 Fructose:Glucose 72 5 1.5 0 0.24 SLC2A5 Glucose 72 5 1 .00 0.01 SLC2A5 Fructose 72 25 1.38 0.26 SLC2A5 Fructose:Glucose 72 25 1.68 0.10 SLC2A5 Glucose 72 25 1.64 0.11 SLC2A5 Fructose 72 100 1.45 0.12 SLC2A5 Fructose:Glucose 72 100 1.84 0.22 SLC2A5 Glucose 72 100 1.42 0.26 SLC2A5 Fructose 72 3 00 1.29 0.03 SLC2A5 Fructose:Glucose 72 300 1.54 0.31 SLC2A5 Glucose 72 300 0.67 0.25 SLC2A5 Mannitol 72 400 1.36 0.10 SLC5A1 Fructose 1 5 1.07 0. 02 SLC5A1 Fructose:Glucose 1 5 1.46 0.09 SLC5A1 Glucose 1 5 1 .00 0.04 SLC5A1 Fructose 1 2 5 1.26 0.15 SLC5A1 Fructose:Glucose 1 25 1.43 0.21 SLC5A1 Glucose 1 25 1.35 0.01 SLC5A1 Fructose 1 100 1.82 0.03 SLC5A1 Fructose:Glucose 1 100 1.25 0.09 SLC5A1 Glucose 1 100 1.48 0.34 SLC5A1 Fructose 1 300 1.28 0.14 SLC5A1 Fructose:Glu cose 1 300 1.29 0.06 SLC5A1 Glucose 1 300 1.31 0.26 SLC5A1 Mannitol 1 400 0.97 0.1 0 SLC5A1 Fructose 6 5 0.69 0.07 SLC5A1 Fructose:Glucose 6 5 1.1 0 0.05 SLC5A1 Glucose 6 5 1.02 0.27 SLC5A1 Fructose 6 25 1.31 0.10 SLC5A1 Fructose:Glucose 6 25 1.14 0.11 SLC5A1 Glucose 6 25 1.38 0.10 SLC5A1 Fructose 6 100 1.36 0.16 SLC5A1 Fructose:Glucose 6 100 1.32 0.12 SLC5A1 Glucose 6 100 1.23 0.19 SLC5A1 Fructose 6 300 1.21 0.11 SLC5A1 Fructose:Glucose 6 300 1.01 0.05 SLC5A1 Glucose 6 300 1.06 0.09 SLC5A1 Mannitol 6 400 0.82 0.12 RQ relative quantity. 5 mM glucose (RQ = 1) was used as the calibrator for each time set. Data shown as mean standard deviation.

PAGE 278

278 Table G 1. Continued. Gene Sugar Incubation Time (H) Concentrati on (mM) RQ SLC5A1 Fructose 24 5 1.16 0.13 SLC5A1 Fructose:Glucose 24 5 1.25 0.18 SLC5A1 Glucose 24 5 1.03 0.29 SLC5A1 Fructose 24 25 1 .00 0.07 SLC5A1 Fructose:Glucose 24 25 1.22 0.15 SLC5A1 Glucose 24 25 1.1 0 0.18 SLC5A1 Fructose 24 100 1.02 0.06 SLC5A1 Fructose:Glucose 24 100 1.32 0.19 SLC5A1 Glucose 24 100 1.52 0.10 SLC5A1 Fructose 24 300 1.47 0.05 SLC5A1 Fructose:Glucose 24 300 1.29 0.18 SLC5A1 Glucose 24 300 0.85 0.01 SLC5A1 Mannitol 24 400 1.01 0.06 SLC5A1 Fructo se 72 5 1.01 0.12 SLC5A1 Fructose:Glucose 72 5 1.19 0.15 SLC5A1 Glucose 72 5 1 .00 0.09 SLC5A1 Fructose 72 25 1.37 0.09 SLC5A1 Fructose:Glucose 72 25 1.26 0.19 SLC5A1 Glucose 72 25 1.46 0.03 SLC5A1 Fructose 72 100 1.22 0.14 SLC5A1 Fruct ose:Glucose 72 100 1.56 0.13 SLC5A1 Glucose 72 100 1.15 0.13 SLC5A1 Fructose 72 300 1.23 0.07 SLC5A1 Fructose:Glucose 72 300 1.14 0.19 SLC5A1 Glucose 72 300 0.43 0.12 SLC5A1 Mannitol 72 400 1.11 0.08 RQ relative quantity. 5 mM glucose (R Q = 1) was used as the calibrator for each time set. Data shown as mean standard deviation.

PAGE 279

279 Figure G 1. Relative quantity (RQ) of KHK expression. Sugar treatments were fructose (F), glucose (G), equimolar fructose:glucose (FG), and mannitol (M).

PAGE 280

280 Figure G 2 Relative quantity (RQ) of SI expression. Sugar treatments were fructose (F), glucose (G), equimolar fructose:glucose (FG), and mannitol (M).

PAGE 281

281 Figure G 3. Relative quantity (RQ) of S LC2A2 expression. Sugar treatments were fructose (F), gluc ose (G), equimolar fructose:glucose (FG), and mannitol (M).

PAGE 282

282 Figure G 4. Relative quantity (RQ) of S LC2A5 expression. Sugar treatments were fructose (F), glucose (G), equimolar fructose:glucose (FG), and mannitol (M).

PAGE 283

283 Figure G 5. Relative quantity ( RQ) of S LC5A1 expression. Sugar treatments were fructose (F), glucose (G), equimolar fructose:glucose (FG), and mannitol (M).

PAGE 284

284 Clinical RT PCR Results Figure G 6. Relative quantity (RQ) of KHK expression after consumption of soft drinks containing either HFCS or sucrose. The expression at the end of the 6 hr study was calibrated to the pretreatment levels. V1 = study visit 1; V2 = study visit 2.

PAGE 285

285 Figure G 7. Relative quantity (RQ) of KHK expression by study visit and treatment. The expression at t he end of the 6 hr study was calibrated to the pretreatment levels.

PAGE 286

286 Figure G 8 Relative quantity (RQ) of KHK expression after consumption of soft drinks containing either HFCS or sucrose. The pretreatment expression level at study visit 1 of subject R 07 was used as the calibrator. V1 = study visit 1; V2 = study visit 2.

PAGE 287

287 Figure G 9. Relative quantity (RQ) of SLC2A5 expression after consumption of soft drinks containing either HFCS or sucrose. The expressions at the end of the 6 hr study were calibr ated to the pretreatment levels. V1 = study visit 1; V2 = study visit 2.

PAGE 288

288 Figure G 10. Relative quantity (RQ) of SLC2A5 expression by study visit and treatment. The expression at the end of the 6 hr study was calibrated to the pretreatment levels.

PAGE 289

289 Fig ure G 11. Relative quantity (RQ) of SLC2A5 expression after consumption of soft drinks containing either HFCS or sucrose. The pretreatment expression level at study visit 1 of subject R07 was used as the calibrator. V1 = study visit 1; V2 = study visit 2.

PAGE 290

290 Figure G 1 2 Correlation of BMI with pretreatment expression levels of A) KHK and B) SLC2A5 .

PAGE 291

291 LIST OF REFERENCES 1. USDA ARS: United States Department of Agriculture Agricultural Research Service. In: USDA National Nutrient Databas e for Standard Reference, Release 21; 2009. 2. Steinmann B, Gitzelman R, Berghe GVd. Disorders of fructose metabolism. In: Scriver C, Beaudet A, Sly W, editors. The Metabolic and Molecular Bases of Inherited Disease. 8th ed. New York: McGraw Hill; 2001. p. 14891520. 3. Cargill Incorporated. Sweeteners. In: Cargill Incorporated; 2009. 4. Archer Daniels Midland Company. Sweeteners. In: Archer Daniels Midland Company; 2009. 5. Le KA, Tappy L. Metabolic effects of fructose. Curr Opin Clin Nutr Metab Care 2006;9(4):46975. 6. Hanover LM, White JS. Manufacturing, composition, and applications of fructose. Am J Clin Nutr 1993;58(5 Suppl):724S 732S. 7. Kaneko T, Takahashi S, Saito K. Characterization of acidstable glucose isomerase from Streptomyces sp., and development of singlestep processes for highfructose corn sweetener (HFCS) production. Biosci Biotechnol Biochem 2000;64(5):9407. 8. United States Department of Agriculture/Economic Research Service. U.S. per capita food consumption: Sugars and Sweeteners. In. 9. Wang Y, Moreno LA, Caballero B, Cole TJ. Limitations of the current world health organization growth references for children and adolescents. Food Nutr Bull 2006;27(4 Suppl Growth Standard):S17588. 10. Bray GA, Nielsen SJ, Popkin BM. Consump tion of highfructose corn syrup in beverages may play a role in the epidemic of obesity. Am J Clin Nutr 2004;79(4):53743. 11. Jacobson MF. Highfructose corn syrup and the obesity epidemic. Am J Clin Nutr 2004;80(4):1081; author reply 10812. 12. Gaby AR. Adverse effects of dietary fructose. Altern Med Rev 2005;10(4):294306. 13. Berg JM, Tymoczko JL, Stryer L. Biochemistry. 5th ed. New York: W.H. Freeman; 2002.

PAGE 292

292 14. Cori CF. The fate of sugar in the animal body. I. The rate of absorption of hexoses and pentoses from the intestinal tract. J. Biol. Chem. 1925;66(2):691715. 15. Cori GT, Cori CF. The fate of sugar in the animal body. VIII. The influence of insulin on the utilization of glucose, fructose, and dihydroxyacetone. Journal of Biological Chem istry 1928;76(3):75595. 16. Groen J. The Absorption of Hexoses from the Upper Part of the Small Intestine in Man. J Clin Invest 1937;16(2):24555. 17. Devlin TM. Textbook of biochemistry with clinical correlations. 5th ed. New York Chichester: Wiley; 2002. 18. Funari VA, Herrera VL, Freeman D, Tolan DR. Genes required for fructose metabolism are expressed in Purkinje cells in the cerebellum. Brain Res Mol Brain Res 2005;142(2):11522. 19. Mazzali M, Hughes J, Kim YG, Jefferson JA, Kang DH, Gordon KL, et al. Elevated uric acid increases blood pressure in the rat by a novel crystal independent mechanism. Hypertension 2001;38(5):11016. 20. Short RA, Tuttle KR. Clinical evidence for the influence of uric acid on hypertension, cardiovascular disease, and k idney disease: a statistical modeling perspective. Semin Nephrol 2005;25(1):2531. 21. Gavin AR, Struthers AD. Hyperuricemia and adverse outcomes in cardiovascular disease: potential for therapeutic intervention. Am J Cardiovasc Drugs 2003;3(5):30914. 2 2. Schachter M. Uric acid and hypertension. Curr Pharm Des 2005;11(32):413943. 23. Nakagawa T, Kang DH, Feig D, Sanchez Lozada LG, Srinivas TR, Sautin Y, et al. Unearthing uric acid: an ancient factor with recently found significance in renal and cardiov ascular disease. Kidney Int 2006;69(10):17225. 24. Chambers RA, Pratt RT. Idiosyncrasy to fructose. Lancet 1956;271(6938):340. 25. Oberhaensli RD, Galloway GJ, Taylor DJ, Bore PJ, Radda GK. Assessment of human liver metabolism by phosphorus 31 magnetic resonance spectroscopy. Br J Radiol 1986;59(703):6959. 26. Terrier F, Vock P, Cotting J, Ladebeck R, Reichen J, Hentschel D. Effect of intravenous fructose on the P 31 MR spectrum of the liver: dose response in healthy volunteers. Radiology 1989;171(2):55763.

PAGE 293

293 27. Bergstrom J, Hultman E, Roch Norlund AE. Lactic acid accumulation in connection with fructose infusion. Acta Med Scand 1968;184(5):35964. 28. Bantle JP, Raatz SK, Thomas W, Georgopoulos A. Effects of dietary fructose on plasma lipids in healt hy subjects. Am J Clin Nutr 2000;72(5):112834. 29. Beck Nielsen H, Pedersen O, Lindskov HO. Impaired cellular insulin binding and insulin sensitivity induced by high fructose feeding in normal subjects. Am J Clin Nutr 1980;33(2):2738. 30. Farah V, Elas ed KM, Chen Y, Key MP, Cunha TS, Irigoyen MC, et al. Nocturnal hypertension in mice consuming a high fructose diet. Auton Neurosci 2006. 31. Teff KL, Elliott SS, Tschop M, Kieffer TJ, Rader D, Heiman M, et al. Dietary fructose reduces circulating insulin and leptin, attenuates postprandial suppression of ghrelin, and increases triglycerides in women. J Clin Endocrinol Metab 2004;89(6):296372. 32. Ludwig DS, Peterson KE, Gortmaker SL. Relation between consumption of sugar sweetened drinks and childhood obesity: a prospective, observational analysis. Lancet 2001;357(9255):5058. 33. Schulze MB, Manson JE, Ludwig DS, Colditz GA, Stampfer MJ, Willett WC, et al. Sugar sweetened beverages, weight gain, and incidence of type 2 diabetes in young and middleaged women. Jama 2004;292(8):92734. 34. Raivio KO, Becker A, Meyer LJ, Greene ML, Nuki G, Seegmiller JE. Stimulation of human purine synthesis de novo by fructose infusion. Metabolism 1975;24(7):8619. 35. Nakagawa T, Hu H, Zharikov S, Tuttle KR, Short RA, G lushakova O, et al. A causal role for uric acid in fructoseinduced metabolic syndrome. Am J Physiol Renal Physiol 2006;290(3):F62531. 36. Sanchez Lozada LG, Tapia E, Jimenez A, Bautista P, Cristobal M, Nepomuceno T, et al. Fructose induced metabolic syn drome is associated with glomerular hypertension and renal microvascular damage in rats. Am J Physiol Renal Physiol 2007;292(1):F4239. 37. Teff KL, Grudziak J, Townsend RR, Dunn TN, Grant RW, Adams SH, et al. Endocrine and metabolic effects of consuming fructose and glucose sweetened beverages with meals in obese men and women: influence of insulin resistance on plasma triglyceride responses. J Clin Endocrinol Metab 2009.

PAGE 294

294 38. Montonen J, Jarvinen R, Knekt P, Heliovaara M, Reunanen A. Consumption of swee tened beverages and intakes of fructose and glucose predict type 2 diabetes occurrence. J Nutr 2007;137(6):144754. 39. Raben A, Vasilaras TH, Moller AC, Astrup A. Sucrose compared with artificial sweeteners: different effects on ad libitum food intake and body weight after 10 wk of supplementation in overweight subjects. Am J Clin Nutr 2002;76(4):7219. 40. Faeh D, Minehira K, Schwarz JM, Periasamy R, Park S, Tappy L. Effect of fructose overfeeding and fish oil administration on hepatic de novo lipogenes is and insulin sensitivity in healthy men. Diabetes 2005;54(7):190713. 41. Dhingra R, Sullivan L, Jacques PF, Wang TJ, Fox CS, Meigs JB, et al. Soft drink consumption and risk of developing cardiometabolic risk factors and the metabolic syndrome in middl e aged adults in the community. Circulation 2007;116(5):4808. 42. Aeberli I, Zimmermann MB, Molinari L, Lehmann R, l'Allemand D, Spinas GA, et al. Fructose intake is a predictor of LDL particle size in overweight schoolchildren. Am J Clin Nutr 2007;86(4) :11748. 43. American Heart Association. Heart Disease and Stroke Statistics 2007 Update. In. Dallas, TX: American Heart Association; 2007. 44. Wharton CM, Hampl JS. Beverage consumption and risk of obesity among Native Americans in Arizona. Nutr Rev 2004;62(4):1539. 45. Choi JW, Ford ES, Gao X, Choi HK. Sugar sweetened soft drinks, diet soft drinks, and serum uric acid level: The third national health and nutrition examination survey. Arthritis Rheum 2007;59(1):109116. 46. Barone S, Fussell SL, Singh AK, Lucas F, Xu J, Kim C, et al. Slc2a5 (Glut5) Is Essential for the Absorption of Fructose in the Intestine and Generation of Fructose induced Hypertension. J Biol Chem 2009;284(8):505666. 47. Fiaschi E, Baggio B, Favaro S, Antonello A, Camerin E, Todesco S, et al. Fructose induced hyperuricemia in essential hypertension. Metabolism 1977;26(11):121923. 48. Hsieh PS. Reversal of fructose induced hypertension and insulin resistance by chronic losartan treatment is independent of AT2 receptor activation in rats. J Hypertens 2005;23(12):220917. 49. Rayssiguier Y, Gueux E, Nowacki W, Rock E, Mazur A. High fructose consumption combined with low dietary magnesium intake may increase the

PAGE 295

295 incidence of the metabolic syndrome by inducing inflammation. Magnes Res 2006;19(4):23743. 50. Kawasaki T, Akanuma H, Yamanouchi T. Increased fructose concentrations in blood and urine in patients with diabetes. Diabetes Care 2002;25(2):3537. 51. Lofgren IE, Herron KL, West KL, Zern TL, Patalay M, Koo SI, et al. Carbohy drate intake is correlated with biomarkers for coronary heart disease in a population of overweight premenopausal women. J Nutr Biochem 2005;16(4):24550. 52. Johnson RJ, Segal MS, Sautin Y, Nakagawa T, Feig DI, Kang DH, et al. Potential role of sugar (fr uctose) in the epidemic of hypertension, obesity and the metabolic syndrome, diabetes, kidney disease, and cardiovascular disease. Am J Clin Nutr 2007;86(4):899906. 53. Manz F, Bickel H, Brodehl J, Feist D, Gellissen K, Gescholl Bauer B, et al. Fanconi B ickel syndrome. Pediatr Nephrol 1987;1(3):50918. 54. Santer R, Schneppenheim R, Dombrowski A, Gotze H, Steinmann B, Schaub J. Mutations in GLUT2, the gene for the liver type glucose transporter, in patients with Fanconi Bickel syndrome. Nat Genet 1997;17(3):3246. 55. Gray GM, Conklin KA, Townley RR. Sucrase isomaltase deficiency. Absence of an inactive enzyme variant. N Engl J Med 1976;294(14):7503. 56. Rhead B, Karolchik D, Kuhn RM, Hinrichs AS, Zweig AS, Fujita PA, et al. The UCSC genome browser dat abase: update 2010. Nucleic Acids Res 2009. 57. Maglott D, Ostell J, Pruitt KD, Tatusova T. Entrez Gene: genecentered information at NCBI. Nucleic Acids Res 2007;35(Database issue):D2631. 58. U.S. per capita food consumption: Sugars and Sweeteners. In: United States Department of Agriculture/Economic Research Service. 59. National Center for Health Statistics Health, United States, 2008 With Chartbook. In. Hyattsville, MD: Center for Disease Control and Prevention. 60. Johnson JA, Boerwinkle E, Zineh I, Chapman AB, Bailey K, Cooper DeHoff RM, et al. Pharmacogenomics of antihypertensive drugs: Rationale and design of the Pharmacogenomic Evaluation of Antihypertensive Responses (PEAR) study. Am Heart J 2009;157(3):4429. 61. Mitchell BD, McArdle PF, She n H, Rampersaud E, Pollin TI, Bielak LF, et al. The genetic response to short term interventions affecting cardiovascular function: rationale and design of the Heredity and Phenotype Intervention (HAPI) Heart Study. Am Heart J 2008;155(5):8238.

PAGE 296

296 62. Chapman AB, Schwartz GL, Boerwinkle E, Turner ST. Predictors of antihypertensive response to a standard dose of hydrochlorothiazide for essential hypertension. Kidney Int 2002;61(3):104755. 63. Canzanello VJ, Baranco Pryor E, Rahbari Oskoui F, Schwartz GL, Boerwinkle E, Turner ST, et al. Predictors of blood pressure response to the angiotensin receptor blocker candesartan in essential hypertension. Am J Hypertens 2008;21(1):616. 64. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, et al. dbSNP : the NCBI database of genetic variation. Nucleic Acids Res 2001;29(1):30811. 65. dbSNP BUILD 128. In: National Center for Biotechnology Information. 66. Conde L, Vaquerizas JM, Dopazo H, Arbiza L, Reumers J, Rousseau F, et al. PupaSuite: finding functi onal single nucleotide polymorphisms for largescale genotyping purposes. Nucleic Acids Res 2006;34(Web Server issue):W6215. 67. Yuan HY, Chiou JJ, Tseng WH, Liu CH, Liu CK, Lin YJ, et al. FASTSNP: an always upto date and extendable service for SNP func tion analysis and prioritization. Nucleic Acids Res 2006;34(Web Server issue):W63541. 68. Asipu A, Hayward BE, O'Reilly J, Bonthron DT. Properties of normal and mutant recombinant human ketohexokinases and implications for the pathogenesis of essential f ructosuria. Diabetes 2003;52(9):242632. 69. Kozak M, Hayward B, Borek D, Bonthron DT, Jaskolski M. Expression, purification and preliminary crystallographic studies of human ketohexokinase. Acta Crystallogr D Biol Crystallogr 2001;57(Pt 4):5868. 70. The International HapMap Project. Nature;426(6968):789796. 71. Thorisson GA, Smith AV, Krishnan L, Stein LD. The International HapMap Project Web site. Genome Res 2005;15(11):15923. 72. The International HapMap Project. Nature 2003;426(6968):78996. 73. Frazer KA, Ballinger DG, Cox DR, Hinds DA, Stuve LL, Gibbs RA, et al. A second generation human haplotype map of over 3.1 million SNPs. Nature 2007;449(7164):85161. 74. Pruitt KD, Tatusova T, Maglott DR. NCBI reference sequences (RefSeq): a curated nonredundant sequence database of genomes, transcripts and proteins. Nucleic Acids Res 2007;35(Database issue):D615.

PAGE 297

297 75. HumanCVD Genotyping BeadChip. In: Illumina, Inc.; 2008. 76. Keating BJ, Tischfield S, Murray SS, Bhangale T, Price TS, Glessner JT, et al. Concept, design and implementation of a cardiovascular genecentric 50 k SNP array for largescale genomic association studies. PLoS ONE 2008;3(10):e3583. 77. Howie BN, Carlson CS, Rieder MJ, Nickerson DA. Efficient selection of tagging singlenucleot ide polymorphisms in multiple populations. Hum Genet 2006;120(1):5868. 78. Steemers FJ, Gunderson KL. Whole genome genotyping technologies on the BeadArray platform. Biotechnol J 2007;2(1):419. 79. Infinium II Assay Workflow. In: Illumina, Inc.; 2006. 80. Lewis I, Oeser S, Chen M, McDaniel T, Yeakley J. Reliable and Accurate HighThroughput SNP Genotyping using the VeraCode GoldenGate Genotyping Assay. In: Illumina, Inc.; 2007. 81. Grundy SM, Cleeman JI, Merz CN, Brewer HB, Jr., Clark LT, Hunninghake DB, et al. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III guidelines. Circulation 2004;110(2):22739. 82. Conen D, Wietlisbach V, Bovet P, Shamlaye C, Riesen W, Paccaud F, et al. Prevalence of hyperuricemia and relation of serum uric acid with cardiovascular risk factors in a developing country. BMC Public Health 2004;4:9. 83. Weiner DE, Tighiouart H, Elsayed EF, Griffith JL, Salem DN, Levey AS. Uric Acid and Incident Kidney Disease in the Community. J Am Soc Nephrol 2008. 84. Gordon DJ, Probstfield JL, Garrison RJ, Neaton JD, Castelli WP, Knoke JD, et al. High density lipoprotein cholesterol and cardiovascular disease. Four prospective American studies. Circulation 1989;79(1):815. 85. Faul F, Erdfelder E, Buchner A, Lang AG. Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods 2009;41(4):114960. 86. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods 2007;39(2):17591. 87. Barroso I, Luan J, Middelberg RP, Harding AH, Franks PW, Jakes RW, et al. Candidate gene association study in type 2 diabetes indicates a role f or genes involved in betacell function as well as insulin action. PLoS Biol 2003;1(1):E20.

PAGE 298

298 88. Taha D, Al Harbi N, Al Sabban E. Hyperglycemia and hypoinsulinemia in patients with Fanconi Bickel syndrome. J Pediatr Endocrinol Metab 2008;21(6):5816. 89. Lewis GF, Rader DJ. New insights into the regulation of HDL metabolism and reverse cholesterol transport. Circ Res 2005;96(12):122132. 90. Laukkanen O, Lindstrom J, Eriksson J, Valle TT, Hamalainen H, IlanneParikka P, et al. Polymorphisms in the SLC2A2 (GLUT2) gene are associated with the conversion from impaired glucose tolerance to type 2 diabetes: the Finnish Diabetes Prevention Study. Diabetes 2005;54(7):225660. 91. Rubins HB, Robins SJ, Collins D, Fye CL, Anderson JW, Elam MB, et al. Gemfibrozil for the secondary prevention of coronary heart disease in men with low levels of highdensity lipoprotein cholesterol. Veterans Affairs HighDensity Lipoprotein Cholesterol Intervention Trial Study Group. N Engl J Med 1999;341(6):4108. 92. Adeli K, Taghib iglou C, Van Iderstine SC, Lewis GF. Mechanisms of hepatic very lowdensity lipoprotein overproduction in insulin resistance. Trends Cardiovasc Med 2001;11(5):1706. 93. Rader DJ. Molecular regulation of HDL metabolism and function: implications for novel therapies. J Clin Invest 2006;116(12):3090100. 94. dbSNP BUILD 130. In: National Center for Biotechnology Information. 95. Kuehn BM. 1000 Genomes Project promises closer look at variation in human genome. JAMA 2008;300(23):2715. 96. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21(2):2635. 97. Lloyd Jones D, Adams RJ, Brown TM, Carnethon M, Dai S, De Simone G, et al. Heart disease and stroke statistics --2010 update: a report from the American Heart Association. Circulation;121(7):e46e215. 98. Fields LE, Burt VL, Cutler JA, Hughes J, Roccella EJ, Sorlie P. The burden of adult hypertension in the United States 1999 to 2000: a rising tide. Hypertension 2004;44(4):398404. 99. Freiburghaus AU, Dubs R, Hadorn B, Gaze H, Hauri HP, Gitzelmann R. The brush border membrane in hereditary sucraseisomaltase deficiency: abnormal protein pattern and presence of immunoreactive enzyme. Eur J Clin Invest 1977;7(5):4559.

PAGE 299

299 100. Ritz V, Alfalah M, Zimmer KP, Schmitz J, Jacob R, Naim HY. Congenital sucrase isomaltase deficiency because of an accumulation of the mutant enzyme in the endoplasmic reticulum. Gastroenterology 2003;125(6):167885. 101. Sander P, Alfalah M, Keiser M, Korponay Szabo I Kovacs JB, Leeb T, et al. Novel mutations in the human sucrase isomaltase gene (SI) that cause congenital carbohydrate malabsorption. Hum Mutat 2006;27(1):119. 102. Chantret I, Rodolosse A, Barbat A, Dussaulx E, Brot Laroche E, Zweibaum A, et al. Differ ential expression of sucrase isomaltase in clones isolated from early and late passages of the cell line Caco2: evidence for glucose dependent negative regulation. J Cell Sci 1994;107 ( Pt 1):21325. 103. Krasinski SD, Van Wering HM, Tannemaat MR, Grand RJ. Differential activation of intestinal gene promoters: functional interactions between GATA 5 and HNF 1 alpha. Am J Physiol Gastrointest Liver Physiol 2001;281(1):G6984. 104. Gu N, Adachi T, Matsunaga T, Tsujimoto G, Ishihara A, Yasuda K, et al. HNF1 alpha participates in glucose regulation of sucrase isomaltase gene expression in epithelial intestinal cells. Biochem Biophys Res Commun 2007;353(3):61722. 105. Schakel SF. Maintaining a nutrient database in a changing marketplace: Keeping pace with changing food products A research perspective. J Food Comp and Anal 2001;14:315322. 106. Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12country reliability and validity. Med S ci Sports Exerc 2003;35(8):138195. 107. IPAQ Research Committee. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ). In; November 2005. p. 15. 108. O'Brien E, Waeber B, Parati G, Staessen J, Myers MG. Blood pressure measuring devices: recommendations of the European Society of Hypertension. BMJ 2001;322(7285):5316. 109. Yahyaoui R, Esteva I, HaroMora JJ, Almaraz MC, Morcillo S, Rojo Martinez G, et al. Effect of longterm administration of cross sex hormone therapy on serum and urinary uric acid in transsexual persons. J Clin Endocrinol Metab 2008;93(6):22303. 110. Shuster J. Design and analysis of experiments. In: Ambrosius W, editor. Topics in Biostatistics. Totowa, NJ: Humana Press; 2007. 111. B ray GA. Soft drink consumption and obesity: it is all about fructose. Curr Opin Lipidol;21(1):517.

PAGE 300

300 112. Nielsen SJ, Popkin BM. Changes in beverage intake between 1977 and 2001. Am J Prev Med 2004;27(3):20510. 113. Parniak MA, Kalant N. Enhancement of gl ycogen concentrations in primary cultures of rat hepatocytes exposed to glucose and fructose. Biochem J 1988;251(3):795802. 114. Yang LY, Kuksis A, Myher JJ, Steiner G. Contribution of de novo fatty acid synthesis to very low density lipoprotein triacylglycerols: evidence from mass isotopomer distribution analysis of fatty acids synthesized from [2H6]ethanol. J Lipid Res 1996;37(2):26274. 115. Mueller NT, Odegaard A, Anderson K, Yuan JM, Gross M, Koh WP, et al. Soft drink and juice consumption and risk of pancreatic cancer: the Singapore Chinese Health Study. Cancer Epidemiol Biomarkers Prev;19(2):44755. 116. Le Gall M, Tobin V, Stolarczyk E, Dalet V, Leturque A, Brot Laroche E. Sugar sensing by enterocytes combines polarity, membrane bound detectors and sugar metabolism. J Cell Physiol 2007;213(3):83443. 117. Truswell AS, Seach JM, Thorburn AW. Incomplete absorption of pure fructose in healthy subjects and the facilitating effect of glucose. Am J Clin Nutr 1988;48(6):142430. 118. Ravich WJ, Bayless TM, Thomas M. Fructose: incomplete intestinal absorption in humans. Gastroenterology 1983;84(1):269. 119. Kneepkens CM, Vonk RJ, Fernandes J. Incomplete intestinal absorption of fructose. Arch Dis Child 1984;59(8):7358. 120. Rumessen JJ, GudmandHoyer E. Absorption capacity of fructose in healthy adults. Comparison with sucrose and its constituent monosaccharides. Gut 1986;27(10):11618. 121. Tsampalieros A, Beauchamp J, Boland M, Mack DR. Dietary fructose intolerance in children and adolescents. Arch Dis Child 2008;93(12):1078. 122. Kellett GL, Brot Laroche E. Apical GLUT2: a major pathway of intestinal sugar absorption. Diabetes 2005;54(10):305662. 123. Gouyon F, Caillaud L, Carriere V, Klein C, Dalet V, Citadelle D, et al. Simplesugar meals target GLUT2 at enterocyte apical membranes to improve sugar absorption: a study in GLUT2null mice. J Physiol 2003;552(Pt 3):82332.

PAGE 301

301 124. Zweibaum A, Triadou N, Kedinger M, Augeron C, RobineLeon S, Pinto M, et al. Sucrase isomaltase: a marker of foetal and malignant epithelial cells of the human colon. Int J Cancer 1983;32(4):40712. 125. Rousset M. The human colon carcinoma cell lines HT29 and Caco 2: two in vitro models for the study of intestinal differentiation. Biochimie 1986;68(9):103540. 126. Hida lgo IJ, Raub TJ, Borchardt RT. Characterization of the human colon carcinoma cell line (Caco 2) as a model system for intestinal epithelial permeability. Gastroenterology 1989;96(3):73649. 127. Mesonero J, Mahraoui L, Matosin M, Rodolosse A, Rousset M, B rot Laroche E. Expression of the hexose transporters GLUT1GLUT5 and SGLT1 in clones of Caco 2 cells. Biochem Soc Trans 1994;22(3):6814. 128. Chantret I, Barbat A, Dussaulx E, Brattain MG, Zweibaum A. Epithelial polarity, villin expression, and enterocyt ic differentiation of cultured human colon carcinoma cells: a survey of twenty cell lines. Cancer Res 1988;48(7):193642. 129. Pappenheimer JR. On the coupling of membrane digestion with intestinal absorption of sugars and amino acids. Am J Physiol 1993;265(3 Pt 1):G40917. 130. Kwon O, Eck P, Chen S, Corpe CP, Lee JH, Kruhlak M, et al. Inhibition of the intestinal glucose transporter GLUT2 by flavonoids. Faseb J 2007;21(2):36677. 131. MatosinMatekalo M, Mesonero JE, Delezay O, Poiree JC, Ilundain AA, Brot Laroche E. Thyroid hormone regulation of the Na+/glucose cotransporter SGLT1 in Caco 2 cells. Biochem J 1998;334 ( Pt 3):63340. 132. Blais A, Aymard P, Lacour B. Paracellular calcium transport across Caco 2 and HT29 cell monolayers. Pflugers Arch 1997;434(3):3005. 133. Markowska M, Oberle R, Juzwin S, Hsu CP, Gryszkiewicz M, Streeter AJ. Optimizing Caco 2 cell monolayers to increase throughput in drug intestinal absorption analysis. J Pharmacol Toxicol Methods 2001;46(1):515. 134. Tang C, Yu J, Y in L, Yin C, Pei Y. Transport of insulin in modified ValiaChien chambers and Caco 2 cell monolayers. Drug Dev Ind Pharm 2007;33(4):44956. 135. Aumklad P. Effect of molecular weight and salt forms of chitosan on epithelial permeability using caco 2 cells Bangkok: SILPAKORN UNIVERSITY; 2006. 136. Hidalgo I. Cultured intestinal epithelial cell models. In: Borchardt RT, Smith, P.L., Wilson, G., editor. Models for assessing drug absorption and metabolism. New York and London: Plenum Press; 1996. p. 3550.

PAGE 302

302 137. Hubatsch I, Ragnarsson EG, Artursson P. Determination of drug permeability and prediction of drug absorption in Caco 2 monolayers. Nat Protoc 2007;2(9):21119. 138. Laitinen L. Caco 2 cell cultures in the assessment of intestinal absorption: Effects of some co administered drugs and natural compounds in biological. Helsinki: University of Helsinki; 2006. 139. Arena A, Phillips J, Blanchard M. Drug transport assays in a 96well system: reproducibility and correlation to human absorption. In. Danvers: Millipore Corporation, Life Science Division; 2003. 140. Mougey EB, Feng H, Castro M, Irvin CG, Lima JJ. Absorption of montelukast is transporter mediated: a common variant of OATP2B1 is associated with reduced plasma concentrations and poor response. Pharmacogenet Genomics 2009;19(2):12938. 141. Kellett GL, Brot Laroche E, Mace OJ, Leturque A. Sugar absorption in the intestine: the role of GLUT2. Annu Rev Nutr 2008;28:35 54. 142. Hirayama BA, Lostao MP, Panayotova Heiermann M, Loo DD, Turk E, Wright EM Kinetic and specificity differences between rat, human, and rabbit Na+ glucose cotransporters (SGLT1). Am J Physiol 1996;270(6 Pt 1):G91926. 143. Khoursandi S, Scharlau D, Herter P, Kuhnen C, Martin D, Kinne RK, et al. Different modes of sodium D gluc ose cotransporter mediated D glucose uptake regulation in Caco 2 cells. Am J Physiol Cell Physiol 2004;287(4):C10417. 144. Kellett GL. The facilitated component of intestinal glucose absorption. J Physiol 2001;531(Pt 3):58595. 145. Leturque A, Brot Lar oche E, Le Gall M, Stolarczyk E, Tobin V. The role of GLUT2 in dietary sugar handling. J Physiol Biochem 2005;61(4):52937. 146. Shi X, Schedl HP, Summers RM, Lambert GP, Chang RT, Xia T, et al. Fructose transport mechanisms in humans. Gastroenterology 19 97;113(4):11719. 147. Miyamoto K, Hase K, Takagi T, Fujii T, Taketani Y, Minami H, et al. Differential responses of intestinal glucose transporter mRNA transcripts to levels of dietary sugars. Biochem J 1993;295 ( Pt 1):2115. 148. Diggle CP, Shires M, Leitch D, Brooke D, Carr IM, Markham AF, et al. Ketohexokinase: expression and localization of the principal fructose metabolizing enzyme. J Histochem Cytochem 2009;57(8):76374.

PAGE 303

303 149. Thorens B, Sarkar HK, Kaback HR, Lodish HF. Cloning and functional expr ession in bacteria of a novel glucose transporter present in liver, intestine, kidney, and betapancreatic islet cells. Cell 1988;55(2):28190. 150. Corpe CP, Basaleh MM, Affleck J, Gould G, Jess TJ, Kellett GL. The regulation of GLUT5 and GLUT2 activity in the adaptation of intestinal brush border fructose transport in diabetes. Pflugers Arch 1996;432(2):192201. 151. Mahraoui L, Takeda J, Mesonero J, Chantret I, Dussaulx E, Bell GI, et al. Regulation of expression of the human fructose transporter (GLUT5) by cyclic AMP. Biochem J 1994;301 ( Pt 1):16975. 152. Sasaki A, Yamaguchi H, Horikoshi Y, Tanaka G, Nakazato Y. Expression of glucose transporter 5 by microglia in human gliomas. Neuropathol Appl Neurobiol 2004;30(5):44755. 153. Schmittgen T, Livak K. Analyzing real time PCR data by the comparative CT method. Nat. Protocols 2008;3(6):11011108. 154. Yuan JS, Reed A, Chen F, Stewart CN, Jr. Statistical analysis of real time PCR data. BMC Bioinformatics 2006;7:85. 155. Schefe JH, Lehmann KE, Buschmann IR, Unger T, Funke Kaiser H. Quantitative real time RT PCR data analysis: current concepts and the novel "gene expression's CT difference" formula. J Mol Med 2006;84(11):90110. 156. Holm S. A Simple Sequentially Rejective Multiple Test Procedure. Scandinavian Journal of Statistics 1979;6(2):6570. 157. CRANE RK. The physiology of the intestinal absorption of sugars. In: James A, Hodges J, editors. Physiological Effects of Food Carbohydrates. Washington, DC; 1975. p. 119. 158. Pappenheimer JR, Reiss K Z. Contribution of solvent drag through intercellular junctions to absorption of nutrients by the small intestine of the rat. J Membr Biol 1987;100(2):12336. 159. Ferraris RP, Yasharpour S, Lloyd KC, Mirzayan R, Diamond JM. Luminal glucose concentrations in the gut under normal conditions. Am J Physiol 1990;259(5 Pt 1):G82237. 160. Brot Laroche E. Differential regulation of the fructose transporters GLUT2 and GLUT5 in the intestinal cell line Caco 2. Proc Nutr Soc 1996;55(1B):2018. 161. Mahraoui L, Rodolosse A, Barbat A, Dussaulx E, Zweibaum A, Rousset M, et al. Presence and differential expression of SGLT1, GLUT1, GLUT2, GLUT3 and

PAGE 304

304 GLUT5 hexose transporter mRNAs in Caco 2 cell clones in relation to cell growth and glucose consumption. Biochem J 1994;298 Pt 3:62933. 162. Rodolosse A, Carriere V, Chantret I, Lacasa M, Zweibaum A, Rousset M. Glucose dependent transcriptional regulation of the human sucrase isomaltase (SI) gene. Biochimie 1997;79(23):11923. 163. Gu N, Adachi T, Takeda J, Aoki N, Tsujim oto G, Ishihara A, et al. Sucrase isomaltase gene expression is inhibited by mutant hepatocyte nuclear factor (HNF) 1alpha and mutant HNF1beta in Caco2 cells. J Nutr Sci Vitaminol (Tokyo) 2006;52(2):10512. 164. Doring A, Gieger C, Mehta D, Gohlke H, Pr okisch H, Coassin S, et al. SLC2A9 influences uric acid concentrations with pronounced sex specific effects. Nat Genet 2008;40(4):4306. 165. Vitart V, Rudan I, Hayward C, Gray NK, Floyd J, Palmer CN, et al. SLC2A9 is a newly identified urate transporter influencing serum urate concentration, urate excretion and gout. Nat Genet 2008;40(4):43742. 166. Cheeseman C. GLUT7: a new intestinal facilitated hexose transporter. Am J Physiol Endocrinol Metab 2008;295(2):E23841. 167. Manolescu A, Salas Burgos AM, Fischbarg J, Cheeseman CI. Identification of a hydrophobic residue as a key determinant of fructose transport by the facilitative hexose transporter SLC2A7 (GLUT7). J Biol Chem 2005;280(52):4297883. 168. Doege H, Bocianski A, Scheepers A, Axer H, Eckel J Joost HG, et al. Characterization of human glucose transporter (GLUT) 11 (encoded by SLC2A11), a novel sugar transport facilitator specifically expressed in heart and skeletal muscle. Biochem J 2001;359(Pt 2):4439. 169. Scheepers A, Schmidt S, Manolesc u A, Cheeseman CI, Bell A, Zahn C, et al. Characterization of the human SLC2A11 (GLUT11) gene: alternative promoter usage, function, expression, and subcellular distribution of three isoforms, and lack of mouse orthologue. Mol Membr Biol 2005;22(4):33951. 170. Ali M, Cox TM. Diverse mutations in the aldolase B gene that underlie the prevalence of hereditary fructose intolerance. Am J Hum Genet 1995;56(4):10025. 171. Davit Spraul A, Costa C, Zater M, Habes D, Berthelot J, Broue P, et al. Hereditary fruct ose intolerance: frequency and spectrum mutations of the aldolase B gene in a large patients cohort from France -identification of eight new mutations. Mol Genet Metab 2008;94(4):4437.

PAGE 305

305 172. Peng SY, Lai PL, Pan HW, Hsiao LP, Hsu HC. Aberrant expression of the glycolytic enzymes aldolase B and type II hexokinase in hepatocellular carcinoma are predictive markers for advanced stage, early recurrence and poor prognosis. Oncol Rep 2008;19(4):104553. 173. Daimon M, Ji G, Saitoh T, Oizumi T, Tominaga M, Nakam ura T, et al. Largescale search of SNPs for type 2 DM susceptibility genes in a Japanese population. Biochem Biophys Res Commun 2003;302(4):7518. 174. Crapo PA, Kolterman OG, Olefsky JM. Effects of oral fructose in normal, diabetic, and impaired glucose tolerance subjects. Diabetes Care 1980;3(5):57582. 175. den Brok MW, Nuijen B, Hillebrand MJ, Grieshaber CK, Harvey MD, Beijnen JH. Development and validation of an LC UV method for the quantification and purity determination of the novel anticancer agent C1311 and its pharmaceutical dosage form. J Pharm Biomed Anal 2005;39(12):4653.

PAGE 306

3 06 BIOGRAPHICAL SKETCH MyPhuong Thi Le was born in Vi t Nam. She received her Bachelor of Art s degree in biology and history from Whitman College in May 1998. She worked at the Genome Center at the University of Washington for two years. Afterwards, she joined Blue Heron Biotechnology, Inc. In August 2005, MyPhuong joined the graduate program in the Department of Pharmaceutics at University of Florida, Gainesville, FL. Under the supervision of her graduate advisor, Dr. Julie Johnson, she completed her dissertation and received her Doctor of Philosophy degree in pharmaceutical sciences in May 2010.