This item is only available as the following downloads:
1 IDENTIFICATION OF BIOMARKER RESPONSES IN HUMANS UNDER EXPERIMENTALLY INDUCED ZINC DEPLETION By MOON SUHN RYU A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQU IREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2011
2 2011 Moon Suh n Ryu
3 To my parents, Ji Chul Ryu and Soon Young Park, and my grandmother J u ng Seo Seo
4 ACKNOWLEDGMENTS The achievements from the past years in my li fe as a doctoral student would not exist without the presence of the support and love from my family far way in South Korea. I thank my parents for their love and belief in me which ha ve been the great est motivation for me to pursue accomplishments and su ccess with my best efforts I also render my gratefulness to my younger sister, Jin Suhn R yu, for being the best supporter in all aspects of my life Additio nally, I thank everyone with whom I shared the moments in the lab, especially Dr. Juan P. Liuzzi Dr. Louis A. Lichten Dr. Liang Guo, Dr. Shou mei Chang, Tolunay Beker Aydemir f or their invaluable advices and help The excitement from our findings w as always the greatest motivation for me to stay in progress I also thank Gregory Guthrie, Alyssa Maki, L uisa Rios and Vanessa Da Silva for their cheer and support, particularly, during the last year of my doctoral research I thank Meena Shankar and the staff members at the General Clinical Research Center for their advices and assistance during the phases of study design and implementation of the dietary regimen. Finally, I fully render my gratefulness to my supervisory committee members, Dr. Robert J. Cousins, Dr. Bobbi Langkamp Henken Dr. James F. Collins and Dr. Nancy D. Denslow for being the b est ment ors and role m odels for my career in the field of nutrition and molecular biology The ir advice ha s been the key for the development of my insights in research and will last long in my minds Especially, I thank Dr. Robert J. Cousins for allowin g me to substantiate my thoughts and ideas by research and for encouraging me to continue my journey in the field of discovery by science. I will truly miss the times under his guidance at the University of Florida, and w ill continue my best with pride o f being a previous member of the Cousins lab.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ .......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 15 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 17 Zinc Biology ................................ ................................ ................................ ............ 17 Regulation of Zinc Homeostasis ................................ ................................ ............. 18 Whole Body Zinc ................................ ................................ .............................. 18 Cellular Zinc ................................ ................................ ................................ ..... 20 Dietary Zinc Deficiency ................................ ................................ ........................... 21 Biomarker Studies of Dietary Zinc Deficiency ................................ ......................... 23 Blood MT and Zinc Transporters ................................ ................................ ...... 24 Cellular Zinc Concentrations ................................ ................................ ............ 25 Other Potential Zinc Indices ................................ ................................ ............. 26 Advances in Techniques for Transcriptome Analysis ................................ ............. 29 Microarray Analysis ................................ ................................ .......................... 29 RNA Sequencing (RNA Seq) ................................ ................................ ........... 30 Sampling and Processing of Biospecimens for Transcriptom e Profiling ................. 31 Study Aims ................................ ................................ ................................ .............. 32 2 MATERIALS AND METHODS ................................ ................................ ................ 34 Human Su bjects ................................ ................................ ................................ ..... 34 Acute Dietary Zinc Depletion ................................ ................................ .................. 34 Sample Collection ................................ ................................ ................................ ... 35 P reparation of Buccal RNA ................................ ................................ ..................... 36 Processing of Blood Samples ................................ ................................ ................. 37 Measures of Serum Zinc Concentrations ................................ ................................ 38 Isolation and Processing of Blood Cell RNA ................................ ........................... 39 RNA Amplification, Hybridization, and Microarray Analysis ................................ .... 39 Microarray Data Analysis ................................ ................................ ........................ 40 Real Time Quantitative PCR ................................ ................................ ................... 41 Western Analysis of RBC Zinc Transporters ................................ ........................... 42 Immunoprecipitation and Mass Spectrometry ................................ ......................... 43
6 Whole Blood Cytokine Assay ................................ ................................ .................. 44 Enzyme Linked Immunosorbent Assay ................................ ................................ ... 44 Isolation and Quantitation of Serum MicroRNA ................................ ...................... 45 Statistical Analysis ................................ ................................ ................................ .. 46 3 IDENTIFICATION OF ZINC RESPONSIVE GENE TRANSCRIPTS IN BUCCAL AND BLOOD CELLS ................................ ................................ .............................. 53 Introductory Remarks ................................ ................................ .............................. 53 Results ................................ ................................ ................................ .................... 55 Experimental Zinc Depletion in Human Subjects ................................ ............. 55 Effects of Zinc Depletion on Zinc Related Tr anscript Levels ............................ 56 Whole Blood RNA Processing and Quality Assessment for Microarray Analysis ................................ ................................ ................................ ......... 57 Global Effect of Zinc Depletion on Whole Blood Transcriptome ....................... 58 Gene Transcripts with Potential of being Biomarkers of Zinc Deficiency .......... 61 Discussion ................................ ................................ ................................ .............. 62 4 EFFECTS OF ACUTE DIETARY ZINC DEPLETION ON WHOLE BLOOD CYTOKINE RELEASE INDUCED BY EX VIVO IMMUNOCHALLENGES ............ 100 Introductory Remarks ................................ ................................ ............................ 100 Results ................................ ................................ ................................ .................. 102 Inflammatory Cytokine Production by Whole Blood ................................ ........ 102 Zinc Transporter Transcript Levels ................................ ................................ 103 Discussion ................................ ................................ ................................ ............ 103 5 SERUM MICRORNA AS BIOMARKERS OF DIETARY ZINC STATUS ............... 114 Introductory Remarks ................................ ................................ ............................ 114 Results ................................ ................................ ................................ .................. 115 Discussion ................................ ................................ ................................ ............ 116 6 ERYTHROCYTE MEMBRANE ZINC TRANSPORTERS AND DEMATIN LEVELS IN HUMANS UNDER SHORT TERM DIETARY ZINC RESTRICTION .. 124 Introductory Remarks ................................ ................................ ............................ 124 Results ................................ ................................ ................................ .................. 125 Erythrocyte Zinc Transporter Expression during Low Zinc Intake .................. 125 Identification of Dematin as a Zinc Responsive Erythrocyte Membrane Protein ................................ ................................ ................................ ......... 125 Discussion ................................ ................................ ................................ ............ 127 7 CONCLUSIONS AND FUTURE DIREC TIONS ................................ .................... 138 APPENDIX A SCREENING QUESTIONNAIRE ................................ ................................ .......... 144
7 B POST PARTICIPATION SURVEY ................................ ................................ ........ 145 C LIST OF GENES DIFFERENTIALLY EXPRESSED BY ACUTE DIETARY ZINC DEPLETION ................................ ................................ ................................ ......... 146 D EFFECTS OF DIETARY ZINC RESTRICTION ON MT AND ZINC TRANSPORTER TRANSCRIPTS IN TONGUE EPITHELIAL CELLS OF MOUSE ................................ ................................ ................................ ................. 157 LIST OF REFERENCES ................................ ................................ ............................. 158 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 171
8 LIST OF TABLES Table page 2 1 Subject characteristics at screening phase (n = 9) ................................ ............. 48 2 2 T he 2 day cycle menu of the acclimation phase ................................ ................. 49 2 3 Dietary components of the liquid diet for the zinc depletion phase ..................... 50 2 4 Mineral contents of the acclimation and zinc repletion diet s measured by inductively coupled plasma optical emission spectrophotometry (ICP OES) ...... 51 3 1 Hematological parameters measured during dietary zinc depletion ................... 73 3 2 Number of differentially expressed genes determined by using various P value and fold change combinations ................................ ................................ .. 84 3 3 Gene ontology (GO) enrichement by genes res ponsive to acute dietary zinc depletion ................................ ................................ ................................ ............. 89 3 4 Cis regulatory elements enriched in the putative promoter regions ( 1000 ~+200) of genes responsive to dietary zinc depletion ................................ ......... 90 3 5 Top 5 functions over represented by all genes differentially expressed after acute dietary zinc depletion ................................ ................................ ................ 92 3 6 Top 5 functions over repres ented by genes up regulated after acute dietary zinc depletion ................................ ................................ ................................ ...... 93 3 7 Top 5 functions over represented by genes down regulated after acute dietary zinc depletion ................................ ................................ .......................... 94 3 8 Genes holding the potential as an indicator of dietary zinc deficiency in individuals ................................ ................................ ................................ ........... 98 5 1 Zinc related genes targeted by the zinc responsive serum m iRNA .................. 123 C 1 List of genes up regulated by acute dietary zinc depltion ranked by fold changes (ZnD/Baseline values) ................................ ................................ ........ 146 C 2 List of genes down regulated by acute dietary zinc depltion ranked by fold changes (ZnD/Baseline values) ................................ ................................ ........ 152
9 LIST OF FIGURES Figure page 2 1 Schematic diagram of sample co llection and handling processes ..................... 52 3 1 Dietary regimen for acute dietary zinc depletion. ................................ ................ 71 3 2 Measures of serum zinc concentrations indicating the effectiveness of the experimental diet for dietary zinc depletion. ................................ ....................... 72 3 3 Effects of acute dietary zinc depletion on buccal MT and ZnT1 transcript abu ndance. ................................ ................................ ................................ ......... 74 3 4 Effects of acute dietary zinc depletion on zinc related gene transcripts in p eripheral blood mononuclear cells (PBMC). ................................ ..................... 75 3 5 Effects of acute dietary zinc depletion on zinc related gene transcripts in circulating reticulocytes. ................................ ................................ ..................... 76 3 6 Association between serum zinc concentrations and transcript abundanc e of the zinc transporters most significantly responsive to dietary zinc depletion. ..... 77 3 7 Effect of acute dietary zinc depletion on whole b lood ZnT1 transcript levels. ..... 78 3 8 Quality assessment of whole blood RNA. ................................ ........................... 79 3 9 RNA recovery after processing for microarray analysis. ................................ ..... 80 3 10 Quality of amplified cRNA assesse d by Agilent 2100 Bioanalyzer. .................... 81 3 11 Quality assessment plots of microarray data generate d by Genome Studio. ..... 82 3 12 Scatter plots of microarray data before and after normalization of pooled whole blood RNA depleted of globin RNA (GRP) (Day 7 vs. Day17) ................ 83 3 13 Differential expression of 328 genes identified by pairwise comparisons at P < 0.005. ................................ ................................ ................................ .............. 85 3 14 Effects of globin RNA reduction on the detection of differentially expressed genes. ................................ ................................ ................................ ................. 86 3 15 Temporal expression pattern of differentially expressed genes during dietary zinc depletion. ................................ ................................ ................................ ..... 87 3 16 Clustering of zinc respons ive genes by expression patterns for functional interpretation. ................................ ................................ ................................ ...... 88
10 3 17 Functional networks of genes differentially expressed after acute dietary zinc depletion. ................................ ................................ ................................ ............ 95 3 18 Functional network enriched by genes up regulated and down regulated after acute dietary zinc depletion. ................................ ................................ ............... 96 3 19 Differential expression of 203 genes identified by unpaired t test at P < 0.005. ................................ ................................ ................................ ................. 97 3 20 Valid ation of microarray data using qPCR ................................ ......................... 99 4 1 Effects of acute dietary zi nc depletion on whole blood cytokine production induced by LPS and PHA in vitro ................................ ................................ ..... 108 4 2 Confirmation of LPS induced monocyte activation in whole blood in vitro ....... 109 4 3 Confirmation of PHA induced lymphocyte activation in whole blood in vitro .... 110 4 4 Confirmation of the repression in LPS and PHA ion by acute dietary zinc depletion. ................................ ................................ ............. 111 4 5 Effects of residual heparin in RNA samples on PCR amplification. .................. 112 4 6 Effects of immunostimulation on zinc related gene transcripts in whole blood. 113 5 1 Serum processing for miRNA isolation. ................................ ............................ 120 5 2 Identi fication of serum miRNA s responsive to acute dietary zinc depletion using a qPCR based array ................................ ................................ .............. 121 5 3 Effects of dietary zinc intake levels on circulating serum miRNA levels .......... 122 6 1 Zinc Transporter Expression in Human Erythrocyte Membrane. ...................... 133 6 2 Effects of acute dietary zinc depletion on zinc transporter expression in human erythrocytes. ................................ ................................ ......................... 134 6 3 Isolation of the peptide producing a non specific band with the Zip8 antibody by immunoprecipitation. ................................ ................................ .................... 135 6 4 Identification of the protein non specifically detected by the hZip8 antibody ... 136 6 5 Confirmation of protein identity determined by mass spectrometry anal ysis ... 137 D 1 Effects of dietary zinc deprivation on zinc transporter mRNA levels in t he tongue epithelium of mouse ................................ ................................ ............ 157
11 LIST OF ABBREVIATION S AAS A tomic A bsorption S pectrophotome try ANOVA Analysis of Variance AP Affinity Purified Antibody AP 2 Activator P rotein 2 Alpha CAMP Cyclic A denosine M onophosphate CBC Com plete Blood Count CDNA Complementary Deoxyribonucleic A cid CGMP Cyclic G uanosine M onophosphate CHCM Cell H emoglobin C oncentration M ean CRNA Complementary Ribonucleic Acid CT T hreshold C ycle d Days DE Differentially Expressed DNA Deoxyribonucleic A cid EDTA Ethylenediaminetetraacetic A cid EKLF Erythroid Krppel L ike Factor ELISA Enzyme L inked I mmunosorbent A ssay ELK 1 ETS L ike G ene 1 EPO E rythropoietin ETF E pidermal G rowth F actor R eceptor S pecific T ranscription F actor EXPANDER E x pression Analyzer and Display er FC Fold Change FDR False Discovery Rate GAPDH Glyceraldehyde 3 P hosphate D ehydrogenase
12 GCRC General Clinical Research Center GO Gene Ontology GRP Globin Transcript Reduced PAXgene Rib onucleic Acid ICP OES I nductively C oupled P lasma O ptical E mission S pectrophotometry IFN Interferon IgG Immunoglobulin G IL Interleukin IPA Ingenuity Pathway Analysis IRB Institutional Review Board k Da Kilodalton KLF4 Krppel L ike Factor 4 LC MS /MS L iquid C hromatography C oupled with T andem M ass S pectrometry LiCl Li thium Chloride LPS L ipopolysaccharides MCHC Mean Corpuscular Hemoglobin Concentration m in Minute m i RNA Micro Ribonucleic A cid MRE Metal Response Element m RNA Messenger Ribonucleic A cid MT Metalloth ionein MTF 1 M etal R egulatory T ranscription F actor 1 MW Molecular Weight NDSR Nutrition Data System for Research NF Nuclear Factor PAX PAXgene Whole Blood Ribonucleic Acid
13 PBMC P eripheral B lood M ononuclear C ell s PCT P robability of C onserved T argeting PDE P hosphodiesterase PHA P hytohemagglutinin PKA Cyclic AMP D ependent P rotein K inase PKC Protein Kinase C Pre m iRNA Precursor MicroRNA PRIMA P r omoter Integration in Microarray Analysis Pri m iRNA Primary MicroRNA QPCR Quantitative Real Time Polymerase Chain Reac tion RDA Recommended Daily Allowance RISC RNA Induced Silencing Complex RNA Ribonucleic A cid RRNA Ribosomal Ribonucleic A cid RT Reverse Transcription SD Standard Deviation SEM Standard Error of the Mean s SLC S olute C arrier TANGO Tool for A n alysis of G ene O ntology E nrichments TEF TEA D omain F amily M ember 2 TNF Tumor Necrosis Factor UTR Untranslated Region VEGF Vascular E ndothelial G rowth F actor v ol Volume wt Weight
14 ZIP Zrt / Irt L ike P rotein ZnA Zinc Adequate ZnD Zinc Depleted ZnT Zinc Transporter
15 Abstract o f Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy I DENTIFICATION OF BIOMARKER RESPONS ES IN HUMANS UNDER EXPERIMENTALLY INDUCED ZINC DEPLETION By Moon Suhn Ryu August 2011 Chair: Robert J. Cousins Major: Nutritional Sciences Zinc, an essential micronutrient, functions as a catalytic, structural and regulatory component for numerous metabolic processes. The estimated prevalen ce of zinc deficiency worldwide remains substantially high at 31%. The lack of a reliable biomarker limits the identification of individuals requir ing zinc interventions and thus a specific and sensitive assessment tool for such condition is greatly need ed. To identify candidate markers holding the potential t o indicate zinc status, a 24 d observational study compris ed of acclimation, zinc depletion a nd repletion phases was conducted wi th healthy male subjects Serum zinc level s were monitored througho ut the 24 d period. On d ay 0, 6 and 10 of zinc depletion, buccal swabs and whole blood were c ollected for various molecular assays. Decreased serum zinc concentrations confirmed the zinc depleted status of each participant along with a reduction in bucca l MT transcript levels. A bundance of MT, ZnT 1, ZnT 4 and ZnT 5 mRNA levels in p eripheral blood mononuclear cells ( PBMC ) and ZnT 1 and Zip 3 transcripts in purified reticulocytes significantly decreased by dietary zinc depletion. The reduction of ZnT 1 mRNA le vels was measu rable in whole blood RNA Additionally, the presence of ZnT 1 and Zip 10 in the plasma membrane of erythrocytes, of which differential expression by
16 zinc deficienc y has been shown in mice was confirmed Increase in membrane dematin levels wa s observed in erythrocytes collected after zinc depletion. Microarray analys i s with globin RNA deple ted whole blood RNA revealed 3 28 genes responsive to acute die tary zinc depletion. Bioinformatic analysis identified enriched pathw ays associated with cel l cycle and immunity by the up regulated and down regulated genes, respectively. Depression i n immunostimulated TNF release by dietary zinc depletion confirmed the functional significance of zinc in immune responses. Furthermore, responses of serum miRN A to dietary zinc levels were measur ed Based on a comprehensive analysis of specimens from zinc depleted human subjects gene transcripts and proteins holding the potential of being zinc bioma rkers were identified The se results also implicate d i mpaire d host defense and predisposition to cancer development as expected outcomes of dietary zinc deficiency T his trial was registered at clinicaltrials.gov as NCT01221129
17 CHAPTER 1 INTRODUCTION Zinc Biology Zinc, an essential micronutrient ubiquitously dis tributed in the human body, is required for numerous metabolic processes in the biological system as a catalytic, structural and regulatory component ( 1 ) The absence of redox chemistry of zinc which is in contrast to the properties of coppe r and iron, allows this trace element to be involved in numerou s physiologic events without introducing the risk of oxidative damage. Such events include cell growth and differentiation, DNA and RNA synthesis, bone formation, cell mediated immune responses and gene expression. Z inc mediates enzymatic reaction s at the catalytic site and /or serves as a key element for the maintenance of structural integrity of various enzyme s The activity of metallo enzymes such as carbonic anhydrase, alkaline phosphatase and carboxypeptidase A has been shown but not consistantly to de crease under dietary zinc deficiency, which implicates the critical rol e of zinc for the function of these enzymes ( 1 2 ) Configurational changes of zinc finger transcription factors produc ed by zinc binding allow for t he DNA binding activity of these proteins. One of the most extensively studied regulatory role s of zinc involves control of the activity of a zinc sensing transcription, metal regulatory transcription factor 1 (MTF 1) ( 3 7 ) Upon interaction with cytosolic zinc, MTF 1 translocates to the nucleus and functions as a transregulator of genes with metal response elements (MRE) at their promo ter region. Such genes include metallothionein (MT) and ZnT1, for both of which expression is up regulated by an increase in zinc availability.
18 Regulation of Zinc Homeostasis The human body holds the capability to maintain the homeostasis of zinc by regu latory mechanisms involving differential expression of zinc transporters by zinc deficient or excessive conditions ( 8 9 ) The zinc transporter family residing in intracellular membranes or the plasma membranes of cells, is composed of 24 genes which can be categorized as 10 SLC30A ( Zinc Transporter; ZnT) and 14 SLC39A ( Zrt / Irt like protein ; Zip ) subfamilies. Th e ZnT transporters mostly mediate the down regulation of zinc availability in the cytosolic region of cells by exporting and retaining zinc to extracellular and vesicular compartments, respectively. Conversely, Zip transporters function as a zinc channel t hat provides zinc in to the cytosolic region of cells by delivering zinc molecules in an opposite manner to that of ZnT The molecular mechanism of homeostatic regulation of zinc in response to zinc restriction or excess involve s the differential expressio n or localization of these transporters as described below. Whole Body Zinc Based on observations from cell based, animal and human studies, the small intestine and pancreas play the major role in the maintenance of body zinc homeostasis by the regulation of absorption and excretion rates under zinc deficiency and excess conditions ( 9 ) The underlying mechanism involves the zinc responsive expression and cellular transloca tion i.e., transcript ional and post translational regulation, respect ively, of the zinc transporters present in enterocytes or pancreatic acinar cells Under conditions which reduce the availability of zinc such as low dietary zinc levels or the presence of zinc chelators intestinal zinc transport system responds to dieta ry zinc
19 restriction towards the direction for enhanced zinc absorption while that of the pancreas is modulated to minimize the loss of zinc via its secretory pathway. Zip 4, Zip 5 and ZnT1 are the major zinc transporters involved in the trafficking of dietar y zinc through the intestinal epithelium ( 9 ) While apical Zip 4 and basolateral ZnT1 mediate the transport of dietary zinc from the intestinal lumen to the bloodstream, basolateral Zip 5 functions as a route of zinc removal from the circulat ion to intestinal cells. When the intestinal cells are exposed to a z inc restricted condition due to low zinc intake or presence of zinc chelators, an increase in Zip 4 gene expression by the induction of Krppel like factor 4 ( KLF4 ) expression and the DNA binding activity of this transactivator to the promoter region of Zip 4 occur ( 10 ) Under such condition s the basolateral importer Zip 5 is internalized and degrad ed, which results in reduced transfer of zinc from the circulation to intestinal cells ( 11 ) Conversely, down regulation of Zip 4 transcription along with the internalization of its protein product has been shown as the regu latory mechanism for reducing the zinc uptake rate by intestinal cells when the dietary zinc bioavailability is in excess. The level of ZnT1, which mediates zinc transport across the b a solateral membrane from intestinal cells to blood, is refractory to the bioavailability of dietary zinc ( 12 ) P ancrea tic acinar cell s play a significant role in th e regulation of fecal zinc excretion, which is the primary route of endogenous zinc loss ( 9 ) The relevant mechanism of zinc excretion involves zinc transporters Zip 5, ZnT1 and ZnT2 that respond to zinc, which enables the regulation of the zinc loss rate by the host s zinc status. Acinar cells mediate the excretion of zinc via the uptake of zinc from the circulation by basolateral Zip 5 ( 11 13 ) and the efflux through ZnT1 at the apical plasma
20 membrane or by a secretory pathway involving vesicular ZnT2 activity ( 7 12 ) Under zinc restricted conditions the route of zinc excretion across acinar cells is compromised via a decrease in baso lateral Zip 5 levels and reduction in apical ZnT1 and secretory vesic ular ZnT2 levels The regulatory mechanism of pancreatic Zip 5 expression mimics that of the intestinal Zip 5, i.e., posttranslational regulation by internalization and subsequent degradation in low zinc status ( 11 ) Recently, MTF 1 has been shown to be the key regulator of both pancreatic acinar zinc exporters involved in the efflux of zinc to the intestinal lumen, ZnT1 and ZnT2, by its zinc respo nsive bin ding activity to MRE present in the promoter region of each respective gene ( 7 ) Consequently, zinc excess, which induces the nuclear translocation of MTF 1, enhances both ZnT1 and ZnT2 gene expression and eventually facilitates the zinc efflux rate by up regulat ing ZnT1 and ZnT2 activities. Cellular Zinc Independent o f the regulatory mechanism of whole bod y zinc, each cell or tissue ha s molecular mechanisms for the adaptation to l ow or excessive zinc conditions I n general, when cells are exposed to a zinc restricted environment, up regulation of Zip transporters along with down regulation of the Zn T exporters occurs as a means to maximize the rete ntion of cytosolic zinc content ( 8 9 ) Typical examples include the increase in Zip10 and the decrease in ZnT1 expression levels on the plasma membrane of erythrocytes, which may contribute to the enhanced cellular zinc uptake rate in mice fed a low zinc diet ( 14 ) Additionally, the decrease in hepatic Zip10, at both transcript and protein levels ( 6 ) and increase in ZnT1 protein levels by oral dosing of zinc ( 15 ) reflect the critical role of these zinc transporters in the regulatory mechanism of cellular zinc homeostasis. However, i t is of note that not all transporters have been shown to be
21 responsive to z inc status and the magnitude or mode of zinc responsiveness can differ by the type of cell or tissue in which the gene is expressed ( 8 ) For instance, while the production of Zip 4 transcripts in the small intestine is markedly induced by dietary zinc depletion in mice no changes in the transcript level occurs in the pancreas ( 12 ) Furthermore, hepatic Zip 4 transcript levels reflect dietary zinc deficiency in a manner opposite to that in enterocytes, i.e., do wn regulation during deficiency ( 6 ) Thus, even though it is a general consensus that differential expression of zinc transporters exerts the adaptive mechanism of cells to modulated zinc conditio ns, cell type specific ity of zinc responsiveness of each zinc transporter is of importance as well Dietary Zinc Deficiency The current recommended daily allowance (RDA) of dietary zinc intake for the adult male and female are 11 and 8 mg/d respectively ( 1 ) RDA for females under pregnancy or lactation are set at higher leve ls (i.e., 11 and 12 mg/d) due to the loss of zinc by fetal accumulation and zinc secretion by milk. Since the first report in the early 1960s, human dietary zinc deficiency has been associated with various clinical manifestations including impaired growth diarrhea, dermatitis, neurological and immunological defects, and cancer development ( 2 16 18 ) Conversely, protective effects of adequate zinc intake on the risk of various clinical outc omes, such as type 2 diabetes ( 19 20 ) and its essential role in immunity ( 21 22 ) have been reported by epidemiologic in vitro and nutrition al biochemistry studies. Even though the importance of adequate zinc intake has been known for around half a century, the global prevalence of zinc deficiency predicted by food based estimates remains a s high as 31% ( 23 ) Additionally, t he d isabi lity adjusted life years a ttributable to zinc deficiency
22 have been estimated to be above 28 million of which cases are mostly associated with diarrhea, pneumonia and malaria deaths ( 23 ) In contrast to iron of which the majority is stored and recycled via the liver and reticuloendothelial system there is no reserve or recycling me chanism for zinc ( 1 ) Thus, adequate uniform dietary zinc intake is essential for the human body to acquire and maintain the syst emic pool size of zinc in an appropriate range. The significance of dietary zinc absorption has been characterized by the deficient zinc status of patient s suffering from acrodermatitis enteropathica (AE) which is due to mutations in a zinc transporter ge ne, Zip 4 ( 24 25 ) This protein mediates zinc transport through the apical membrane of the intestinal cell. The sever e zinc deficiency produced by this autosomal recessive disorder include s symptoms as dermatological lesions, diarrhea lac k of weight gain, and impairment in immune and reproductive systems ( 26 27 ) It is of note that these symptoms can be co rrected by supplemental zinc, through which absorption occurs by paracellular transport or via other l ower affinity zinc trans porters functioning in the intestinal epithelium ( 28 ) Extensive consumption of diets with hig h contents of phytic acid ( Inositol hexa phosphate ) may lead to dietary zinc deficiency even in cases when adequate zinc is present in the diet. The tenacious zinc binding nature of phytic acid binding leads to formation of an insoluble zinc phytate comple x which decreases the bioavailability of ingested zinc in the small intestine ( 29 ) Thus, the molar ratio between phytate and zinc may serve as a useful index of zin c bioavailability At a phytate to zinc molar ratio of 10, the solubility of zinc has been shown to be reduced by 98% in vitro and the detrimental effect of phytate on zinc absorption has been observed by a significant decrease in
23 plasma zinc concentratio ns of rats ( 30 ) A decrease in zinc absorption resulting in a negative zinc balance has been reported in human males fed a diet with a phytate/zinc molar ratio of 15 ( 29 ) Recently, a mathematic model for the estimation of quantitative effects of dietary phytate on the bioavailability of zinc was developed based on previous dietary studies reporting the phytate content of diet and measure s of the fractional zinc absorbed ( 31 ) Based on this model, the absorption of zinc in the absence of phytates is estimated to be approximately 4.5 mg/d when the amount of dietary zinc meets the current RDA for males ( 32 ) By the presence of 1000 mg/d of phytic acid, the estimated physiological requirement becomes 2 fold that of a phytate free diet. This implies the necessity of suffic ient zinc intake by individuals on a diet with high phytate contents, such as vegetarians. Biomarker Studies of Dietary Zinc Deficiency Serum (or plasma) zinc concentration is currently the only biomolecular tool recommended by the World Health Organizatio n (WHO) for the community based assessment of dietary zinc status in epidemi ological nutrition studies ( 33 ) As supported by numerous dietary human studies, seru m zinc concentrations can successfully reflect the ingested levels of zinc in a dose dependent manner ( 34 37 ) T he currently suggested lower cut off value s for heal thy male and female adults un der morning fasting sta t e are 74 and 70 g/dL respectively ( 38 ) However, due to its lack of specificity, the use of this indicator is generally limited to experimental conditions of which subjects are in a healthy sta t e and under well controlled condition s preventing other variables that can affect serum zinc measures. For instance, a decrease in serum zinc levels can be caused by inflammation and infection which are prevalent conditions in most sites where nutri ent status assessment is needed ( 33 ) In other words serum
24 zinc concentration remains inadequate for the diagnosis of dietary z inc deficiency in indi viduals and the identification of a n alternative reliable laboratory biomarker of zinc status remains as an essential task for the field of zinc nutrition. Blood MT and Zinc Transporters D ata from a c onsiderable amount of research with in vitro and in viv o models suggest th at gene products related to zinc metabolism can reflect the host s dietary zinc status. MT transcript and protein levels, respectively, respond to the cellular zinc availability by a transcriptional and posttranslational regulatory mech anism and thus hold the potential as biomarker s of zinc deficiency. Und er the zinc depleted condition gene expression of MT decreases due to reduction in translocation of its transactivator, MTF 1, to the nucleus ( 3 ) Both marginal and supplemental intake s of zinc have been shown to decrease and increase blood MT transcript levels in human s respectively ( 39 41 ) Posttranslational regulation of MT by zinc depletion involves the conformational change by metal release which leads an increase in its susceptibility to proteoly sis In accordance to this, the MT protein level i n human erythrocytes was shown to decrease after subjects were fed a zinc restricted diet ( 42 43 ) Additional ly, erythrocyte lysates at the zinc depleted sta t e exhibited higher degradation capabilities for MT protein in vitro ( 42 ) In addition to MT, zinc transporter transcript levels in blood cel ls have been shown t o respond to supplemental zinc Higher levels of ZnT1 transcripts along with lower abundance in Zip 3 mRNA were measured in whole blood RNA isolated from dried blood spots of male adult under zinc supplementation ( 41 ) Additionally, d own regulation in Zip 1 mRNA of peripheral leukocytes has been reported as a consequence of dietary zinc supplementat ion to elderly women ( 44 ) However, due to the absence of
25 experimental confirmation of the effects of dietary zinc depletion on these zinc transporter s in human s the potential of these transporters to identify dietary zinc depletion remains unclear. Cellula r Zinc Concentrations Extensive research o n the prop erties of cellular zinc content of these cells have been conducted for the evaluation of their potential of being assessment tools of zinc status Comprehensive analysis of the effects of diet ary zinc de pletion on the zinc concentration of blood cells indicate these parameters lack the ability to serve as indices of zinc deficiency ( 45 ) No significant changes in erythrocyte, neutrophil or platel et zinc content were identified while effects of the experimental zinc treatment (0. 6 mg/d, 1 week; 4 mg/d, 6 weeks) on other biological indices such as plasma zinc concentrations were present However, the effect of dietary zinc depletion on e rythrocyte zinc remains controversial. When healthy male subjects were fed a low zinc diet (0.55 mg/d, 12 days) a significant decrease in erythrocyte zinc was produced regardless of the amount of zinc ingest ed during the preceding phase of the study ( 43 ) The discrepancy between these two studies may be due to the dietary zinc conten t of the diet provided during the depletion period It is also of importance to not e that citric acid, which cause s osmotic redistribution of fluid between plasma and cells and thus modulate s plasma or cellular metabolite conc en trations ( 46 ) was used in the first study while the latter used heparin for anticoagulation. Another population based study characterizing zinc parameters with heparinized blood from elderly subjects showed lower erythrocyte zinc content in subjects with zinc scores (= freq uency x quantity x zinc content calculated by a food frequency questionnaire) below 134 ( 47 ) Increa se in the red cell zinc content was observed after zinc supplementation as well Thus, further
26 validation studies focused on the selection of anticoagulants are required to determine whe the r the erythrocyte zinc concentration is a valid index of zin c deficiency The effects of dietary zinc depletion and supplementation on white blood ce lls remain controversial due to inconsistent results Increase s in lymphocyte and granulocyte zinc by prolonged zi nc supplementation (3 years) have been observed in i ndividuals with sickle cell disease ( 48 ) Additionally, dietary zinc restriction resulted in a decrease in both lymphocyte and granulocyte zinc concentrations of healthy male adults ( 49 ) However, no effects of dietary zinc content on mononuclear white cells and neutrophils in postmenopausal women were identified after consump tion of a zinc restricted diet for 6 months ( 50 ) T he practicality of using isolated blood cel ls for zinc measures is limited due to the technical complexity of sample processing F ractionation of leukocytes from whole blood requires extensive exposure of the ce lls to exogenous conditions which may influence the cellular zinc trafficking and reten tion ex vivo Additionally, recent research has sh own the effects of immune response s on zinc transporter expression of T cells ( 51 ) and dendritic cells ( 51 52 ) which impl y redistribution of the cellu lar zinc contents during inflammation and infection. Thus, as for the case of serum zinc concentrations, the responsiveness of cellular zinc of the white blood cells may not be specific to dietary zinc deficiency. Ot her Potential Zinc Indices Zinc is essential for the activity of various enzymes and, thus, it has been hypothesized that these metalloenzymes may reflect the host s zinc status. The enzymatic activity of serum a lkaline phosphatase, which requires zinc as a catalytic component, has been shown to be affected by dietary zinc deficiency in rats ( 53 54 )
27 Data from human studies partially support this observation with a decrease in the metalloenzyme activity by low zinc status indicated by serum zinc concentrations ( 55 56 ) However, the sensitivity of this index to identify zinc deficiency is questionable due to the lack of responses in experimental models with human s under mild dietary zinc restriction ( 34 ) Additionally, because of its responsiveness to absolute zinc deficie ncy caused by inflammation ( 57 ) other complementary means for the discrimination of zinc deficiency by low intake levels from other cases leading to a decrease in serum zinc is required T he validity of other metalloenzymes, e.g. D mannosidase ( 34 45 ) 5 nucl e otidase ( 8 34 58 ) and superoxide dismutase ( 59 60 ) as indicator s of dietary zinc deficiency has been investigated with animal and human models as well. However, according to the inconsistency in the r esults, there is no consensus for any zin c dependent enzyme as a biomarker for dietary zinc deficiency as yet One of the most broadly studied properties of zinc is its regulatory function in immune and inflammatory responses. The recently discovered essentiality of zinc transporters for the act ivation of immune cells, such as Zip 8 and Zip 6 for T cells and dendritic cells, respectively, imply the critical role of zinc in immunity ( 51 52 ) In accordance to this, increase s in zinc availability has been shown to result in induction of interleukin 1 beta ( IL 1 ) and tumor necrosis factor alpha ( TNF ) release by PBMC in vitro ( 61 62 ) It is of note that none of the other metals s haring structural similarity with zinc (i.e., calcium, magnesium, cobalt nickel and mercury) were able to s ignificantly induce the release of the se cytokine s by PBMC ( 62 ) Modulation in serum cytokine profiles by experimental depletion of diet ary zinc in human suggests immunoregulatory properties of zinc in vivo ( 63 ) Association of IL 2 levels with serum zinc concentrations
28 ( 64 ) and its responsiveness to dietary zinc supplementation in elderly subjects ( 65 ) implicate the properties of this cytokine as a diagn ostic tool f or dietary zinc deficiency. Additionally, d ietary zinc supplementation to healthy male adults enhance s the capability of immune cells to be activate d in response to immunostimulation ( 41 ) Thus, functional analyse s focusing on immune responses of blood cells may serve as means for identifying dietary zinc depletion, particularly when responses to low zinc are opposite to those produced by infection or inflammation. Another biological function of zinc is its involvement in metabolic processes related to cancer development. Overexpression or repression of certain zinc transporters has been identified as characteristics of various cancers and their mod ulated expression levels have been associated with the metastatic stage of each cancer type ( 8 9 ) These imply the therapeutic properties of zinc as well as the feasibility to apply approaches for cancer biomarker research onto the identification of diagnostic tools of dietary zinc deficiency. Recently, multiple zinc responsive defense mechanisms against oxidativ e stress and DNA damage have been identified by using a zinc deficient animal model ( 66 ) or human prostate epithelial cells ( 67 ) Additionally, DNA integrity measured by c omet assays, w as low er in PBMC s collected after the host human subjects were fed a diet leading to mild dietary zinc deficiency than in those collected at the baseline ( 68 ) The effect of zinc depletion was reversed by dietary zinc supplementation. Thus, such a technique would be applicable to the identification of subclinical zinc deficie ncy in healthy individuals that have been confirmed to be free from cancer development (which may be assessed by other cancer specific diagnostic tools)
29 Advances in Techniques for Transcriptome Analysis The initial phase of b iomarker research includes the identification of biomolecules that differentially respond to the clinical or nutritional condition of interest. Human genome project and consecutive and continuous updates to the genome database provide the list of gene s expressed in human cells. Novel approaches allow comprehensive analysis of whole transcriptome for a simultaneous detection of multiple responsive gene products in a cost effective and high throughput manner Microarray Analysis DNA microarray technology has been considered as the gold standard approach for a comprehensive analysis of the whole gene expression pattern ( 69 ) The technology is based on the detection of transcripts mediated by their hybridization to gene specific probes placed on glass plat es, beads or nitrocellulose etc., and the validity of this approach has been investigated extensively, e.g., by the Microarray Quality Control project ( 70 ) The principle of detecting differe ntial ly expressed genes within the transcriptome depends on the number of f l uorescent dyes used during visualization Single channel microarray allows comparisons among transcript profiles of more than two samples Signal intensities of transcripts in eac h RNA sample are produced by a single dye and are measure d individually. Thus, appropriate means of normalization prior to comparison is required to minimize the variables that may have been introduced during the sample processing and measurement steps In contrast to the single dye approach, the dual channel microarray system is suitable for comparison between two samples Each sample is labeled with d ifferent detection dyes and combin ed prior to hybridization to the microarray chip. Simultaneous asses sment of signals produced by each dye labeling the treatment and control sample enables the
30 identification of affected transcripts by detecting higher or lower intensities originating from the dye labeling of treatment samples Even t hough the use of thi s technique led to success in identification of numerous genes responding to experimental treatments under laboratory environments the application to clinical settings has been limited due to its high cost and labor intensive nature. A recently introduce d bead based platform by Illumina has been suggested to hold advantages over other platforms with regards of its practicality for the use in clinical fields due to its relatively lower cost per sample and its multiplexing feature which allows simultaneous scanning of up to 12 samples ( 71 73 ) Other key features of the beadchip system versus the classical Affymetrix platform include a lower requirement of starting m aterial (50 ng vs. 5 g total RNA) and its higher stringency (50 mer vs. 25 mer probes) in detection. R ecent research for the development of a standardized protocol for blood based biomarker discovery su pports the use of beadchip arrays due to its outperformance over other mi croarray platforms ( 6 ) RNA Sequencing (RNA S eq) The currently most advanced technique for transcriptome profiling is based on direct sequencing of RNA molecul e s present i n the sample of interest and holds multiple advantages over the hybridization based approaches for gene expression analysis (i.e., DNA microarray) ( 74 ) The principle steps of deep sequencing technology are 1) generation of fragmented cDNA with adaptor ligation, 2) next generation sequencing and 3) mapping by seq uence alignment Due to its nature of direct sequencing, RNA Seq allows absolute quantita t ion of indiv idual transcripts enabling direct comparison among multiple samples in the absence of a ny reference s or standards. Other novel features include identifi cation of polymorphisms within the
31 coding sequence of each gene that may lead clinical manifestations due to defective protein products. In other w ords, RNA seq allows the charact erization of transcriptome in posttranscriptional perspectives in addition t o that at the transcription level. However, due to its novelty and active stage of development substantial costs are yet required which limits its practicality to laboratory experiments with small sample sizes. Sampling and Processing of Biospecimens for Transcriptome Profiling Two of the most noninvasive collection means of biological sampling from human, and thus widely used approach for clinical diagnosis, are blood draws and buccal swabs ( 75 ) However, limitations in the use of these samples for transcript ana lysis exist due to the requirement of extensive care during the processing steps after sampling. In the absence of RNA stabilization methods, the transcriptome profile of blood cells can be modulated by multiple technical factors such as the processing or storage temperature and time of exposure to the respective temperature ( 76 77 ) Additionally, high content of RNase in saliva limit the use of buccal swabs as a source of RNA samples ( 78 79 ) In other words, immedi ate lab processing and a cold chain system for storage, processing and transportation have been considered as prerequisites for reliable transcript measures with minimal ex vivo effects, i.e., preservation of the in vivo transcriptome profile. Sites at whi ch nutrient status assessment would be conducted, e.g., communities of a developing country, mostly lack the availability of certain facilities that are required for immediate processing or sample stabilization ( 75 ) Particularly for cases of multiple center clinic al practices, the time lag between sample collection and processing may vary and thus contribute to the variable of transcript measures. Thus, optimal methods for biomarker studies would be those that allow maximum stability of the target molecule
32 of int erest for an extensive period of time under ambient temperature. The PAXgene Blood RNA system developed by PreAnalytiX allows the preservation of blood samples for up to 3 days at room temperature or 5 days at 2~8 C without compromising RNA integrity ( 76 ) By direct collection of blood into tubes with PAXgene reagent, the transcriptome profile of whole blood is immediately stabilized and thus minimal ex vivo effects are introduce after blood collection. Another recently developed technique for nucleic a cid stabilization is based on a filter paper composed of microfibers that denature proteins, including nucleases ( 80 ) Even though the major application of this platform is for preservation of DNA contents in plant or mammalian samples it has been su ccessfully implement ed as templates of dried blood spots that can be used for whole blood RNA assays ( 40 41 ) It is of note that globin mRNA, highly abundant in whole blood RNA (>70%), has been shown to mask the presence of transcripts, particularly those at low abundance, during gene expression analysis ( 71 73 81 ) In order to avoid the effects of globin RNA on transcriptome analysis, fractionation of PBMC from whole blood may be conducted in cases of experim ents c arried out in laboratories ( 77 ) However, this approach may not be practical for field applications due to the requirement of immediate processing and the u se a refrigerated centrifuge machine. Alternatively, strategies to reduce globin transcript levels in whole blood RNA have been developed ( 82 ) and validation studi es of the combination of PAXgene reagent and globin RNA methods strongly support the practicality of this approach for biomarker studies ( 71 73 ) Study Aims Based o n the essentiality of zinc and the absence of a reliable biomarker for its status assessment, it is of importance to characterize biomolecules that are affected by
33 dietary zinc depletion in human s The major aim of this study wa s to identify differential responses of RNA and proteins that occ ur when the dietary intake of zinc is acutely reduced below the dietary requirement for a period of ten days Biospecimens were collected in a noninvasive manner and novel techniques for high throughput analysis with cost effectiveness were selected to allow the applicati on of the findings to the field where nutrient assessment is conducted
34 CHAPTER 2 MATERIALS AND METHOD S Human Subjects Male subjects 21 35 years of age, weighing at least 50 kg, were recruited t o participate (Table 2 1) Exclusion criteria for the dietary regimen included; current cigarette smoking, alcohol abuse, routine consumption of medications, chronic use of denture cream or dietary supplements containing z inc and history of any chronic d isease or allergic reaction (Appendix A) Upon enrollment, the z inc contents in the habitual diet of each sub ject w ere determined by a 24 h diet recall followed by calculations with the Nutrition Data System for Research (NDSR ) and blood was collected fo r serum zinc measures The study was reviewed and approved by the University of Florida Institutional Review Board and the General Clinical Research Center (GCRC) at the University of Florida and was registered at clinicaltrials.gov as NCT01221129 All subjects were asked for a written informed consent prior to enrollment. Acute Dietary Zinc Depletion The study design was a 24 day observational study comprised of three phases of dietary treatment and each subject served as one s own control. In order t o establish a defined baseline condition (acclimation) prior to dietary z inc depletion, subjects consumed meals composed of a basal mixed diet (2 d cycle menu; Table 3 1), providing 2,700 kcal and ~11 mg Zn/d and adequate amount of z inc free energy supple ment for additional calories required for body weight maintenance. After 7 days of acclimation, the subjects consumed white based liquid formula which provided <0.5 mg Zn/d for 10 days. The caloric and mineral
35 conten ts of t he liquid formula ( Table 2 3 ) we re similar to those used in previous dietary z inc studies. It is of note that additional care was taken for the current study to minimize the difference in the daily intake levels of each mineral between the two phas es with the exception of z inc This wa s confirmed by analyses using inductively coupled plasma optical emission spectrophotometry (ICP OES) at Michigan State University ( Table 2 4 ). Four pieces of starburst fruitchews (< 2.50 ug z inc /g by ICP OES; Mars, Inc) and a supplemental energy shake, used in phase 1, were provided for energy adjustment. To minimize the bioavailability of z inc 1.4 g/d of sodium phytate from rice (Sigma) was supplemented to the formula. Carboxymethyl cellulose (2 g/d; TIC Gums) wa s added to prevent bowel discomfort that could be caused by the liquid diet ( 83 ) Supplemental biotin (2 mg/d) was provided to ensure sufficient biotin absorption under the presence of excessive avidin originating from egg white. A multivitamin supplement (CVS) was given as a source of other vitamins. Distilled water (Zephyhills), Diet Pepsi and Sierra Mist (Pepsico), of which z inc contents were undetectable by fla me atomic absorption spectrophotometry (AAS), were provided throughout the first two phases. Upon completion of the second phase (on day 17), subjects returned to their self selected diet and 15 mg Zn/d (Jarrow) was provided for the repletion of their z in c status. On the last day of participation, an anonymous questionnaire of compliance was provided to each subject in order to identify any major deviations from the protocol (Appendix B ) Serum z inc concentrations were also monitored throughout the study to assess whether the dietary effect was present Sample Collection All samples were collected at the GCRC by an on site nurse under overnight fasting status of each subject. On day 7, 13 and 17, two buccal swabs and 33.5 mL of
36 whole blood were collected Each buccal mucosa sample was immediately transferred to a Whatman FTA filter paper sampling card to stabilize its RNA contents. All samples, except for 3 mL of whole blood used for complete blood counting (CBC) at Shands Hospita l at the University of Florida were transported to the Food Science and Human Nutrition D epartment for further processing. A Styrofoam chamber with an internal temperature of 5~10C, maintained with wet ice sealed in leak proof plastic bags, was used for the transportation of whole blood samples collected in PAXgene tubes and EDTA treated tubes. Other samples were transported under room temperature. Additional blood draws (5 mL) were conducted on the first day and at day 10, day15, day 20 and day 24 for serum samples Overa ll scheme of the sample handling and processing work flow is shown in Figure 2 1, and described in details below. Preparation of Buccal RNA Buccal RNA samples were prepared by using the protocol previously developed for whole blood RNA extraction from drie d blood sp ots. FTA paper cards (Whatman) with buccal samples were air dried for 3 ~ 4 h at room temperature and stored at 20C until processed. To isolate RNA, filter papers with the buccal samples were cut into small parts using sterile surgical scissors and were incubated in 1.5 mL of TRI reagent (Ambion) at room temperature for 15 min. Samples were agitated every 5 min during this period. After centrifugation at 12,000 x g x 30 min, 4C, the lysate was transferred to a new microcentrifuge tube for RN A isolation. When residual filter paper was observed in the RNA solution, samples were treated with 1 mL of TRI reagent and the phenol chloroform extraction step was repeated. All samples were further purified by the sodium acetate isopropanol precipitat ion, and the final RNA solution was stored at 80C until analyzed.
37 Processing of Blood Samples Sera were isolated from 5 mL of whole blood collected in red top serum tubes (BD Vacutainer). Blood was allowed to clot at room temperature for 60 min after co llection, and placed on ice for no longer that 2 h until processed. Clotted blood was centrifuged at 2,000 x g x 10 min, 4C, and serum was collected in 500 L aliquots. One aliquot was treated with 1% (vol/vol) protease inhibitor cocktail (Pierce) and s tored at 80C. Untreated samples were stored in 4C and 80C until processed for AAS. PBMC and circulating erythroid cells (erythrocytes and reticulocytes) were fractionated from whole blood collected in Vacutainer tubes pretreated with K 3 EDTA. For P BMC isolation, 9 mL of whole blood was placed on an equal vol of Histopaque 1077 (Sigma) and centrifuged at 400 x g x 30 min at room temperature. The PBMC layer was collected and washed with PBS at 250 x g x 10 min, twice, and the RBC pellet diluted with 1 vol of PBS was stored at 4C until processed for reticulocyte RNA preparation. After counting the cells, half of the recovered PBMC was treated with 1 mL of TRI reagent for RNA isolation while the other was cryopreserved as previously described ( 47 ) Briefly, cells were placed in heat inactivated FBS supplemented with 10% DMSO and were stored in a Nalgene freezing container s overnight at 80C, and transfe rred to liquid nitrogen for extensive storage. Remnant erythroid cells from the PBMC isolation procedure (with 9 mL of whole blood) were processed for reticulocyte RNA preparation. To achieve a pure population of erythroid cells, leukocyte depletion was carried out by using cellulose columns ( 84 ) Enriched RBC s were suspended in 1 vol of HEPES buffer (154 mM NaCl, 10 mM HEPES, 1 g/L BSA) and filtered t cellulose (Sigma) and Sigmacell Type 50 microcrystalline cellulose (Sigma) in a 1:1 (wt:wt) ratio.
38 After RBC preparations were loaded into the cellulose column, 2 vol of HEPES buffer was applied for the elution of R BC. After being collected from the eluate by 2,000 x g x 5 min at 4C, the cells were washed two times with equal vol of ice cold PBS at 200 x g for 10 min to eliminate platelets. Purified RBC s packed by centrifugation at 2,000 x g x 5 min at 4C were tr eated with 20 mL of TRI reagent for reticulocyte RNA isolation and purification by the phenol chloroform extraction and sodium acetate isopropanol precipitation, respectively. Plasma membrane of RBC (ghost cells) was prepared from 6 mL of whole blood, c ollected in a K 3 EDTA treated Vacutainer. After leukocytes and platelets were removed as described above, packed RBC s were suspended in an equal vol of ice cold PBS. After 3 washes with PBS, the RBC s were lysed by 5 vol of ice cold hypotonic buffer [5 mM Na 2 HPO 4 (pH 7.4)] containing protease inhibitor cocktail (P ierce ) and the membranes were collected by centrifugation at 12,000 x g x 10 min, 4C, unti l the supernatant and pellet w ere clarified. After a final wash at 20,000 x g, membrane pellets were sol ubilized in 5 mM Tris HCl, 0.5% Triton X 100 with protease inhibitor s and stored at 80C. Measures of Serum Zinc Concentrations Serum zinc conce ntrations were determined by AAS on the day of sample collection. Sera were diluted 1/4 with Milli Q water, o f which zinc content w as undetectable by AAS prior to each analysis To monitor the consistency among each respective measure, a reference zinc solution was prepared at a concentration of 0.6 mg Zn/L with zinc sulfate and was subject to AAS along with the diluted serum samples. After the completion of participation of each subject, measures from each collection day were confirmed by repeating AAS with all samples collected throughout the study period.
39 Isolation and Processing of Blood Cell RNA With the ex ception of RNA from whole blood collected in PAXgene tubes, all blood RNA samples were prepared by phenol chloroform extraction using either TRI reagent or TRI reagent BD, and treated with Turbo DNA free reagents (Ambion) to remove residual DNA contaminati on. For RNA isolated from samples of whole blood assays, additional RNA precipitation using 2.5 M lithium chloride ( LiCl ) was conducted to minimize the inhibitory effects of heparin on reverse transcription and PCR reactions ( 85 ) Stabilized whole blood RNA was prepared from 7.5 mL of blood collected in PAXgene blood collection tubes (BD) by the manual procedure described in the rotocol. Briefly, whole blood lysates treated with proteinase K were homogenized and RNA was isolated by using silica gel membrane columns (Qiagen). Samples were treated with DNase I (Qiagen) prior to elution. A genome wide gene expression analysis for th e detection of transcripts responsive to dietary z inc depletion was conducted with stabilized whole blood RNA from PAXgene samples G lobin RNA depletion from the whole blood RNA was done by using GLOBINclear (Am bion). Globin transcripts in PAXgene RNA (3 g) was hybridized to biotinylated oligonucleotides and captured by streptavidin magnetic beads. The supernatant was further purified by a poly T oligos conjugated to magnetic bead s prior to downstream processing for microarray analysis. All RNA samples were stored at 80C until further processed. Integrity and quality of RNA were as sessed by using 2100 Bioanalyzer (Agilent) and Nanodrop 1000 ( Thermo Scientific ), respectively. RNA Amplification, Hybridization, and Microarray Analysis The Illumina BeadC hip platform (HumanHT 12 v4) was selected for the assessment of global effect of dietary z inc depletion on the blood transcriptome, due to
40 its high throughput a nd cost effective nature PAXgene whole blood RNA depleted of globin transcripts (200 ng of tot al RNA) was amplified by the Illumina TotalPrep RNA Amplification Kit (Ambion) for the array analysis. After reverse transcription using T7 Oligo(dT) primers and second strand cDNA synthesis, biotinylated cRNA was synthesized by in vitro transcription with T7 RNA polymerase and biotin UTP. The yield and quality of cRNA were assessed with the NanoDrop spectrophotometer and Agilent 2100 bioanalyzer, respectively. For the detection of differential gene expression, labeled cRNA (750 ng) w as loaded to beadchip s, hybridized for 14 ~ 20 h at 58C, and stained with Cy3 streptavidin. Fluorescence signals from Cy3 were detected by a BeadArray Reader (Illumina Bead station 500GX) and signal intensities were exported by using GenomeStudio software (Illumina). The hybri dization and scanning processes were conducted at the Gene Expression core at the Interdisciplinary Center for Biotechnology Research University of Florida. Microarray Data Analysis Probes with detection P values lower than 0.05 in all samples were excl uded from the dataset prior to analyses. Ra w data were quantile normalized and log transformed for statistical analyses. D ifferentially expressed (DE) genes wer e determined by comparison s between baseline and post zinc depletion levels using BRB ArrayTool s developed by Dr. Richard Simon and BRB ArrayTools Development Team (permutation = 10,000) Genes differentially affected by acute zinc depletion were determined by a pairwise comparison at P < 0.005. For the visualization of gene expression patterns by using a heat map, i ntensity values were standardized by subtracting the mean value of each gene across all arrays and division by the standard deviation of each respective gene (resulting in mean = 0, standard deviation = 1). After unsupervised clustering by
41 expression patterns using the k means algorithm (k = 2), c omputational analyses based on gene ontology and putati v e transcription factor motifs were conducted to identify the highly associated functional aspects of the DE genes by using EXpression Anal yzer and DisplayER (EXPANDER) 5.2 software ( 86 ) Ingenuity Pathway A na lysis (IPA) with the list of DE genes was conducted to identify biological n etwork s betwee n the responsive genes and to predict the associated function s diseases and disorders to the modulated gene expression by dietary zinc depletion R eal Time Quantitative PCR Tr anscript abundance of genes known to be directly involved in the regulation of c ellular z inc homeostasis, i.e ., z inc transporters ( ZnT1, ZnT2, ZnT4, ZnT5, ZnT6, ZnT7, Zip1, Zip2, Zip4, Zip5, Zip8, Zip10, and Zip14 ) and MT, were measured by assays designed and validated previously (9). Primers and probe sets for the detection of Zip6 (forward, 5' AGGCTGGCATGACCGTTAAG 3' ; reverse, 5' AAAATTCCTGTTGCCATTCCA 3 ; probe, 5 FAM CCTTTATAATGCATTGTCAGCCATGCTGG BHQ1 3 ) and GAPDH transcripts (forward, 5' GAAGGTGAAGGTCGGAGTC 3' ; reverse, 5' GAAGATGGTGATGGGATTTC 3 ; probe, 5 FAM CAAGCTTCCCGTTCTCA GCC BHQ1 3 ) were designed by using PRIMER EXPRESS 3.0 software (Applied Biosystem) and validated by Primer3plus. Quantitation of CDC20 TXNDC5 MZB1 and IGJ transcripts were conducted using TaqMan gene expression assays from Applied Biosystems. Real tim e quantitative reverse transcriptase polymerase chain reaction (q PCR ) assays were done with cDNA, generated by using high capacity cDNA reverse transcription kit (Applied Biosystems), and the Taq M an fast universal PCR master mix (Applied Biosystems). For assays with the RNA samples from the whole blood cytokine assay, Taqman one step RT PCR
42 master mix reagents were used in order to minimize the residual contents of heparin that was present in the reaction mixture. Glyceraldehyde 3 phosphate dehydrogenase ( GAPDH ) mRNA and 18S rRNA levels were measured as internal controls, and amplification and detection were performed with the StepOnePlus real time PCR system (Applied Biosystems). Western Analysis of RBC Zinc Transporters We have previously reported the pr esence of ZnT 1, Z ip 8 and Z ip 10 in murine erythrocytes ( 14 ) A decrease and increase in RBC ZnT 1 and Z ip 10, respectively, was observed when mice were fed a low zinc d iet. T he presence of these tran sporters in human RBC was confirmed by western analysis using affinity purified rabbit polyclonal antibodies All primary antibodies besides that for the detection of dematin (Abcam ab89161 ), were designed for previous stu dies ( 14 15 51 ) Specificity of signals produced by each i n house made antibody was determined by preabsorption with respective antigenic peptides specific to the target proteins of interest. Glycosylation status of each protein was evaluated by incubation of protein samples with PNGase F at 37 C for 2 h prior t o western analysis. Erythrocyte proteins (total of 20 g) were separated by 7.5~10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE) and were transferred to nitrocellulose membranes. Efficient transfer and equal loading were visualize d by Ponceau staining. Membranes were blocked by Tris buffered saline (TBS T) containing 5% skim milk for 1 h, and primary antibodies were added at a final concentration of 1~2 g/mL After incubation with the primary antibody for 1~2 h, blots were washed with TBS T and then treated with anti IgG antibody co njugated to horseradish peroxidase (1:2,000~1:10,000 of antibody in 5% blocking solution) for 1 h. Signals indicating abundance of the protein were visualized by using
43 an enhanced chemiluminescent substrate SuperSignal WestPico (Pierce) and autoradiograph ic films. Blots were incubated in Restore PLUS Western Blot stripping buffer (Thermo Scientific) for 15 min when reprobing with subsequent primary antibody was needed. Immunoprecipitation and Mass Spectrometry Erythrocyte membrane fraction was immunoprec ipitated by using anti human Zip8 polyclonal antibodies to enrich the abundance of proteins targeted by the respective antibody. Briefly, 1 mg of protein was incubated in 1 vol of RIPA buffer containing 4 g of antibody at 4 C overnight, and purified by using protein A/G agarose conjugates (Pierce) After extensive wash steps, protein bound to the agarose beads were eluted by 10 min incubation at 100 C with Laemmli buffer. Immunoprecipitated proteins were loaded on wells of a 7.5% polyacrylamide gel for the separation by electrophoresis After the g el was divided into two halves eac h was subject to gel staining and western analysis, respectively. Proteins were stained by incubation in Coomassie Blue staining solution (0.1% Coomassie Brilliant Blue R 2 50, 10% acetic acid, 50% methanol and 40% water) for 1 h at room temperature, and background stains were removed by a destaining solution composed of 10% acetic acid, 50% methanol and 40% water. Stained gel was stored in the destain in g solution diluted 1/ 4 with water until processed for mass spectrometry. Western analysis was conducted for the identification of IgG oriented bands and the protein producing a non specific band by the anti human Zip8 primary antibody. Thereafter, the western image was match ed with the stained gel to determine the position of the non specific band in the gel, which was excised wi th methanol treated blades for liquid chromatography coupled with tandem mass spectrometry (LC MS /MS ) using a hybrid quadrupole /time of flight mass s pectrometer
44 (QSTAR E lite, Applied Biosystems ) Protein digestion and LC MS / MS were conducted at the proteomics core at the Interdisciplinary Center for Biotechnology Research University of Florida. The nature of identi fied peptides was determined with Mascot version 2.2.2 (Matrix Science ) and the IPI human database Scaffold (Proteome Software Inc ) was used to validate MS/MS based peptide identifications. Whole Blood Cytokine Assay Whole blood collected in heparin Vacutainers (5 mL) w as used to determ ine the effects of dietary z inc on the production of cytokines by LPS or PHA stimulation. In vitro induction of cytokine release were conducted with 20% of whole blood in phenol free RPMI 1640 medium supplemented with 2 mM L glutamine, 100 kU penicillin/L and 100 mg streptomycin/L, with or without LPS (1 mg/L) or PHA (10 mg/L) treatment. Cells were incubated in 24 well ultra low attachment plates (Corning) at 37C in 5% CO2 for up to 24 h. For measures of cytokine release, cell free supernatant w as colle cted by centrifugation at 600 x g x 5 min, 4C, and stored with protease inhibitor cocktail at 80C. Whole blood cell pellets were treated with TRI reagent BD (Molecular Research Center) supplemented with acetic acid, and stored at 80C for RNA isolatio n. Enzyme Linked Immunosorbent Assay In vitro z inc treatment has be en shown to increase IL interferon gamma ( IFN ) expression levels of PBMC. A previous study from our lab has shown the effects of dietary z inc su pplementation on the IL expression of immune cells activated in vitro ( 41 ) To test whether dietary z inc depletion causes a modulation in LPS or PHA induced cytokin e production, cytokine levels in pooled cell free medium collected from the whole blood culture were measured by using a multi analyte ELISA array kit for inflammatory cytokines (IL 2,
45 IL 4, IL 6, IL 8, I L 10, IL 12, IL CSF; SA Bioscienc es). The effects on IL using single analyte ELISArray kits from SA Biosciences. Briefly, protease inhibitor treated cell free medium were added to each well of a pre coa ted capture antibody microplate loaded with assay buffer. With extensive washes in between each step, captured cytokines were probed with relevant detection antibodies and avidin horse radish peroxidase (HRP). After subsequent treatments with the develop ment and stop solution, absorbance at 450 nm was measured by a multi mode microplate reader (SpectraMax M5; Molecular Devices) for quantification. Isolation and Quantitation of Serum Micro RNA S erum was treated with 5 volumes of QIAzol (Qiagen) for the dena turation of protein contents and subsequent isolation of RNA. Due to the absence of a known housekeeping microRNA (miRNA) in human serum, 25 fmol of s ynthetic cel miR 39 ( 5' UCACCGGGUGUAAAUCAGCUUG 3' ) was added as a means of normalization ( 87 ) After adding 1 volume of chloroform, aqueous and organic phases were separated by centrifugation at 12,000 x g x 15 min at 4 C The aqueous phase was treated with 1.5 volumes of 100% ethanol and loaded to a miRNeasy Mini column for RNA extraction. Aft er extensive washes, p urified RNA was eluted from the column membrane with 50 L of nuclease free water and stored at 80 C until analyzed. T he human serum miRNA PCR array from SABiosciences was used for the identification of circulating miRNA responsive t o dietary zinc levels Serum RNA was isolated from 400 L of pooled s e ra at th e initial screening step. Day 7, 17 and 24 samples were selected as those representing baseline, zinc depleted and zinc repl eted conditions, respectively. Serum RNA was polyad enylated and converted to cDNA by
46 using RT 2 miRNA First Strand kit (SABiosciences) containing universal reverse transcription (RT) primers targeting poly A tails. After a 1/4 dilution with nuclease free water RT products were distributed across the miRNA qPCR arrays with SYBR Green qPCR reagents (SABiosciences) for amplification and quantitation. Melt curve analyses with amplified products were conducted to determine the specificity of each primer set. C riteria for determin ing the zinc responsive miRNA were ; 1) single melting temperature 2) fold change > 1.5 by dietary zinc depletion 3) threshold cycle (Ct) < 35, and 4) response to zinc repletion in a direction opposite to that of zinc depletion Zinc responsiveness of selected miRNA identified in po ole d samples, was confirmed with RNA from 200 L of individual se rum samples collected at day 7, 13, 17, 20 and 24 by using the miScript PCR system (Qiagen) Reverse transcription products of miRNA were diluted by equal volume of nuclease free water, and 2 L of the diluted cDNA was used as templates for SYBRgreen based qPCR assays at a total reaction volume of 10 L. All values of miRNA qPCR experiments were normalized to the abundance of th eir respective cel miR 39, of which values were constant across all RNA preparations. Statistical Analysis Bas ed on measures from previous dietary z inc studies with humans ( 41 42 ) a complete dataset from 9 subjects is adequate to detect a within subject difference in transcript levels of z inc responsive gene with 80% power at P < 0.05 two sided. Values from samples of Day 7 (baseline) served as the control for comparisons. test or r epeated measures of ANO VA followed by a Student Newman Keuls multiple comparisons test were conducted for pairwise comparisons. Linear association between transcript levels and serum zinc concentrations was determined by the linear regression m ethod All statistical analyses w ere conducted using the InStat 3 software
47 (GraphPad). The level of significance was set at P < 0.05 for all analyses except for those of microarray data
48 Table 2 1 Subject characteristics at screening phase (n = 9) Subject Age Height BWt Caloric Needs S erum Zn 24 h Recall (years) (cm) (kg) (kcal/d) (ug/dL) (mg Zn/day) 1 31 179.5 86.3 2,950 96 20.6 2 26 163 .0 62.4 2,854 114 8.8 3 29 178.6 88.5 3,201 87 8.6 4 25 177.4 66.8 2,760 111 10.5 5 25 179.1 96.6 3,161 96 11.3 6 23 182.5 72.7 2,819 99 17.9 7 22 165 .0 85.1 3,236 72 28.4 8 22 171.5 64.7 3,000 84 12.6 9 24 176.1 87.8 3,091 104 21.2 Mean 25 174.7 79.0 3008 95.9 15.5 ( SD ) ( 3 ) ( 6.8 ) ( 12.4 ) ( 175 ) ( 13.3 ) ( 6.8 )
49 Table 2 2 The 2 day cycle menu of the acclimation phase Meal Menu 1 Menu 2 Brea kfast (8:00 AM) English muffin (63 g) Grape nuts cereal (56 g) Margarine (11 g) Milk (135 g) Jelly (15 g) Low fat yogurt (238 g) Low fat yogurt (146 g) Apple juice (215 g) Lunch (12:00 PM) Roast beef (120 g) Chicken breast (203 g) White bread (63 g) Gravy (23 g) Mayonnaise (12 g) Corn (89 g) Canned peach (126 g) Margarine (9 g) Pudding (126 g) Canned peach (176 g) Cool whip (6 g) Cookies(38 g) Snack (3:00 PM) Power bar (39 g) Short bread(47 g) Canned pear (170 g) Peanut butter (20 g ) Jell O (196 g) Crackers (15 g) Dinner (5:00 PM) Chicken breast (139 g) Turkey breast (155 g) Peas (120 g) Gravy (27 g) Margarine (7 g) Carrots (108 g) Icecream (107 g) Margarine (11 g) Wafers (31 g) White bread (34 g) Apple juice 200 g) Ice cream (114 g) Snack (8:00 PM) Almonds (30 g) Raisins (51 g) Sherbet (104 g) Power bar (64 g) Lemonade (200 g) Jell O (166 g)
50 Table 2 3 Dietary components of the liquid di et for the zinc depletion phase 1,2 Component Mineral content Amount per k g diet Egg white Cornstarch Maltose dextrin Sucrose Corn oil Cellulose Mineral mixture Sodium chloride Calcium carbonate Potassium phosphate, monobasic a Potassium phosphate, dibasic b Potassium chloride c Magnesium carbonate 5 H 2 O Ferric ci trate 6 H 2 O Copper sulfate5 H 2 O Potassium Iodate Manganese chloride Glucose 129 g 300 g 300 g 60 g 151 g 10 g 9 8 g Na 10 .0 g Ca 3 8 g P a+b 12 .0 g K a+b+c 1 28 g Mg 57 6 mg Fe 7 6 mg Cu 0. 50 mg I 11 4 mg Mn 16 8 g 1 Vita mins and extra biotin were provided as separate supplements. 2 Sufficient energy intakes were ensured by supplemental mineral free energy shakes.
51 Table 2 4 Mineral contents of the acclimation and zinc repletion diets measured by inductively coupled plas ma optical emission spectrophotometry (ICP OES) Minerals per day Acclimation phase Depletion phase g Calcium 1.134 2.994 g Phosphorus 2.048 1.390 g Magnesium 0.343 0.509 g Potassium 3.259 4.417 g Sodium 3.208 5.038 g Sulfur 1.302 1.319 mg Copper 1.8 18 1.539 mg Iron 25.014 25.988 mg Zinc 10.415 0.296 mg Manganese 3.533 3.828 mg Molybdenum ND 1 ND 1 mg Cobalt ND 1 ND 1 1 Not detected: M easures were below the detection limit of ICP OES
52 Figure 2 1. Schematic diagram of sample collection and handl ing processes. Relevant materials and methods were selected based on their practicality for being use d in the field for nutrient assessment. Availability of me thods for sample stabilization allowed the completion of the entire processing by a single rese archer.
53 CHAPTER 3 IDENTIFICATION OF ZI NC RESPONSIVE GENE TRAN SCRIPTS IN BUCCAL AN D BLOOD CELLS Introduct ory Remarks Zinc as a catalytic, structural and regulatory component is involved in various physiological events in the human body ( 1 ) Documented outcomes of i n adequate zinc ingestion in clude growth retardation diarrhea dermatitis, hypogonadism, visceromegaly, hematological abnormalities and impaired immunity ( 2 16 ) Recent evidence of the protective role of zinc against DNA oxidation ( 66 68 ) which may result in p redisposition to cancer development, indicates the detrimental long term effect of low zin c consumption T he rapeutic properties of supplemental zin c against infectious diseases ( 23 ) and diabetes mellitus ( 19 ) substantiate the values of this nutrient for the maintenance of human health Even though the biological and biochemical properties of zinc has been intensively explored since the first characterization of its essentiality ( 18 ) a reliable diagnostic tool to assess the dietary zinc status of individual h umans or populations is in absence Dietary zinc deficiency, with a predict ed worldwide prevalence of 31% ( 23 ) has been estimated to be responsible for approximately 450,000 global deaths in children under 5 years of age ( 88 ) Consequently, development of a biomarker defining zinc status is needed, particularly f or the identification of individuals or communities which may benefit from zinc intervent ions Serum or plasma zinc concentrations are widely used for the assessment of zinc status in experimental animal and human models. However, due to its lack of specificity e.g., responsiveness to stress starvation or immunological conditions its capab ility to serve as a diagnostic tool for zinc deficiency is limited Past research has shown that both metallothionein gene expression and specific zinc transporter genes in blood
54 leukocyte subsets are directly proportional to dietary zinc intake ( 40 41 ) Additionally specific zinc transporter genes in monocyte s, T cells, and granulocytes have been shown to be als o zinc responsive. ZnT1 and Zip3 transcript levels of whole blood corresponded to dietary zinc supplementation in human subjects with an increase and decrease, respectively ( 41 ) Transcripts for specific cytokines produced when these cells are activated in vitro were also zinc responsive over a 16 day dietary protocol that included 10 days of zinc supplementation at 15 mg/day or a placebo ( 41 ) These findings reflectin g the regulatory role of zinc in gene expression ha ve provided the hypoth esis that these specific transcript changes can be used for the identification of zinc deficiency Transcriptome profiling using cDNA array analysis and q PCR technology enable s the identification of genes most sensitive to a biological variable of interest, and thus have been considered as the gold standard approaches for developing ge ne transcript based diagnostic methods ( 69 ) I n formation related to the zinc effect on whole genome expression of human cells is limited to those from in vitro treatments ( 44 89 90 ) Our previous microarray experiments with monocytic/macrophage THP 1 cells under zinc deprived and excess conditions iden tified s everal hundreds of zinc responsive genes ( 89 ) Functional grouping of the differentially expressed genes indicated the regulatory role of zinc in immune/ cytokine function and signal transduction. Another study on various cell lines each representing T cells, B cells and monocytes, identified a list of differentially expressed genes coordinately associated with immunity and cellular survival ( 90 ) Although these clearly indicate the presence of zinc effects on the whole
55 genome expression of blood cells, assessment of the in vivo effect via a dietary study is requir ed to fully understand the physiological role of zinc in human biology In a manner to pursue this a human dietary study of experimentally induced acute zinc depletion was conducted and tr anscriptome analyses of whole blood were undertaken using oligonucl eotide microarrays T ranscripts holding the potential of being an assessment tool of dietary zinc status were ide ntified by quantitative measures of buccal and blood RNA as well Specific aims of this part of study include; 1) Evaluat ion of the effects of a cute dietary zinc depletion on MT and zinc transporter transcript leve ls in buccal epithelial cells. 2) Evaluat ion of the effects of acute diet ary zinc depletion on levels of zinc transpo rter and MT transcripts in leukocyte s and reticulocytes 3) Identif ication of genes responding to dietary zinc restriction through transcriptome profiling of whole blood RN A by cDNA microarray analysis. Results Experimental Zinc Depletion in Human Subjects The experimental model of acute dietary zinc depletion used in this study was developed based on previously standardized methods successfully implemented by others ( 34 37 42 43 ) Analytic measures by OES ICP confirmed the zinc contents of the acclimation diet s as 10. 4 mg zinc per da il y serving which is approximate to the current R DA of zinc for adult males (Figure 3 1) The low zinc (0.3 mg/d by OES ICP) and high phytate contents (1.4 g/d) of the depletion diet yields a phytate to zinc molar ratio of 462 indicating low bioavailability of zinc. In accordance to these aspects of t he diet, subjects had a significant decrease in their serum zinc concentrations after 8 and 10 d after consuming the zinc depleted diet, and this was reversed by subsequent zinc
56 supplementation (Figure 3 2) It is of note that the inter subject var iabilit y of serum zinc levels decreased by the end of the acclimation phase All participants had a serum zinc concentration above the suggested lower cutoff value for males, 74 g/dL ( 38 ) at the baseline state No hematological effects of dietary zinc depletion w ere observed by complete blood counts ( Table 3 1 ) of which values were all residing within their respective suggested normal ranges. Effects of Zinc Depletion o n Zinc Related Transcript Levels A panel of gene expression assays targe ting zinc related gene products was used for the quantitation of respective mRNA levels in the biological samples collected. Buccal RNA samples were selected as an optimal approach of sampling due to its noninvasiveness and the yield of RNA isolated from FTA filter papers were 989 492 ng per oral swab A mong the zinc related transcripts in buccal RNA only those of MT and ZnT1 were stably detected by qPCR (Figure 3 3A) Criteria f or determining the reliability of these assays include; 1) low analytical variation, 2) amplification curve parallel to that produced from reference RNA, and 3) Ct values below 35. Human Reference RNA composed of total RNA from various cancer cell lines ( Stratagene) was used as a positive control of amplification. A significant decrease in MT mRNA abundance was observed in samples collected after 10 d of dietary zinc depletion, while no effect was present in the ZnT1 transcript levels (Figure 3 3B) Cell type specific detection and responsiveness to dietary zinc of MT and zinc transporter transc ript levels were observed by qPCR with PBMC and reticulocyte RNA. MT, Zn T1, ZnT4 and ZnT5 mRNA levels significantly decreas ed in PBMC (Figure 3 4) while reduction in ZnT1 and Zip3 transcripts of purified reticulocytes (Figure 3 5 ) occurred by acute dietary zinc depletion. Among the differentially expressed zinc
57 transporter transcripts, ZnT1 and Zip3 were the most significantly affected ones (with lowest P value ) in PBMC and reticulocytes, respectively. L inear regression analysis revealed a significant c orrelation between serum zinc concentrations and PBMC Z nT1 transcript abundance with a correlation coefficient ( R ) of 0.6 900 and a P value of 0.0015 However, t he R between serum zinc concentrations and reticulocyte Zip3 levels was not significantly different from 1 ( P = 0.0778) indicating the absence of association between these two indices in this study The reduction of ZnT1 mRNA levels observed by both PBMC and reticulocyte RNA, was measurable in whole blood RNA as well. As with the PBMC ZnT1, a significant correlation of serum zinc concentrations with whole blood ZnT1 mRNA levels was observed ( R = 0.5 558 P = 0.002 6 ). Whole Blood RNA Processing and Quality As sessment for Microarray Analysis Because of the requirement of high quality RNA for reliable microarray results stringent qualit y assessment was conducted throughout the sample pre processing steps i.e., from RNA isolation to hybridization to the array c hips Whole blood RNA isolated by using the PAXgene system w as highly intact as indicated by RNA integrity numbers (RIN) above 8 (Figure 3 8). A260/A230 and A260/A280 ratios were above 2.00 in all samples implying the high purity of RNA in each preparati on. The yield of RNA after globin transcript depletion was approximately 75% of the starting material (Figure 3 9 A ), and the purity of these products were confirmed by A260/A230 and A260/A280 ratios at 2.12~2.31 and 2.19~2.30, respectively. There were no differences in the cRNA recovery between RNA prepared from whole blood collected at day 7 (baseline) and day 17 (ZnD) after biotinylation and amplification (Figure 3 9B). When the post amplification yield between RNA samples that were untreated (PAX) and treated for globin RNA reduction (GRP) was compared, higher recovery was observed
58 in PAX samples (Figure 3 9C) which is due to the presence of redundant globin RNA The amplification efficiency indicated by folds of the starting amount are comparable to reference values provided by the manufacturer, which are 148 and 61 for PAX and GRP, respectively (Figure 3 9D) E lectropherogram profile s of the amplified PAX RNA and GRP RNA confirm the successful removal of globin transcripts from the processed whole b lood RNA (Figure 3 10) The prominent peak produced by the highly abundant globin RNA content was only present in the PAX samples after RNA amplification. Global Effect of Zinc Depletion on Whole Blood Transcriptome The Illumina bead array chip was selec ted for the transcriptome analysis due to its multiplexing feature and lower per sample cost when compared to platforms from Affymetrix and Agilent ( 71 72 ) Prior to the data preprocessing, microarray data quality control was conducted by using control metrics generated by the GenomStudio software. Values of each control parameter corresponded to the reference va lues provided by Illumina in all arrays and thus, verif ied the validity of gene expression data obtained (Figure 3 11). Quantile normalization producing a best fit slope near 1 b etween the distribution of signals from each probe by pooled GRP samples fr om day 7 and 17, was selected for the downstream data analyses (Figure 3 12) Lists of differentially expressed genes were determined by using various combinations of P value and fold change as filtering thresholds (Table 3 2). It is of note that fewer g ene s were considered as differentially expressed by unpaired comparisons than by paired t tests, which indicates the presence of inter subject variability in the genes of which expression before and after the acute dietary zinc depletion. The list of diffe rentially expressed genes determined by a pairwise comparison at P < 0.005 (328 genes) were further analyzed by bioinformatic tools for characterization
59 and the identification of biological events associated with dietary zinc depletion. Among the genes id entified, 192 had higher expression levels (Table C 1 ) while 136 showed less express ion (Table C 2 ) after the consumption of a low zinc diet. Unsupervised a verage linkage hierarchical clustering of the se genes with Pear s on correlation metric identified tw o major clusters each composed of 192 and 136 genes showing trends of overexpression and repression by zinc depletion respectively (Figure 3 13A) While no clustering by baseline and post zinc depletion condition s was observed when hierarchical clusterin g of the individual samples was conducted with all genes (Figure 3 13B), a partial separation between the baseline and post zinc depletion condition was identified by using the 328 differentially expressed genes for clustering (Figure 3 13 C ). To determin e the effect of globin RNA reduction on the detection of differentially expressed genes, average signals produced by pooled PAX and GRP samples from day 7, 13 and 17 were compared. B ased on a detection P value of 0.05 as a upper cutoff threshold a total of 6,861 genes were considered absent when the RNA samples were not processed for globin RNA reduction (Figure 3 14A). Among th o se masked genes, 51 were among the 328 genes determined to be differentially expressed by dietary zinc intake levels. As indic ated in Figure 3 14B, the rela tive signal intensities of these gene transcript s were higher when RNA samples were processed for globin RNA reduction, implying the importance of the removal of this highly abundant RNA prior to whole blood transc riptome anal yse s By using the data obtained from the pooled samples, the temporal expression pattern of the 328 differentially expressed genes were determined as well. A gradual increase and decrease throughout the 10 day
60 depletion period was observed by the up reg ulated and down regulated genes, respectively (Figure 3 15). Advances in bioinformatic tools allowing the integration of transcriptome profiles into molecular and network databases enable the prediction of physiological effects attributed to the different ial gene expression by changes in clinical or nutritional conditions. By using the EXPANDER an d IPA software packages f unctional analyses were conducted to d iscover the implications of differential expression of genes caused by dietary zinc depletion G ene ontolog y enrichment analysis of gene clusters categorized based on their expression pattern as analyzed with the k means algorithm (Figure 3 16) by using the Tool for Analysis of GO enrichment (TANGO) of EXPANDER ( 86 ) identified overrepresentation of functional categories associated with cell cycle regulation and ATP binding by the up regulated genes (Table 3 3 ). P r omoter Integration in Microarray Analysis (PRIMA) predictions based on the cis regulatory elements of the resp onsive genes identified NF Y, AP 2 and ETF as the transcription factors med iating dietary zinc effects on the up regulated genes and El k 1 and TEF as those involved in the repression of the down regulated genes (Table 3 4 ) Further characterization of all 328 differentially expressed genes using the Ingenuity Pathway Analysis (IPA) confirmed the TANGO results by identifying enrichment in functional categories related to cell proliferation (Table 3 5 ) It is of note that cancer was identified as the top disease and disorder associated with the differential gene expression caused by the dietary restriction regimen. While the enriched fun ctional categories identified using the list of up re gulated genes for IPA functional analysis corresponded to those by all responsive genes (Table 3 6 ), genes
61 down regulated by acute dietary zinc depletion were shown to be associated with biological events related to cell death, cell mediated immune respons e, and cellular development and function (Table 3 7 ) T o further unravel the physiological and mechanistic implications of our microarray data, molecular networks composed of connections between functionally related genes were generated by analyzing the l ist of differentially expressed genes with IPA In accordance to the results from functional analyses above, the top two networks identified by all differentially expressed genes were composed of m olecular interactions related to cell cycle and cellular g rowth, respectively (Figure 3 17) Overrepresented functional networks by the up regulated and down regulated genes, respectively, imply enhancement in cell proliferative events and repression in cell death, cell mediated immunity and cellular development as physiological outcome s of inadequate zinc intake (Figure 3 18) Gene Transcripts with Potential of being Biomarkers of Zinc Deficiency Pairwise comparisons between before and after treatment values of each individual are suitable for the identificat ion of indices responsive to the treatment. However, in order to id entify molecular responses with potentials of being a biomarker, inter individual variability should be taken into account. As shown in Figure 3 19, a clear discrimination between baselin e and post zinc depletion conditions was produced by hierarchical clustering with the 203 differentially expressed genes determined by paired t test ( P < 0.005). After filtering by fold changes above 2, a total of 8 well characterized genes among the 14 f iltered genes including LOC649923, LOC651751, IGJ, CDC20, MZB1, TXNDC5, IGLL1, CD38, LOC642113, LOC647506, LOC652493, GLDC, LOC647450 and TNFRSF17 were determined as candidate molecules for the status assessment of dietary zinc intake levels (Table 3 8 ) q PCR of IGJ, CDC20,
62 MZB1 and TXNDC5 mRNAs confirmed the responsiveness of these gene transcripts to acute dietary zinc depletion identified by microarray analysis (Figure 3 20 ). Discussion Zinc transporter and MT expression w ere the primary target s of in terest due to their direct involvement in zinc trafficking and homeostatic regulation and likelihood to respond to lower zinc availability by dietary restriction ( 8 9 ) It was of importance to implement noninvasive means for biopsy collection for future applications to human patients in the field of nutrient assessment ( 75 ) Thus, we selected buccal swabs and blood draws a s the sources of RNA for the current study T he method we utilized for RNA isolation from oral epithelial cells was developed based on swabbing approach commonly used for DNA based diagnosis and profiling in clinical and forensic settings. The major con cern of this technique was the high contents of RNase in saliva which may lead impairment of RNA quality during and after the collection process ( 78 79 ) By using Whatman FTA filter papers originally designed for the stabilization of nucleic acid isolated from plant specimens and which has also been used for RNA preparation from whole blood ( pre processed as dri ed blood spots) ( 41 80 ) we attempted to eliminate the ex vivo effects on RNA integrity. Amplification c urves indicated s uccessful acquire ment of RNA sufficient for the detection of buccal MT and Zn T1 transcripts by qPCR A marked decrease in transcript levels of MT, the prototypical zinc regulated gene, by low zinc ingestion was measured in oral epithelium indicating its potential of being a zinc biomarker Quantitative real time PCR identifies the presence of transcripts by amplifying a short region of its template cDNA, i.e., amplicon, which is generally composed of less than 100 nucleotides Thus, th e reliability of qPCR results depends on the integrity of
63 the amplicon site Consequently, o ur results do not exclude the presence of partial RNA fragmentation due to RNase activity. T he expression of some zinc transporter transcripts may be present in t he buccal mucosa in vivo al though they were not detected during our screening step Further optimization in the experimental approach for buccal RNA preparation may allow us to identify other zinc homeostatic genes responsive to the host s zinc status P revious findings o f the effects of zinc supplementation on the blood zinc transporter expression in human s ( 41 ) led us to the characterization of such indices of individuals under dietary zinc restriction The distinction between the differentially expressed transporter genes in PBMC and reticulocyte RNA indicate the cell type specific regulation of these genes ( 8 ) Only ZnT1, which is known to be ubiquitously expressed on the plasma membrane of various cell types and tissues ( 8 ) was commonly affected by the acute zinc depletion regimen in these blood cells. This zinc responsiveness of ZnT1 was also detected in whole blood RNA by using qPCR However, the differential expression was not identified by our microarray experiments indicating the l ow sensitivity, particularly for transcripts at low abundance and minimal fold change respon ses, of microarray technology Th e mode of ZnT1 response to dietary zinc depletion agrees with its role in cellular zinc export ( 8 ) The data of the present study also corresponds to our previous observation of ZnT1 responding in an opposite mode to supplemental zinc ( 41 ) indicating its property t o reflect the levels of dietary zinc bioavailability in both zinc deprived and excess conditions. Serum zinc concentratio ns were measured as an evaluation tool of com pliance and the effectiveness of our dietary zinc depletion protocol. Lower variation in the
64 serum zinc levels after acclimation indicate s the equilibrated zinc status of each individual at the baseline level. Its s ignificant decrease by zinc depletion and its reversal by repletion, respectively, imply the value of this i ndex to be indicat ive of dietary zinc in gestion levels. However, due to its responsiveness to other conditions such as acute infection or starvation ( 1 33 ) the use of thi s serum zinc measures may be limited to healthy individuals under well controlled experimental settings. W e identified significant association serum zinc measures with ZnT1 levels in PBMC and whole blood sample s. This further validates the potential of this zinc transporter transcript to function as a diagnostic too l for dietary zinc deficiency. A recent study reported the lack of association between PBMC ZnT1 transcript abundance and serum zinc levels of huma n subjects ( 91 ) It is of note that this was based on measures from individuals under self selected diets, mostly providing daily zinc above the estimated average requirement. A dditionally, a s indicated by the absence of co rrelation between plasma zinc concentrations and the estimated dietary intake levels (determined by food record or food frequency questionnaire) the plasma levels were not successfully reflecting the dietary zinc status of each individual Another notabl e difference of this study from the present study is the use of values from a sample group composed of both female s and male s and thus possible gender effects on plasma zinc or ZnT1 expression may have confounded the statistics Data of the present study were produced from men only. These imply the correlation between serum zinc concentrations and blood ZnT1 transcripts may be present only when 1) a defined group representing zinc deficiency is included, 2) variables affecting serum zinc
65 (besides of diet ary zinc level) are minimal, and 3) the study population is composed of a single gender. Among the transporters assessed with the blood cells, reticulocyte Zip3 was the only importer transcript affected after the 10 day dietary zinc depletion phase, how ever in an opposite direction to the general response mode of Zip genes to zinc deprived cond itions, i.e., up regulation ( 8 ) It is of note that the presence of Zip3 was not detected in the plasma membrane f raction of mature erythrocytes ( 14 ) which implies the intracellul ar localization of this protein in differentiating erythoid progenitor cells. Decrease in its expression by low zinc availability suggests its function mediating zinc removal from a cellular compartment at conditions of higher zinc availability. Recently a cellular zinc exporter ZnT2 has been characterized to function as a mitochondrial zinc importer ( 92 ) This leads to the speculation of Zip3 functioning as a zinc exporter for the mitochondr ion, of which zinc content need s to be tightly regulated for optimal hemoglobin synthesis during the terminal erythroid differentiation ( 93 94 ) Thus, relevant mechanistic studies for the characterization of erythroid Zip3 function are required. The global effects of zinc on the transcriptome of blood cells have been previously characterized by us and others ( 44 89 90 ) W e identified total of 1,045 zinc responsi ve genes in a h uman acute monocytic leukemia cell line THP 1, under zinc excess or depleted condition in vitro ( 89 ) Among those, expression levels of 283 genes were significantly altered by both zinc excess and deprived conditions, of which 104 and 86 gene s responded to zinc levels in a positive and negative mode, respectively. The organ specific effect of zinc on the transcriptome was identified by microarray analyses
66 of Jurkat, Raji and THP 1 cell li nes representing the T cell B cell and monocyte population, respectively, under different zinc conditions ( 90 ) Even t hough the expression of only 7 genes was generally modulated by zinc in all cell types, the functional networks identified by the zinc responsive genes in each cell type were commonly r elated to inflammatory response and cellular survival. In the pres ent study, we identified whole blood genes having modulated expression levels after acute dietary zinc depletion using an extensively validated pla tform originally designed for cancer biomarker research ( 71 ) A combination of PAXgene reagent and globin RNA reduction allow s the preservation of in vivo transcriptome profile and an increase in detection sensitivity of gene transcripts at low abundance ( 71 72 ) The capability to minimize ex vivo effects on the original gene expression profile is essential for transcriptome analyses particularly, when immediate sample processing is impossible ( 75 ) The negative effect of globin RNA, representing ~70% of the total whole blood RNA, on transcript detection was clearly observed by comparisons between pre and post globin RNA reduction samples from the current study Additionally, among the 328 genes differentially expressed by zinc depletion, fifty one were determin ed as absent due to the masking effect of globin RNA. The s e data further support the importance of globin transcript removal from whole blood RNA prior to microarray based biomarker discovery ( 81 82 ) As a n approach to predict physiological outcome attributed to the overexpression and repression of the differentially regulated genes by low zinc intake we implemen ted functional enrichment analyses in a fashion similar to that of Haase H et al. ( 90 ) T he overrepresentation of genes involved in cell cycle and cell mediated immune response
67 identified by the up and down regulated genes in the whole blood RNA, respectiv ely, remarkably aligns with the finding from the in vitro study of zinc status with a combination of various immune cell models ( 90 ) Such observation s indicate t h at whole blood transcriptome is compatible for the identification of biological zinc effect s on its subsets i.e., monocytes and lymphocytes, which coordinately function and interact during circulation in vivo The functional implications of the differe ntially expressed genes we identified correspond to the well characterized clinical outcomes of prolonged zinc deficiency i.e., impaired immunity ( 95 97 ) and predi sposition to cancer development ( 98 99 ) Of considerable interest is the appearance of v ascular endothelial growt h factor ( VEGF ) in the functional network s enriched by the overexpressed gene s. VEGF has been considered to be a therapeutic target for anti cancer treatment due to its pro tumor i genic feature ( 100 101 ) Recently, intracellular zinc deprivation has been shown to result in increased production of VEGF by prostate cancer cells ( 102 ) No change of VEGF gene expression in the blood cell population was identified by our microarray experiment s indicating that modulated production or secretion of this growth factor holding both autocrine and paracrine functions may have occurred in a different organ during zinc depletion Low serum zinc has been characterized in patients of various types of cancer ( 103 105 ) Impairment in repair systems for DNA damage has been suggested as the underlying mechanism of higher risks of cancer by zinc deficiency ( 98 ) Here we prop ose another possible regulatory factor VEGF as a mediat or of an effect of zinc on tumor development and cancer progression.
68 Nuclear factor kappa B (NF B ) was identified in the functional network composed of down regulated genes indicating its impaired function under acute dietary zinc depletion The immunosuppressive effect of suboptimal zinc conditions has been associated with i mpaired DNA binding activity of this transcription factor ( 63 ) Relevant mechanism s include the zinc dependence of the transactivation capability of NF B indicated by lower affinity to its binding motif under zinc deprivation ( 106 108 ) G lucocorticoid another key regulator of immunity and its nuclear recepto rs can decrease NF B activity by up regulating its inhibitor I B or by competitively binding to coactivators of NF B induced gene expression ( 109 ) I ncrease d plasma corticosterone has been foun d in zinc deficient mice ( 110 ) However, the contributi on of this hormon a l change to immune dys function by low zinc intake was determined to be minimal because the zinc effect was yet present in adrenalec tomized animals ( 111 ) These imply that low zinc ingestion leads to a modu lation in host defense by affecting gene expression associated with immune response via a dire ct effect on NF B activity. Functional assays confirming the dietary zinc effect on cytokine production will be further described in the following chapter. Transcription factors potentially mediating the zinc responsive regulation were determin ed by computational analys is identify ing the enriched cis regulatory elements present in the differentially expressed gene promoter s. The results suggest modulated activity of transcription factors, NF Y, AP 2 ETF, Elk 1 and TEF, during zinc depletion. A r ecent promoter analysi s of miRNA identified c Myb, NF Y, Sp 1, MTF 1 and AP 2 as ma ster regulators of their gene expression ( 112 ) The inclusion of NF Y AP 2 and the zinc sensing tra nscription factor MTF 1 suggests a putative role of zinc on miRNA
69 gene regulation. Even though probes against miRNA were present we could not test this hypothesis by our microarray experiments as the PAXgene system used was incompatible for small RNA iso lation Alternatively, circulating miRNA in serum samples were assessed and relevant info rmation will be described in C hapter 5 C omparisons between before and after tre atment values paired by subjects enable the removal of inter person variability, and t hus are suitable for the identification of indices responsive to the treatment. However, a reliable diagnostic tool requires low variance among individuals at a healthy status. The lower numbers of differentially expressed genes determined by unpaired te sting indicate high variation among the baseline levels of those identifi ed by pairwise compa rison. Thus, the genes suggested as potential biomarkers were suggested by statistically aff ected genes based on unpaired t tests. By applying a st ringent filter ing criterion we identified eight well characterized gene transcripts as candidate indices of dietary zinc deprivation. All of these were up regulated by the dietary zinc restriction protocol. Whether other clinical or nutritional conditions affect each of th ese gene transcripts cannot be determined by the current study, and thus requires further exploration. H owever, a simultaneous measurement of these highly respon sive genes as a signature profile may sufficiently serve as a diagnostic approach for di etary zinc deficiency Based on our knowledge, this is the first study to conduct whole genome expression analysis and measure zinc transporter transcripts in hu man subjects under experimental zinc restriction The two major purposes of the current study was to identify mRNAs that hold the potential to reflect the host s zinc intake levels, and to evaluate the physiological effects of short term zinc depletion based on whole gene
70 expression profiling. Our data suggest genes involved in the regulation of z inc homeostasis, cell proliferation and immune response as biomolecules applicable to the diagnosis of zinc deficiency. The microarray data presented provide target molecules for future research for the understanding of molecular mechanism of zinc effects on immune response and cancer development.
71 Figure 3 1. Dietary regimen for acute dietary zinc depletion. Supplemental phytate (1.4 g/d) was added to the diet during the depletion to limit the bioavailability of zinc. a Analytical measures of zinc co nt ent by inductively coupled plasma optical emission spectrophotometry with sample diets. b Estimated zinc content based on calculations by adding the zinc content of the supplement used (15 mg) and the average zinc level of sel f selected diets from 24 h die t recalls (14.5 mg).
72 Figure 3 2. Measures of s erum zinc concentrations indicatin g the effectiveness of the experimental diet for dietary zinc depletion Serum was diluted with 3 volumes of Milli Q water and zinc content was m easured by atomic absorption spectrophotometry Values are expressed as mean standard deviation (SD), and those significantly differe nt compared to baseline (Day 7) levels are noted by *, P < 0.05; ***, P < 0.001 (n = 9 subjects )
73 Table 3 1 Hematolog ical parameters measured during dietary zinc depletion Measures Normal Range Days of Depletion 0 6 10 Mean (SD) Mean (SD) Mean (SD) WBC (thousand/mm 3 ) 4.0~10.0 5.78 (1.43) 5.8 (1.39) 5.97 (1.50) Hemoblobin (g/dL) 13.0~16.5 15.9 (0.98) 15.44 (1.09) 15.41 (1.11) Hematocrit (%) 39.0~49.0 44.83 (2.49) 44.08 (3.02) 43.18 (2.97) Platelet Count (thousand/mm 3 ) 150~450 282.9 (29.4) 266.7 (32.7) 249.9 (37.9) RBC (million/mm 3 ) 4.5~5.9 5.379 (0.46) 5.279 (0.49) 5.179 (0.45) Mean Cell Volume (micron 3 ) 78.0~1 00.0 83.64 (5.09) 83.84 (5.12) 83.56 (5.15) Mean Cell Hemoglobin (pg) 26.0~34.0 29.64 (2.01) 29.35 (1.68) 29.8 (1.62) MCHC (g/dL) 1 31~37 35.41 (0.77) 35.03 (0.68) 35.7 (1.15) CHCM (g/dL) 2 32.0~38.0 36.34 (1.21) 35.59 (1.06) 36.38 (1.49) RBC Distance Wi dth (%) 11.0~14.0 12.96 (1.20) 13.15 (1.19) 12.79 (1.22) Mean Platelet Volume (fL) 6.0~10.0 7.91 (0.41) 8.07 (0.83) 8.84 (0.64) 1 MCHC, m ean corpuscular h emoglobin c oncentration 2 C HCM, c ell hemoglobin concentration mean
74 Figure 3 3. Effects of acute dietary zinc depletion on buccal MT and ZnT1 transcript abundance. ( A) Amplification curve of MT and ZnT1 transcripts in human reference RNA (Stratagene) and buccal RNA produced by qPCR. These w ere the only zinc related gene transcript s being stably detected, i.e., with 1) low variation, 2) amplification curves parallel to th ose from h uman reference RNA, and 3) Ct values lower than 35 in buccal RNA ( B ) Relative MT and ZnT1 mRNA abundance during dietary zinc depletion. Values were norm alized to 18S rRNA levels and baseline level s for each individual were set at 1. Data are expressed as mean standard error of the means (SEM). indicates significant difference to baseline levels at P < 0.05 (n = 9 subjects). A B
75 Figure 3 4. Effects of acute dietary zinc depletion on zinc related gene transcripts in p eripheral blood mononuclear cells (PBMC). Values were normalized to GAPD H mRNA levels and baseline level s for each individual were set at 1. Data are expressed as mean SD. Values sig nificantly different to respective baseline levels are *, P < 0.05; **, P < 0.01; ***, P < 0.001 (n = 9 subjects).
76 Figure 3 5. Effects of acute dietary zinc depletion on zinc related gene transcripts in circulating reticulocytes. ( A) Cellulose columns made in house were used for the purification of erythroid cells by leukocyte depletion ( B) Relative MT and zinc transporter mRNA abundance before and after dietary zinc restriction. Values were normalized to GAPD H mRNA levels and base line level s for each individual were set at 1. Data are expressed as mean SD. Values significantly different to respective baseline levels are **, P < 0.01; ***, P < 0.001 (n = 9 subjects). A B
77 Figure 3 6 Association between serum zinc concentrati ons and transcript abundance of the zinc transporters most significantly responsive to dietary zinc depletion. (A) PBMC ZnT1 and (B) reticulocyte Zip 3 transcript abundance were normalized to their respective GAPDH mRNA levels. Correlation with serum zinc levels was determined by correlation coefficient s ( R ) determined by linear regression analys e s White, baseline levels; black, measures after 10 d of zinc depletion. A B
78 Figure 3 7 Effect of acute dietary zinc depletion on whole blood ZnT1 transcript levels. (A) Reduction in whole blood ZnT1 transcript levels by zinc restriction. ZnT1 mRNA abundance decreased in both PBMC and reticulocyt es by zinc restriction. Values were normalized to GAPD H mRNA levels Data are expressed as mean SD ** indicates significant differenc e to baseline levels at P < 0.001 (n = 9 subjects). (B) Correlation with serum zinc levels was determined by the correlation coefficient ( R ) determined by linear regression analysis. White, baseline levels; grey, m easures after 6 d of zinc depletion; black, measures after 10 d of zinc depletion. A B
79 Figure 3 8 Quality assessment of whole blood RNA. Purity of RNA was determined by A260/A230 and A260/A280 ratios from the Nanodrop 1000 Ratios were 2 .05~ 2 .15 and 2.13~ 2.21 respectively PAXgene system was used for RNA stabilization and isolation. Integrity of RNA was analyzed by using the Agilent 2100 Bioanalyzer (A) Bioanalyzer electropherograms of RNA from whole blood collecte d on day 7 and 17 from s ubject 1 are presented as representative examples. (B) RNA integrity numbers (RIN) of whole blood RNA prepared by using the PAXgene system. A ll RNA samples were highly intact (RIN > 8.0) A B
80 Figure 3 9 RNA recovery after processing for microar ray analysis. (A) RNA recovery after globin RNA reduction by using GLOBINclear (Ambion). A260/A230 and A260/A280 ratios of the cRNA products of RNA amplification were at 2.12~ 2.31 and 2.1 9 ~2. 30, respectively. (B) Yields of RNA after RNA amplification of individual whole blood RNA samples treated for globin RNA reduction. (C) RNA recovery of amplified pooled RNA samples processed with out (PAX) or with the globin reduction procedure (GRP) (D) Amplification efficiency shown as folds of starting amount of individual and pooled GRP, and pooled PAX samples A D B C
81 Figure 3 10 Quality of amplified cRNA assessed by Agilent 2100 Bioanal yzer. Representative electrophero gram of amplified (A) whole blood RNA (PAX) and (B) globin RNA reduced whole blood RNA (GRP) are shown. The pea k only present in the electrophro gram of PAX indicates the highly abundant globin RNA content. Removal of this peak was confirmed by all globin RNA reduced samples. A B
82 Figure 3 11 Quality assessment plo ts of microarray data generated by Genome Studio. (A) H ybridizatio n controls, high > medium > low ( B ) N egative controls (background and noise), low ( C ) L ow stringency control perfect match > mismatch ( D ) G ene intensity (housekeeping and all genes), higher than background (housekeeping > all genes) (E) B iotin high All parameters corresponded to (F ) the expec ted values provided by Illumina Data are expressed as mean SD (n = 24 arrays). A B C D E F
83 Figure 3 1 2 Scatter plots of microarray data b efore and after normalization of pooled whole blood RNA depleted of globin RNA ( GRP) ( Day 7 vs. Day 17) (A) No normalization. (B) Average normalization. (C) Quantile normalization. Qua ntile normalization producing a highly symmetric distribution betw een day 7 and 17 values, was selected for the downstream data analyses. A B C
84 Table 3 2 Number of differential ly express ed genes determined by using various P value and fold change combinations All Genes 1 P < 0.05 P < 0.01 P < 0.005 P < 0.001 P < 0.0001 Pai red by Subjects 2 All FC 20,7 52 2,565 631 328 105 41 FC > 1.1 2,035 625 328 105 41 FC > 1.2 532 249 162 87 41 FC > 1.5 51 28 25 22 21 FC > 2.0 15 15 15 14 14 Unpaired 3 All FC 20,7 52 1,256 221 108 44 10 FC > 1.1 1,017 213 107 44 10 FC > 1.2 29 2 106 73 38 10 FC > 1.5 29 24 23 18 9 FC > 2.0 15 15 14 14 8 1 Genes of which detection P value was lower than 0.05 in at least one sample. 2 Pairwise comparison was conducted to identify genes of which transcript level was changed after dietary z inc restriction. 3 Less differentially expressed genes were detected when inter subject variability was taken into account by unpaired class comparison
85 Figure 3 13 Differential expression of 328 genes identified by pairwise comparis ons at P < 0.005 (A) Differentially expressed ( DE ) genes were cluste red by a verage linkage hierarchical clustering with Pearson correlation metric Clustering of conditions by (B) all genes and (C) DE genes Baseline and zinc depleted condit ions were p artially separated when clustered by DE genes Numbers indicated each individual subject (n = 9). A B C
86 Figure 3 1 4 Effects of globin RNA reduction on t he detection of differentially expressed genes (A) Venn diagram showing the numb er of zinc responsive genes undetected under the presence of globin RNA. RNA from whole blood collected on day 7, 13 and 17 were pooled, respectively. Signal intensity and detection P values of microarray data from whole blood RNA (PAX) and globin RNA de pleted RNA (GRP) were compared. Transcripts were considered present when detection P value was lower than 0.05 in at least one sample of each group Differentially expressed genes (baseline vs. ZnD) were determined as for Figure 3 13. ( B ) S ignal intensi ties of zinc responsive gene transcripts affected by globin RNA in PAX and GRP samples S olid and dashed lines across bars indicate median and mean values, respectively. A B
87 Figure 3 1 5 Temporal expression pattern of differentially expressed genes durin g dietary zinc depletion. RNA from whole blood collected on day 7, 13 and 17 were pooled, respectively, and treated for globin RNA depletion prior to microarray analysis. The up regulated and down regulated genes identified from day 7 and 17 samples from each individual show a temporal trend of increase and decrease respectively, during the dietary zinc depletion phase.
88 Figure 3 1 6 Clustering of zinc responsive genes by expression patterns for functional interpretation. Differentiall y expressed (DE) genes identified by paired t test at P < 0.005 (3 28 genes) were selected for functional enrichment analyses. K means algorithm ( k = 2) was used to cluster the genes by their mode of responsiveness o acute dietary zinc depletion. Cluster 1, up regulated genes; cluster 2, down regulated genes.
89 Table 3 3 Gene ontology (GO) enrichement by genes responsive to acute dietary zinc depletion 1 Gene Ontology Category GO Number P Value Corrected P Value 2 Frequency in Cluster (%) 3 Gene List Up reg ulated Genes (Cluster 1) Nuclear D ivision GO:0000280 3.90E 15 0.001 9.27 DLGAP5, TPX2, NUSAP1, CDC20, BIRC5, AURKA, AURKB, UBE2C, DCTN1, SPC24, KIF2C, CDCA8, CCNB2, RCC2, NCAPG, BUB1, CCNA2, CDCA5 Establishment of O rganelle L ocalization GO:0051656 3.69E 06 0.007 3.09 DLGAP5, NUSAP1, BIRC5, COPG, MYH9, CDCA5 Positive R egulation of C ell C ycle P rocess GO:0090068 8.15E 06 0.014 2.06 DLGAP5, NUSAP1, BIRC5, UBE2C ATP B inding GO:0005524 1.89E 05 0.029 13.4 AURKA, ABCA1, AURKB, TK1, KIF2C, PTK2B, TAP1, B UB1, TRIP13, NLRP7, ABCB9, CAMK1G, AARS, TPX2, PIM2, MCM2, MYH9, MCM4, ABCG1, NLRP1, HYOU1, HSP90B1, UBA1, FBXO18, SEMA4A, KIF20A Regulation of M itotic C ell C ycle GO:0007346 2.46E 05 0.032 3.6 DLGAP5, BUB1, NUSAP1, BIRC5, UBE2C, CCNA2, CDT1 1 Identified by Tool for A n alysis of GO enrichments (TANGO ) of EXPANDER with zinc responsive genes determined by a pairwise comparison ( P < 0.005). Threshold P value for significance was set at 0.005. 2 Empirical P values corrected for multiple comparison computed b y TANGO algorithm 3 Percentage of genes related to each functional class within the clusters generated by the K means algorithm.
90 Table 3 4 Cis regulatory elements enriched in the putative p romoter regions ( 1000~+200) of genes responsive to dietary zi nc depletion 1 Transcription Factor TRANSFAC ID P Value Enrichment Factor 2 Target Gene (Location of Putative TF Binding Site) Up regulated Genes (Cluster 1) NF Y M00287 5.98E 7 1.692 PDIA4( 61, 41,104), SPC24( 213, 115), NUSAP1( 59), CDCA5( 745), CIC( 765, 716), CDC45L(158), AARS( 51), TPX2(99,135), AURKA( 11,21), RRBP1( 60), SLC38A10( 85), SHMT1( 39), SDF2L1( 138, 19), UHRF1( 88), ALDH6A1( 226, 134, 91, 33, 5), FAM46A( 737), TK1(98,129), KIF20A(41), HSP90B1( 171, 118), SRPR( 257, 156), MCM4( 59), CDCA 8(17), CXXC1( 923), MGC29506( 50), NARF( 122, 82, 46, 6), PRR11( 479, 345, 254, 133, 105, 39), PIM2( 124), UBE2C( 68, 37), SLC1A4( 181), CCNA2( 82), BUB1( 61, 35), GLDC( 97), SORT1( 189, 128), P4HB( 256, 140), MAGT1( 216, 169, 79), AURKB( 92), OTUD5( 847, 480), CCNB2( 100, 67), MLEC( 647, 129, 73), OXCT2( 294), TRPM3( 125), CDC20( 79, 34), SPCS2( 52), FBXO18( 97), PTK2B(133), DLGAP5(1,66), GMPPB( 190, 124, 51) AP M00469 0.0020 1.605 HYOU1( 240), ALOX5( 137), CLPTM1L( 811, 416), NUSAP1( 362), CIC( 771, 527, 490, 63), CDC45L( 423), RRBP1( 523), TXNDC11(15), TRAM2( 444, 395, 247), SLC38A10( 184), SDF2L1(55), ARID3A( 432, 237), FAM46A( 639), C19ORF10( 66), KIF1B( 403), DUSP5( 61), UBE2J1( 193), HSP90B1( 820), SRPR( 116), NARF( 539), PIM2(106), CDT1( 334), NLRP1(76), UBE2C( 336), SLC1A4( 656,60), SORT1( 296), P4HB( 703), MCM2( 133), OTUD5( 334), OXCT2(36), TLN1( 323), PTK2B( 277), DLGAP5( 158), APOBEC3B( 125), LAMA 5( 213), NFKB2( 435), GMPPA( 678), GMPPB( 510) ETF M00695 9.4E 4 1.374 PDIA4( 126,50), ALOX5( 911, 708,52), MTF1( 518), TYMS( 51,52,80,114), BIRC5( 52,110), ZYX( 541,15), CDC45L( 300, 285,199), TPX2(200), MGAT3(97,142), TRAM2( 469), SEC23B( 95), SDF2L1(29 ), UHRF1(18), SEC13(108), COL18A1(10), UBE2J1( 92), SRPR( 9), SPATS2( 430, 414), CDT1(46), PIM2( 401), SLC1A4( 208, 195), FAHD2B( 146,126,135), BIK(43), SLC35B1( 416,27), P4HB( 459), KIAA0226( 246,25), EHBP1L1(104), ADAM19(53), DENND5B( 114,37,148,157), KI F2C( 37), OTUD5( 408), TRPM3( 211, 91), FBXO18( 279, 60, 38), C4ORF28( 106), LAMA5( 969,15), ITM2C(52), SORL1( 723,86), HYOU1( 131), TRIP13( 315, 295,64), CKAP4( 211, 116), FLOT2( 132), CLPTM1L( 204, 87, 68), CIC( 745, 724, 468), AARS( 76), STT3A( 732), RR BP1( 118,193), TXNDC11(22), SLC38A10(4), SHMT1( 101,92,142), SIL1(17,47), IFNAR2(164), TK1(116), KIF1B( 928, 846, 732, 445), DUSP5(105), HSP90B1( 318), ZNF341(101), MCM4( 724), CDCA8( 56), ABCG1( 110), NARF( 438, 222, 213,126), BTBD12(10), PRR11( 709, 550) CCNA2( 30), BUB1( 39,34), UBA1(86,151), BMP8B( 243,198), MLEC( 227, 37), MYH9(89), CDC20( 305, 139), POLR2A( 388), GMPPA( 867), SEC61A1(176), GMPPB( 486)
91 Table 3 4 Continued Transcription Factor TRANSFAC ID P Value Enrichment Factor 2 Target Gene (Loc ation of Putative TF Binding Site) Down regulated Genes (Cluster 2 ) Elk 1 M00025 0.0020 1.628 LEMD3( 38), TMEM9B( 2), CREBZF( 905), ZNF256( 149), VTA1( 45), KCTD5( 124,9), CRYZ( 273), C9ORF78(6), ZNF594( 176,12), PQLC3( 731), LEPROTL1( 21), SLFN12( 7 48), RAB11A( 732, 223, 17), C1ORF52( 22, 1), PALB2( 116), SENP7( 123, 93), TAF7( 257), FLJ14213( 411), C20ORF30( 39), ZCCHC7( 6), COX7A2L( 778), ZNF550( 23), CRIPT( 3), MIS12( 92, 74) TEF M00672 0.0030 1.621 C6ORF190( 38), ZNF187( 202, 140), PTGER2( 968, 827), KLRD1( 605), MS4A1( 547), GZMK( 451, 47), C9ORF78( 906), CCNB1IP1( 86), ZNF594( 951), TSPAN13( 553), RAB11A( 845), TAF7( 981, 876), RGS18( 306,49), SETMAR( 155), ERP27( 836, 181), FASLG( 286), AKR1C3( 172, 10), GNG2( 969), HOPX( 548), GK5( 667) 1 I dentified by P r omoter Integration in Microarray Analysis (PRIMA ) of EXPANDER with zinc responsive genes determined by a pairwise comparison ( P < 0.005) Threshold P value for significance was set at 0.005. 2 Ratio of prevalence of TF hits in t he cluster to that in all genes
92 Table 3 5 Top 5 functions over represented by all genes differentially expressed after acute dietary zinc depletion 1 Functional Category P Value 2 Gene List Cellular Assembly and Organization 1.02E 08 ~ 2.38E 02 DLGAP5, CDT1, KIF1B, NCAPG, CCNB2, AURKB, TAP1, ABCA1, MCM4, BIRC5, CCNA2, TOP2A, HJURP, KLHDC5, KIF2C, FASLG, RRBP1, NUSAP1, AURKA, LILRB1, RPA2, BUB1, WAS, DCTN1, MYH9, MIS12 DNA Replication, Recombination, and Repair 1.02E 08 ~ 2.5E 02 DLGAP5, PTK2B, CDT1, NCAPG, HSF2, CCNB 2, AURKB, MCM4, BIRC5, INTS3, CCNA2, BMI1, RARA, TOP2A, HJURP, CDCA5, KIF2C, FASLG, TYMS, SETMAR, TRIP13, CDC45, NUSAP1, DCK, SLX4, AURKA, RPA2, BUB1, MCM2, PPIA, BIK, MIS12, ALOX5, TK1 Cell Cycle 7.74E 08 ~ 2.23E 02 KIF20A, DLGAP5, CDC20, CDT1, NCAPG, HSF2 CCNB2, AURKB, BIRC5, CCNA2, BMI1, CCNB1IP1, RARA, TOP2A, CD38, HJURP, CDCA5, KIF2C, TYMS, CDC45, TRIP13, IL4R, NUSAP1, CDCA8, SMARCE1, IRF9, AURKA, TPX2, MST4, ARID3A, BUB1, RPL5, MCM2, WAS, BIK, STX16, MYH9, RAB11A, DCTN1, MIS12, PIM2, UBE2C Cellular M ovement 1.89E 06 ~ 2.23E 02 KIF20A, LTBP2, PTK2B, CDC20, PTGDR, POU2AF1, HSF2, TLN1, AURKB, BIRC5, CD226, TOP2A, CD38, FASLG, LAMA5, NUSAP1, NFKB2, AURKA, CCR9, WAS, CXCR7, PPIA, CCL3L1/CCL3L3, MYH9, STX16, RAB11A, ALOX5 Cancer 8.23E 06 ~ 2.37E 02 DLGAP5, AKR 1C3, KIF1B, NLRP7, NCAPG, NLRP1, CCNB2, ABCG1, AURKB, BIRC5, BMI1, RARA, VPS8, FASLG, SLC2A5, CLNS1A, TYMS, IL4R, P4HB, CDCA8, EIF4G3, AURKA, NFKB2, IFNAR2, TPX2, PALB2, CCR9, CXCR7, ZYX, TRIAP1, PTGER2, ALOX5, TK1, UBE2C, CRYZ, KIF20A, TCF4, SLC1A4, CDC20 MS4A1, SOCS2, HYOU1, SPC24, SIL1, MCM4, IGLL1/IGLL5, MGAT3, CCNA2, DUSP5, HSP90B1, POLR2A, F5, PHGDH, TOP2A, CD38, CDCA5, SORT1, COL18A1, LAMA5, TRIP13, NUSAP1, IRF4, GZMK, UHRF1, DCK, MST4, ARID3A, SERPINE2, BUB1, HOPX, SDF2L1, MCM2, IGJ, PPIA, UBA1, TN FRSF13B, PIM2 1 Identified by Ingenuity Pathway Analysis with zinc responsive genes determined by a pairwise comparison ( P < 0.005). 2 Range of F isher s e xact test P value s of functional annotations assigned to each respective functional catagory
93 T able 3 6 Top 5 functions over represented by genes up regulated after acute dietary zinc depletion 1 Functional Category P Value 2 Gene List Cell Cycle 1.03E 08 ~ 1.64E 02 KIF20A, DLGAP5, CDC20, CDT1, NCAPG, CCNB2, AURKB, BIRC5, CCNA2, RARA, TOP2A, CD38, H JURP, CDCA5, KIF2C, TYMS, CDC45, TRIP13, NUSAP1, IL4R, CDCA8, IRF9, AURKA, TPX2, ARID3A, BUB1, MCM2, WAS, BIK, MYH9, DCTN1, PIM2, UBE2C Cellular Assembly and Organization 1.9E 08 ~ 1.64E 02 DLGAP5, PTK2B, CDT1, KIF1B, NCAPG, CCNB2, AURKB, TAP1, MCM4, ABCA1 BIRC5, CCNA2, TOP2A, HJURP, KIF2C, COL18A1, RRBP1, NUSAP1, UHRF1, AURKA, LILRB1, TPX2, BUB1, WAS, DCTN1, MYH9, ZYX DNA Replication, Recombination, and Repair 1.9E 08 ~ 1.64E 02 DLGAP5, PTK2B, CDT1, NCAPG, CCNB2, MCM4, BIRC5, INTS3, CCNA2, RARA, TOP2A, CD CA5, HJURP, KIF2C, TYMS, CDC45, TRIP13, NUSAP1, SLX4, AURKA, TPX2, BUB1, MCM2, BIK, ALOX5, TK1 Cancer 1.88E 07 ~ 1.75E 02 DLGAP5, KIF1B, NLRP7, NCAPG, NLRP1, CCNB2, ABCG1, AURKB, BIRC5, RARA, VPS8, SLC2A5, TYMS, IL4R, P4HB, CDCA8, EIF4G3, AURKA, NFKB2, IFN AR2, TPX2, CCR9, ZYX, ALOX5, TK1, UBE2C, KIF20A, TCF4, SLC1A4, CDC20, HYOU1, SPC24, SIL1, MCM4, IGLL1/IGLL5, MGAT3, CCNA2, DUSP5, HSP90B1, POLR2A, F5, PHGDH, TOP2A, CD38, CDCA5, SORT1, COL18A1, TRIP13, LAMA5, NUSAP1, IRF4, UHRF1, ARID3A, BUB1, SDF2L1, MCM2 IGJ, UBA1, TNFRSF13B, PIM2 Genetic Disorder 7.69E 07 ~ 1.72E 02 DLGAP5, KIF1B, SLC38A10, NLRP1, CCNB2, ABCG1, TLN1, AURKB, ABCA1, BIRC5, TNFRSF17, INTS3, RARA, ALDH6A1, TYMS, PDIA5, IL4R, P4HB, CDCA8, SORL1, MAGT1, EIF4G3, AURKA, ZBTB43, NFKB2, IFNAR2, LILRB1, CCR9, FLOT2, ZYX, ALDH3B1, ALOX5, PACRGL, TK1, UBE2C, TCF4, SLC1A4, PTK2B, CDC20, CKAP4, SIL1, MCM4, IGLL1/IGLL5, TRPM3, CCNA2, DUSP5, HSP90B1, SEC61A1, POLR2A, F5, SEC23B, ADAM19, TOP2A, PHGDH, CD38, SORT1, HJURP, CDCA5, SEMA4A, COL18A1, SEMA3E, T RIP13, CDC45, CIC, LAMA5, IRF4, UHRF1, SLC35B1, COPG, SDF2L1, MCM2, WAS, IGJ, MYH9, DCTN1, SLC12A9, SPATS2, UBA1, TNFRSF13B, GLDC 1 Identified by Ingenuity Pathway Analysis with zinc responsive genes determined by a pairwise comparison ( P < 0.005). 2 R ange of Fisher s exact test P values of functional annotations assigned to each respective functional cat e gory.
94 Table 3 7 Top 5 functions over represented by genes down regulated after acute dietary zinc depletion 1 Functional Category P Value 2 Gene Li st Cell Death 1.02E 04 ~ 4.93E 02 CLNS1A, SETMAR, GZMK, CD160, DCK, PTGDR, KLRD1, SOCS2, HSF2, MS4A1, SMARCE1, SEDLP, SERPINE2, MST4, BCL2L13, FAU, BMI1, PPIA, CD226, CCL3L1/CCL3L3, GNG2, TRIAP1, PTGER2, FASLG Cell mediated Immune Response 1.02E 04 ~ 4.05E 02 KLRD1, CD226, CCL3L1/CCL3L3, FASLG Cellular Development 1.02E 04 ~ 4.55E 02 AKR1C3, KLRD1, STK39, SMARCE1, TSPAN13, TAF7, MST4, BMI1, CXCR7, PPIA, CD226, PTGER2, FASLG Cellular Function and Maintenance 1.02E 04 ~ 4.55E 02 KLRD1, PPIA, HSF2, CRIPT, SMAR CE1, CD226, PTGER2, FASLG Hematological System Development and Function 1.02E 04 ~ 4.55E 02 CLNS1A, BMI1, PTGDR, CXCR7, KLRD1, PPIA, SOCS2, CD226, CCL3L1/CCL3L3, PTGER2, FASLG, SERPINE2 1 Identified by Ingenuity Pathway Analysis with zinc responsive gene s determined by a pairwise comparison ( P < 0.005). 2 Range of Fisher s exact test P values of functional annotations assigned to each respective functional cate gory.
95 Figure 3 17 Functional networks of genes differentially expressed after acute dietary zinc depletion. Top two networks identified by Ingenuity Pathway Analysis Each is associated with (A) c ell c ycle, c ellular m ovement, c ellular a ssembly and o rganization and (B) c ellular g rowth and p roliferation, h ematological s ystem d evelopment and f unction, c ellular d evelopment respectively. Red up r e gulated by zinc restriction; green down regulated by zinc restriction A B
96 Figure 3 18. Functional network enriched by genes up regulated and down regulated after acute dietary zinc depletion. (A) Top n etwork of up regulated genes associated with c ellular a ssembly and o rganization, DNA r eplication, r ecombination and r epair, and c ell cycle (B) Top network of down regulated genes associated with c ell d eath, c ell mediated i mmune r esponse and c ellular d evelopment Red up re gulated by zinc restriction; green down regulated by zinc restriction A B
97 Figure 3 19 Differential expression of 203 genes identified by unp aired t test at P < 0.005. Differentially expressed (DE) genes wer e clustered by average linkage hierarchical clustering wi th Pearson correlation metric. A clear distinction between baseline and zinc depleted condition is shown when clustering is conducted with the DE genes.
98 Table 3 8 Genes holding the potential as an indicat or of dietary zinc deficiency in individuals 1 Gene Symbol Entrez ID Gene Name Fold Change (ZnD/Baseline) P Value FDR CDC20 991 cell division cycle 20 homolog (S. cerevisiae) 2.69 2.00E 07 0.00124 TXNDC5 81567 thioredoxin domain containing 5 (en doplasmic reticulum) 2.73 5.00E 07 0.00173 MZB1 51237 marginal zone B and B1 cell specific protein 2.07 7.00E 07 0.00182 IGJ 3512 immunoglobulin J polypeptide, linker protein for immunoglobulin alpha and mu polypeptides 4.22 1.00E 06 0.00189 IGLL1 3543 immunoglobulin lambda like polypeptide 1 2.77 1.00E 06 0.00189 CD38 952 CD38 molecule 2.03 2.07E 05 0.0178 0 GLDC 2731 glycine dehydrogenase (decarboxylating) 2.49 2.30E 06 0.00367 TNFRSF17 608 tumor necrosis factor receptor superfamily, member 17 2.55 1.85E 05 0.0174 0 1 Baseline and post zinc depletion values were grouped as to represent zinc adequate (ZnA) and zinc deficient (ZnD) conditions. Total of 8 well characterized genes were identified to be significantly different between each group ( P < 0. 005 by t test without pairing) with a fold change larger than 2 (n = 9). FDR f alse discovery rate.
99 Figure 3 20 Valida tion of microarray data using qPCR Relative transcript levels of (A) CDC20 (B) TXNDC5 (C) MZB1 ( MGC29506 ) and (D) IGJ in whole blood RNA during dietary zinc depletion were measured by qPCR Values were normalized to GAPDH mRNA levels. Solid and dashed lines across bars indicate median and mean values, respectively. Values significantly different to respective b aseline levels are *, P < 0.05; **, P < 0.01; ***, P < 0.001 (n = 9 subjects). A B C D
100 CHAPTER 4 EFFECTS OF ACUTE DIE TARY ZINC DEPLETION ON WHOLE BLOOD CYTOKINE RELEASE INDUCED BY EX VIVO IMMUNOCHALLENGES Introductory Remarks The essentiality of adequate zinc inta ke has been derived from its involvement in various physiological events including immunity. The association of dietary zinc deficiency with c linical symptoms reflecting imp aired immunity, such as thymic atrophy and increased risks of bacterial, viral and fungal infections, imply the significance of sufficient zinc supply for maintenance of the immune system ( 1 2 22 ) Additionally, low serum zinc levels have been associated with a decrease in cytokine produc tion by leukocytes in human s ( 63 96 ) The beneficial effects of zinc supplementation on th e activation of immune cells, i.e., granulocytes, lymphocytes and monocytes have been shown by increases in cytokine production induced by ex vivo immunostimulation with human subjects ( 41 ) Recent in vitro and animal experiments indicate the necessity of zinc and zinc transporter functi on for the expression of genes responsible for immune responses ( 22 ) Direct involvement of zinc molecules in the signaling pathways mediating in activation and cytokine production of T cells and dendritic cells has been identified ( 51 52 ) Consequently inadequate dietary intake of zinc lead s to a modulation in immunity, and thus functional assessment of these features may provide means for the diagnosis of dietary zinc deficiency Whole blood cytokine assays, which involve ex vivo challenge of cells to either lipopolysaccharides (LPS) or phytohemagglutinin (PHA), have been employed as a meth od to evaluate the in vivo effects of various conditions on immune response ( 113 ) Research with whole blood assays impli es the appropriateness of this methodology as a diagnostic tool for various conditi ons such as tuberculosis ( 114 ) sepsis ( 115 ) multiple
101 sclerosis ( 116 ) and HIV infection ( 117 ) Additionally, association of immunity with vitamin A store level s ( 118 ) and serum high density lipoportein levels ( 119 ) was identi fied by using this approach. T he classical approach for the identification of responses of a protein to a biological effect is based on immunological labeling of the target protein and visualization by an enzyme conjugated protein that produces coloromet ric, autoradiographic or chemiluminescent signals upon substrate treatment. Relevant techniques include western analysis and enzyme linked immunosorbent assay (ELISA). Multi analyte ELISA arrays have been used for the identification of molecular effects of a treatment on proteins involved in certain pathways such as cytokines and oth er markers of immune response The platform provides ELISA assays for the quantitation of multiple proteins on a 96 well plate template and thus allows the simultaneous eval uation of the effects of a biological variable on the protein expression levels T he limitation of these techniques are that the proteins of interest should be identified prior to the experiment is conducted. However, when researchers have an idea of the affected list of proteins based on previous findings from relevant research, these tools enable multiplexing which marke dly reduces labor and cost intensity. Additionally, by utilizing commercially available platforms along with standard peptides, the i nter laboratory variable can be minimized particularly with regards of the nature of the technique allowing absolute quantitation. T he aim of the study was to determine the effects of acute dietary zinc depletion on the levels of whole blood cytokine relea se induced by ex vivo exposure to LP S or PHA using the recently introduced multi analyte ELISA array platform The validity of
102 this approach for the diagnosis of zinc deficiency was further assessed by absolute quantitation of secreted cytokine levels by using single analyte ELISA arrays as well. Additionally, differential expression of zinc transporters by ex vivo activation and the effect of the hosts zinc intake level on its profile were assessed by qPCR. Results Inflammatory Cytokine Production by Whole Blood W hole blood cytoki ne assay w as implemented as an approach to test if the modulating effects of zinc on the capability of immune cell activation can be used as a tool for identifying dietary zinc deficiency T o identify the cytokines of which p roduction induced by immunostimulation is affected by dietary zinc depletion, cell free supernatants collected on day 7 and 17 were pooled and analyzed using a mu lti analyte ELISA array designed for human inflammatory cytokines Among the twelve cytokines tested, only TNF showed a noticeable effect of dietary zinc depletion on its production induced by challenges with LPS and PHA (Figure 4 1) In order to confirm this effect of dietary z inc deprivation, supernatant collec ted from individual whole blood incubations were s ubject to ELISA specific for IL 1 IFN and TNF quantitation. Prominent induction of the monokine IL 1 by LPS confirmed the activation of monocytes present in the whole blood samples during the ex vivo immunostimulation (Figure 4 2) A substantial inc rease in lymphokine IFN secretion indicate d the successful activation of lymphocytes in the PHA treated whole blood samples as well (Figure 4 3). As observed by the multi analyte screening with pooled sampl es, there were no differences between whole bloo d samples collected before (baseline) and after dietary zinc depletion (Zn depleted) by IL 1 and IFN responses to LPS or PHA. In contrast, whole blood samples collected at the post depletion phase secreted significantly lower TNF in
103 response to immunos timulation than those collect ed at the baseline (Figure 4 4), indicating the presence of a zinc effect on a pathway specific to the production of this cytokine. Zinc Transporter Transcript Levels T he critical role s of zinc and its transpor ters in immune re sponse implicate the possibilities of zinc transporter expression induced by immune cell activation to respond to the host s dietary zinc status. Hepa rin was used as an anticoagulant for the whole blood samples for the cytokine assays. It is of note that residual heparin present in the whole blood RNA can inhibit reverse transcriptase and polymerase activity ( 120 121 ) and thus RNA purification by LiCl prior to qPCR assays was essential (Figure 4 5). Among the zinc homeostatic gene transcripts measured, only MT mRNA levels showed a statistically significant response to LPS challenge, while MT, ZnT6, Zip1, Zip3 Zip6, Zip8 and Zip14 transcripts were up regulated in response to activation by PHA (Figure 4 6) Only PHA induced Zip8 exp ression was responsive to acute dietary zinc depletion with an approximate decrease of 50% Discussion Impaired immunity is one of the most extensively characterized outcome s of zinc deficiency by various types of research ( 95 97 ) T he clinical significance of adequate zinc intake has been su bstantiated by epidemiological data identifying higher risks of morbidity and mortality from infectious diseases including pneumonia and malaria in populations with high prevalence of zinc deficiency ( 23 88 ) Previously, we have shown the synergistic effects of zinc supplementation on the expression of cytokine genes induced by immune cell activation ( 41 ) The dependence of these genes on zinc transporter activity ( 51 52 ) also underlines the importance of constant zinc levels in the
104 cells conferring immune response. C orresponding to these aspects of zinc in immunity f unctional network analysis of our microarray data p resented in the preceding chapter, identified an overrepresentation of transcripts for genes down regulated during zinc depletion that are associated with cell mediated immune response. Thus, a s an approach to functionally assess the effect s of low zinc i ngestion on the host s immune response we implemented a standardized method, i.e., w hole blood cytokine assay involving an ex vivo challe nge and quantitative measures of cytokine release ( 113 ) Successes in discrimination between patients of various diseases and healthy individuals by using this means implicat e its practicality for molecular diagnosis ( 114 117 ) The results of the present study indicate the potential of TNF production levels induced by blood cell activation as an indicative marker of the host s zinc status The in vitro effect of zinc on TNF secretion fro m blood cells induced by immunoactivators including LPS and PHA has been well documented elsewhere ( 122 ) It is of note that zinc alone can function as an immunoactivator, particularly, by inducing TNF production ( 62 ) The in vivo e ffects of dietary zinc levels on immune cell activation, measured by relevant cytokine transcript level s were recently shown by a d ietary human study with regimen of supplemental zinc Consumption of zinc supplements for a 10 day period resulted in enhancement in activation induced TNF IFN and IL 1 gene expression of monocytes, T lymphocytes and granulocytes, respectively ( 41 ) T he increase in TNF expression by high zinc intake agrees wit h the repression by zinc depletion observed in the present study. In agreement to the observations in the current study, t he effects of zinc on cytokine production were limited to the cells exposed to ex vivo
105 immunoactivators in the previous study on zinc supplementation Th is activation dependent zinc responsiveness of cytokine expressi on emphasizes the significance of sufficient dietary zinc intake for populations particularly at higher risks of infect ion or immunological disorders A previous microa rray study with immune cell lines i dentified tumor necrosis factors as central regulators of the biological events associated with the genes influenced by zinc availability ( 90 ) Additionally, NF B a major regulator of TNF gene expression, was present in the functional network composed of the down regulated genes identified by our transcriptome analysis. The se bioinformatic data suggest th a t impaired TNF production upon activation may be a con sequence of impairment in NF B activity under low zinc conditions. Mechanistic studies with various immune and cancer cell models further support our hypothesis of the involvement of NF B in conferring the zinc effects on immunoactivator induced TNF gene expression ( 106 108 ) It is of note that two TNF receptor superfamily genes, TNFRSF17 and TNFRSF13B w ere identified as zinc responsive by the microarray experi ments of the current study Both receptor genes produced higher transcript levels after the cons umption of the zinc de ficient diet. Upon recognition by TNF receptors, TNF triggers a feedback inhibitory mechanism of TNF gene expression by attenuating the transactivation potential of its inducer NF B ( 123 ) These suggest that the negative feedback regulation enhanced by higher recep tor activities may account for the reduced LPS or PHA induced TNF prod uction by blood cells from zinc depleted individuals The whole blood model has been implemented as a means to evaluate the immune system of patients under various clinical conditions ( 113 ) The use of whole
106 blood for biomolecular diagnostics holds several advantages over the use of isolated blood mononu clear cells The lack of extensive processing for cellular fractionation enables the implementation of assays in a timely efficient manner and under such conditions where specific laboratory equipment or reagents are absent Additionally, because whole b lood culturing conveys all blood components to the ex vivo cellular environment, the cytokine assay results allow the prediction of the in vivo outcome of immune cell activation under the physiological condition of interest The use of inappropriate antic oagulants can markedly affect the results, and thus relevant precaution during blood sampling was taken for the current study Even though, other anticoagulants hold the advantage for PCR based assays ( 85 ) heparin was selected as an anticoagulant for the pre sented experiments. EDTA and citric acid based reagents were excluded due to their respective potential to chelate zinc ( 124 ) and to modulate plasma or cellular metabolite concentrations ( 46 ) which would potentially mo dulate the zinc bio availabili ty during in vitro incubation The limitation of heparin inhibiting reverse transcriptase and DNA polymerase activities during qPCR ( 120 121 ) was overcome by implementing an additional step of RNA purification by using lithium chloride ( 85 ) The release of zinc ions compartmentalized in intracellular vesicles to the cytosol has been suggested as a regulatory mechanism of various signaling pathways mediating immune responses ( 21 22 ) In addition to its effects on the DNA binding activity of transcription factors, zinc molecules exploit their regulatory role as inhibitors or activators o f enzymes involved in the regulation of signaling molecules, such as phosphatases and kinases ( 51 125 129 ) Z inc transporters Zip8 and Zip6 have been shown to mediate the redistribution of cellular zinc during the activation of T
107 lymphocytes ( 51 ) and dendritic ce lls ( 52 ) respectively Among the responses of zinc transporter transcripts to ex vivo challenging, that of Zip8 mRNA to PHA treatment was most prominent in th e current study These r esults from the whole blood model were comparable to those from a previous screening carried out with isolated human primary T cells ( 51 ) RNA interference of lymphocyte Zip8 result ed in impaired activation of T lymphocytes as indicated by a reduction in the expression levels of an activation marker gene IFN ( 51 ) Potentiating effect of zinc on the activation of T cells was also shown by this study. The NF B mediated regulation of Zip8 has been suggested by its re sponsiveness to TNF ( 130 ) Even though the depressing effect of dietary zinc depletion on the activation induced TNF levels was also identified by LPS treat ed blood the down regulation of Zip8 under such condition s was o nly present in PHA treated whole blood culture Thus, it is unlikely that TNF is mediating the down regulation of Zip8 by the host s suboptimal zinc intake. Consequently, future research f or the identification of a zinc responsive regulatory factor specifically mediating the attenuation of Zip8 responses to T lymphocyte activation is warranted
108 Figure 4 1. Effects of acute dietary zinc depletion on whole blood cytokine production induce d by LPS and PHA in vitro Heprainized whole blood collected on day 0 (baseline) and day 10 (ZnD) of zinc depletion were incubated with LPS (1 g/mL) or PHA (10 g/mL) for 24 h. Cell free supernatant s were pooled for the screening measures by a cytokine focused multi analyte ELISA array. Values of zinc depleted status are expressed as ratios relative to the baseline levels in a log 2 scale (Poo led samples from n = 8 subjects)
109 Figure 4 2 Confirmation of LPS induced monocyte activation in whole blood in vitro Cell free supernatant was collected from whole blood exposed to LPS or PHA for 24 h. Levels of a m onokine IL 1 in each individua l sample were measured by using a single analyte ELISA array. A bsolute concentrations deduced from a standard curve generated with synthetic IL 1 peptides. Data are expressed as mean SD (n = 8 subjects).
110 Figure 4 3 Confirmation of PHA induced lymph ocyte activation in whole blood in vitro Cell free supernatant was collected from whole blood exposed to LPS or PHA for 24 h. Levels of a lymphokine, IFN in each individual sample w ere measured by using a single analyte ELISA array. A bsolute concentr ations were deduced from a standard curve generated with synthetic IFN peptides. Data are expressed as mean SD (n = 8 subjects ).
111 Figure 4 4 Confirmation of the repression in LPS and PHA induced TNF production by acute dietary zinc depletion. Cell free supernatant was collected from whole blood exposed to LPS or PHA for 24 h. TNF levels of each individual sample were measured by using a single analyte ELISA array. A bsolute concentrations were deduced from a standard curve generated with syntheti c TNF peptides. Data are expressed as mean SD. Values significantly different to respective baseline levels are **, P < 0.01; ***, P < 0.001 (n = 8 subjects).
112 Figure 4 5 Effects of residual heparin in RNA samples on PCR amplification. (A) Amp lification plot of MT transcripts in standards prepared by 1/5 serial dilution of RNA isolated from heparinized whole blood. Amplification was detected only when standards were sufficiently diluted in nuclease free water. (B) Amplification plot of MT tra nscripts in standards prepared by 1/5 serial dilution of RNA purified by LiCl 2 after isolated from heparinized whole blood. All standards produced amplification products by qPCR. A B
113 Figure 4 6 Effects of immunostimulation on zinc related gene transcr ipts in whole blood (A) Responsive ness of whole blood MT and zinc transporter transcripts to LPS or PHA induced activation. V alues were normalized to 18S r RNA levels and control level s for each individual were set at 1. (B) Repression in PHA induced Z ip 8 transcript levels by acute dietary zinc depletion. Da ta are expressed as mean SD. Values significantly different to respective baseline levels are *, P < 0.0 5 ; **, P < 0.01; ***, P < 0.001 (n = 3 subjects). A B
114 CHAPTER 5 SERUM MICRORNA AS BIOMARKER S OF DIETARY ZINC STATUS Introductory Remarks The r ecent discovery of microRN A (miRNA) has introduced a promoter independent regulatory mechanism of gene expression. Target transcripts of these small non protein conding RNA are determined by conserved sequen ces present in the 3 untranslated region (3 UTR) which are complementary to the sequence of the functioning miRNA ( 131 ) The biogenesis of functional miRNA from their gene transcript (primary miRNA ; pri miRNA ) requires two major steps of processing each mediated by the M icroprocessor ( D rosha DGCR8 complex) and D icer which generate and trim precursor miRNA (pre miRNA), respectively ( 132 133 ) The single stranded mature form of miRNA in c orporate s into the RNA induced silencing complex (RISC) via Argonaute (Ago) proteins and binds to its recognition s ite at the targeted transcript I nhibition of gene expression occurs by the induction of gene trans cript decay through deadenylation pathway s and translational repression by obstructing the initiation and elongation of peptide synthesis ( 131 ) S erum microRNA (miRNA) profiles have been shown to reflect the developmental and metastatic stage of various cancer types ( 134 ) and thus have become an emerging t arget for cancer biomarker research. The use of serum miRNA as a diagnostic tool has been supported by its high stability a nd low variability among individuals in a healthy sta t e ( 135 137 ) The highly stable nature of miRNA in serum, which contains a substantial amount of RNase activit y is maintained by its presence in cell derived exosomes ( 138 139 ) Even though the complete physiological role of circulating m iRNA is yet to be known its function as a mediator of intercellular communication has
115 been suggest ed ( 139 ) Serum miRNA s are secreted from cells via an energy dependent export mechanism ( 140 ) and their sources are not limit ed to blood cells ( 136 137 ) In other words the profile of circulating miRNA s r eflect s the effect of cl inical conditions on the miRNA profile of various tissues which enables the detection of cancers also unrelated to the hematological system Thus characterization of serum miRNA would be a feasible approach for biomarker discovery in not only cancer rese arch but also the nutrition field. W e hypothesized that miRNA levels in serum may be modulated by dietary zinc depletion and thus hold the potential of being a diagnostic tool for zinc deficiency As the first attempt to comprehensive ly assess the zinc effect on miRNA in human our data introduce miRNA s likely to mediate post transcriptional regulation of genes under dietary zinc deprivation Results Major advantages of using serum miRNA include its resistance against ex vivo temperature fluctuation and its low inter individual variability ( 135 137 ) By using a standardized protocol for the isola tion of total RNA from serum and a qPCR based tool for miRNA detecti on ( 87 ) we implemented the concept developed by cancer biomarker research to the field of nutrient status assessment. Due to the relatively short history of research related to serum miRNA, there is no known endogenous h ousekeeping transcript that can serve as a means of normalization. Thus, equal amount of synthetic cel miR 39 was added into the samples prior to the RNA isolation process. Ct values acquired by the amplification of this exogenous control validate the experimental protocol by indicating consistent yield of serum RNA recovery ( Figure 5 1).
116 T o identify the miRNA circulating in serum that can serve as a biomarker of zinc deficie ncy, total RNA from pooled sera collected before and after dietary zinc depletion w as assessed by using a qPCR array focused on those identified to be present in human serum. Among the serum miRNA s measured total of twenty were show n to respond to the dietary zinc depletion regimen by fold changes above 1.5 and with C t values below 35 (Figure 5 2A) It i s of note that the majority of di fferentially regulated miRNA showed a trend of decrease d abundance in their response to acute dietary zinc depletion (Figure 5 2B) Further analysis with sera collect ed after zinc repletion was conducted to determine if th e effect s of zinc deprivation can be reversed by dietary zinc replenishment. Consequently, we identified n ine miRNA ( miR 10b, miR 155, miR 200b, miR 296 5p, miR 375, miR 92a, miR 145, miR 204, miR 211 ) responding to dietary zinc deprivation and s upplement ation in opposite modes (Figure 5 3). Discussion In 2005, the estimated number of miRNA s in human genome was approximately 800 ( 141 ) Currently, there are 1,424 human miRNAs deposited i n the miRBase database (release 17) ( 142 ) which indicates the contribution of this small miRNA family to the regulation of gene expression to be greater than that expected at the initial stage of discovery More than 45,000 target sites are co nserved within UTRs of human gene products, and above 60% of protein coding transcript s are estimated to be under the regulation by miRNAs ( 143 ) Und erstanding of the role of miRNA in the regulation of genes related to nutrient metabolism is fairly limit ed ( 144 146 ) To da te published experimental evidence of the effec t s of zinc on miRNA expression is nonexistant and rele vant information was identified only through conference reports and as a part of a book chapter, noted as unpublished data T he enhanced miR 34a, miR 1274a miR
117 140 and miR 1949 expression in the small intestine of zinc deficient mice identified by microa rray analysis is of most relevance ( 147 ) T he presence of two zinc responsive miRNA involved in the regulation of Zip5 by supplemental zinc has also been described, however without deta ils ( 148 ) D ietary zinc deprivation has been shown to cause u p regulation of miR 31 and miR 183 and down regulation of miR 183 in precancerous esophagus of rats ( 99 ) Modulation in the miRNA profile was suggested by t he authors as an underlying mechanism by which zinc exerts its anti tumorigenic property. Based on our knowledge, the evaluation of serum miRNA profiles during differential dietary zinc intakes conducted by the current study i s the first attempt to identi fy the effects of zinc on miRNA metabolism in human s The transcriptional machinery of m icroRNA gene expression, generally mediated by RNA polymerase II, shares its regulatory components with that of protein coding transcripts ( 132 ) A recent computational analysis of the regulatory elements at the promoter region of miRNA genes ident ified putative transcription factors involved in the regulation of miRNA gene expression ( 112 ) The zinc sensing transcription factor, MTF 1 was predicted as one of the five master regulators of human pre miRNA expression It is of note that two other predicted master regulators, NF Y and AP 2 were identified as putative transcription factors elicit ing the effects of dietary zinc depletion on gene expression by the microarray dataset of the current study These findings suggest the possible regulatory role of zinc in multiple miRNA gene expression via modulated transcription factor activity As an approach to identify the miRNA responsive to zinc, we utilized a commercially developed PCR array platform focused on miRNA known to be present in
118 serum. Due to the activ ity of serum miRNA research in the cancer field, most of the miRNA s assessed have been identified as regulators of oncogenes or tumor suppressors ( 149 150 ) The normal biological role of circulating miRNA remains unclear. However, its energy depe ndent export by tissues and stable form in the blood circulation suggest its potential to function as an inter cellular or organ messenger ( 138 140 ) Delivery of miRNA eliciting protumo r igen ic effects, secreted from cancerous tissues to healthy tissues may partially contribute to the metastasis of cancer. Conversely, miRNA with antitumoigenic effects may mediate the defense mechanism against carcinogenesis. A mong the miRNA screened, miR 204 and miR 296 5p showed the highest responsiveness of down regulation to dietary zinc depletion. It is of note that both of these miRNA have been recently characterized to hold suppressing effects on oncogene expression. Th e adhesion, migration and invasion of squamous cell carcinoma of the head and neck were reduc ed by restoration of miR 204 in vitro ( 151 ) Ectopic expression of miR 296 resulted in the repression of an oncogene, high motility group AT hook gene 1, in various types of prostate cancer cells and suppressed cell proliferation and invasion were observed as consequences of miR 296 activity ( 152 ) These findings suggest the therapeutic properties of these zinc responsive miRNA on tumor progression, and imply their involvement in the mechanism of the predisposition to cancer by zinc deficiency. T he tissues producing the modulated seru m miRNA profile by zinc deficiency were not identified by the current study, and thus the effect of zinc on tissue miRNA expression needs to be further explored. As identified by TargetScan (Table 5 1) several zinc related genes are potential targets of the differentially expressed serum
119 miRNA. Thus, the zinc responsive serum miRNA provided here may function as candidate miRNAs for fu ture studies focusing on the role of miRNA s in the regulation of biological events affected by the levels of dietary zinc intake
120 Figure 5 1 Serum processing for miRNA isolation (A) Schematic workflow of serum RNA isolation and data analysis. (B) Yield of cel miR 39 spiked into serum prior to RNA isolation. Synthetic Caenorhabditis elegans miR 39 was incorporated into serum samples prior to RNA isolation and served as a norm alization means during the downstream quantitative analyses. Data of sera from subjects at day 7, day 17 and day 24 are shown. A B
121 Figure 5 2 Identification of serum miRN A s responsive to acute dietary zinc depletion using a qPCR based array Circulating miRNA were isolated from pooled sera collected on baseline (day 7) and post depletion phase (day 17), and were quantified by using a qPCR array focused on miRNA known to b e present in human serum. (A) A sc atter plot indicating the m iRNA of which levels were modulated by fold changes above 1.5 under dietary zinc restriction Red, up re gulated ; green, down regulated (B) Relative abundance of serum miRNA s affected by dieta ry zinc depletion. Values were normalized to cel miR 39 levels A B
122 Figure 5 3 Effects of dietary zinc intake levels on circulating serum miRNA levels. Serum miRNA respond ing to zinc restriction and repletion in opposite directi ons were selected as can didate biomarkers reflecting dietary zinc intake levels V alues were normalized to cel miR 39 levels and baseline levels were set at 1
123 Table 5 1 Zinc related genes targeted by the zinc responsive serum miRNA 1 Representative miRNA Target gene Conserved sites Poorly conserved sites Total Context Score Aggregate P CT 2 total 8mer 7mer m8 7mer 1A total 8mer 7mer m8 7mer 1A hsa miR 10b MTF1 1 0 0 1 1 0 1 0 0.19 0.41 hsa miR 92a MTF1 2 1 0 1 1 0 0 1 0.31 >0.99 hsa miR 145 MTF1 1 0 1 0 1 0 0 1 0.35 0.69 hsa miR 375 KLF4 1 0 0 1 0 0 0 0 0.13 0.38 hsa miR 145 KLF4 1 0 0 1 0 0 0 0 0.09 0.61 hsa miR 296 5p SLC30A3 1 0 1 0 1 0 1 0 0.27 N/A hsa miR 155 SLC39A10 1 1 0 0 0 0 0 0 0.25 <0.10 hsa miR 200b SLC39A14 1 0 1 0 0 0 0 0 0.29 0.70 1 Predicted by TargetScan release 5.1. 2 P CT the probability of conserved targeting
124 CHAPTER 6 ERYT HROCYTE MEMBRANE ZIN C TRANSPORTERS AND D EMATIN LEVELS IN HUMANS UNDER SHORT TERM DIETARY ZINC RE STRICTION Introductory Remarks The homeostatic regulation of zinc is crucial during the maturation of erythroid progenitor cells. More than 90% of erythrocyte zinc function s as a component essential for the activity of z inc metalloenzymes, such as carbonic anhydrase and Cu/Zn s u peroxide dismutase ( 15 3 ) Master transcription factors involve d in the erythropoietin (EPO) induced gene expression include zinc finger transcription factors GATA 1 and erythroid Krppel l ike factor (EKLF) ( 154 155 ) Zinc is also important for the maintenance of the membrane integrity of erythrocytes as indicated by higher osmotic fragility of red cells from zinc deficient ani mals ( 156 ) However, an excess in cellular zinc at the terminal phase of erythropoiesi s can be detrimental due to its interference with iron molecules during hemoglobin biosynthesis ( 93 94 ) D i fferential expression of m etallothionein (MT) and zinc transporter s ha ve been identified as regulatory mechanism s of cellular zinc homeostasis ( 8 9 ) Reduction in red cell MT protein levels by acute dietary zinc depletion has been previously shown ( 42 43 ) Recently, we have identified the presence of zinc transporters ZnT1, Zip8 and Zip10 in the plasma membrane fraction of mouse erythrocytes ( 14 ) Temporal expression patterns of these transporter genes indicated higher zinc import and export during the early and late stage of terminal erythroid differentiation, respectively The zinc transpor ter system was shown to be influenced b y the host s zinc status in mice After the mice were fed a zinc deficient diet for three weeks, an increase in Zip10 and a decrease in ZnT1 protein levels occurred in the erythrocyte membrane. Additionally h igher 65 Zn uptake was observed in erythrocyte s colle cted from zinc restricted animal s
125 T his study was conducted to determine i f the erythroid ZnT1, Zip8 and Zip10 expression is present in humans and to assess their potential of being a status assessment tool of human dietary zinc deficiency. Additi onally, levels of a protein recognized non specifically by the Zip8 antibody in the plasma membrane were identified as zinc responsive, indicating its potential as a zinc biomarker Results Erythrocyte Zinc Transporter Expression during Low Zinc Intake Wes tern analyses by using in house designed antibodies against human ZnT1, Zip8 and Zip10 successfully produced positive signals from erythrocyte samples (Figure 6 1) confirming the screening results from the previous animal experiments ( 14 ) Specificity of signals was determined by preabsorption controls, indicating the estimated molecular weight of human ZnT1, Zip8 and Zip10 in erythrocytes to be 130~150 kDa, 150 kDa and 40 kDa, respectively. The prominent band at 50 kDa produced by the Zip8 ant ibody did not disappear by peptide competition and thus was considered non specific A shift in the migration of ZnT1 by PNGase F treatment indicates the glycosylation of thi s transporter in human erythrocyte. In contrast to our findings from the mouse experiments, there were no significant changes in the zinc transporters by dietary zinc depletion (Figure 6 2) However, signal intensities from the non specific band produced by the Zip8 antibody were significantly higher in the membrane fraction of erythrocytes collected after the depletion phase. Identification of Dematin as a Zinc Responsive Erythrocyte Membrane Protein The zinc responsive protein non specifically detected by the human Zip8 antibody was further characterized by proteomic approaches including immunoprecipitation and mass spectrometry. After enriching t he protein of interest by immunoprecipitation with
126 the human Zip8 antibody, protein samples were subject to SDS PAGE and subsequent gel staining or western analysis Due to its molecular weight at approximately 50 kDa based on the known size of IgG heavy chain (55 kDa) it was of importance to confirm the separation between these two proteins which were highly abundant in the immunoprecipitated samples prior to band excision for protein identification (Figure 6 3 A ) Western analysis using the secondary antibody against rabbit IgG iden tified three bands originating from the Zip8 antibody used for immunoprecipita tion, of which two representing heavy chains (40 and 55 kDa) and another identifying the light chain (25 kDa) R eprobing the membrane with the Zip8 antibody and its respective secondary antibody enabled the discrimination of the signal produced by t he pro tein of interest (50 kDa) from that originating from the heavy chain of IgG. By matching with the western blot based on molecular weight markers the region corresponding to the size of the non specific band was excised from the stained gel for protein id entification by l iquid c hromatography m ass s pectrometry (Figure 6 3 B). The protein profile identified from the mass spectrum of the digested sample was composed of eleven proteins (Figure 6 4 A). Among these, the long isof orm of dematin was detect ed with the highest normalized spectrum counts The predi cted molecular weight of 46 kDa was closest to that of the non specific band size detected by western anal y sis using the Zip8 antibody (50 kDa). A spectr um representative of those exclusive ly corresponding to the amino acid sequence of dematin is presented in Figure 6 4 B. Approximately 20% of the complete amino acid sequence of dematin was covered by seven associated tryptic peptide s identified by the mass spectrum (Figure 6 4 C) This coverage rate was hi ghest among all proteins d etected suggesting the
127 nature of the unknown protein producing non specific band by Zip8 antibody to be dematin. T he protein identity determined by mass spectrometry and the presence of dematin in the membrane fraction of human e rythrocyte was validated by western analysis of samples immunoprecip itated with a n antibody designed to target dematin A s trong signal corresponding to a molecular weight of 50 kDa was produced by probing the immunoprecipitated samples by the human Zip8 antibody (Figure 6 5 A ). Western analyses o f samples from blood collected before and after dietary zinc depletion confirmed the zinc responsiveness of membrane dematin le vels in human erythrocytes (Figure 6 5 B). A significant increase with a fold change o f 2 was observed by low zinc ingestion The estimated molecular weigh t and the magnitude of response determine d by the dematin antibody were comparable to those detected by the non specific signals from the human Zip8 antibody No change in the Zip8 leve ls by the dietary treatment confirms the observation shown in Figure 6 2 Discussion The presence of ZnT1, Zip8 and Zip10 in the plasma membrane of mouse erythrocytes was previously identified by using a battery of in house made antibodies targeting zinc t ransporter s ( 14 ) Differential expression of these transporter genes during terminal erythroid differentiation was shown by using a primary cell model inducible by E PO treatment in vitro ( 14 ) Up regulation of the importers, Zip8 and Zip10, preced ed the induction of ZnT1 by EPO These findings implicate the involvement of zinc transporter activity in the development of optimal zinc balance required for the transactivation of genes by zinc finger transcription factors ( 154 155 ) and hemoglobin synthesis during the differentiation process ( 93 94 ) The results from the current study
128 confirm the expression of these transporter proteins in the plasma membrane of human red cells. The enhancement in erythrocyte z inc uptake by low dietary levels of zinc has been identified in animal models ( 14 157 159 ) However, the underlying m olecular mechanism remains unclear The down regulation of ZnT1 and the up r egulation of Zip10 observed in erythrocytes of zinc deficient mice ( 14 ) suggest t hese transporters as factors exerting this effect of zinc deficiency. It is of note that MTF 1 has been shown to mediate the zinc responsive transcription of both ZnT1 and Zip10 however, towards the opposite mode of response ( 4 6 ) Upon zinc sen sing, nuclear translocation of MTF 1 and its binding to the MRE motif occurs for the activation and repression of ZnT1 and Zip10, respectively The recently discover ed mechanism for Zip10 repression by MTF 1 involves its binding to an MRE down stream from the transcription start site resulting in the inhibition of the movement of RNA polymerase I I ( 6 ) Based on these, we postulate d that the differential express ion of ZnT1 and Zip10 obse rved in mouse erythrocytes is meditated by the zinc dependent activation of MTF 1 during their gene expression at the terminal stage of erythropoiesis Even though the observations from the m ouse model agree with the mode of MTF 1 meditated regulation of ZnT1 and Zip10 by zinc, there were no changes in the zinc transporter proteins of human erythrocytes by acute dietary zinc depl e tion. During the differentiation of erythroid progenitors to reticulocytes, enucleation occurs and th us the capability of these cells to carry out gene regulation is lost. Thus, the erythrocyte proteome is dependent on that formed during the preceding differe ntiation stage. T he estimated life span of human ery throcytes and mouse red cells are 120 d and 40 d
129 respectively ( 160 ) The dietary zinc regimen of the current human study for zinc depletion was limited to 10 d, which covers less than 10% of the life span of circulating erythrocytes. Thus, a substantial portion of circulating eryth rocytes collected after the zinc depletion phase would be those formed u nder adequate z inc conditions. The length of the zinc depletion period in mice approximated 50% of the life span of their red cells ( 14 ) consequently a higher portion of erythrocy tes in the bloodstream are produced during zinc deprivation Additionally, as indicated by changes in plasma/serum zinc levels observed in the mouse and human subjects (approximately 60% vs. 20% reduction), the severity of zinc deficiency induced by the 2 1 d dietary zinc deprivation in mice was greater than that produced by the 10 d dietary regimen applied to the human subjects. This indicates the murine model of zinc deficiency is close to moderate or severe deficiency while the h uman model used represe nts condition s of short term modest dietary zinc deprivation. Consequently, the potential of erythrocyte ZnT1 and Zip10 levels for the diagnosis of chronic dietary zinc deficiency in human needs to be further investigated. Among the zinc transporters o f which proteins were detected in the plasma membrane of red cells, only ZnT1 transcripts were identified to reflect the host s zinc intake levels in reticulocytes. Reticulocyte transcript levels reflect a gene expression profile at the very late stage of erythropoiesis. Induction of Zip10 expression by EPO was only observed at the early stage of terminal erythroid differentiation in mice ( 14 ) In cont rast, repressi on of ZnT1 tran scripts by zinc chelation was present at a later stage of the 24 h differentiation period ( 14 ) The s e results indicate the effect of zinc on Zip10 tra n scription in human erythroid progenitor cells may be limited to the early phase of
130 EPO induced terminal differentia tion as reflected by the absence of differences between the baseline and post depletion levels of Zip10 mRNA in reticulocytes D ematin a pr otein initi ally identified as protein 4.9 in the membrane of h uman red cells ( 161 ) functions as an actin bun d l ing protein located at the juncti onal complex i.e., spectrin actin junction, of the erythrocyte membrane skeleton ( 162 ) Along with other core constituent s of the membrane skeleton such as spect rin, actin adducin and protein 4.1 ( 163 164 ) dematin has been shown to be essential for the maintenance of the cellular morphology motility and membrane structural integrity ( 165 166 ) In the current study, the level of erythrocyte dematin in the plasma membrane was shown to be highly sensitive to the host s zinc status. Its rapid response implicates the presence of a posttranslational regulatory mechanism m ediating the effects of zinc on this protein. Two pro tein kinases, cyclic adenosine monophosphate dependent protein kinase (PKA) and protein kinase C (PKC), are involved in the regulation of the actin cytoskeleton, and both have been shown to phosphorylate dematin in vitro ( 167 ) In particular phosphorylation by PKA of which activity in red cells is evident ( 168 ) exerts an inhibitory effect on the actin bundling activity of dematin by causing a conformational change in the headpiece domain ( 167 169 171 ) Zinc has been shown to predominantly inhibit the hyd rolysis of cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP) by phosphodiesterase (PDE) acitivity in vitro ( 172 ) Additionally this inhibitory effect on PDE activity has been recently suggested as an indirect mechanism for zinc to enhance PKA activity in immune cells ( 129 ) Conversely, suboptimal zinc conditions may result in lower PKA activity, and thus lead to less phosphorylation of dematin. The inhibitory effect of phosphorylation on the a ctin
131 binding of dematin can be reversed by p h osphatase treatment ( 167 ) The capability of zinc to inhibit phosphatase activities has been identified in various signaling pathways ( 51 125 127 ) Taken these together with the observations of the present study, dietary zinc depletion may increase the translocation of cytosolic dematin to the plasma membrane by disrup ti ng the balance between PKA and phosphatase activitie s P hosphorylation of the component proteins associated with the junctional complex by modulated PKC activity can also lead to the formation of an instable membrane skeleton ( 173 ) However, t he phosphorylation of dematin by PKC has been shown to have no effects on its actin bundling activity ( 167 ) I t is of note that cytosolic translocation of membrane a dducin, whi ch is another actin binding protein a nchoring the junctional complex to the plasma membrane by PKC activity resulted in the redistribution of another membrane PKC independent skeletal protein, spectrin to the cytoplasm ( 174 ) Thus, impaired PKC activity can also lead to the internalization of membrane dematin in an indirect manner In contrast to PKA, PKC can be dire ctly regulated by zinc ( 128 ) Upon binding to its metal binding site, zinc increase s the activity of PKC and also induces the t ranslocation of cytosolic PKC to the plasma membrane This mode of regul ation agrees with the present observation of decreased membrane levels of a cytoskeletal protein under a zinc depleted condition and thus suggests PKC as a nother possible mediator of the effects of dietary zinc depletion on membrane dematin levels Impair ed cellular membrane stability and increased osmotic fragility have been identified in red cells of zinc deficient animals ( 156 ) Whether excess of dematin in the membrane complex produces beneficial or detrimental effects to the structural integrity
132 of normal erythrocytes is unclear. Overexpression of dematin has been shown to alter the c ellular phenotype of prostate cancer cells towards that of normal pro state epithelia l cells ( 175 ) It is of note that, respective overexpressio n of the core a nd headpiece domain of dematin also caused phenotypic changes such as cytoplasmic shrinkage xtensions, and inlamellopodial extensions of the cancer cells Thus the increase in membrane dematin levels may contribute to the effect of zinc deficiency on red cell integrity, particularly, by causing a dysfunctional assembly of t he actin based erythrocyte cytoskelet on
133 Figure 6 1 Zinc Transporter Expression in Human Erythrocyte Membrane. The presence of (A) ZnT1, (B) ZIP8 and (C) ZIP10 in erythrocyte ghosts was detected by western analy sis. Specificity of signal s w as determined by preincubating each primary antibody with respective antigenic peptides. (D) Glycosylation of ZnT1 was identified by PNGase F treatment. A B C D
134 Figure 6 2 Effects of acute dietary zinc depletion on zinc trans porter expression in human erythrocytes. (A) The expression of each transporter was identified by western analysis with respective antibodies. N/S indicates the non specific band that is zinc responsive and is detected with the Zip8 antibody. (B) Signal intensities quantified by densitometr ic analysis Data are expressed as mea n SD and s tatistically significant differenc es are P < 0.0 5 (n = 3 subjects in each pooled sample) A B
135 Figure 6 3 Isolation of the peptide producing a non specific band with the Z ip 8 antibody by immunoprecipitation (A) S taining and labeling of immunoprecipitated erythrocyte membrane proteins by using Coomassie Blue and respective antibodies HC and LC indicate heavy chain and light chain, respectively. Position of th e non specific band is annotated by a red asterisk. (B) Excision of the region migrating at the size of the non specific protein for mass spectrometry analysis A B
136 Figure 6 4 Identification of the protein non sp ecifically dete cted by the hZ ip 8 antibody. Protein samples were trypsinized and were subject to l iquid chromatography coupled with tandem mass spectrometry (LC MS /MS ) (A) Dematin, with an expected molecular weight (MW) of 46 kDa, showing the highest normalized spectru m count among the twelve identified proteins. ( B ) Representative mass spectrum of dematin. ( C ) Identified peptide sequences unique for dematin by LC MS /MS A B C
137 Figure 6 5 Confirmation of protein identity determined by mass spectrometry analysis (A) Western analysis with a rabbit anti Zip8 polyclonal antibody of erythrocyte membrane proteins immunoprecipitated by a mouse polyclonal anti dematin antibody (B) Responsiveness of erythrocyte membrane dematin to acute dietary zinc depletio n Signal intensities quantified by densitometr ic analysis Data are expressed as mea n SD and s tatistically significant differenc es are P < 0.0 5 (n = 3 subjects in each pooled sample) A B
138 CHAPTER 7 CONCLUSIONS AND FUTURE DIRECTION S The presence o f a reliable biomarker is a prerequisite for employing a preventive or pr ognositic policy on a population suffering from clinical outcomes associated wi th malnutrition Identification of a specific and sensitive assessment tool for individuals with zinc d eficiency has be en a long term goal in zinc research The current study was conducted primarily to identify biomolecules holding the potential to reflect the host s zinc status in blood and buccal specimens. Given the function of metallothionein and zinc transporters mediating the regulation of cellular zinc homeostasis, the effects of acute dietary zinc depletion on the transcript abundance of these zinc homeostatic genes in blood cells and buccal epithelial mucosa were determined. MT, ZnT1, ZnT 4, ZnT 5 and Z ip3 responded to a short term zinc depletion regimen in a cell type specific manner. Based on the significance in change s by zinc depletion and their correlation with serum zinc concentrations, PBMC and whole blood Z nT1 levels were identified as can didate molecule s holding potentials of being an assessment tool for dietary zinc status The well known regulatory mechanism of ZnT1 by the zinc sensing transcription factor MTF 1 ( 4 ) further validates the findings of the present study As a standard approach for identifying gene transcripts responding to modulated metabolic or clinical conditions, microarray analysis with stabilized whole blood was conducted. It is of interest that the gene products responsive to acute dietary zinc depletion were mostly associated to cell cycle regulation and immune response. VEGF, a therapeutic target molecule for cancer treatment, and NF B, a transcription factor mediating gen e expression related to cell proliferation and immune response, were suggested as putative regulatory factors mediating the effects of low zinc ingestion on
139 the whole blood transcriptome profile. These correspond to previous findings describing the signif icance of zinc in the maintenance of host defense ( 95 97 ) and its protective role against cancer development ( 98 99 ) Among those differentially expressed during zinc depletion, eight well characterized gene transcripts IGJ, CDC20, MZB1, IGLL 1, TXNDC5, CD38, GLDC and TNFRSF17 were determined as genes highly responsive to dietary zinc restriction The molecular mechanism exerting the modulation of these genes by zinc restriction, and the specificity of their responsiveness to zinc should be determin ed by future research The functional significance of zinc in pathways mediating immune response was confirmed by the depression of LPS and PHA induced TNF release from whole blood by acute dietary zinc restriction. The mode of TNF response by the activated whole blood samples t o suboptimal zinc ingestion corresponds to 1) the implications of impaired NF B activity by the down regulated genes in the microarray dataset of the current study and 2) the enhancement of TNF gene expression observed in a previous dietary study focused on the effects of zinc supplementation ( 41 ) It is of note that a standardized commercial ELISA kit w as used for the absolute quantit ation of cytokines in order to enable comparisons among data produced by the current study and those from future studies requiring dietary zinc assessment Erythrocyte ZnT1 and Zip10 respond to zinc depletion in mice however, in an opposite manner ( 14 ) In vitro experiments with mouse ery t h roid progenitor cells indicate the regulation of these genes to be present during terminal erythroid differentiation. The mode of responses of erythroid ZnT1 and Zip10 to dietary zinc depletion agree with the previous findings of MTF 1 meditated activation of ZnT1 and
140 repression of Zip10 gene expression under high zinc conditions ( 4 6 ) In contrast to the observations in the animal study, no changes in erythrocyte ZnT1 and Zip10 protein levels were identified in humans under dietary zinc restriction. Differences in the length of experimental zinc depletion, life s pan of circulation red cells, and the magnitude in reduction of serum zinc concentrations between the human and mouse model indicate the former to represent conditions of acute short term zinc depletion while the latter mimics exten ded dietary zinc defici ency resulting in higher severity T herefore detection of the zinc responsiveness of erythrocyte ZnT1 and Zip10 proteins may be limited to conditions of chronic and severe zinc deficiency. Future human studies with experimental models o f long term dieta ry zinc restriction or supplementations are required to further confirm this hypothesis The erythrocyte membrane level of dematin, a protein functioning as a core component of the cytoskeletal complex w as identified to be responsive to short term dietary zinc depletion. As described above there were no changes in the levels of ZnT1 and Zip10 proteins in erythrocytes. Both are known to be regulated by zinc at the transcript level ( 4 6 ) Taking this into account the presence of a posttranslational regulatory mechanism exerting the effect of zinc on dematin can be speculated. I nvolvement of p rotein kinases and phosphatases in the regulation of the actin bundling activity of cytoskeletal proteins including dematin, is of particular relevance ( 167 173 ) It i s of note that zinc has been shown to enhance PKA and PKC activity ( 128 129 ) and also inhibit various phosphatases ( 125 127 ) Collectively the observations of higher membrane dematin in red cells under the zinc restricted condition may be due to a modulatio n in the phosphorylation status of dematin or other proteins involved in the
141 assembly of the cykoskelatal complex on the intracellular side of the plasma membrane. Future in vitro mechanistic studies focusing on the effects of zinc on 1) the activities of erythrocyte PKA, PKC, and phosphatases, 2) the phosphorylation degree of cytoskeletal proteins and 3) the cellular distribu tion of releva nt proteins will allow the identification of a novel role of zinc contributing to the morphology, mo tility and integr i ty of circulating erythrocytes ( 156 ) Serum miRNA profiles of cancer patients have b een extensively characterized in cancer biomarker studies, particularly, due to its highly stable nature ex vivo ( 135 137 ) Modulation in the profile of serum miR NA s was identified as a consequence of dietary zinc depletion by the current biomarker study Among the responsive miRNAs, miR 204 and miR 296 5p showed the highest responsiveness to zinc restriction. Reversal effects of supplemental zinc on these miRNA levels further potentiates their property as indicators of the host s dietary zinc status. It is of note that both miR 204 and miR 296 5p have been shown to target genes involved in tumor development and progression ( 151 152 ) The differential response s of circulating miRNAs to dietary zinc levels imply their possible role in the post transcriptional regulation of protein coding gene products related to zinc metabolism as well. Even though several hypothesis driven in vitro and in vivo studies have suggested the potential of zinc homeostatic gene products and metalloenzyme activities as zinc biomarkers their lim ited practicality for larger sample sizes hinders the implementation of relevant population based validation studies Based on an in depth analysis of the transcriptome of blood cells and serum by using novel molecular techniques enabling sample stabiliza tion and high throughput analysis, a plethora of biomolecules were
142 identified as potential diagnostic indices of suboptimal zinc consumption. It is of note that the following aspects were taken into consideration during the experimental design to allow th e application of the current findings to future verification studies and, eventually, to the field of nutrient assessment : 1) n oninvasiveness of the sampl ing process, 2) reasonable cost of the implemented technique for quantitative measures (i.e., cost eff ectiveness), 3) minimal requirement of skills and laboratory devices for sample processing and analysis, 4) stability of samples or target molecules, particularly, during transportation after the collection or processing, and 5) availability of a s tandardi zed protocol covering the processes from sample collection to analytical assessments. In conclusion the present study on the effects of acute dietary zinc depletion pr oduced a plethora of indicative molecular indices implying the physiological effects of dietary zinc in the human body Transcripts of blood genes associated with cell cycle regulation, host defense and the regulation of zinc homeostasis hold the potential to indicate the dietary zinc status of individuals. TNF release from whole blood under immunostimulation was clearly depressed by suboptimal zinc intake. These strongly agree with the biological roles of zinc in immunity and cell proliferation characterized by previous in vitro and in vivo experiments. T he clinical implications and relevant molecular mechanisms of the modulation in erythrocyte membrane dematin and se rum miRNAs by dietary zinc restriction remain unclear, and thus require further exploration B y using novel techniques enabling sample stabiliz ation and high throughput data analyses along with cost effectiveness, this study may also provide insight s into the
143 standardization of experimental designs for future nutritional or clinical biomarker studies.
144 APPENDIX A SCREENING QUESTIONNA IRE
145 APPENDI X B P OST PARTICIPATION SURVEY
146 APPENDIX C LIST OF GENES DIFFER ENTIALLY EXPRESSED B Y ACUTE DIETARY ZINC DEPLETION Table C 1 List of genes up regulated by acute dietary zinc depltion ranked by fold changes (ZnD/Baseline values) 1 Gene Symbol Entrez ID Gen e Name Fold Change Parametric P Value FDR LOC651751 651751 similar to Ig kappa chain V II region RPMI 6410 precursor 4.32 < 1e 07 < 1e 07 IGJ 3512 immunoglobulin J polypeptide, linker protein for immunoglobulin alpha and mu polypeptides 4.22 1.00E 06 0 .00189 LOC649923 649923 similar to Ig gamma 2 chain C region 3.51 3.00E 07 0.00124 LOC642113 642113 Ig kappa chain V III region HAH like 3.05 < 1e 07 < 1e 07 LOC652493 652493 ig kappa chain V I region HK102 like 3.05 6.00E 07 0.00178 LOC647450 647450 similar to Ig kappa chain V I region HK101 precursor 2.87 3.00E 07 0.00124 LOC647506 647506 similar to Ig kappa chain V I region HK101 precursor 2.77 9.00E 07 0.00189 IGLL1 3543 immunoglobulin lambda like polypeptide 1 2.77 1.00E 06 0.00189 TXNDC5 8156 7 thioredoxin domain containing 5 (endoplasmic reticulum) 2.73 5.00E 07 0.00173 CDC20 991 cell division cycle 20 homolog (S. cerevisiae) 2.69 2.00E 07 0.00124 TNFRSF17 608 tumor necrosis factor receptor superfamily, member 17 2.55 0.0000185 0.0174 GLDC 2731 glycine dehydrogenase (decarboxylating) 2.49 2.30E 06 0.00367 LOC652102 652102 similar to Ig heavy chain V I region HG3 precursor 2.46 1.07E 03 0.208 MZB1 51237 marginal zone B and B1 cell specific protein 2.07 0.0000007 0.00182 CD38 952 CD38 molec ule 2.03 2.07E 05 0.0178 ABCB9 23457 ATP binding cassette, sub family B (MDR/TAP), member 9 1.85 2.20E 06 0.00367 IGLL3 91353 immunoglobulin lambda like polypeptide 3, pseudogene 1.85 0.0001122 0.0541 LOC652775 652775 similar to Ig kappa chain V V regio n L7 precursor 1.81 0.0014877 0.232 LOC649210 649210 similar to Ig lambda chain V region 4A precursor 1.73 3.13E 03 0.273 CCNB2 9133 cyclin B2 1.70 0.0000182 0.0174 ITM2C 81618 integral membrane protein 2C 1.70 2.23E 05 0.0178 ABCA1 19 ATP binding cass ette, sub family A (ABC1), member 1 1.69 0.0000061 0.00791 GPRC5D 55507 G protein coupled receptor, family C, group 5, member D 1.67 0.000022 0.0178 UBE2C 11065 ubiquitin conjugating enzyme E2C 1.59 0.0000044 0.00652 NLRP7 199713 NLR family, pyrin domai n containing 7 1.51 8.19E 05 0.0447 LOC728741 728741 hypothetical LOC728741 1.50 0.0000118 0.0129 APOBEC3B 9582 apolipoprotein B mRNA editing enzyme, catalytic polypeptide like 3B 1.49 0.0007272 0.168 LOC652694 652694 similar to Ig kappa chain V I regi on HK102 precursor 1.46 0.0025921 0.265 AURKB 9212 aurora kinase B 1.45 0.0000497 0.0316 TYMS 7298 thymidylate synthetase 1.44 1.90E 04 0.0787 CDCA5 113130 cell division cycle associated 5 1.43 0.0000691 0.0387
147 Table C 1 Continued Gene Symbol Entre z ID Gene Name Fold Change Parametric P Value FDR LOC651612 651612 hypothetical protein LOC651612 1.43 0.0025996 0.265 POU2AF1 5450 POU class 2 associating factor 1 1.42 0.0000075 0.00915 SORL1 6653 sortilin related receptor, L(DLR class) A repeats cont aining 1.42 1.65E 03 0.245 HSP90B1 7184 heat shock protein 90kDa beta (Grp94), member 1 1.42 0.0017993 0.248 HS.520591 1.42 1.87E 03 0.248 SEMA4A 64218 sema domain, immunoglobulin domain (Ig), transmembrane domain (TM) and short cytoplasmic domain, (semaphorin) 4A 1.42 0.0034833 0.278 TNFRSF13B 23495 tumor necrosis factor receptor superfamily, member 13B 1.41 0.0002692 0.102 TK1 7083 thymidine kinase 1, soluble 1.40 0.0003142 0.107 CDC45L 8318 cell division cycle 45 homolog (S. cerevisiae) 1.40 3 .64E 04 0.118 FKBP11 51303 FK506 binding protein 11, 19 kDa 1.40 0.0003749 0.12 RRBP1 6238 ribosome binding protein 1 homolog 180kDa (dog) 1.37 0.0000325 0.0241 TOP2A 7153 topoisomerase (DNA) II alpha 170kDa 1.36 0.0000993 0.0503 NT5DC2 64943 5' nucle otidase domain containing 2 1.36 0.0002826 0.103 NCAPG 64151 non SMC condensin I complex, subunit G 1.36 0.0017575 0.247 BIK 638 BCL2 interacting killer (apoptosis inducing) 1.35 0.0006896 0.163 SDF2L1 23753 stromal cell derived factor 2 like 1 1.35 0.0 009311 0.191 F5 2153 coagulation factor V (proaccelerin, labile factor) 1.35 0.0044828 0.305 UHRF1 29128 ubiquitin like with PHD and ring finger domains 1 1.34 3.62E 05 0.0259 C19ORF10 56005 chromosome 19 open reading frame 10 1.34 0.000042 0.0281 LOC1 00131727 10013172 7 hypothetical protein LOC100131727 1.34 0.0002553 0.0984 SLC35A5 55032 solute carrier family 35, member A5 1.33 1.09E 04 0.0538 CAMK1G 57172 calcium/calmodulin dependent protein kinase IG 1.33 0.0001167 0.055 IRF4 3662 interferon regul atory factor 4 1.32 0.0000085 0.0098 DENND5B 160518 DENN/MADD domain containing 5B 1.32 0.0000583 0.0336 CCNA2 890 cyclin A2 1.32 0.0009848 0.195 ASAP1IT1 29065 ASAP1 intronic transcript 1 (non protein coding) 1.32 0.0041321 0.295 LOC100129905 10012990 5 ribosomal protein S27 pseudogene 19 1.31 0.0043965 0.305 PTTG3P 26255 pituitary tumor transforming 3, pseudogene 1.30 0.000091 0.0472 SLC38A10 124565 solute carrier family 38, member 10 1.29 0.0000394 0.0272 GMPPB 29925 GDP mannose pyrophosphorylase B 1.29 0.0000578 0.0336 BUB1 699 budding uninhibited by benzimidazoles 1 homolog (yeast) 1.29 0.0028315 0.269 OXCT2 64064 3 oxoacid CoA transferase 2 1.28 0.0000521 0.0318 MCM4 4173 minichromosome maintenance complex component 4 1.28 0.000169 0.0746 P HGDH 26227 phosphoglycerate dehydrogenase 1.28 0.0002838 0.103
148 Table C 1 Continued Gene Symbol Entrez ID Gene Name Fold Change Parametric P Value FDR BIRC5 332 baculoviral IAP repeat containing 5 1.28 0.0004178 0.129 TLN1 7094 talin 1 1.28 0.0013616 0.231 HS.184721 1.28 0.0024909 0.265 LOC441124 441124 hypothetical gene supported by AK093729; BX647918 1.28 0.0045759 0.305 CKAP4 10970 cytoskeleton associated protein 4 1.28 0.0047783 0.309 CCR9 10803 chemokine (C C motif) receptor 9 1.27 0.0000 053 0.00733 PDIA5 10954 protein disulfide isomerase family A, member 5 1.27 0.0002362 0.0942 COL18A1 80781 collagen, type XVIII, alpha 1 1.27 0.0024194 0.265 ABCG1 9619 ATP binding cassette, sub family G (WHITE), member 1 1.26 0.0000169 0.0174 SLC2A5 6 518 solute carrier family 2 (facilitated glucose/fructose transporter), member 5 1.26 0.0003109 0.107 MTF1 4520 metal regulatory transcription factor 1 1.26 0.0007353 0.168 CDT1 81620 chromatin licensing and DNA replication factor 1 1.26 0.0020703 0.258 STT3A 3703 STT3, subunit of the oligosaccharyltransferase complex, homolog A (S. cerevisiae) 1.25 0.0001593 0.0726 TRAM2 9697 translocation associated membrane protein 2 1.25 0.0004935 0.142 AURKA 6790 aurora kinase A 1.25 0.0006783 0.163 HS.157344 1.25 0.0009777 0.195 POLR2A 5430 polymerase (RNA) II (DNA directed) polypeptide A, 220kDa 1.25 0.0020028 0.258 LOC642131 642131 putative V set and immunoglobulin domain containing protein 6 like 1.25 0.0030141 0.27 SPC24 147841 SPC24, NDC80 kinetochore complex component, homolog (S. cerevisiae) 1.25 0.0044094 0.305 DLGAP5 9787 discs, large (Drosophila) homolog associated protein 5 1.24 0.0015724 0.24 NUSAP1 51203 nucleolar and spindle associated protein 1 1.24 0.0029244 0.27 UBE2J1 51465 ubiquitin co njugating enzyme E2, J1 (UBC6 homolog, yeast) 1.23 0.0000268 0.0206 HS.352677 1.23 0.0001848 0.0782 TRIP13 9319 thyroid hormone receptor interactor 13 1.23 0.0006141 0.159 WAS 7454 Wiskott Aldrich syndrome (eczema thrombocytopenia) 1.23 0.0011971 0 .224 ARID3A 1820 AT rich interactive domain 3A (BRIGHT like) 1.23 0.0018194 0.248 CENPM 79019 centromere protein M 1.23 0.0024777 0.265 UBA1 7317 ubiquitin like modifier activating enzyme 1 1.23 0.0032981 0.276 BMP8B 656 bone morphogenetic protein 8b 1 .22 0.0006654 0.163 SPATS2 65244 spermatogenesis associated, serine rich 2 1.22 0.0008565 0.187 SLC12A9 56996 solute carrier family 12 (potassium/chloride transporters), member 9 1.22 0.00092 0.191 SLC1A4 6509 solute carrier family 1 (glutamate/neutral amino acid transporter), member 4 1.22 0.0012287 0.228
149 Table C 1 Continued Gene Symbol Entrez ID Gene Name Fold Change Parametric P Value FDR TXNDC11 51061 thioredoxin domain containing 11 1.22 0.002231 0.265 LOC440348 440348 nuclear pore complex in teracting protein like 2 1.22 0.0023746 0.265 HJURP 55355 Holliday junction recognition protein 1.22 0.0032498 0.276 ZNF341 84905 zinc finger protein 341 1.22 0.0034433 0.278 ZYX 7791 zyxin 1.22 0.0043195 0.302 LOC100128269 100128269 hypothetical LOC1 00128269 1.22 0.0049398 0.314 ZBTB43 23099 zinc finger and BTB domain containing 43 1.21 0.0004846 0.142 KIF1B 23095 kinesin family member 1B 1.21 0.0007647 0.171 SNORA11B 100124539 small nucleolar RNA, H/ACA box 11B (retrotransposed) 1.21 0.0009366 0.191 KIF2C 11004 kinesin family member 2C 1.21 0.0009411 0.191 MAGT1 84061 magnesium transporter 1 1.21 0.0011432 0.218 PRR11 55771 proline rich 11 1.21 0.0013451 0.231 MCM2 4171 minichromosome maintenance complex component 2 1.21 0.0017097 0.247 AL OX5 240 arachidonate 5 lipoxygenase 1.21 0.002618 0.265 TCF4 6925 transcription factor 4 1.21 0.0029275 0.27 NARF 26502 nuclear prelamin A recognition factor 1.21 0.0036279 0.285 SORT1 6272 sortilin 1 1.21 0.0045313 0.305 CDCA8 55143 cell division cy cle associated 8 1.20 0.001591 0.241 LAMA5 3911 laminin, alpha 5 1.20 0.0023538 0.265 HYOU1 10525 hypoxia up regulated 1 1.20 0.0026657 0.266 HS.143018 1.20 0.0035111 0.278 FAHD2B 151313 fumarylacetoacetate hydrolase domain containing 2B 1.20 0.004 7638 0.309 ZBTB32 27033 zinc finger and BTB domain containing 32 1.19 0.0003531 0.116 MGAT3 4248 mannosyl (beta 1,4 ) glycoprotein beta 1,4 N acetylglucosaminyltransferase 1.19 0.000672 0.163 ADAM19 8728 ADAM metallopeptidase domain 19 1.19 0.0008079 0.178 DCTN1 1639 dynactin 1 1.19 0.0014801 0.232 SEMA3E 9723 sema domain, immunoglobulin domain (Ig), short basic domain, secreted, (semaphorin) 3E 1.19 0.0015156 0.235 GMPPA 29926 GDP mannose pyrophosphorylase A 1.19 0.0016243 0.244 LOC651816 651816 s imilar to Ubiquitin conjugating enzyme E2S (Ubiquitin conjugating enzyme E2 24 kDa) (Ubiquitin protein ligase) (Ubiquitin carrier protein) (E2 EPF5) 1.19 0.0023823 0.265 SEC61A1 29927 Sec61 alpha 1 subunit (S. cerevisiae) 1.19 0.002402 0.265 EHBP1L1 2541 02 EH domain binding protein 1 like 1 1.19 0.0024833 0.265
150 Table C 1 Continued Gene Symbol Entrez ID Gene Name Fold Change Parametric P Value FDR CIC 23152 capicua homolog (Drosophila) 1.19 0.0028768 0.27 FLOT2 2319 flotillin 2 1.19 0.0032535 0.276 RARA 5914 retinoic acid receptor, alpha 1.19 0.0033469 0.276 DUSP5 1847 dual specificity phosphatase 5 1.19 0.0043166 0.302 OR2AG1 144125 olfactory receptor, family 2, subfamily AG, member 1 1.18 0.0004701 0.141 LOC399900 399900 hypothetical LOC399900 1.18 0.000509 0.144 SPCS2 9789 signal peptidase complex subunit 2 homolog (S. cerevisiae) 1.18 0.0005436 0.146 ZBP1 81030 Z DNA binding protein 1 1.18 0.0009123 0.191 LOC399491 399491 GPS, PLAT and transmembrane domain containing protein 1.18 0.001041 1 0.204 KIF20A 10112 kinesin family member 20A 1.18 0.0026819 0.266 COPG 22820 coatomer protein complex, subunit gamma 1.18 0.0027871 0.269 OTUD5 55593 OTU domain containing 5 1.18 0.0043392 0.302 MYH9 4627 myosin, heavy chain 9, non muscle 1.18 0.004 8114 0.309 SEC13 6396 SEC13 homolog (S. cerevisiae) 1.17 0.0014051 0.231 VPS8 23355 vacuolar protein sorting 8 homolog (S. cerevisiae) 1.17 0.0014412 0.232 PARM1 25849 prostate androgen regulated mucin like protein 1 1.17 0.0014727 0.232 SRPR 6734 sign al recognition particle receptor (docking protein) 1.17 0.0016889 0.247 PIM2 11040 pim 2 oncogene 1.17 0.0020115 0.258 TPX2 22974 TPX2, microtubule associated, homolog (Xenopus laevis) 1.17 0.0021016 0.258 LOC100130168 100130168 hypothetical protein LO C100130168 1.17 0.0023524 0.265 HS.562118 1.17 0.0025938 0.265 NLRP1 22861 NLR family, pyrin domain containing 1 1.17 0.0028257 0.269 HS.569823 1.17 0.0028932 0.27 KIAA0226 9711 KIAA0226 1.17 0.0029328 0.27 TAP1 6890 transporter 1, ATP bindi ng cassette, sub family B (MDR/TAP) 1.17 0.0029947 0.27 MLEC 9761 malectin 1.17 0.0032293 0.276 CXXC1 30827 CXXC finger protein 1 1.17 0.0033412 0.276 AARS 16 alanyl tRNA synthetase 1.17 0.0036372 0.285 C1QB 713 complement component 1, q subcomponent, B chain 1.17 0.0040823 0.295 KIAA0492 57238 KIAA0492 protein 1.17 0.0041365 0.295 HS.447737 1.17 0.0045788 0.305 HS.559654 1.16 0.0014509 0.232 LOC728790 728790 hypothetical LOC728790 1.16 0.0021339 0.26 ALDH6A1 4329 aldehyde dehydrogenase 6 family, member A1 1.16 0.0026201 0.265 IL4R 3566 interleukin 4 receptor 1.16 0.0029048 0.27 PDIA4 9601 protein disulfide isomerase family A, member 4 1.16 0.0041813 0.296
151 Table C 1 Continued Gene Symbol Entrez ID Gene Name Fold Change Parametric P Value FDR ALDH3B1 221 aldehyde dehydrogenase 3 family, member B1 1.16 0.0046729 0.307 ZNF486 90649 zinc finger protein 486 1.16 0.0046782 0.307 IRF9 10379 interferon regulatory factor 9 1.16 0.0047607 0.309 TMEM106A 113277 transmembrane protein 10 6A 1.16 0.0047759 0.309 INTS3 65123 integrator complex subunit 3 1.15 0.0011132 0.214 SIL1 64374 SIL1 homolog, endoplasmic reticulum chaperone (S. cerevisiae) 1.15 0.0020861 0.258 FKBP2 2286 FK506 binding protein 2, 13kDa 1.15 0.0023331 0.265 BTBD12 84464 SLX4 structure specific endonuclease subunit homolog (S. cerevisiae) 1.15 0.0023915 0.265 NFKB2 4791 nuclear factor of kappa light polypeptide gene enhancer in B cells 2 (p49/p100) 1.15 0.0033623 0.276 SLC35B1 10237 solute carrier family 35, member B1 1.15 0.004008 0.295 C4ORF28 133015 PARK2 co regulated like 1.15 0.0045292 0.305 P4HB 5034 prolyl 4 hydroxylase, beta polypeptide 1.15 0.0047315 0.309 PTK2B 2185 PTK2B protein tyrosine kinase 2 beta 1.15 0.0049749 0.315 FBXO18 84893 F box protein helicase, 18 1.14 0.0005795 0.154 LILRB1 10859 leukocyte immunoglobulin like receptor, subfamily B (with TM and ITIM domains), member 1 1.14 0.0020484 0.258 PCYOX1 51449 prenylcysteine oxidase 1 1.14 0.0025229 0.265 CLPTM1L 81037 CLPTM1 like 1.14 0.0 030232 0.27 FAM46A 55603 family with sequence similarity 46, member A 1.14 0.0034408 0.278 CAPN11 11131 calpain 11 1.14 0.0040412 0.295 EIF4G3 8672 eukaryotic translation initiation factor 4 gamma, 3 1.14 0.0045359 0.305 HS.188979 1.13 0.0013004 0.231 TRPM3 80036 transient receptor potential cation channel, subfamily M, member 3 1.13 0.0018613 0.248 FLJ44124 641737 hypothetical LOC641737 1.13 0.0022495 0.265 FLJ20254 54867 transmembrane protein 214 1.13 0.0039144 0.295 IFNAR2 3455 interferon (alpha, beta and omega) receptor 2 1.12 0.0021891 0.265 RCC2 55920 regulator of chromosome condensation 2 1.12 0.0038441 0.295 SEC23B 10483 Sec23 homolog B (S. cerevisiae) 1.11 0.0041106 0.295 SHMT1 6470 serine hydroxymethyltransferase 1 (soluble) 1.10 0.0036726 0.286 1 Determined by paired t test at P < 0.005 and a permutation of 10,000 (n = 9). FDR f alse discovery rate.
152 Table C 2 List of genes down regulated by acute dietary zinc depltion ranked by fold changes (ZnD/Baseline values) 1 Gene Symbol E ntrez ID Gene Name Fold Change Parametric P Value FDR HOPX 84525 HOP homeobox 1.43 0.0001609 0.0726 AKR1C3 8644 aldo keto reductase family 1, member C3 (3 alpha hydroxysteroid dehydrogenase, type II) 1.41 0.0048328 0.309 POP5 51367 processing of prec ursor 5, ribonuclease P/MRP subunit (S. cerevisiae) 1.37 0.0000503 0.0316 RPL5 6125 ribosomal protein L5 1.35 0.003269 0.276 LOC100133185 100133185 hypothetical LOC100133185 1.33 0.0004762 0.141 LOC338870 338870 ribosomal protein S12 pseudogene 23 1.32 0.0000862 0.0459 C6ORF190 387357 thymocyte selection associated 1.32 0.0004691 0.141 LOC391370 391370 ribosomal protein S12 pseudogene 4 1.32 0.0006724 0.163 CD160 11126 CD160 molecule 1.32 2.00E 03 0.258 ENPP4 22875 ectonucleotide pyrophospha tase/phosphodiesterase 4 (putative) 1.32 0.0023961 0.265 GZMK 3003 granzyme K (granzyme 3; tryptase II) 1.30 0.0023638 0.265 RCN2 5955 reticulocalbin 2, EF hand calcium binding domain 1.30 0.0033195 0.276 TRAPPC2P1 10597 trafficking protein particle complex 2 pseudogene 1 1.30 0.0041531 0.295 CRYZ 1429 crystallin, zeta (quinone reductase) 1.30 0.0044254 0.305 TMEM106B 54664 transmembrane protein 106B 1.28 3.93E 04 0.123 LOC441506 441506 similar to laminin receptor 1 1.28 9.06E 04 0.191 BMI1 648 BMI1 polycomb ring finger oncogene 1.28 0.0023516 0.265 C17ORF45 125144 non protein coding RNA 188 1.28 0.0025859 0.265 ZNF594 84622 zinc finger protein 594 1.27 0.0005382 0.146 FAU 2197 Finkel Biskis Reilly murine sarcoma virus (FBR MuSV) ubiqu itously expressed 1.27 0.0013918 0.231 LAIR2 3904 leukocyte associated immunoglobulin like receptor 2 1.27 0.0020925 0.258 RGS18 64407 regulator of G protein signaling 18 1.27 0.0024428 0.265 ANKRD46 157567 ankyrin repeat domain 46 1.27 0.003684 0. 286 C12ORF41 54934 chromosome 12 open reading frame 41 1.25 0.0002883 0.103 LEPROTL1 23484 leptin receptor overlapping transcript like 1 1.25 5.22E 04 0.144 KLRD1 3824 killer cell lectin like receptor subfamily D, member 1 1.25 0.0006852 0.163 DCK 1633 deoxycytidine kinase 1.25 0.0029953 0.27 ZCCHC7 84186 zinc finger, CCHC domain containing 7 1.25 0.0032017 0.274 LOC100133662 100133662 hypothetical protein LOC100133662 1.23 0.0005223 0.144 LOC388789 388789 hypothetical LOC388789 1.23 1.83E 0 3 0.248 TGDS 23483 TDP glucose 4,6 dehydratase 1.23 2.73E 03 0.267 TRIAP1 51499 TP53 regulated inhibitor of apoptosis 1 1.23 0.0031114 0.273 CREBZF 58487 CREB/ATF bZIP transcription factor 1.23 0.0031873 0.274
153 Table C 2 Continued Gene Symbol Ent rez ID Gene Name Fold Change Parametric P Value FDR TAF7 6879 TAF7 RNA polymerase II, TATA box binding protein (TBP) associated factor, 55kDa 1.23 0.0039695 0.295 PTGDR 5729 prostaglandin D2 receptor (DP) 1.22 0.0002983 0.105 PTGER2 5732 prostaglandi n E receptor 2 (subtype EP2), 53kDa 1.22 0.0009335 0.191 CLNS1A 1207 chloride channel, nucleotide sensitive, 1A 1.22 0.0016481 0.245 GNG2 54331 guanine nucleotide binding protein (G protein), gamma 2 1.22 0.0019601 0.257 ZNF256 10172 zinc finger pr otein 256 1.22 3.28E 03 0.276 LOC648249 648249 similar to 40S ribosomal protein SA (p40) (34/67 kDa laminin receptor) (Colon carcinoma laminin binding protein) (NEM/1CHD4) (Multidrug resistance associated protein MGr1 Ag) 1.22 0.0040863 0.295 CCL4L2 3 88372 chemokine (C C motif) ligand 4 like 2 1.20 0.0000194 0.0175 C16ORF63 123811 FGFR1OP N terminal like 1.20 0.0017643 0.247 C9ORF78 51759 chromosome 9 open reading frame 78 1.20 0.0018299 0.248 LOC440055 440055 ribosomal protein S12 pseudogene 22 1.20 0.0022964 0.265 MOCS2 4338 molybdenum cofactor synthesis 2 1.20 0.0025874 0.265 BBS10 79738 Bardet Biedl syndrome 10 1.20 0.0030162 0.27 ZNF32 7580 zinc finger protein 32 1.20 0.003189 0.274 TSPAN13 27075 tetraspanin 13 1.20 0.0039988 0.295 ZNF480 147657 zinc finger protein 480 1.19 0.0001729 0.0747 MED30 90390 mediator complex subunit 30 1.19 0.0003468 0.116 AQP12B 653437 aquaporin 12B 1.19 0.0009562 0.193 FAM179A 165186 family with sequence similarity 179, member A 1.19 0.0011852 0.224 ZNF550 162972 zinc finger protein 550 1.19 0.0024386 0.265 COX7A2L 9167 cytochrome c oxidase subunit VIIa polypeptide 2 like 1.19 0.0025752 0.265 FLJ14213 79899 proline rich 5 like 1.19 0.0028075 0.269 HSF2 3298 heat shock transcription fact or 2 1.19 0.002991 0.27 ARL6IP5 10550 ADP ribosylation like factor 6 interacting protein 5 1.19 0.0033878 0.277 LEMD3 23592 LEM domain containing 3 1.19 0.0034091 0.277 CXCR7 57007 chemokine (C X C motif) receptor 7 1.19 0.0034953 0.278 SNORD21 608 3 small nucleolar RNA, C/D box 21 1.19 0.0040171 0.295 C1ORF52 148423 chromosome 1 open reading frame 52 1.19 0.0046506 0.307 KCTD5 54442 potassium channel tetramerisation domain containing 5 1.18 0.0002354 0.0942 SMARCE1 6605 SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily e, member 1 1.18 0.0002561 0.0984 C7ORF70 84792 chromosome 7 open reading frame 70 1.18 0.0007107 0.166 CYP2R1 120227 cytochrome P450, family 2, subfamily R, polypeptide 1 1.18 0.0007449 0 .168 HSF5 124535 heat shock transcription factor family member 5 1.18 0.0014013 0.231
154 Table C 2 Continued Gene Symbol Entrez ID Gene Name Fold Change Parametric P Value FDR LOC641844 641844 SDHDP2 succinate dehydrogenase complex, subunit D, integral membrane protein pseudogene 2 1.18 0.0014446 0.232 CD226 10666 CD226 molecule 1.18 0.0017396 0.247 C1ORF75 55248 transmembrane protein 206 1.18 0.0017485 0.247 STK39 27347 serine threonine kinase 39 1.18 0.0020663 0.258 METTL4 64863 methyltransfe rase like 4 1.18 0.0020985 0.258 LOC401206 401206 ribosomal protein S25 pseudogene 6 1.18 0.0026953 0.266 MIS12 79003 MIS12, MIND kinetochore complex component, homolog (S. pombe) 1.18 0.0031138 0.273 C3ORF10 55845 chromosome 3 open reading frame 10 1.18 0.0038643 0.295 CRIPT 9419 cysteine rich PDZ binding protein 1.18 0.0039907 0.295 RPS13 6207 ribosomal protein S13 1.18 0.0039929 0.295 SENP7 57337 SUMO1/sentrin specific peptidase 7 1.18 0.0040921 0.295 SLFN12 55106 schlafen family member 12 1.16 0.0005855 0.154 SLC27A5 10998 solute carrier family 27 (fatty acid transporter), member 5 1.16 0.0006825 0.163 LOC100130154 100130154 similar to thymosin, beta 10 1.16 0.001321 0.231 PRKAB2 5565 protein kinase, AMP activated, beta 2 non cataly tic subunit 1.16 0.0013488 0.231 RPA2 6118 replication protein A2, 32kDa 1.16 0.0014052 0.231 MED4 29079 mediator complex subunit 4 1.16 0.0018549 0.248 FASLG 356 Fas ligand (TNF superfamily, member 6) 1.16 0.001886 0.249 MS4A1 931 membrane spannin g 4 domains, subfamily A, member 1 1.16 0.002495 0.265 HS.25318 1.16 0.002686 0.266 LOC100127918 100127918 similar to small ubiquitin related modifier 2 1.16 0.0035131 0.278 RAB11A 8766 RAB11A, member RAS oncogene family 1.16 0.0040014 0.295 KP NA5 3841 karyopherin alpha 5 (importin alpha 6) 1.16 0.0044516 0.305 SETMAR 6419 SET domain and mariner transposase fusion gene 1.16 0.004666 0.307 SMYD2 56950 SET and MYND domain containing 2 1.15 0.0006797 0.163 KLHDC5 57542 kelch domain containin g 5 1.15 0.0014674 0.232 TMEM9B 56674 TMEM9 domain family, member B 1.15 0.0017569 0.247 ZNF304 57343 zinc finger protein 304 1.15 0.0018168 0.248 GK5 256356 glycerol kinase 5 (putative) 1.15 0.0025517 0.265 SOCS2 8835 suppressor of cytokine signal ing 2 1.15 0.0026346 0.265 LOC347292 347292 ribosomal protein L36 pseudogene 14 1.15 0.0027187 0.267 C11ORF46 120534 chromosome 11 open reading frame 46 1.15 0.0033249 0.276 GK5 256356 glycerol kinase 5 (putative) 1.15 0.0025517 0.265 SOCS2 8835 s uppressor of cytokine signaling 2 1.15 0.0026346 0.265 LOC347292 347292 ribosomal protein L36 pseudogene 14 1.15 0.0027187 0.267 C11ORF46 120534 chromosome 11 open reading frame 46 1.15 0.0033249 0.276
155 Table C 2 Continued Gene Symbol Entrez ID Ge ne Name Fold Change Parametric P Value FDR STX16 8675 syntaxin 16 1.15 0.0037438 0.29 VAMP4 8674 vesicle associated membrane protein 4 1.15 0.0039966 0.295 C1ORF131 128061 chromosome 1 open reading frame 131 1.15 0.0040033 0.295 LTBP2 4053 latent tr ansforming growth factor beta binding protein 2 1.15 0.0040103 0.295 PQLC3 130814 PQ loop repeat containing 3 1.15 0.0040677 0.295 ZNF187 7741 zinc finger protein 187 1.15 0.004258 0.3 NGRN 51335 neugrin, neurite outgrowth associated 1.15 0.0043007 0.302 RHEB 6009 Ras homolog enriched in brain 1.15 0.00447 0.305 HS.133324 1.15 0.0045816 0.305 SNORA76 677842 small nucleolar RNA, H/ACA box 76 1.15 0.0045932 0.305 SERPINE2 5270 serpin peptidase inhibitor, clade E (nexin, plasminogen activato r inhibitor type 1), member 2 1.14 0.0012791 0.231 ERP27 121506 endoplasmic reticulum protein 27 1.14 0.0013217 0.231 CCL3L1 6349 chemokine (C C motif) ligand 3 like 1 1.14 0.0013283 0.231 MLLT11 10962 myeloid/lymphoid or mixed lineage leukemia (tri thorax homolog, Drosophila); translocated to, 11 1.14 0.0013952 0.231 AASDHPPT 60496 aminoadipate semialdehyde dehydrogenase phosphopantetheinyl transferase 1.14 0.00153 0.235 C20ORF30 29058 chromosome 20 open reading frame 30 1.14 0.0016723 0.246 L OC642341 642341 hypothetical LOC642341 1.14 0.0019922 0.258 VTA1 51534 Vps20 associated 1 homolog (S. cerevisiae) 1.14 0.0024282 0.265 MFSD6 54842 major facilitator superfamily domain containing 6 1.14 0.0024388 0.265 KBTBD6 89890 kelch repeat and BT B (POZ) domain containing 6 1.14 0.0025852 0.265 LOC731915 731915 similar to ATP binding cassette sub family D member 1 (Adrenoleukodystrophy protein) (ALDP) 1.14 0.0026105 0.265 CCL3L3 414062 chemokine (C C motif) ligand 3 like 3 1.14 0.002803 0.269 LOC100130633 100130633 hypothetical protein LOC100130633 1.14 0.0029335 0.27 CCNB1IP1 57820 cyclin B1 interacting protein 1, E3 ubiquitin protein ligase 1.14 0.0036227 0.285 PPIA 5478 peptidylprolyl isomerase A (cyclophilin A) 1.14 0.0039574 0.295 MST4 51765 serine/threonine protein kinase MST4 1.14 0.0049485 0.314 ZXDB 158586 zinc finger, X linked, duplicated B 1.12 0.0013386 0.231 NUDT11 55190 nudix (nucleoside diphosphate linked moiety X) type motif 11 1.12 0.0028079 0.269 BCL2L13 23786 BCL2 like 13 (apoptosis facilitator) 1.12 0.0030176 0.27 PALB2 79728 partner and localizer of BRCA2 1.12 0.0030502 0.272 C9ORF85 138241 chromosome 9 open reading frame 85 1.12 0.0030864 0.273 MRPS36 92259 mitochondrial ribosomal protein S36 1.12 0 .0031178 0.273 LYRM4 57128 LYR motif containing 4 1.12 0.0031671 0.274
156 Table C 2 Continued Gene Symbol Entrez ID Gene Name Fold Change Parametric P Value FDR MFF 56947 mitochondrial fission factor 1.12 0.0034594 0.278 MRPS17 51373 mitochondrial r ibosomal protein S17 1.12 0.0038929 0.295 HS.581615 1.12 0.004807 0.309 DNAJA4 55466 DnaJ (Hsp40) homolog, subfamily A, member 4 1.12 0.004935 0.314 FBXO3 26273 F box protein 3 1.11 0.004473 0.305 1 Determined by paired t test at P < 0.005 a nd a permutation of 10,000 (n = 9). FDR, false discovery rate.
157 APPENDIX D EFFECTS OF DIETARY Z INC RESTRICTION ON M T AND ZINC TRANSPORT ER TRANSCRIPTS IN TONGU E EPITHELIAL CELLS O F MOUSE Figure D 1 Effects of dietary zinc deprivation on zinc transporter mRNA levels in the tongue epithelium of mouse. Mice were fed a zinc adequate or low zinc diet for 21 d. The whole tongue was collected and washed with calcium free medium Mixture of collagenase ( 2 mg/mL), dispase (1.5 unit s /mL) and trypsin inhibitor (0. 5 mg/mL) was injected between the epithelium and muscle layers with a 27G needle. After incubation for 15 min the epithelium was peeled off from the cut end of the tongue with a fine forceps T he epithelial sheet was stored in RNAlater, 4 C before treat ed with TRI reagent for total RNA isolation. Relative mRNA abundance was quantified by qPCR and normalized to 18S rRNA levels. Data are expressed as mean SD (n=3 animals)
158 LIST OF REFERENCES 1. Cousins RJ (2006) Zinc. Present Knowl edge in Nutrition eds Bowman BA, Russel RM (Int. Life Sci. Inst. Nutr. Foundation, Washington, DC), 9th Ed, pp 445 457. 2. King J, Cousins RJ (2005) Zinc. Modern Nutrition in Health and Disease eds Shils ME, Shike M, Ross AC, Caballero B, Cousins RJ (Lip pincott Williams and Wilkins, Baltimore), 10th Ed, pp 271 285. 3. Radtke F et al. (1993) Cloned transcription factor MTF 1 activates the mouse metallothionein I promoter. Embo J 12(4):1355 1362 4. Langmade SJ, Ravindra R, Daniels PJ, Andrews GK (2000) Th e transcription factor MTF 1 mediates metal regulation of the mouse ZnT1 gene. J Biol Chem 275(44):34803 34809 5. Zheng D, Feeney GP, Kille P, Hogstrand C (2008) Regulation of ZIP and ZnT zinc transporters in zebrafish gill: zinc repression of ZIP10 trans cription by an intronic MRE cluster. Physiol Genomics 34(2):205 214 6. Lichten LA, Ryu MS, Guo L, Embury JE, Cousins RJ (2011) MTF 1 mediated repression of the zinc transporter Zip10 is alleviated by zinc restriction. PLoS One (in press). 7. Guo L et al. (2010) STAT5 glucocorticoid receptor interaction and MTF 1 regulate the expression of ZnT2 (Slc30a2) in pancreatic acinar cells. Proc Natl Acad Sci U S A 107(7):2818 2823 8. Lichten LA, Cousins RJ (2009) Mammalian zinc transporters: nutritional and physi ologic regulation. Annu Rev Nutr 29:153 176 9. Cousins RJ, Liuzzi JP, Lichten LA (2006) Mammalian zinc transport, trafficking, and signals. J Biol Chem 281(34):24085 24089 10. Liuzzi JP, Guo L, Chang SM, Cousins RJ (2009) Kruppel like factor 4 regulates adaptive expression of the zinc transporter Zip4 in mouse small intestine. Am J Physiol Gastrointest Liver Physiol 296(3):G517 523 11. Dufner Beattie J, Kuo YM, Gitschier J, Andrews GK (2004) The adaptive response to dietary zinc in mice involves the diff erential cellular localization and zinc regulation of the zinc transporters ZIP4 and ZIP5. J Biol Chem 279(47):49082 49090 12. Liuzzi JP, Bobo JA, Lichten LA, Samuelson DA, Cousins RJ (2004) Responsive transporter genes within the murine intestinal pancre atic axis form a basis of zinc homeostasis. Proc Natl Acad Sci U S A 101(40):14355 14360
159 13. Wang F, Kim BE, Petris MJ, Eide DJ (2004) The mammalian Zip5 protein is a zinc transporter that localizes to the basolateral surface of polarized cells. J Biol Ch em 279(49):51433 51441 14. Ryu MS, Lichten LA, Liuzzi JP, Cousins RJ (2008) Zinc transporters ZnT1 (Slc30a1), Zip8 (Slc39a8), and Zip10 (Slc39a10) in mouse red blood cells are differentially regulated during erythroid development and by dietary zinc defic iency. J Nutr 138(11):2076 2083 15. McMahon RJ, Cousins RJ (1998) Regulation of the zinc transporter ZnT 1 by dietary zinc. Proc Natl Acad Sci U S A 95(9):4841 4846 16. Prasad AS (1985) Clinical manifestations of zinc deficiency. Annu Rev Nutr 5:341 363 17. Hambidge M (2000) Human zinc deficiency. J Nutr 130(5S Suppl):1344S 1349S 18. Prasad AS (2001) Recognition of zinc deficiency syndrome. Nutrition 17(1):67 69 19. Jansen J, Karges W, Rink L (2009) Zinc and diabetes -clinical links and molecular mecha nisms. J Nutr Biochem 20(6):399 417 20. Sun Q, van Dam RM, Willett WC, Hu FB (2009) Prospective study of zinc intake and risk of type 2 diabetes in women. Diabetes Care 32(4):629 634 21. Haase H, Rink L (2009) Functional significance of zinc related sign aling pathways in immune cells. Annu Rev Nutr 29:133 152 22. Rink L, Haase H (2007) Zinc homeostasis and immunity. Trends Immunol 28(1):1 4 23. Caulfield LE, Black RE (2004) Zinc deficiency. Comparative quantification of health risks eds Ezzati M, Lopez AD, Rodgers A, Murray CJ (World Health Organization, Geneva, Switzerland), Vol 1, pp 257 280. 24. Dufner Beattie J et al. (2003) The acrodermatitis enteropathica gene ZIP4 encodes a tissue specific, zinc regulated zinc transporter in mice. J Biol Chem 27 8(35):33474 33481 25. Kury S et al. (2002) Identification of SLC39A4, a gene involved in acrodermatitis enteropathica. Nat Genet 31(3):239 240 26. Moynahan EJ (1974) Letter: Acrodermatitis enteropathica: a lethal inherited human zinc deficiency disorder Lancet 2(7877):399 400 27. Chandra RK (1980) Acrodermatitis enteropathica: zinc levels and cell mediated immunity. Pediatrics 66(5):789 791
160 28. Neldner KH, Hambidge KM (1975) Zinc therapy of acrodermatitis enteropathica. N Engl J Med 292(17):879 882 2 9. Turnlund JR, King JC, Keyes WR, Gong B, Michel MC (1984) A stable isotope study of zinc absorption in young men: effects of phytate and alpha cellulose. Am J Clin Nutr 40(5):1071 1077 30. Davies NT, Olpin SE (1979) Studies on the phytate: zinc molar co ntents in diets as a determinant of Zn availability to young rats. Br J Nutr 41(3):590 603 31. Miller LV, Krebs NF, Hambidge KM (2007) A mathematical model of zinc absorption in humans as a function of dietary zinc and phytate. J Nutr 137(1):135 141 32. Hambidge KM, Miller LV, Westcott JE, Krebs NF (2008) Dietary reference intakes for zinc may require adjustment for phytate intake based upon model predictions. J Nutr 138(12):2363 2366 33. Gibson RS, Hess SY, Hotz C, Brown KH (2008) Indicators of zinc sta tus at the population level: a review of the evidence. Br J Nutr 99 Suppl 3:S14 23 34. Ruz M et al. (1991) Development of a dietary model for the study of mild zinc deficiency in humans and evaluation of some biochemical and functional indices of zinc st atus. Am J Clin Nutr 53(5):1295 1303 35. Wessells KR et al. (2010) Plasma zinc concentration responds rapidly to the initiation and discontinuation of short term zinc supplementation in healthy men. J Nutr 140(12):2128 2133 36. Chung CS et al. (2008) C urrent dietary zinc intake has a greater effect on fractional zinc absorption than does longer term zinc consumption in healthy adult men. Am J Clin Nutr 87(5):1224 1229 37. Gordon PR, Woodruff CW, Anderson HL, O'Dell BL (1982) Effect of acute zinc depriv ation on plasma zinc and platelet aggregation in adult males. Am J Clin Nutr 35(1):113 119 38. Hotz C, Peerson JM, Brown KH (2003) Suggested lower cutoffs of serum zinc concentrations for assessing zinc status: reanalysis of the second National Health and Nutrition Examination Survey data (1976 1980). Am J Clin Nutr 78(4):756 764 39. Allan AK et al. (2000) Lymphocyte metallothionein mRNA responds to marginal zinc intake in human volunteers. Br J Nutr 84(5):747 756 40. Cao J, Cousins RJ (2000) Metallothi onein mRNA in monocytes and peripheral blood mononuclear cells and in cells from dried blood spots increases after zinc supplementation of men. J Nutr 130(9):2180 2187
161 41. Aydemir TB, Blanchard RK, Cousins RJ (2006) Zinc supplementation of young men alter s metallothionein, zinc transporter, and cytokine gene expression in leukocyte populations. Proc Natl Acad Sci U S A 103(6):1699 1704 42. Grider A, Bailey LB, Cousins RJ (1990) Erythrocyte metallothionein as an index of zinc status in humans. Proc Natl Ac ad Sci U S A 87(4):1259 1262 43. Thomas EA, Bailey LB, Kauwell GA, Lee DY, Cousins RJ (1992) Erythrocyte metallothionein response to dietary zinc in humans. J Nutr 122(12):2408 2414 44. Andree KB et al. (2004) Investigation of lymphocyte gene expression for use as biomarkers for zinc status in humans. J Nutr 134(7):1716 1723 45. Ruz M, Cavan KR, Bettger WJ, Gibson RS (1992) Erythrocytes, erythrocyte membranes, neutrophils and platelets as biopsy materials for the assessment of zinc status in humans. Br J Nutr 68(2):515 527 46. Smith JC, Jr., Lewis S, Holbrook J, Seidel K, Rose A (1987) Effect of heparin and citrate on measured concentrations of various analytes in plasma. Clin Chem 33(6):814 816 47. Mocchegiani E et al. (2008) Zinc deficiency and IL 6 174G/C polymorphism in old people from different European countries: effect of zinc supplementation. ZINCAGE study. Exp Gerontol 43(5):433 444 48. Prasad AS et al. (1999) Effect of zinc supplementation on incidence of infections and hospital admissions in sickle cell disease (SCD). Am J Hematol 61(3):194 202 49. Prasad AS, Mantzoros CS, Beck FW, Hess JW, Brewer GJ (1996) Zinc status and serum testosterone levels of healthy adults. Nutrition 12(5):344 348 50. Milne DB, Canfield WK, Gallagher SK, Hunt J R, Klevay LM (1987) Ethanol metabolism in postmenopausal women fed a diet marginal in zinc. Am J Clin Nutr 46(4):688 693 51. Aydemir TB, Liuzzi JP, McClellan S, Cousins RJ (2009) Zinc transporter ZIP8 (SLC39A8) and zinc influence IFN gamma expression in a ctivated human T cells. J Leukoc Biol 86(2):337 348 52. Kitamura H et al. (2006) Toll like receptor mediated regulation of zinc homeostasis influences dendritic cell function. Nat Immunol 7(9):971 977 53. Adeniyi FA, Heaton FW (1980) The effect of zinc deficiency on alkaline phosphatase (EC 126.96.36.199) and its isoenzymes. Br J Nutr 43(3):561 569 54. Luecke RW, Olman ME, Baltzer BV (1968) Zinc deficiency in the rat: effect on serum and intestinal alkaline phosphatase activities. J Nutr 94(3):344 350
162 55. Ka sarskis EJ, Schuna A (1980) Serum alkaline phosphatase after treatment of zinc deficiency in humans. Am J Clin Nutr 33(12):2609 2612 56. Weismann K, Hoyer H (1985) Serum alkaline phosphatase and serum zinc levels in the diagnosis and exclusion of zinc def iciency in man. Am J Clin Nutr 41(6):1214 1219 57. Naber TH, Baadenhuysen H, Jansen JB, van den Hamer CJ, van den Broek W (1996) Serum alkaline phosphatase activity during zinc deficiency and long term inflammatory stress. Clin Chim Acta 249(1 2):109 127 58. Blostein Fujii A, DiSilvestro RA, Frid D, Katz C, Malarkey W (1997) Short term zinc supplementation in women with non insulin dependent diabetes mellitus: effects on plasma 5' nucleotidase activities, insulin like growth factor I concentrations, and l ipoprotein oxidation rates in vitro. Am J Clin Nutr 66(3):639 642 59. Davis CD, Milne DB, Nielsen FH (2000) Changes in dietary zinc and copper affect zinc status indicators of postmenopausal women, notably, extracellular superoxide dismutase and amyloid p recursor proteins. Am J Clin Nutr 71(3):781 788 60. Paik HY et al. (1999) Serum extracellular superoxide dismutase activity as an indicator of zinc status in humans. Biol Trace Elem Res 69(1):45 57 61. Driessen C, Hirv K, Kirchner H, Rink L (1995) Zinc regulates cytokine induction by superantigens and lipopolysaccharide. Immunology 84(2):272 277 62. Wellinghausen N, Driessen C, Rink L (1996) Stimulation of human peripheral blood mononuclear cells by zinc and related cations. Cytokine 8(10):767 771 63. Prasad AS (2000) Effects of zinc deficiency on Th1 and Th2 cytokine shifts. J Infect Dis 182 Suppl 1:S62 68 64. Prasad AS, Bao B, Beck FW, Sarkar FH (2006) Correction of interleukin 2 gene expression by in vitro zinc addition to mononuclear cells from zin c deficient human subjects: a specific test for zinc deficiency in humans. Transl Res 148(6):325 333 65. Prasad AS et al. (2007) Zinc supplementation decreases incidence of infections in the elderly: effect of zinc on generation of cytokines and oxidativ e stress. Am J Clin Nutr 85(3):837 844 66. Song Y, Leonard SW, Traber MG, Ho E (2009) Zinc deficiency affects DNA damage, oxidative stress, antioxidant defenses, and DNA repair in rats. J Nutr 139(9):1626 1631 67. Yan M, Song Y, Wong CP, Hardin K, Ho E ( 2008) Zinc deficiency alters DNA damage response genes in normal human prostate epithelial cells. J Nutr 138(4):667 673
163 68. Song Y et al. (2009) Dietary zinc restriction and repletion affects DNA integrity in healthy men. Am J Clin Nutr 90(2):321 328 69 Sunde RA (2010) mRNA transcripts as molecular biomarkers in medicine and nutrition. J Nutr Biochem 21(8):665 670 70. Shi L et al. (2006) The MicroArray Quality Control (MAQC) project shows inter and intraplatform reproducibility of gene expression mea surements. Nat Biotechnol 24(9):1151 1161 71. Tian Z et al. (2009) A practical platform for blood biomarker study by using global gene expression profiling of peripheral whole blood. PLoS One 4(4):e5157 72. Debey S et al. (2006) A highly standardized, robust, and cost effective method for genome wide transcriptome analysis of peripheral blood applicable to large scale clinical trials. Genomics 87(5):653 664 73. Debey Pascher S, Eggle D, Schultze JL (2009) RNA stabilization of peripheral blood and profi ling by bead chip analysis. Methods Mol Biol 496:175 210 74. Wang Z, Gerstein M, Snyder M (2009) RNA Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10(1):57 63 75. Holland NT, Smith MT, Eskenazi B, Bastaki M (2003) Biological sample collect ion and processing for molecular epidemiological studies. Mutat Res 543(3):217 234 76. Rainen L et al. (2002) Stabilization of mRNA expression in whole blood samples. Clin Chem 48(11):1883 1890 77. Debey S et al. (2004) Comparison of different isolatio n techniques prior gene expression profiling of blood derived cells: impact on physiological responses, on overall expression and the role of different cell types. Pharmacogenomics J 4(3):193 207 78. Spira A et al. (2004) Noninvasive method for obtaining RNA from buccal mucosa epithelial cells for gene expression profiling. Biotechniques 36(3):484 487 79. Sridhar S et al. (2008) Smoking induced gene expression changes in the bronchial airway are reflected in nasal and buccal epithelium. BMC Genomics 9:2 59 80. Natarajan P, Trinh T, Mertz L, Goldsborough M, Fox DK (2000) Paper based archiving of mammalian and plant samples for RNA analysis. Biotechniques 29(6):1328 1333 81. Field LA et al. (2007) Functional identity of genes detectable in expression pro filing assays following globin mRNA reduction of peripheral blood samples. Clin Biochem 40(7):499 502
164 82. Liu J, Walter E, Stenger D, Thach D (2006) Effects of globin mRNA reduction methods on gene expression profiles from whole blood. J Mol Diagn 8(5):55 1 558 83. King JC et al. (2001) Effect of acute zinc depletion on zinc homeostasis and plasma zinc kinetics in men. Am J Clin Nutr 74(1):116 124 84. Bonafoux B, Commes T (2009) Serial analysis of gene expression adapted for downsized extracts (SAGE/SADE ) analysis in reticulocytes. Methods Mol Biol 496:299 311 85. Jung R, Lubcke C, Wagener C, Neumaier M (1997) Reversal of RT PCR inhibition observed in heparinized clinical specimens. Biotechniques 23(1):24, 26, 28 86. Ulitsky I et al. (2010) Expander: f rom expression microarrays to networks and functions. Nat Protoc 5(2):303 322 87. Kroh EM, Parkin RK, Mitchell PS, Tewari M (2010) Analysis of circulating microRNA biomarkers in plasma and serum using quantitative reverse transcription PCR (qRT PCR). Meth ods 50(4):298 301 88. Black RE et al. (2008) Maternal and child undernutrition: global and regional exposures and health consequences. Lancet 371(9608):243 260 89. Cousins RJ et al. (2003) A global view of the selectivity of zinc deprivation and excess on genes expressed in human THP 1 mononuclear cells. Proc Natl Acad Sci U S A 100(12):6952 6957 90. Haase H et al. (2007) Differential gene expression after zinc supplementation and deprivation in human leukocyte subsets. Mol Med 13(7 8):362 370 91. Fo ster M, Hancock D, Petocz P, Samman S (2011) Zinc transporter genes are coordinately expressed in men and women independently of dietary or plasma zinc. J Nutr 141(6):1195 1201 92. Seo YA, Lopez V, Kelleher SL (2011) A histidine rich motif mediates mitoch ondrial localization of ZnT2 to modulate mitochondrial function. Am J Physiol Cell Physiol 300(6):C1479 1489 93. Bloomer JR, Reuter RJ, Morton KO, Wehner JM (1983) Enzymatic formation of zinc protoporphyrin by rat liver and its potential effect on hepatic heme metabolism. Gastroenterology 85(3):663 668 94. Fiske DN, McCoy HE, 3rd, Kitchens CS (1994) Zinc induced sideroblastic anemia: report of a case, review of the literature, and description of the hematologic syndrome. Am J Hematol 46(2):147 150 95. Ke en CL, Gershwin ME (1990) Zinc deficiency and immune function. Annu Rev Nutr 10:415 431
165 96. Rink L, Kirchner H (2000) Zinc altered immune function and cytokine production. J Nutr 130(5S Suppl):1407S 1411S 97. Fraker PJ, King LE (2004) Reprogramming of th e immune system during zinc deficiency. Annu Rev Nutr 24:277 298 98. Ho E (2004) Zinc deficiency, DNA damage and cancer risk. J Nutr Biochem 15(10):572 578 99. Fong LY (2010) Zinc in Cancer Development and Prevention. Bioactive Compounds and Cancer, Nutr ition and Health, eds Milner JA, Romagnolo DF (Humana Press, New York), 1st Ed, pp 497 531. 100. Hasan MR, Ho SH, Owen D, Tai IT (2011) VEGF inhibitors induce cellular senescence in colorectal cancer cells. Int J Cancer 101. Wu W, Shu X, Hovsepyan H, Mos teller RD, Broek D (2003) VEGF receptor expression and signaling in human bladder tumors. Oncogene 22(22):3361 3370 102. Golovine K et al. (2008) Depletion of intracellular zinc increases expression of tumorigenic cytokines VEGF, IL 6 and IL 8 in prostat e cancer cells via NF kappaB dependent pathway. Prostate 68(13):1443 1449 103. Issell BF et al. (1981) Serum zinc levels in lung cancer patients. Cancer 47(7):1845 1848 104. Buntzel J et al. (2007) Zinc concentrations in serum during head and neck canc er progression. Anticancer Res 27(4A):1941 1943 105. Martin Lagos F, Navarro Alarcon M, Terres Martos C, Lopez GdlSH, Lopez Martinez MC (1997) Serum copper and zinc concentrations in serum from patients with cancer and cardiovascular disease. Sci Total En viron 204(1):27 35 106. Prasad AS, Bao B, Beck FW, Sarkar FH (2001) Zinc activates NF kappaB in HUT 78 cells. J Lab Clin Med 138(4):250 256 107. Mackenzie GG, Zago MP, Keen CL, Oteiza PI (2002) Low intracellular zinc impairs the translocation of activate d NF kappa B to the nuclei in human neuroblastoma IMR 32 cells. J Biol Chem 277(37):34610 34617 108. Oteiza PI, Clegg MS, Keen CL (2001) Short term zinc deficiency affects nuclear factor kappab nuclear binding activity in rat testes. J Nutr 131(1):21 26 109. Yamamoto Y, Gaynor RB (2001) Therapeutic potential of inhibition of the NF kappaB pathway in the treatment of inflammation and cancer. J Clin Invest 107(2):135 142
166 110. DePasquale Jardieu P, Fraker PJ (1979) The role of corticosterone in the loss in immune function in the zinc deficient A/J mouse. J Nutr 109(11):1847 1855 111. DePasquale Jardieu P, Fraker PJ (1980) Further characterization of the role of corticosterone in the loss of humoral immunity in zinc deficient A/J mice as determined by adrena lectomy. J Immunol 124(6):2650 2655 112. Lee J, Li Z, Brower Sinning R, John B (2007) Regulatory circuit of human microRNA biogenesis. PLoS Comput Biol 3(4):e67 113. Remick DG, Newcomb DE, Friedland JS (2000) Whole blood assays for cytokine production. M ethods Mol Med 36:101 112 114. Hussain R et al. (2002) Cytokine profiles using whole blood assays can discriminate between tuberculosis patients and healthy endemic controls in a BCG vaccinated population. J Immunol Methods 264(1 2):95 108 115. Giamarel los Bourboulis EJ et al. (2010) Ex vivo effect of dexamethasone on cytokine production from whole blood of septic patients: correlation with disease severity. Cytokine 49(1):89 94 116. Schluep M et al. (1998) In vitro cytokine profiles as indicators of relapse activity and clinical course in multiple sclerosis. Mult Scler 4(3):198 202 117. Wallis RS et al. (1998) Measurement of induced cytokines in AIDS clinical trials using whole blood: a preliminary report. ACTG Inducible Cytokines Focus Group. AIDS Clinical Trials Group. Clin Diagn Lab Immunol 5(4):556 560 118. Ahmad SM, Haskell MJ, Raqib R, Stephensen CB (2009) Markers of innate immune function are associated with vitamin a stores in men. J Nutr 139(2):377 385 119. Eggesbo JB, Hjermann I, Hostmark AT, Kierulf P (1996) LPS induced release of IL 1 beta, IL 6, IL 8 and TNF alpha in EDTA or heparin anticoagulated whole blood from persons with high or low levels of serum HDL. Cytokine 8(2):152 160 120. Bai X, Fischer S, Keshavjee S, Liu M (2000) Hepari n interference with reverse transcriptase polymerase chain reaction of RNA extracted from lungs after ischemia reperfusion. Transpl Int 13(2):146 150 121. Beutler E, Gelbart T, Kuhl W (1990) Interference of heparin with the polymerase chain reaction. Biot echniques 9(2):166 122. Wellinghausen N, Kirchner H, Rink L (1997) The immunobiology of zinc. Immunol Today 18(11):519 521 123. Gilmore TD (2006) Introduction to NF kappaB: players, pathways, perspectives. Oncogene 25(51):6680 6684
167 124. Frederickson CJ et al. (2002) Depletion of intracellular zinc from neurons by use of an extracellular chelator in vivo and in vitro. J Histochem Cytochem 50(12):1659 1662 125. Brautigan DL, Bornstein P, Gallis B (1981) Phosphotyrosyl protein phosphatase. Specific inhibi tion by Zn. J Biol Chem 256(13):6519 6522 126. Ho Y et al. (2008) Selective inhibition of mitogen activated protein kinase phosphatases by zinc accounts for extracellular signal regulated kinase 1/2 dependent oxidative neuronal cell death. Mol Pharmacol 74(4):1141 1151 127. Yamasaki S et al. (2007) Zinc is a novel intracellular second messenger. J Cell Biol 177(4):637 645 128. Csermely P, Szamel M, Resch K, Somogyi J (1988) Zinc can increase the activity of protein kinase C and contributes to its bindi ng to plasma membranes in T lymphocytes. J Biol Chem 263(14):6487 6490 129. von Bulow V et al. (2007) Zinc dependent suppression of TNF alpha production is mediated by protein kinase A induced inhibition of Raf 1, I kappa B kinase beta, and NF kappa B. J Immunol 179(6):4180 4186 130. Besecker B et al. (2008) The human zinc transporter SLC39A8 (Zip8) is critical in zinc mediated cytoprotection in lung epithelia. Am J Physiol Lung Cell Mol Physiol 294(6):L1127 1136 131. Bartel DP (2009) MicroRNAs: target recognition and regulatory functions. Cell 136(2):215 233 132. Bartel DP (2004) MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 116(2):281 297 133. Siomi H, Siomi MC (2010) Posttranscriptional regulation of microRNA biogenesis in animals. Mol Cell 38(3):323 332 134. Wittmann J, Jack HM (2010) Serum microRNAs as powerful cancer biomarkers. Biochim Biophys Acta 1806(2):200 207 135. Chen X et al. (2008) Characterization of microRNAs in serum: a novel class of biomarkers for diagnosis of ca ncer and other diseases. Cell Res 18(10):997 1006 136. Mitchell PS et al. (2008) Circulating microRNAs as stable blood based markers for cancer detection. Proc Natl Acad Sci U S A 105(30):10513 10518 137. Gilad S et al. (2008) Serum microRNAs are promi sing novel biomarkers. PLoS One 3(9):e3148
168 138. Hunter MP et al. (2008) Detection of microRNA expression in human peripheral blood microvesicles. PLoS One 3(11):e3694 139. Valadi H et al. (2007) Exosome mediated transfer of mRNAs and microRNAs is a nov el mechanism of genetic exchange between cells. Nat Cell Biol 9(6):654 659 140. Wang K, Zhang S, Weber J, Baxter D, Galas DJ (2010) Export of microRNAs and microRNA protective protein by mammalian cells. Nucleic Acids Res 38(20):7248 7259 141. Bentwich I et al. (2005) Identification of hundreds of conserved and nonconserved human microRNAs. Nat Genet 37(7):766 770 142. Kozomara A, Griffiths Jones S (2011) miRBase: integrating microRNA annotation and deep sequencing data. Nucleic Acids Res 39(Database is sue):D152 157 143. Friedman RC, Farh KK, Burge CB, Bartel DP (2009) Most mammalian mRNAs are conserved targets of microRNAs. Genome Res 19(1):92 105 144. Najafi Shoushtari SH et al. (2010) MicroRNA 33 and the SREBP host genes cooperate to control choles terol homeostasis. Science 328(5985):1566 1569 145. Rayner KJ et al. (2010) MiR 33 contributes to the regulation of cholesterol homeostasis. Science 328(5985):1570 1573 146. Castoldi M et al. (2011) The liver specific microRNA miR 122 controls systemic iron homeostasis in mice. J Clin Invest 121(4):1386 1396 147. Liuzzi JP, Valencia K, Cao J, Gonzalez A (2011) Zinc deficiency increases miR 34a expression in mice. Faseb J 25:977.971. 148. Hershfinkel M, Aizenman E, Andrews G, Sekler I (2010) Zinc bells rang in Jerusalem! Sci Signal 3(129):mr2 149. Esquela Kerscher A, Slack FJ (2006) Oncomirs microRNAs with a role in cancer. Nat Rev Cancer 6(4):259 269 150. Chin LJ, Slack FJ (2008) A truth serum for cancer -microRNAs have major potential as cancer bio markers. Cell Res 18(10):983 984 151. Lee Y et al. (2010) Network modeling identifies molecular functions targeted by miR 204 to suppress head and neck tumor metastasis. PLoS Comput Biol 6(4):e1000730 152. Wei JJ et al. (2011) Regulation of HMGA1 expre ssion by microRNA 296 affects prostate cancer growth and invasion. Clin Cancer Res 17(6):1297 1305
169 153. Ohno H et al. (1985) A study of zinc distribution in erythrocytes of normal humans. Blut 50(2):113 116 154. Hodge D et al. (2006) A global role for EKLF in definitive and primitive erythropoiesis. Blood 107(8):3359 3370 155. Ferreira R, Ohneda K, Yamamoto M, Philipsen S (2005) GATA1 function, a paradigm for transcription factors in hematopoiesis. Mol Cell Biol 25(4):1215 1227 156. O'Dell BL, Brownin g JD, Reeves PG (1987) Zinc deficiency increases the osmotic fragility of rat erythrocytes. J Nutr 117(11):1883 1889 157. Naber TH, van den Hamer CJ, van den Broek WJ, van Tongeren JH (1992) Zinc uptake by blood cells of rats in zinc deficiency and inflam mation. Biol Trace Elem Res 35(2):137 152 158. Van Wouwe JP, Veldhuizen M, De Goeij JJ, Van den Hamer CJ (1991) Laboratory assessment of early dietary, subclinical zinc deficiency: a model study on weaning rats. Pediatr Res 29(4 Pt 1):391 395 159. Cheste rs JK, Will M (1978) The assessment of zinc status of an animal from the uptake of 65Zn by the cells of whole blood in vitro. Br J Nutr 39(2):297 306 160. Glader B (2008) Destruction of Erythrocyte. Wintrobe's Clinical Hematology eds Greer JP, Foerster J Rodgers GM, Paraskevas F, Glader B (Lippincott Williams and Wilkins, Philadelphia), 12th Ed, pp 156 169. 161. Siegel DL, Branton D (1985) Partial purification and characterization of an actin bundling protein, band 4.9, from human erythrocytes. J Cell Bi ol 100(3):775 785 162. Rana AP, Ruff P, Maalouf GJ, Speicher DW, Chishti AH (1993) Cloning of human erythroid dematin reveals another member of the villin family. Proc Natl Acad Sci U S A 90(14):6651 6655 163. Salomao M et al. (2008) Protein 4.1R depend ent multiprotein complex: new insights into the structural organization of the red blood cell membrane. Proc Natl Acad Sci U S A 105(23):8026 8031 164. Anong WA et al. (2009) Adducin forms a bridge between the erythrocyte membrane and its cytoskeleton an d regulates membrane cohesion. Blood 114(9):1904 1912 165. Khanna R et al. (2002) Headpiece domain of dematin is required for the stability of the erythrocyte membrane. Proc Natl Acad Sci U S A 99(10):6637 6642 166. Chen H et al. (2007) Combined deleti on of mouse dematin headpiece and beta adducin exerts a novel effect on the spectrin actin junctions leading to erythrocyte fragility and hemolytic anemia. J Biol Chem 282(6):4124 4135
170 167. Husain Chishti A, Levin A, Branton D (1988) Abolition of actin bu ndling by phosphorylation of human erythrocyte protein 4.9. Nature 334(6184):718 721 168. Tsukamoto T, Suyama K, Germann P, Sonenberg M (1980) Adenosine cyclic 3',5' monophosphate uptake and regulation of membrane protein kinase in intact human erythrocyt es. Biochemistry 19(5):918 924 169. Husain Chishti A, Faquin W, Wu CC, Branton D (1989) Purification of erythrocyte dematin (protein 4.9) reveals an endogenous protein kinase that modulates actin bundling activity. J Biol Chem 264(15):8985 8991 170. Fran k BS, Vardar D, Chishti AH, McKnight CJ (2004) The NMR structure of dematin headpiece reveals a dynamic loop that is conformationally altered upon phosphorylation at a distal site. J Biol Chem 279(9):7909 7916 171. Vugmeyster L, McKnight CJ (2009) Phospho rylation induced changes in backbone dynamics of the dematin headpiece C terminal domain. J Biomol NMR 43(1):39 50 172. Donnelly TE, Jr. (1978) Effects of zinc chloride on the hydrolysis of cyclic GMP and cyclic AMP by the activator dependent cyclic nucle otide phosphodiesterase from bovine heart. Biochim Biophys Acta 522(1):151 160 173. Larsson C (2006) Protein kinase C and the regulation of the actin cytoskeleton. Cell Signal 18(3):276 284 174. Matsuoka Y, Li X, Bennett V (1998) Adducin is an in vivo su bstrate for protein kinase C: phosphorylation in the MARCKS related domain inhibits activity in promoting spectrin actin complexes and occurs in many cells, including dendritic spines of neurons. J Cell Biol 142(2):485 497 175. Lutchman M et al. (1999) L oss of heterozygosity on 8p in prostate cancer implicates a role for dematin in tumor progression. Cancer Genet Cytogenet 115(1):65 69
171 BIOGRAPHICAL SKETCH Moon Suhn Ryu was born in Seoul, South Korea. He received his Bachelor of Science degree in biot echnology from Yonsei University in 2001. Moon Suhn joined the Republic of Korea Air Force for his military service from 2001 to 2003. In 2004, Moon Suhn started his career as an assistant manager of the overseas business unit of L otte Chilsung Beverage Company Limited, in S outh Korea He came to the United States in the fall of 2005 to start his master s studies with Dr. Cousins at the University of Florida. After r eceiving his Master of Science degree in 2005, Moon Suhn continued his graduate studies by entering the doctorate program for nutritional sciences at the University of Florida.