This item is only available as the following downloads:
1 CHARACTERIZATION OF ZIP11 WITHIN THE MURINE GASTROINTESTINAL TRACT By ALYSSA BROOKE MAKI A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGRE E OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Alyssa Brooke Maki
3 To my husband, Tanner Martin and my sisters, Britni Ruppert and Kelsey Nimmer
4 ACKNOWLEDGMENTS I would like to thank my fa mily and friends for fully supporting me through the past years as a doctoral student. It has been a tough road, and they have all stood by me through it all. To my husband, whom I love dearly, thank you for being my rock through this all. I also could not have come this far without the love from my Lord and Savior, Jesus Christ. I could always look to him for peace and guidance through the toughest times. Next, I would like to thank my mentor, Dr. Cousins for giving me this great opportunity to work to in such an amazing lab. This opportunity as a graduate student in his lab will be a time in my life I will never forget. I owe much gratitude to Dr. Cousins for everything I have learned and for helping me become a better scientist. I would like to als o thank my committee members, Dr. James F. Collins, Dr. Bobbi Langkamp Henken and Dr. Don A. Samuelson for their invaluable advice and guidance during my years here as a doctoral student. Lastly, I would like to thank those people from my lab that have p rovided much needed advice and guidance throughout these years: Dr. Tolunay Beker Aydemir, Dr. Shou Mei Chang, Gregory Guthrie, Dr. Moon Suhn Ryu, Dr. Liang Guo Dr. Louis Lichten, Dr. C atalina Troche, and Dr. Inga Wess els. I would also like to thanks oth ers from the FSHN department that were always there for me with a helpful hand or kind words : Dr. Karla Shelnutt, Dr. Vanessa DaSilvia, Cindy Montero, and April Kim.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 12 1 INTRODUCTION ................................ ................................ ................................ .... 14 Zinc and Zinc Transporters ................................ ................................ ..................... 14 Gastric Functionality and Signaling ................................ ................................ ......... 16 Colonic Functionality and Signaling ................................ ................................ ........ 18 The Role of Zinc in Gastric and Colonic Tissues ................................ .................... 20 2 MATERIALS AND METHODS ................................ ................................ ................ 27 Animals ................................ ................................ ................................ ................... 27 Husbandry, Dietary Treatments, and LPS Administration ................................ ....... 27 Genotyping ................................ ................................ ................................ ............. 28 Tissue Collection ................................ ................................ ................................ .... 31 Measurement of Serum and Tissue Zinc Concentrations ................................ ....... 32 RNA Isolation ................................ ................................ ................................ .......... 32 Zinc Transporter Antibodies ................................ ................................ .................... 34 Immunoblotting ................................ ................................ ................................ ....... 34 Immunohistochemistry ................................ ................................ ............................ 36 Immunoperoxidase Staining ................................ ................................ ............. 36 Immunofluorescence ................................ ................................ ........................ 37 Isolation and Analysis of Serum microRNA ................................ ............................ 38 Statistical Analysis ................................ ................................ ................................ .. 39 3 ZIP11 KNOCK OUT MOUSE MODEL CHARACTERIZATION ............................... 42 Introductory Remarks ................................ ................................ .............................. 42 Results ................................ ................................ ................................ .................... 45 Understanding the Zip11 Knockout Model ................................ ........................ 45 A Significant Decrease in the Zip11 RNA Expression in KO Mice .................... 4 9 No Change in ZIP11 Protein Expression in KO Mice ................................ ....... 50 Dietary Zinc Did Not Affect the Functionality of KO Mice ................................ 51 Discussion ................................ ................................ ................................ .............. 53
6 4 EFFECT OF DIETARY ZINC ON EXPRESSION OF ZIP11 AND OTHER ZINC TRANSPORTERS WITHIN THE GASTROINTESTINAL TRACT ........................... 63 Introductory Remarks ................................ ................................ .............................. 63 Results ................................ ................................ ................................ .................... 65 Features of the Zip11 Gene and the Predicted ZIP11 Topology ...................... 65 Zip11 RNA and Protein Expression Tissue Distribution ................................ ... 65 Tissue Zinc Concentrations of Stomach, Intestines, and Colon ....................... 67 Zip11 mRNA Decreases During Zinc Deficiency ................................ .............. 68 ZIP11 Protein Expression Is Variable Among Tissues During Zinc Deficiency ................................ ................................ ................................ ...... 69 ZIP11 Localizes to the Nuclei in the Gastric Tissue ................................ ......... 71 ZIP11 Localizes to the Colonic Epithelial Cells ................................ ................. 71 ZIP4 Increases in the Colon During Zinc Deficiency ................................ ........ 72 Zip5 Increases in the Stomach and Colon During Zinc Supplementation ......... 73 Several microRNAs Increase During Zinc Deficiency ................................ ....... 73 Discussion ................................ ................................ ................................ .............. 74 5 CHARACTERIZATION OF ZIP11, ZIP4, AND ZIP5 E XPRESSION IN THE MURINE COLON DURING ACUTE INFLAMMATION ................................ ............ 97 Introductory Remarks ................................ ................................ .............................. 97 Results ................................ ................................ ................................ .................... 99 Colonic Tissue Zinc Decreases After an LPS Challenge ................................ .. 99 Colonic Metallothionein mRNA Expression Increases in Response to LPS ..... 99 Zip11 Colonic mRNA Expression Decreases in Response to LPS ................... 99 Zip4 Colonic Expression Increases in Response to LPS ................................ 100 Zip5 Colonic Expression Decreases in Response to LPS .............................. 100 Discussion ................................ ................................ ................................ ............ 100 6 CONCLUSIONS AND FUTURE DIRECTION ................................ ....................... 108 LIST OF REFERENCES ................................ ................................ ............................. 110 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 117
7 LIST OF TABLES Table p age 2 1 Zip Transporter Primer/Probe Sequences ................................ .......................... 40 2 2 ZnT Transporter Primer/Probe Sequences ................................ ......................... 40 2 3 Other Primer/Probe Sequences ................................ ................................ .......... 40 2 4 ZIP Transp orter Antibody Sequences ................................ ................................ 41
8 LIST OF FIGURES Figure page 3 1 Diagrams of the mouse Zip11 gene, gene trap, and alternative slicing ............... 56 3 2 Genotyping tail sample DNA ................................ ................................ ............... 57 3 3 Confirming genotype s using different primer sets ................................ ............... 58 3 4 Knock out Zip11 transcript exp ression compared to wild type. .......................... 59 3 5 Wild type versus knock out ZIP11 tissue distribution. ................................ ........ 60 3 6 Plasma and tissue zinc concentrations of K O mice on dietary restriction ........... 61 3 7 Effects of zinc depletion and supplementation on zinc related gene transcripts in the sto mach and colon of the KO mice. ................................ ......... 62 4 1 Features of the murine Zip11 gene. ................................ ................................ ... 81 4 2 Tissue distribution of the Zip11 ge ne transcript in C57BL/6 mice. ...................... 82 4 3 ZIP11 protein tissue dis tribution in the murine model ................................ ......... 83 4 4 Wild type tissue zinc concentrations during ZnA, Z nD, or ZnS dietary conditions. ................................ ................................ ................................ .......... 84 4 5 Effects of dietary zinc depleti on and repletion (ZnR) on serum zi nc and GI tract MT expression. ................................ ................................ ........................... 85 4 6 Effects of dietary zinc depletion and repletion on the Zip11 mRNA throughout the GI tract. ................................ ................................ ................................ ........ 86 4 7 Effects of dietary zinc depletion and repletion on the murine ZIP11 protein expression in the stomach. ................................ ................................ ................ 87 4 8 Effects of dietary zinc depletion a nd repletion on the murine ZIP11 pro tein expression in the colon. ................................ ................................ ...................... 88 4 9 Effects of dietary zinc depletion and repletion on the murine ZIP11 protein expression in the cecum and small intestin e. ................................ ..................... 89 4 10 Visualizing the murine ZIP11 protein with immunoperoxidase staining in the stom ach. ................................ ................................ ................................ ............. 90 4 11 Visualizing the murine ZIP11 protein with immunoperoxidase st aining in the colon and cecum. ................................ ................................ .............................. 91
9 4 12 Immunofluorescence imaging o f ZIP11 in the murine stomach. ......................... 92 4 13 Immunofluorescence imaging of ZIP11 in the murine colon. .............................. 93 4 14 The effect of ZnD or ZnR on ZIP4 expression in murine GI tract. ...................... 94 4 15 The effect of ZnS and ZnR on Zip5 expr ession in the murine GI tract. ............... 95 4 16 Identification of serum miRNAs responsive to dietary zinc deficiency in mice using a qPCR based array. ................................ ................................ ............... 96 5 1 The effects of a ZnA or ZnS diet on murine colonic tissue zinc concentrations and MT mRNA expression after LPS administration.. ................................ ...... 104 5 2 The effect of LPS administration on colonic Zip11 expression when mice were fed either a ZnA or ZnS diet. ................................ ................................ ... 105 5 3 The effect ZnA and ZnS diets have on murine Zip4 expression in the c olon after LPS administration. ................................ ................................ ................. 106 5 4 The effect of LPS administration on Zip5 colonic expression when mice were either fed a ZnA or ZnS diet ................................ ................................ ............ 107
10 LIST OF ABBREVIATIONS AAS Atomic Absorption Spectrophotometry AE Acrodermatits Enteropathica cDNA Complementary Deoxyribonucleic Acid C T Threshold Cycle DNA Deoxyribonucleic Acid ES Embryonic Stem Cell GI Gastrointestinal Tract HET Heterozygous HOM Homozygous Null IF Immunofluorescence IgG Immunoglobulin G IHC Immunohistochemistry ISC Intestinal Stem Cell KO Knock Out LPS Lipopolysaccharide miRNA MicroRNA MRE Metal Response Element mRNA Messenger Ribonucleic Acid MT Metallothionein M TF 1 Metal Response Element Binding Transcription Factor 1 NTC No Template Control PBS Phosphate Buffered Saline qPCR Quantitative Polymerase Chain Reaction RNA Ribonucleic Acid
11 TBP Tata Binding Protein TM Transmembrane TV Tubulovesicle WT Wild Type Z IP Zrt Irt like Zinc Transporter Superfamily Zip 11 Zip11 gene Zip11 Zip 11 mRNA ZIP11 Zip11 protein ZnA Zinc Adequate ZnD Zinc Deficient ZnR Zinc Repletion ZnS Zinc Supplementation ZnT Zinc Transporter
12 Abstract of Dissertation Presented to the Graduate S chool of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy CHARACTERIZATION OF ZIP11 WITHIN THE MURINE GASTROINTESTINAL TRACT By Alyssa Brooke Maki August 2013 Chair: Robert J. C ousins Major: Nutritional Sciences Zinc transporters have been characterized to further understand the absorption and metabolism of dietary zinc. Two transporter families have been identified: the ZnT and the Zrt Irt like (Zip) transporters. A previou sly uncharacterized Zip transporter, Zip11 was found through microarray analysis to be highly expressed in the murine stomach, cecum, and colon. The objective was to characterize Zip11 within the murine gastrointestinal tract and to determine if dietary zi nc and inflammation regulated Zip11 Through q PCR and western blot analysis, Zip11 was shown to be down regulat ed during zinc deficiency in murine stomach tissue but appeared to be unaffected in the large and small intestine. Immunohistochemistry (IHC) r evealed high ZIP11 expres sion in the nucleus of gastric and colonic epithelial cells, while some staining was seen in the cytoplasm of these cells. IHC analysis of the colon also revealed an increase in ZIP 11 staining during zinc deficiency. Since inflam mation can affect the colon Zip11 expression was analyzed during acute inflammation. A lipopolysaccharide (LPS ) murine model was utilized to create acute inflammatory responses for that purpose. ZIP11 protein i ncreased after LPS administration. A Zip11 knock out murine model was
13 obtained to further analyze functional roles of Zip11 Due to the genetrap used a hypo morphic al lele was created, resulting in no change to ZIP11 protein expression The high expression of this specific zinc transporter in the stomach and colon relates to recent o bservations related to functions of zinc in these organs For example, zinc deficiency is known to damage gastric mucosa and total gastric acidity increases with supplementation Research has also shown zinc accumulation in the parietal cell when the pr oton pump was inhibited, thus linking acid output and cellular zinc levels The colon absorbs fractional zinc during healthy states; however, when zinc absorption is compromised during malnutrition, the colon appears to have an increased efficiency for ab sorption. In conclusion, Zip11 is highly expressed within the murine stomach and large intestine and appears to be partially regulated by zinc deficiency within the se tissues. Zip11 may play a role in zinc homeostasis within these tissues, helping to mai ntain mucosal integrity and function.
14 CHAPTER 1 INTRODUCTION Zinc and Zinc Transporters Essential nutrition in living organisms requires the balance of many macronutrients and micronutrients. One particular micronutrient, zinc, has been shown to be req uired for proper growth and development in living organisms. This trace mineral is found throughout the human body, but the tissues with the high percentages are skeletal muscle and bone (57% a nd 29% of total body zinc respectively) ( 13, 38 ). Zinc can p lay roles in three different biological functions: catalytic, structural, and regulatory. More than 300 enzymes require zinc for proper functionality, and a loss of function is observed when zinc is removed (13 ) Some examples of zinc metalloenzymes are carbonic anhydrase, alkaline phosphastase, RNA polymerases, and reverse transcriptase. Zinc is also required for structural stability of proteins binding to such motifs as the zin c finger DNA binding domains (38, 70 ) Newer evidence in the zinc field ha s also shown that zinc may influence cell signaling pathways involving STAT3 CREB and protein kinases and phosphatases (12, 26 ) An inducible metalloprotein known as metallothionein ( MT ) act s as a metal detoxif ier and a zinc exchanger. Changes in dietary zinc concentration will influence MT F or example during zinc deficiency, M T RNA expression will be significantly decreased in the kidney and liver. Zinc and other metals stabilize MT from degradation. Meta l response element(s) (MRE) have been identified on the promoter region of the MT gene and several other genes. A protein that binds this MRE region, known as metal reg ulatory transcription factor (MTF 1), senses the intracellular zinc concentration, bind s up Zn, and travels into the nucleus where it will recognize the MRE on the MT
15 gene and several other genes ( 38, 70 ). Zinc may also play regulatory roles involving 38 ), as with T cell activation or regulation of i nsulin like growth factor. With the high abundance of zinc in the central nervous system (CNS), several regulatory mechanisms must be in place to ensure zinc homeostasis. Both the regulation of MT and zinc transporters need to be working properly to ensu re zinc does not accumulate in the CNS. Several zinc transporter proteins have been identified and characterized over the years, and currently two families divide up these transporters. The ZnT family (Slc30A) is comprised of ten members, whose main objec tive is lowering the intracellular zinc concentration, whether by sequestering zinc into organelles or the efflux of zinc from cells. Fourteen membe rs have been placed into the Zrt Irt like protein (ZIP) family (Slc39A), and their main function is to incr ease the intracellular zinc concentration of cells by transporting zinc into the cytoplasm from outside of the cell or from the organelles. The ZnT transporter family of proteins is predicted to have six transmembrane domains, while the Zip transporter pr otein family is predicted to contain eight transmembrane domains. Both zinc families contain a histidine rich loop that is thought to contain a metal binding site (12 1 9 ). These transporters not only facilitate the flux of zinc, but may also other eleme nts such as iron, cadmium, and manganese (29, 4 9 ) Several zinc transporters have been characterized, and much is known about their cellular location, what nutrients they are shuttling, and if they play a role in zinc related nutritional issues. The first zinc transporter to be discovered was ZnT1 and its expression has been shown to be distributed within several tissues. ZnT1 may play a role in zinc transfer from the enterocytes, recycling zinc from the kidney, and has also
16 been shown, with a knock out (KO) model, to be essential for the transfer of maternal zinc to the embryo (5 ) ZnT5 has also been shown to have varying responses due to zinc supplementation and restriction, depending on the cell type being an alyzed (44 ), and through a ZnT5 KO model that showed reduced growth and decreased bone density in mice ( 31 ), this transporter has been shown to be vital to zinc homeostasis. ZIP 4 is most well known for its mutation that causes the disease acrodermatitis enteropathica (AE) (44, 72 ) ZIP 4 can be found highly expressed in the gastrointestinal tract (GI), including the three regions of the small intestine, and to a lesser ext ent in the stomach and colon (18 ). The ZIP 4 mutation causes zinc deficiency within these AE patients by aff ecting the absorption of zinc by the intestines. Those suffering from AE, however, can overcome the deficiency by zinc supplementation (44 ) suggesting other zinc transporters must exist within the intestine. Zip 5 unlike the expr ession of Zip4 on the apical membrane, is expressed on the basolateral membrane of the entero cyte. During zinc repletion ZIP 5 localizes to the basolateral surface, but during deficiency it is internalized and degraded in enterocytes ( 44, 64 71 ). Zip 14 i s another zinc transporter found to be important in zinc homeostasis in the liver (7 ) and it is also expressed within the small intestine, where it could potentially play a role during inflammation ( 11 ). Gastric Functionality and Signaling Even t hough th e stomach may not be a major site for nutrient absorption, this organ is a secretory tissue that has many important functions to prepare nutrients to be absorbed in the small intestine Within the three major regions of the stomach ( non glandular body, a nd pylorus), the gastric mucosa harbors the secretion glands and gastric pits. The different gastric secretions consist of hydrochloric acid, pepsinogen, bicarbonate, gastrin, intrinsic factor, mucus, and water and these products can be
17 involved in sever al functions of the stomach, including endocrine and motility functions (35, 40, 65 ) The gastric glands of the upper portion of the stomach contain four cell types found in different regions of the gland. These cell types consist of the chief cells, parietal cells, enterochromaffin like (ECL) cells and neck cells, moving from the base towards the opening of the gland respectively. Cells found within the pyloric region of the stomach are known as G cells and D c ells. These cells secrete either gastrin or somatostatin, respectively, and these secretions are also involved in acid secretion of parietal cells (65 ) All of the cells within the stomach have specific function s to maintain a protective mucosal barrier, control the pH, or aid with digestion. The parietal cell has many ion transporters and channels located on the apical and basolateral membranes. These transporters are important to mainta in the proper ion balance within the parietal cell. When acid secretion occurs, the intracellular lumen of a parietal cell is depleted of chloride ions, and bicarbonate ions begin to accumulate (40, 65 ) Transpor ters such as H,K ATPase, Slc26a9, and possibly NHE3 localize to the apical surfa ce; while, the Slc4a2, NHE1,2,4; Na, K ATPase, and possibly NKCC1 localize to the basolateral surface (40 ) Many factors stimulate or repress the secretion of acid from the parietal cell. Gastrin, histamine, and acetylcho line (ACh) are the major stimulators of acid secretion. Gastrin will either act on the parietal cell directly, binding to the CCK B receptors, increasing intracellular Ca 2+ or indirectly, binding to the ECL cells. The ECL cell releases histamine after gast rin activation or neural stimulation by pituitary adenylate cyclase activating polypeptide (PACAP) (40 61, 65 ) Histamine will then induce the cAMP mediated PKA pathway by binding to the H 2 receptors on the basol ateral surface. Lastly, ACh is released by the vagus nerve due to another neural
18 stimulation and binds to the M 3 receptor. ACh binding also stimulates an increase in intracellular Ca 2+ (40, 61 ) The stimulation of the parietal cell by increasing intracellular calcium and cAMP causes the inactive tubulovesicle (TV) to move to the apical membrane. The TVs store the H,K ATPase proton pumps, and upon stimulation TVs are trafficked to the apical membrane where acid c an now be secreted out of the parietal cell (40, 61, 79 ) The stomach has a complex network of mechanisms involved in creating a toxic environment for harmful bacteria, maintaining proper pH for digestion, and beg inning the process of breaking down nutrients. Without these networks working properly, digestion can be severely altered, the integrity of the mucosal layers will be disrupted, and issues such as acid reflux may occur. Colonic Functionality and Signaling The large intestine consists of the cecum, colon, rectum and anus, but for this absorption of water, Na + and other minerals, and vitamins, such as Vitamin K. The colon is the last region for these molecules to be recaptured before excretion. Unlike the small intestine, the colon does not have villi present at the luminal surface, rather colonic crypts are present. The cells present in the colonic crypt are similar to t hose found in the intestinal mucosa; however, paneth cells have not been detected in the lower regions of the colonic crypts A region of intestinal stem cells (ISCs) has been identified at the base of the crypt containing crypt base columnar cells (CBCC s) and +4 stem cells. These ISCs will start to differentiate into secretory goblet and enteroendocrine cells or enterocyte absorptive cells as they move up the colonic crypt ( 51 ). These cells are responsible for the secretory and absorptive properties of the colon.
19 The transport of NaCl in the colon is through electroneutral absorption of ions. The cell surface exchangers, Na/H and Cl/HCO 3 are located on the apical membrane, as in the stomach, with two exceptions being the NHE1 and a Cl/HCO 3 exchanger located on the basolateral membrane. Evidence has shown that these exchangers can be affected by the cystic fibrosis transmembrane conductance regulator (CFTR), and mutations in this gene can have a detrimental effect on fluid absorption in the colon. An other pathway for Na + uptake is also found in the distal colon and is known as the epithelial Na + channel (ENaC). This channel is localized to the apical membrane, and the Na+ absorption coincides with Cl absorption through a Cl channel or paracellular diffusion. These ions exit through the Na/K ATPase pump, Cl channels, or the Cl/HNO 3 exchanger located on the basolateral membrane ( 22 ). important for maintaining fluid bala nce. The secretory pathway is thought to assist the transport of mucus out of the crypts, while retaining the hydration of the mucus. This balance of secretion and absorption is critical for maintaining barrier function and avoiding secretory diarrhea in conjunction with a large loss of electrolytes ( 22 ). Secretory channels & transporters found within the colon consist of the Na + 2Cl K cotransporter type 1 (NKCC1), Cl and K channels, and a variety of pathways that are involved in secreting bicarbonate. The NKCC1 transporter is found on the basolateral membrane of colonocytes and research with NKCC1 knock out mice showed Cl secretion reduction in the colon. The Cl channels involved in secretion consist of the CFTR protein (apical membrane), Ca activa ted Cl channels, and in some animals basolateral Cl channels, though this channel has not been confirmed in humans ( 22 ).
20 As with the chloride channels, potassium channels are also located on both the apical and basolateral membranes. Two important regul ators of ion secretion in the colon are the cAMP and cGMP dependent secretions. Concentration changes of cAMP and cGMP can both result in Cl flux; depending on the protein they are activating ( 22 ). Lastly, a calcium sensing receptor (CaSR) has been iden tified in the colon on both the apical and basolateral membranes. This receptor is important for altering fluid secretion by detecting changes in calcium level s thus reacting by increasing protein kinase C activity, which in turn augments the breakdown o f cAMP or cGMP. The destruction of cAMP or cGMP will cause a decrease in fluid secretion. By increasing calcium delivery to the receptor, CaSR could be a potential therapeutic target during disease states (IBD, diarrhea, etc) to help alter secretion and reabsorption of electrolytes (22 ). In recent years, the scientific community has seen a significant rise in the number of publications studying the effects of commensal microbiota on gut health and the health of an individual. The colon contains the highe st concentration, 10 12 cells/gram of wet weight of gut microbes compared to the other regions of the GI tract, such as the stomach which harbors around 10 1 cells/gram of wet weight ( 37 62 ). Several factors can vary the amount, type, or location of the c ommensal bacteria in the host, such as age, diet, antibiotic usage, or environmental exposures ( 62 ). Colonic zinc concentration could be an important factor when examining commensal microbiota health and diseases that affect this microbial community. The Role of Zinc in Gastric and Colonic Tissues Zinc metabolism and regulation is found throug hout various tissues such as ZIP 4 in the enterocytes ( 18, 46 ) ZnT2 present in pancreatic acinar cells ( 25, 46 ) and Zip14 in the liver ( 48 ) The stoma ch does not appear to be a site of zinc absorption ( 36 ) but
21 z inc regulation within the gastric mucosa has been shown to be important during the beginning of digestion. The accumulation of zinc into the tubulove sicle (TV) membrane and possibly regulated by the secretion of HCl from the parietal cell. When the H,K ATPase pump is inhibited, the concentration of Zn 2+ increases within the cytoplasm of the cell due to immobilization into the TV compartment. Along with the able to show that zinc may be regulating the back diffusion of hydrogen ions from the TVs ( 24, 54 ) Zinc could thus be an important factor in tubulovesicle maturation and maintaining the integrity of a mature parietal call. Studies looking at dietary zinc status have revealed findings that may also show the i mportance of zinc for acid secretion. Stomachs of zinc deficient rats show signs of gastric damage when compared to control rats, and these deficient rats also had a decrease in acidic output. Some explanations for this decrease in acid could be the redu ction of carbonic anhydrase activity or an increase of hydrogen ions being reabsorbed by the eroded gastric mucosa. Carbonic anhydrase a zinc metalloenzyme is activated by zinc and is required for carbonic acid production from water and carbon dioxide in the parietal cells ( 56 ) The effect of zinc supplementation has also been analyzed within rat stomachs by Yama guchi et al Zinc sulfate provided in water or as an intraperitoneal ( ip ) injection resulted in an increase in acid secretion. In the ip injected rats, the pylorus was ligated one hour after the zinc was administered and total gastric acidity was measure d 3 hours after ligation. The total gastric acidity increased when 3 g of zinc /100 g of body weight
22 up to 50 g of zinc /100 g of body weight was ip administered; however total acidity reached a maximum at 50 g Zn/100 g of body weight ( 77 ) Yamaguchi et al. also wanted to determine the effects of known stimulato rs of acid secretion (histamine, gastrin, ACC) on the total acidity of gastric secretion after rats were injected with zinc (77) Rats were pretreated with an ip injection of zinc and 1 hour later were injected with histamine, tetragastrin (a peptide frag ment of gastrin), and ACC. The combination of histamine or tetragastrin injections after ip zinc administration produced a significant increase in acid secretion when compared to injection of zinc alone; however, the combination of ACC and zinc did not pr oduce a change in gastric acidity. Fu rther studies showed that total acidity of gastric secretion was significantly decreased when zinc and an ACC inhibitor were both administered ; however, when ACC was administered along with zinc and the inhibitor, gast ric acid secretion levels returned to increased levels seen in ACC stimulated control rats. ACC could be an underlying factor of the zinc stimulating mechanism of acid secretion ( 77 ) This previous study did not examine the integrity of the gastric mucosa, but other research has shown that a disruption in the leve l of zinc within a parietal cell results in decreased cell viability (39) Monochloramine is an oxidant produced by bacterial and host interactions within the stomach. This oxidant has been shown to disrupt calcium homeostasis and mucosal integrity. Mo nochloramine has also been shown to negatively influence the intracellular levels of zinc within a parietal cell, and this increase in intracellular zinc reduces cell viability ( 39 ) If monocholoramine disrupts ca lcium homeostasis then this oxidant could also be interrupting the stimulation of the H,K ATPase transporter by calcium As mentioned above if this transporter of HCl is blocked the concentration of
23 zinc within the cytoplasm increases causing a decrease in cell viability. Maintaining zinc homeostasis within the gastric mucosa and parietal cells appears to be an important facto r in normal gastric functions. S ince zinc could be affecting gastric mucosa integrity and acid secretion, zinc transporter prote ins may be involved in regulating the zinc concentrations to maintain these functions Using IHC, ZnT4 and ZnT5 were visualized within the cytoplasm of the murine parietal cell. ZnT6 was detected within the murine chief cells, while ZnT7 was expressed wi thin the murine neck cells ( 79 ). These transporters could be involved in sequestering zinc into the cytoplasm to aid in gastric secretions. 65 Zn has been used for decades to study zinc absorption by the gastrointestinal tract ( 14 55 ) Of particular not e is a study where rats were fed a zinc deficient diet two days prior to measuring zinc absorption. This approach was used so the zinc concentration within the gut would not interfere with the specific activity of the 65 Zn dose administered ( 14 ). 65 Zn wa s injected into the different regions of the small intestine and large intestine of fasted rats 65 Zn was measured daily with a whole body gamma counter. 65 Zn retained by each tissue and the contribution to overall zinc absorption by each section were al so measured These isotope studies showed that zinc absorption decreases moving down the small intestine with the duodenum contributing to the highest absorption and the large intestine, including the cecum and colon revealing minimal zinc absorption ( 14 64 ). The large intestine has been shown to be incapable of absorbing sufficient zinc to meet requiremen ts for proper growth in rats (28 ). Hara et al. also analyzed how zinc absorption in the large intestine was affected by a gastric acid inhibitor (omep razole), along with removing the large intestine. Their results indicated
24 that the large intestine has increased efficiency to absorb zinc when omeprazole was provided to the rats. Rats with their cecum and colon removed, showed a decrease in zinc absorp tion, and the zinc absorption compensation seen in control groups, with their large intestines intact, was lost due to the re moval of the large intestine (28 ). Though the stud i es mentioned above reported the colon to have little zinc absorptive properties Nevah et al. were able to show that the colon could play a larger role in zinc absorption in rats Zinc absorption studies were completed by injecting 65 Zn into ligated colonic segments ( 55 ) Uptake, mucosal retention, and absorption were calculated af ter the tissue and luminal 65 Zn content was measured by a gamma spectro meter. This group was able to show that the colon could absorb 14% of the 65 Zn injected into the ligated colon within 15 minutes ( 55 ). Changes in the large intestine can be attributa ble to differences in species, age, or the commensal bacterial population which could be causing variations among research studies. The large intestine may have a low e fficiency for zinc absorption in healthy individuals; however, the large intestine can be a site of zinc absorption when gastric a cid secretion or small intestinal zinc absorption is impaired. Another key factor in examining zinc absorption with 65 Zn is t he amount of time a dose is in contact with a particular segment ; i.e. the transit time of each segment will affect results ( 14 ). Taking transit time into consideration, the colon could be a significant contributor to zinc absorption when colonic transit time is low, such as during constipation. The colon has the potential to be a large c ontributor to zinc absorption during different disease states where gastric acid production is altered, small intestinal zinc absorption is decreased, or during times when transit time is significantly altered.
25 Expression of only a few Zip and ZnT transp orters has been characterized within the colon. Z IP 4 localizes to the apical membrane, while Z IP 5 localizes to the basolateral membrane of the colonic epithelial cells ( 16 71 ) Zip4 expression in the colon, as with the small intestine, increases during ZnD, while Zip5 does not seem to be altered by ZnD. ZnT6 and ZnT1 are also expressed within the cecum and colon, and could be involved in zinc import ( 44, 64, 79 ). Other ZnT transporters shown to be strongly expressed in the colon were ZnT4 and ZnT7, whi ch both showed expression in the epithelial and goblet cells ( 44, 79 ). Some of these zinc transporters may be involved with importing zinc into the colonic cells and others may be involved in trafficking zinc throughout the colonic cells. Endogenous zinc may play a role in regulating Cl secretion from the colonic cells. Physiological levels of zinc (~100 M) have been shown to inhibit Cl secretion (64 ), which could be detrimental for the En dogenous zinc secretion by the c olon is not well understood. M ore research is required to understand how differing zin c concentrations can affect mu cus secretion pH, the microbial community, and mucosal integrity. With all of the information discussed above, obtaining some preliminary pilot study data, and the availability of a knock out mouse model, a hypothesis was formed that Zip11 is highly expressed within the murine gastrointestinal tract and influences zinc metabolism within the GI tract, mainly in the stomach and colon. To investigate this hypothesis, three specific aims were proposed in this dissertation project: Examine the phenotypic effects of the Zip11 / mutation and analyze zinc metabolism within this mouse model
26 Characterize expression of Zip11 and other zinc transporters in the gastrointestinal tract, particularly the stomach and colon and selected other o rgans during control or dietary zinc studies. Characterize expression of Zip11 and other zinc transporters in colon during acute in flammation.
27 CHAPTER 2 MATERIALS AND METHODS Animals A random genetrap insertion in the Zip11 gene (between exons 4 and 5) was created by Bay Genomics MMRRC in the mouse line 129P2/OlaHsd derived from embryonic stem cells This clone was sent to UC Da vis, University of California, where chimeras were generated. The chimeras were bred with C57BL/6 mice to generate F1 Zip11 /+ mice. Two heterozygous breeding pairs were obtained and further breeding was completed at the Genetic and Cancer Research Compl ex (GCRC), University of Florida. Once a knock out line had been bred from heterozygous mice, confirmed by genotyping, knock out breeders were placed together to provide a sufficient number of knock out animals for the experiments. The phenotype produced by the specific knock out line did not lead to lethality. Husbandr y, Dietary Treatments and LPS Administration Mice were given free access to tap water and received commercial rodent diets ( Harlan Teklad 7912) with 12 hour light and dark cycle. Wild t ype, heterozygous, and knock out mice generated from the founder mice were all used at different times for different experiments during this dissertation project. Both male and female young adult (8 12 weeks) mice were used at different points of this r esearch. Different dietary treatments were used. The murine purified diet consisted of the AIN76 diet with egg white protein and either low zinc (ZnD, <1 mg Zn/kg diet), adequate zinc (ZnA, 30 mg Zn/kg diet), or high zinc (180 mg Zn/kg diet) ( 46, 53 ). W hen mice were fed the purified zinc diets they were placed in hanging stainless steel cages and received free access to Mill Q water and food. Mice were acclimated to the hanging wire cages for one
28 week while receiving free access to water and ZnA food. After acclimation, mice were switched to new hanging wire cages and provided the appropriate experimental diet and water. Mice were placed on the varying zinc diets from 1 3 weeks depending on the dietary experiment. W eights and food intake were meas ured weekly After dietary experiments, mice were anesthetized using Isoflu rane and sacrificed by cardiac puncture and exsanguination. Blood and tissues (stomach, small intestine, large intestine, spleen, pancreas, liver, kidney and brain) were immediate ly excised for later analyses. Wild type mice from another in house knock out mouse model ( Zip14 / ), with a C57BL6 background, were also used in a separate dietary experiment and lipopolysaccharide (LPS) experiments. For LPS experiments, mice were place d on ZnA or high zinc (ZnS) diets for one week. An intraperitone al (ip ) injection of LPS (2 mg/kg) was administered to the mice. The mice were anesthetized and sacrificed 18 hours post injection. Blood and tissues were collected for later analyses. Pro tocols were approved by the University of Florida Institutional Animal Care and Use Committee. Genotyping Genomic DNA was extracted from mouse tail samples using the p rep GEM tissue kit (ZyGEM). After tips were cut in half with a razor blade and pl aced i n a 0.2 mL PCR tube, 89 L of DNA free water, 10 L of 10x Buffer GOLD, and 1 L of prep GEM enzyme were added to the tube. Samples were vortexed and centrifuged briefly. Next, the samples were placed in a thermal cycler and incubated at 75C for 15 minu tes and 95C for 5 minutes. The DNA containing supernatants were collected and concentrations were measured by optical density at 260 nm (Nanodrop ). The s amples were diluted to and used as templates in the polymerase chain reaction (PCR). Two primer/probe sets were used in a multiplex PCR reaction. The first set was
29 designed to amplify a region of the LacZ gene which was inserted into the knock out genetrap (sense pri ATCAGGATATGTGGCGGATGA TGATTTG TGTAGTCGGTTTATGCA FAM CGCCCACGCGATGGGTAACAG BHQ region that contained a housekeeping gene, UC GTCATCAAGTGAGAAAGACATCC T CATCATGAATTTTGATAAGCCCATT HEX CTCCTGGCTGCCTGGCTGGC BHQ probes in the reaction mixture was 1400 nM and 390 nM, respectively. Master mix and template DNA (100 ng/L ) we re placed into each well on a 96 well plate. The genotyping assay used a StepOnePlus sequence detection (Applied Biosystems) using the comparative C T method for analysis. This method required triplicate wells for the comparative C T calculations performed genotyped samples were used to verify assay data produced by unknown samples. The C T values for the triplicates were averaged, and the C T of each sample was calculated by subtracting the normalizing gene C T (C TNORM ) from the gene of interest C T (C T GOI ). For this genotyping experiment, UC329 was used as the normalizing gene and LacZ was the gene of interest. The calculation for genotypi ng the Zip11 knock out line was as follows: C T = LacZ C T GOI UC329 C T NORM The wild type (WT) always had a large C T ranging from 3 18, when compared to the heterozygous (HET) mice and the homozygous (HOM) null mice. The HOM and HET mice had very l ow or sometimes negative C T values because of the LacZ gene amplifying early i n the cycles. Depending on the previously genotyped controls the HOM null C T could range
30 from 0.6 2.2, while the HET C T values could range from 1.0 1.0. Genotyping r TGTTCTCTGCCCTAGCACCT CACTCCAACCTCCGCAAACTC) that would amplify a genetrap region only found in the HOM null and HET genotypes. Through a series of PCRs, different forward primers were used to find a region in the intron between exons 4 and 5 that would amplify with the genetrap reverse primer. The final concentration of each primer was 0.5 M, and these primers were added to the Phusion High Fidelity PCR master mix with GC buffer (NEB) which was diluted in PCR water to a concentration o f 1X. Each well contained 19 L of ma ster mix and 1 L of template DNA. Reaction mixtures were incubated at 98C for 3 0 seconds, amplified for 35 cycles at 98 C for 1 0s, 5 8C for 15s, and 72C for 30s, and lastly a final extension s tep at 72C for 7 minutes. Four L of 6X Orange DNA loading dye (Thermo) was added to each sample before pl acement into the gel well. Five L 1kb DNA ladder was added to a we ll to help determine the size of the amplified product. No template control samples, using PCR wat er as the template, were also ru n to show there was no contamination. The PCR samples were run on a 1% agarose gel made of agarose, 1X TAE, and ethidium bro mide (EtBr, for visualization) for 45 minutes at 75 volts. UV light was used to visualize the PCR products on the gel, using a FluorChem E Imager (Protein Simple). The Zip11 transcript was also analyzed in the WT and HOM samples. The cDNA samples of seve ral tissues were used in a similar experiment to the genotyping PCR. The primer set used in this experiment was the Zip11 RNA sequences mentioned in Table 2 1. The Phusion master mix protocol mentioned above was the same one used
31 for this experiment. Th e Zip11 forward and reverse primers were diluted to a working solution of 3 M and further diluted in the master mix to a final concentration of 0.5 M. The cycling instructions were also the same as mentioned abo ve for the genotyping PCR. 4 L of 6X Ora nge DNA loading dye was added to each PCR product and these ose gel (agarose, 0.5X TBE, 1 L EtBr) for 45 60 minutes at 75V. The gel was exposed to UV light (FluorChem E Imager) t o visualize the PCR products. Following the same protocol for the Phusion master mix and PCR run method, presence of the genetrap was CAAGGTTACAGCTCCGTGGT GACAGTA TCGGCCTCAGGAAGATCG amplify a region (~ 620 bp ) only in KO and HET mice. The only change to the run method provided previous ly was the annealing temperature being set to 60C. The PCR products were run on a 1% ag aro se gel (agarose, 1X TAE, 1 L EtBr) for 45 minutes at 75V. The gel was exposed to a UV light as mentioned above for visualization. Tissue Collection For this dissertation research several tissues were collected from different parts of the mice. After b lood was collected, tissues (stomach, small intestine, large intestine, spleen, pancreas, liver, kidney and brain) were removed and depending on the later analyses were either placed in RNA later (Ambion), flash frozen with liquid nitrogen or fixed in 10 % formalin (Fisher). Specifically, concerning the GI tract, the stomach, small intestine, cecum, and colon were flushed with ice cold 1X phosphate buffered saline (PBS), containing 1X Halt Protease Inhibitor (Pierce) solution, several times to remove
32 food particles and feces. Pieces of each tissue were carefully cut and placed into the appropriate container for further analysis of RNA and protein. Measurement of Serum and Tissue Zi nc Concentrations Blood was collected by cardiac puncture with a 25 gauge needle (Becton Dickinson). Each blood sample was placed in a CAPIJECT capillary blood collection tube (Terumo) and inverted several times before being placed on ice for one hour. The samples were centrifuged at 2,000 x g for 10 minutes at 4C to allow c omplete separation of the serum from the red blood cells. Serum samples were carefully removed and diluted 1:5 in Milli Q water Several different tissue samples were co llected and digested in 1 2 mL of HNO 3 for 2 hours at 90C. Acid digested samples were diluted in 1:3 Milli Q water Serum and tissue zinc concentrations were measured by flame atomic absorption spectrophotometry (AAS). Values for tissue zinc were normalized to the wet tissue weight. RNA Isolation Tissues were excised from mi ce an d immediately placed in RNA later (Ambion). Tissues were subsequently removed from RNA later and homogenized (Brinkmann Polytron Instruments) in TRI Reagent (Molecular Research Center). The homogenate was allowed to rest at room temperature to allow com plete disassociation. C hloroform ) was then added to 1 mL of the TRI Reagent solution. The solution was shaken for 15 seconds and allowed to incubate for 2 3 minutes at room temperature. The samples were centrifuged at 12,000 x g for 15 minutes at 4C. The upper clear phase was carefully removed and placed in a new labeled tube. A 100% isopropanol solution was added to each tube with the aqueous phase. Mixtures were vortexed and allowed to incubate for 10 minutes at room temperature. The samples were centrifuged
33 at 1 2,000 x g for 10 minutes at 4C. The p ellets formed after centrifugation, were washed two tim es with 1 mL of 75% ethanol and centrifuged at 7,500 x g for 5 minutes at 4C. The ethanol wash was discarded after each spin. After the second spin, the ethan ol was removed and pe llets were allowed to air dry. Once the ethanol was c ompletely evaporated, 30 50 L of PCR water, depending on pellet size, was added to each tube. Tubes were placed in a 37C water bath for 10 minutes to allow pellet to dissolve i n the water. To remove any residual DNA contamination, all RNA samples were treated with TURBO DNase reagents (Ambion) as described by the manufacturer. Total RNA concentrations were measured using a Nanodrop spectrophotometer (Thermo Scientific). For on e step PCR, DNase treated samples were diluted to 10 ng/L The primer and probe sequences are shown in Table 2 1, 2 2, and 2 3. All primers and probes were used at concentrations of 900 nm and 250 nm respectively. Master mix ) and 1 L of dil uted RNA were combined for the PCR reaction. Standards were created using a serial dilution from the stock RNA. R eaction mixtures were incubated at 48C for 30 minutes followed by 95C for 15 minutes and amplification of 45 cycles at 95C for 15 s econds and then 60 C f or 60 seconds. For two step q PCR, RNA samples were first converted to cDNA following the High Capacity cDNA Reverse Transcription protocol (Applied Biosystems). cDNA samples were diluted from the stock at a 1: 15 dilution in PCR water. Standards were created using the stock cDNA by making a serial dilution. Reaction mixtures were incubated at 95C for 20 seconds and amplification of 45 cycles at 95C for 1 second and 60C for 20 seconds. F luorescence was measured
34 with a StepOnePlus Real Time PCR System (Applied Biosystems). Relative quantitation for assays used 18S or TATA binding protein (TBP) mRNA as normalizer s Zinc Transporter Antibodies Peptides listed on Table 2 4 were used to generate polyclonal a ntibodies against ZIP5 and ZIP 11. Each peptide was synthesized ( Biosynthesis ) with an additional N terminal cysteine to facilitate conjugation to maleimide activated keyhole limpet hemocyanin (Pierce). The conjugated peptides were used to produce polyclo nal antibodies in rabbits (Cocalico Biologicals) and IgG fractions were affinity purified (Pierce). Immunoblotting Several tissues were flash frozen in liquid nitrogen immediately after collection. Tissues were triturated, using the Polytron homogenizer, on ice in a Tris Triton buffer (1M Tris HCl, pH 7.4, 1% Triton X 100 (Sigma), 1X Halt Protease Inhibitor). The homogenates were incubated in 4C on a nutator for one hour. After incubation, samples were centrifuged at 12,000 x g for 20 minutes at 4C. The supernatant was collected from tubes, leaving the pellet behind. Protein samples were aliquot ed to reduce the number of freeze/thaw cycles. Total cell lysate protein concentrations were measured by a RC/DC assay kit (BioRad), following the manufactur only deviation to the RC/DC protocol was reducing the amount of reagents used by 1/4. For example, the protocol says to use of s of standards and samples were used. Depending on protein abundance 25 total protein was denatured in a 6X sample loading buffer (35 0 mM Tris, pH 6.8, 600 mM dithiothreitol (DTT), 10 % sodiu m dodecyl sul fate (SDS), 0.0012% Bromophenol blue and 3 0% Glycerol) for 15 minutes at 37C. Dena tured proteins were separated by
35 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS PAGE). Proteins were transferred to a Protran nitrocellulose membrane ( Whatman) overnight at 30 volts. Protein transfer was confirmed by Ponceau Red staining (0.25% Ponceau stain red, 40% Methanol and 15% Glacial acetic acid) Ponceau stained membranes were washed with TBS T (Tris Buffered Saline, containing 0.05% Tween 20) to remove any traces of the Ponceau stain. After washing, membranes were blocked in 5% nonfat dry milk in TBS T for one hour on an orbital shaker at room temperature. Incubation times and primary antibody concentrations were different between the housek eeping protein, tubulin (Abcam) and th e zinc transporter proteins ZIP4 (Thermo Scientific), ZIP5 (in house), and ZIP 11 (ProSci or in house). The ProSci ZIP11 antibody bound to a protein at ~ 25 kD and 70kD. The in house ZIP11 antibody bound to a protein a t 35 kD (the predicted size of the ZIP11 protein). Tubulin was diluted 1:5,000 in 5% nonfat dry milk in TBS T and incubated for one hour at room temperature on an orbital shaker. ZIP4, ZIP5, and ZIP 11 were individually diluted to a final concentration of in 5% non fat dry milk in TBS T. The milk/antibody solution also contained 0.02% sodium azide. The membranes were incubated in this milk/a ntibody solution overnight on a nutator at 4C. To test the specificity of an antibody, primary antibodies were pre incubated with the specific peptides used as the antigen when creating the antibody in the rabbit s After primary antibody incubation, membranes were washed four times with TBS T for five minutes. Next, the membranes were incubated with an anti rabbit secondary antibody (GE Healthcare) for 45 minutes to one hour. Membranes incubated with the tubulin primary antibody were incubated with the secondary antibody conjugated to horseradish peroxidase at a 1:10,000 dilution in 5% nonfat dry milk in TBS T.
36 Membranes incubated with the Zip transporter primary antibodies were incubated with the secondary antibody conjugated to horseradish peroxidase at a 1:5,000 dilution in 5% nonfat dry milk in TBS T. Following secondary antibody incubation, membranes w ere washed two times with TBS T and two times with 1X TBS. Immunoreactivity was visualized by enhanced chemiluminescence (Thermo Scientific) and a FluorChem E Imager. Immunohistochemistry Stomach and colon tissue pieces were collected and placed in 10% f ormalin for 24 hours at room temperature. Tissues were carefully removed from the formalin and placed in cold 1X PBS. Samples were placed in a 4C cooler until ready for paraffin embedding. A mouse tissue array slide (US Biomax) w as also purchased to ex amine ZIP 11 staining in several tissue slices. The same protocol mentioned below was used with the tissue array slide. Slides were deparaffinized by immersing them 2X in 100% xylene for 5 minutes, 3X in 100% ethanol for 3 minutes, 2X in 95% ethanol for 3 minutes, 70% ethanol for 3 minutes, and lastly tap water for 5 minutes. Next, antigen retrieval was performed by immersing slides in a 95C, 10 mM sodium citrate solution (pH 6.0) for 10 12 minutes. After antigen retrieval slides were washed in tap w ater. Two different IHC methods were used during this research, an immunoperoxidase staining technique and immunofluorescence. Immunoperoxidase Staining For this procedure, the Vectastain Elite ABC (Vector Laboratories) pr otocol was followed. B riefly, slides were immersed in 3% hydrogen peroxide diluted in methanol for 10 minutes to block any endogenous peroxidase activity. During incubation periods, slides were placed in a humidified chamber. After a wash, tissues were incubated w ith
37 normal goat bloc king serum provided for 20 30 minutes. Tissues were next incubat ed overnight at 4C with the ZIP 1 1 primary antibody (ProSci, 2.5 g /mL ) diluted in the blocking serum. To test the specificity of the antibody, the primary antibody solution was pre incub ated with the peptide, used as the antigen. Tissues were washed with 1X PBS and incubated with the biotinylated se condary antibody provided for 30 minutes at room temperature. After another wash, tissues were incubated at room temperature with the Vectas tain Elite ABC Reagent for 30 minutes. Tissues were washed to remove ABC reagent, and then incubated with a peroxidase substrate solution (ImmPACT DAB, Vector Laboratories) for 1 3 minutes, depending on tissue expression of ZIP11 The substrate was rin sed off with tap water and to ensure the tissues did not dry out during the last steps, water droplets were carefully placed onto each tissue piece. Lastly, the slides were stained with hematoxylin (Santa Cruz) to visualize the nuclei. Slides were then p assed through the deparaffinization solutions in reverse order, starting with 70% ethanol and ending with 100% xylene, to dehydrate the tissues. After coverslips were mounted over tissue sections with Cytoseal TM 60 (Thermo), staining was viewed with a mic roscope ( Carl Zeiss, Axiovert S100 ). Immunofluorescence As mentioned above, a humidified chamber was used during incubation periods. After deparaffinization, tissues were incubated with a 3% BSA (bovine serum albumin) in TBS T (1X TBS, 0.05% Tween 20 ) blo cking buffer to prevent nonspecific binding and to reduce background interference After blocking, tissues were incubated overnight in 4C with a mixture of the in house Z IP 11 (10 g/mL ) produced from rabbits and H,K ATPase (1:2000, Novus), produced fro m mice, primary antibodies diluted in 3% BSA. Slid es were rinsed before incubation at room temperature with a mixture of secondary
38 antibodies in 3% BSA: donkey, anti rabbit IgG, labeled with AlexaFluor 594 (1:300, Invitrogen) and goat, anti mouse IgG, lab eled with AlexFluor 488 (1:400, Invitrogen). After slides were rinsed, they were incubated at room temperature with DAPI ( 4',6 diamidino 2 phenylindole, 1:200, Molecular Probes) for 30 minutes. Samples were visualized on a Laser Scanning Confocal Fluores cent Microscope (Leica TCS SP2) and a Spinning Disk Confocal Fluorescent Microscope (Olympus IX2 DSU) at the Cell and Tissue Core Facility within the McKnight Brain Institute at the University of Florida Isolation and Analysis of Serum microRNA Serum RNA was isolated from the mice fed either a zinc adequate (ZnA) or a zinc deficient (ZnD) diet the protocol. Synthetic C. elegans miRNA ( Syn cel miR 39 Qiagen) was added to pooled serum samples as a normalizer during analyses (42 ) Total RNA concentrations were measured with the Nanodrop spectrophotometry Reverse transcription (RT) of the miRNA in the total RNA sample was amplified using the miScript II RT Kit (Qiagen), following s protocol. HiSpec buffer, miScript nucleic acid mix, miScript reverse transcriptase mix and RNase free water made up the RT master mix. The rea ction mixture consisted of 11 L of master mix and 9 L (~160 ng) of template RNA. Reaction mixtures were inc ubated for 60 minutes at 37C and then at 95C for 5 minutes to inactivate the transcriptase. The cDNA mixtur es were diluted by adding 200 L o f RNase free water to each 20 L reaction mixture. These diluted mixture s were separated into two 110 L aliquo ts and stored at 20C. Using the mouse s erum and plasma miRNA PCR array (Qiagen) and a SYBR Green miScript PCR kit (Qiagen), real time PCR was performed with the serum miRNA cDNA samples, using the on mixture consisted of 1,375 L of 2X
39 Quantitect S YBR Green PCR master mix, 275 L of 10X mi Script Universal Primer, 1000 L of RNase free water, and 100 L (0.5 1 ng) of diluted template cDNA. The reaction mixture ) was placed into wells of the PCR array p late. Reaction mixtures were incubated at 95C for 15 minutes and amplification of 40 cycles at 94C for 15 seconds, 55C for 30 seconds, and 70C for 30 seconds. A melting curve was generated to ensure that one product was amplified during the reactions Data from PCR runs were analyzed using miScript miRNA software provided by Qiagen. Statistical Analysis Data are presented as the means SD Significant changes between groups were analyzed by unpaired two tail test. S tatistical significance was set to p < 0.05.
40 Table 2 1 Zip Transport er Primer/Probe S equences Transporter Sequence Zip4 Forward Reverse Probe CTCTGCAGCTGGCACCAA CACCAAGTCTGAACGAGAGCTTT FAM CAATCTCCGACAGTCCAAACAGACCCAT BHQ1 Zip5 Forward Reverse Prob e Zip11 Forward Reverse Probe GGGCAGCCTCATGTTTACCA CCACATCAGCCGTCAGGAA FAM CCCTATTGGAGGAGCAGCTAGTGCCC BHQ1 CACTGAGTGGAAGGCATCTTTCT TGAGGTGTTGAAGTTGAGTCTAGTGA FAM TCGAGGCTAACCCCTACTTGTCCCACC BHQ1 3 Table 2 2 ZnT Transporter Primer/Probe S equences Transporter Sequence ZnT4 ZnT5 ZnT6 Forward Reverse Probe Forward Reverse Probe Forward Probe Reverse GCTGAAGCAGAGGAAGGTGAA TCTCCGATCATGAAAAGCAAGTAG FAM CAGGCTGACCATCGCTGCCGT BHQ1 CTGCTCGGCTTTGGTCATG CGGCCATACCCATAGGAGGA TTTGCTGCCCTGATGAGCCGC BHQ1 TCCCAGGACTCAGCAGTATCTTC 3 GCCCCAGCAAGATCAATCAG FAM TGCCCCGCATGAATCCGTTTG BHQ1 3 Table 2 3 Other Primer/Probe S equences Transporter Sequence 18S Forward Revers e Probe AGTCCCTGCCCTTTGTACACA GATCCGAGGGCCTCACTAAAC CGCCCGTCGCTACTACCGATTGG TBP MT Forward Reverse Probe Forward Reverse Probe TCTGCGGTCGCGTCATT GGGTTATCTTCACACACCATGAAA FAM TCTCCGCAGTGCCCAGCATCA BHQ1 GCTGTGCCTGATGTGACGAA AGGAAGACGCTGGGTTGGT AGCGCTGCCACCACGTGTAAATAGTATCG BHQ1 3
41 Table 2 4 ZIP Transporter Antibody Sequences Antibody Sequence ZIP 5 ZIP 11 CRNKRDLGEPNPD CPALMKKSDPRDPTS
42 CHAPTER 3 ZIP11 KNOCK OUT MOUSE MODEL CHARACTERIZATION Introductory Remarks Knock out mice are created by two different strategies, one being a gene targeting method and the other, a gene trapping method. Knock out mouse models are important when studying the functionality of a specific gene. The gene targeting method is also known as homologous recombination. This method introduces a target ing vector into an embryonic stem (ES ) cell that has an identical existing gene sequence present. The vector will insert itself both upstream and downstream of the existing gene, and the nd remove the existing gene and replace it with the artificial DNA sequence, found in the target vector ( 1 ). The artificial DNA is inactive and will cause the gene function to be knocked out. A conditional KO mouse model has also been creat ed to knock ou t gene function in a specific tissue or organ. This model utilizes a Cre lox system which uses the Cre recombinase enzyme and loxP sites in the DNA for a specific deletion in the target tissue. Usually two genetically altered mouse lines are used for thi s method, one line containing loxP sites flanking the target gene, and the second containing the Cre recombinase in a specific tissue. The breeding of these two mice will produce offspring that will inherit both genes. The DNA piece bordered by the loxP sites will be removed and inactivated only in tissues that express the Cre transgene ( 41 ). The second conventional KO strategy is the gene trapping method. This method uses a piece of artificial DNA with a reporter gene to be randomly inserted into the in tron of any gene in an ES cell. The reporter genes within the vectors are used to identify cell lines where the gene has been successfully interrupted. Gene trapping
43 vectors usually have an important piece of the transcriptional component removed, which could be the enhancer, promoter, or polyade nylation (polyA) signal (1, 66 ). The vector will be transcriptionally activated only when inserted into the endogenouse gene (1). A gene trap vector can contain its own promoter and require the endogenous enhance r to drive the transcription of the artificial DNA. The second type of gene trap vector involves a vector containing a splice acceptor sequence, followed by a reporter gene, and ending with a polyA signal. The vector is inserted downstream of the endogen ous promoter, allowing for the expression of the reporter gene fused to the upstream exons. The third gene trap vector involves a vector with its own promoter and a spice donor sequence, inserting itself upstream of the endogenous polyadenylation site ( 1, 66 ). The reporter gene vector will be transcribed and fused to the downstream f the gene. Another concern with the random gene trap insertion method is the possibility of the gene trap inactivating multiple genes or genes coded on the opposite strand also being inactivated by the insertion. Lastly, another disadvantage of using random genetrap vectors is the poss ibility of alternative slicing producing hypomorphic alleles ( 66 ), in which the altered gene products are found to have a decreased level of activity. Several zinc transporter genes have been knocked out in order to study their functionality in the mouse. ZnT1 / and Zip4 / mouse models both produce an embryonic lethal mutation in mice ( 5, 17, 44 ). H eterozygous ZnT1 mice developed normally, but female embryos did not develop properly if the dam was zinc deficient ( 5 ). The ZnT1 KO model revealed that ZnT 1 was an important protein involved in the transport of
44 maternal zinc to the embryo ( 44 ). Though the Zip4 KO model is embryonic lethal, the mutation is not lethal to AE patients that have lost symptoms can be relieved by z inc supplementation ( 44 ). Zn deficiency affected the embryonic development in Zip4 heter o zygotes, while Zn supplementation helped protect the heterozygous embryo, but did not aid in preventing lethality in Zip4 homozygous null mice ( 17 ). An obvious diffe rence between species occurs with the loss of function of ZIP4, with humans having another zinc transport system available to provide zinc to the embryo. The Zip1 KO model does not show any phenotypic differences when zinc levels are adequate; however, d uring ZnD the Zip1 KO negatively affects embryonic development. The double Zip1/Zip3 KO model revealed that both of these proteins are essential for proper embryonic development during times of zinc deficiency ( 15 44 ). After ZnT3 was knocked down in the mouse model, the amount of labile zinc within certain area s of the brain was decreased (44 ). Understanding zinc transporter functionality in the brain is important considering that zinc cannot diffuse across the blood brain barrier. Both the ZnT5 and Zn T7 KO models have re vealed poor growth in mice. U sing a genetrap method to knock out the function of ZnT5 resulted in irregular bone development due to the key involvement of ZnT5 in osteoblast maturation ( 31 44 ) The genetrap method was also used to c reate ZnT7 / mice which resulted in irregular zinc absorption in the gut and a d ecrease in body fat composition ( 30, 44 ). Investigators have also found abnormalities with the bone, teeth, and connective tissue w ith the Zip13 KO mouse model (20, 44 ). A recent investigation by Beker et al, of a Zip14 / mouse model has revealed new information into key roles Zip14 may be playing
45 in intestinal zinc absorption and glycolysis in the liver The Zip14 / mice had an increase in body fat, lower blood glucose, and higher insulin levels when compared to the wild type controls. The Zip14 knock out model also revealed an increase in iron regulated genes, such as DMT1, hepcidin, and ferritin, in the liver (7 ). These key findings elucidate further information into the complex roles Zip14 plays in zinc, iron, and glucose homeostasis. Knock out models are an important aspect of zinc transport study because they provide an in vivo system to analyze the functionality of a specific gene. Not every knockout model will p rovide phenotypic differences; however, when a KO model is made available, each will be able to shed new light on that specific zinc transporter. A KO model could reveal functionality results about Zip11 Therefore, the aim of this chapter was to examin e the phenotypic effects of the Zip11 / mutation and analyze zinc metabolism within this mouse model. Results Understand ing the Zip11 Knockout Model A pGT0lxf genetrap construct (RRJ574) was created by the Bay Genomics MMRRC in the mouse embryonic stem ( ES) cell line 129P2/OlaHsd. The construct of geo reporter galactosidase and neomycin transferase) and a polyadenylation (pA) sequence downstream of the reporter gene. This con struct was randomly inserted into an intron region of the genomic DNA within the ES cell. Confirmation of the genetrap insertion was completed by Bay Genomics. The confirmed clone was sent to UC Davis, where chimeras were generated by injecting the ES ce ll line into C57BL/6 host
46 blastocysts. The chimeras were bred with C57BL/6 mice to generate heterozygous breeders. Working in collaboration with Dr. Fudi Wang ( Zhejiang University China), our lab obtained two pairs of heterozygous breeders from UC Davis The offspring from the heterozygous breeding pairs were genotyped using tail clippings obtained shortly after weaning. Since the genetrap was inserted randomly, conventional genotyping could not be used since primers around the genetrap were not avail able. Originally the genetrap was identified to be located between exons 3 and 5 using the UCSC Genome Browser. Using this computer program, the location of the genetrap was identified on the anti se nse strand of chromosome 11 at 113244853 113650079 T Consortium (IGTC), was blasted against the mRNA sequence of both isoforms of m ouse Zip11, using NCBI blast. The alignment of the blasted sequences stopped at the end of exon 4, which meant the ge netrap was located in the large intron between exons 4 and 5. This intron has an approximate size of 59,000 base pairs which was determined using the blasted sequences and the base pair numbers provided by NCBI. Diagrams of the mouse Zip11 gene, the gene trap, and the transcripts produced by alternative splicing can be found in Figure 3 1 A The positioning of the random genetrap is unfavorable, due to the genetrap insertion in the middle of the gene, rather than towards the first and second exons of the g ene. A genetrap insertion towards the the fusion protein. Another issue arising with this genetrap is the alternative splicing activity that occurs with in the Zip11 gene between exon 5 and 6 formi ng two isoforms. Issues arising with alternative splicing and a randoml y inserted genetrap can produce
47 hypomorphic alleles, where in some instances full functioning transcrip t and the mutant transcript would be produced (Fi gure 3 1 B ). Genotyping the offspring of the heterozygous breeding pairs was completed using a multiplex, comparative C T method. The protocol, followed for genotyping these mice, is discussed in the methods and materials section of this dissertation. The data collected from the comparative C T C T and averages the sample triplicates into a C T mean. The LacZ gene is found within the genetrap and is considered the gene of interest. An ultraconserved gene, U C329 was used as the no rmalizing gene for genotyping. A calculation (LacZ C T GOI UC329 C T NORM ) was used to determine the C T of each sample. The wild type samples did not express LacZ, so the C T for the LacZ was usually 35 45, which indicates no LacZ is present for amplification. For samples with no LacZ amplification, the cycle number, 45, was arbitrarily set as the LacZ C T After the C T means were determined for the unknown samples, they were sorted from smallest to largest, along with the previo usly genotyped samples for WT, HET, and HOM null The values were also analyzed as a graph (Figure 3 2 A ) to better represent the data. The WT samples had large C T means ranging from 3 18, when compared to the HET or HOM null mice. The HOM null and HET mice had very low or sometimes negative C T values because of t he LacZ gene amplifying early in the cycles. Depending on the previously genotyped controls the HOM C T could range from 0.6 2.2, while the HET C T values could range from 1.0 + 1.0. The ranges overlap due to the control samples determining which samples were considered HET or HOM null. The LacZ gene was either present
48 as one copy (HET) or two copies (HOM null ), so the C T mean value s for the HET mice were usually half of the value for HOM null mice. To ensure that the comparative C T method was providing accurate genotypes, I also ran PCRs with a primer set that would amplify a region in the genetrap DNA only found in the HOM null and HET mice. I needed to run these PCRs because our RNA and protein data from the KO model (to be discussed below) were not showing the results t hat would be expected from a KO. The PCR products were run on an agarose gel and amplified regions were visualized using UV light. The amplified regions appeare d as a bright band ( ~2.2 kb ) for the HOM samples and a fairly faint almost undetectable band for the HET samples. The WT and no template control (NTC) samples did not have any amplification (Figure 3 2 B ). Another confirmation used to analyze the HOM null mice was using cDNA from several tissues of both the WT and KO mice. For this experiment, I used a primer set that would amplify a region of the Zip11 transcript region was approximate ensure this region had been removed during the genetrap splicing. Unfortunately, the results from these PCR experiments revealed that the Zip11 transcript was still being amplified in all tissue s evaluated from the KO model (Figure 3 3 A ). After seeing the unexpected Zip11 amplification in all tissues from WT or KO mice, another primer set was used to amplify a region (~620 bp) in the tissue cDNA from exon gal portion of the genetrap. The tissue samples used for analysis were from a male and female mouse. This amplified region should only appear in the KO tissues, which was confirmed in Figure 3 3 B The faint band appearing in the WT female cec um tissue was
49 not expected, but the stomach and liver tissues were from the same female mouse and both WT samples show no amplification. The KO mice, whether male or female, did not have any phenotypic differences compared to the WT mice. Mouse weights we re consistent among the genotypes. No apparent difference wa s seen with percentages of KO versus WT offspring. Since this genetrap seems to not have knocked out the Zip11 gene, a new knock out line would need to be established to evaluate phenotypes in m ice without Zip 11 A Significant Decrease in the Zip11 RNA Expression in KO Mice Once mice were genotyped and confirmed to be HOM null mice, we started breeding a KO line. WT mice that were offspring of the HET breeders were separated and bred together to have a WT line to use as controls for experiments. Tissue from WT and KO mice was extracted and processed to isolate RNA. PCR results revealed a 50 80% decrease in most tissues (Figure 3 4 A ). The C T values for the mice ranged from approximately 28 30 The WT mice had C T values on the lower side of the range, while the KO mice amplified around cycle 30. The C T differences revealed about a 2 fold difference between the WT and KO mice. The duodenum did not have a significant decrease due to the combin ation of male and female data. This tissue was the only tissue to have a significant difference in Zip11 mRNA expression between male and female mice. The stomach tissue of WT and KO mice was divided into three regions ( non glandular body, and pylorus) and the Zip11 mRNA expression of each region was analyzed. The three regions were looked at individually since they each have different cell types and funct ion. As with the other tissue Zip11 mRNA expression, a 50% decrease in Zip11 expression was seen i n the KO tissue versus WT tissue in the
50 non glandular and body sections, but not the pylorus section (Figure 3 4 B ). If the Zip11 gene had been properly knocked out, the results would have shown us little to no Zip11 RNA expression in all tissues. These r esults revealed that the genetrap was knocking down some of the Zip11 gene transcripts with mutant transcripts but full functioning transcripts were still being produced with the genetrap present. With the reduction in mRNA expression but not a complete loss of expression, the genetrap represents a hypomorphic rather than a null allele ( 66, 75 ). No Change in ZIP11 Protein Expression in KO Mice After seeing the significant reduction in the expression of Zip11 mRNA in the KO mice compared to the WT mice, a d ecrease in the protein should be the expected result. E xpression of ZIP1 1 based on western analysis data, did not consistently show a decrease in any of the tissues evaluat ed (Figure 3 5). The ProSci ZIP 11 antibody bound to a protein expressed at 25 kD, and no change in the ZIP11 expression was seen within the stomach or cecum (Figure 3 5A) ZIP11 expression was also visualized using an in house ZIP11 antibody. This antibody bound to a protein with a size of 35 kD. The commercial ZIP11 antibody (Pro Sc i) bound to a protein at a size lower than the expected 35 kD size. This Pro Sci antibody was used at the beginning of this dissertation research, but after issues with the KO mouse model, an in house ZIP11 antibody was isolated and purified and used for the remaining research sections. The colon tissue appeared to have lower ZIP11 expression, but the expression was inconsistent (Figure 3 5B) With the ZIP11 expression prominent in all protein samples, the results again indicate the genetrap is producing hypomorphic alleles ( 75 ). These hypomorphic al leles can produce both the full functioning Zip11 transcript and the
51 truncated fusion transcript, which would translate into a functioning ZIP11 protein and the non functional fusion protein respectively Di etary Zinc Did Not Affect the Functionality of KO Mice A dietary study with ZnA, ZnD, and ZnS diets was set up to analyze if varying zinc concentrations in diets would affect the functionality of the KO mice. Each dietary group had 9 10 mice. The group s were broken down further by genotype (WT and KO n = 4 5) and sex (male n = 2 3 and female n = 1 2). The mice were acclimated in hanging wire cages for 1 week and changed to the experimental diets for 3 weeks. Mouse weight & food intake fluctuated as expected ( 46, 53 ). No phenotypic differences were seen between the genotypes on each diet. Female mice seem ed to fluctuate with weight gain/loss more than the male mice, but no significant changes were noted To confirm the efficiency of the dietary study plasma samples were collected from the blood samples. No significant changes in the plasma zinc concentrations were seen between the KO and WT samples, so data from the WT and KO plasma zinc concentrations were combined to analyze concentrations b etween diets. The plasma samples from ZnD samples showed a significant decrease in zinc concentration while the ZnS diet resulted in no change compared to the ZnA which has been seen previously in zinc restricted dietary studies ( 52 ) (Figure 3 6 A ). Sev eral tissue samples were ex cised to analyze the tissue zinc concentrations. The KO tissue samples did not have significant zinc concentration differences when comparing the different diet effects (Figure 3 6 B ). The duodenum was the only tissue to show a significant difference in zinc concentration during ZnD. No significant changes were seen between th e WT and KO samples; however the stomach, colon and liver of KO mice samples had a trend of
52 lower zinc concentrations when compared to the WT from mice th at were on the ZnA diet. The KO stomach and colon tissues were analyzed to determine further the effects of zinc depletion and supplementation on zinc related gene transcripts. The zinc scavenger metallothionein (MT) can differ in expression depending o n available metals in tissue and blood. MT expression has been previously described to decrease during per iods of zinc deficiency, while expression increases during periods of high zinc ( 12 59 ). T he KO stomach and colon tissue MT expression decreased si gnificantly during ZnD, which is another confirmation that the deficient diet worked properly (Figure 3 7) The MT expression increased noticeably in ZnS samples when compared to the ZnA samples in the colon; however the change was not found to be signif icant. The stomach tissue also showed an increase in MT during ZnS but was not significant. When comparing the KO relative MT mRNA data to the WT data, the colon MT values had no significant changes noted between the KO and WT samples; however, the knock out stomach ZnD samples appeared to respond better than the WT ZnD samples with a significa nt decrease in relative MT mRNA. This significant decrease of MT would suggest a lower level of certain metals (cadmium, copper, etc) in the stomach; however, zinc concentration did not decrease in the stomach during zinc deficiency. Other metal concentrations such as copper or cadmium, may be affected in the stomach during zinc deficiency in this KO mouse model. Zip4 expression in the colon also responded as prev iously described ( 46 ), by increasing significantly du ring ZnD and having no change during ZnS when compared to ZnA (Figure 3 7 B ). Expression of Zip11 mRNA in the stomach and colon of KO mice
53 was unresponsive to ZnD, while ZnS caused an increase in expressi on with in the colon expressing a significantly higher expression in the ZnS group versus the ZnA group (Figure 3 7 B ). Together these results proved that the Zip11 genetrap did not knock out the gene correctly, creating a hypomorph allele. Since the gene was not knocked out, KO data from the dietary study did not appear to have any significant changes when compared to WT data. Discussion The usage of a random genetrap insertion in the mouse model used for this dissertation research proved to be unfavorabl e, by creating a hypomorphic allele. Hypomorphic allele production by a genetrap has been previously discussed ( 66, 75 ). Br iefly, since the genetrap vector inserts into an intron of the gene, alternative splicing can be affected, resulting in altered lev els of the gene to be knocked out. In the case of the Zip11 knock out model, the mice were genotyped as null mice ( Zip11 / ) because they had two copies of the LacZ gene; however, when RNA was transcribed from the DNA, alternative splicing may have create d both wild type and knock out transcripts resultin g in the failure to observe the Zip11 transcript and protein to be knocked out in the null samples Some animals that have a hypomorphic allele still reveal differences between the knock out and wild typ e animals, such as in the research by Wilson et al (75) Even though their genetrap did not knock out the activity of i nositol 1,3,4 trisphosphate 5/6 kinase (ITPK1), expression of the protein was decreased when co mpared to the WT or HET mice. Mice with the hypomorphic allele had an increase of neural tube defect s when compared to the WT ( 75 ), which meant that the decrease in the enzyme affected the KO mouse even though the gene was no t completely knocked
54 out. The knock out model unfortunately did not sh ow any signs of being affected by a partial knocked out allele. The breeders produce offspring at the same rate as the WT, along with the KO mice having simila r weights and eating patterns. The remaining Zip and ZnT transporters were also analyzed to det ermine if mRNA expression levels were different between the WT and KD animals. Of the other thirteen Zip transporters, only Zip4 mRNA showed an increase expression in the KD animal when compared to WT levels in the murine stomach (data not shown) An inc rease in Zip1 mRNA expression in KD samples compared to WT was only seen in the murine colon (data not shown). The ZnT transporter mRNA expression did not show any changes when comparing KD and WT samples (data not shown). When knock out models are used i n research, the hope is that the genetrap worked correctly. The results from the comparative C T genotyping were consistently providing evidence of WT, HET, and HOM null mice. The transcript results revealed a 50 80% decrease in Zip11 mRNA, which was the first evidence that the genetrap did not work properly. Since the genetrap is suppose to produce a fusion transcript ( 66 ), Zip11 transcript should not have been detected in the HOM null mice. The primer/probe set used for Zip11 was specifically designed to be located at the end of the transcript in exon 10. The genetrap should have produced a fusion transcript that only contained exons 1 4. Using commercial ZIP 11 antibodies an d a previously made in house ZIP 11 antibody, the protein data never consiste ntly showed that the ZIP11 activity was reduced or lost. Once a new ZIP 11 antibody was made from a peptide, specifically designed to be located after the genetrap, the hope was to see protein results consistent with the transcript results. The HOM null m ice only showed a slight decrease
55 in ZIP11 protein in the colon, which was not consistent with the stomach protein expression results. Since the transcript showed decreased expression and the protein had varying expression among tissues, the genetrap was considered a hypomorph ( 66, 75 ). As other research with KO mice has shown ( 7, 17 ), the genetrap should cause a complete loss of Zip11 transcript and protein expression; however, since this particular KO model used a random genetrap insertion, the unfavora ble outcomes that could happen, did happen. The purpose of this aim was to examine the phenotypic effects of the Zip11 / mutation and analyze zinc metabolism within this mouse model, and the results concluded that the Zip11 gene was not knocked out prope rly and the mice did not show any differing pheno types or gene activity. R esearch with the KO mice was discontinued, and further research was only completed with the WT mice.
56 Figure 3 1. Diagrams of the mouse Zip11 gene, genetrap and alternative slicing. (A) The Zip11 gene including the introns and exons, along with an expanded view of the random genetrap insertion between exons 4 and 5. Alternative mRNA splicing of the Zip11 genetrap produces a full functioning WT transcript ( B) and a Zip11 The represents splicing events. C B A
57 A B F igure 3 2. Genotyping tail sample DNA (A) The sample data are graphically T means. The darkest bar s represent the HOM null samples, the gray bars the HET samples, and the light gray the WT samples. Values shown are means SE (n=3 wells per sample). (B) The amplified region of the Zip11 g ene with the genetrap present. Bands highlighted by the white box represent the 2.2 kb reg ion present in the HOM Null DNA. This amplified region is not pr esent in the WT and NTC samples Homozygous Null Heterozygous Wild type HOM Null HOM Null HOM Null HOM Null WT HET 2.2 kb NTC
58 Figure 3 3. Confirming genotypes using diffe rent primer sets (A) The 98 bp band represents a region in the Zip11 mRNA in exon 10. Both the WT and KO samples revealed Zip11 tran script in all tissues presented, though the KO samples appear ed to have less Zip11 expression (B) The region represente d by the band at ~620 bp was amplified using primer sequences specific for the KO samples. The reverse primer could only be found in the gal region of the genetrap vector and the forward primer was made from the BayGenomics sequence tag. Cecum Stomach Cecum Kidney Duodenum Wild type Knock out Stomach Cecum Kidney Duodenum 98 bp 98 bp 620 bp Stomach NTC WT KO WT KO WT KO 620 bp Cecum Liver WT KO W T KO WT KO A B
59 Figure 3 4. Knock out Zip11 transcript expression compared to wild type The tissues were extracted an d used for q PCR analysis of Zip11 expression (A) E xpression of KO compared to WT Zip11 mRNA in several tissues. ( B) Zip11 mRNA expression of KO compared to WT in each region of the mouse stomach. Values were normalized to 18S and WT levels were set to 1. Data are expressed as SD (*=P<0.05, **=P<0.01, ***=P<0.001) (n=4 for each tissue sample). ** B ** ** ** ** ** ** *** A
60 A B Figu re 3 5. Wild type versus knock out ZIP11 tissue d istribution. W estern blo t analysis of ZIP11 expression T ubulin wa s used a loading control (55k D). (A) The presence of ZIP 11 (~25kD) in the stomach and cecum was visualized with the commercial ProSci antibody. (B) The presence of ZIP 11 (35kD) in the stomach and colon was visualized with an in house antibody. Female Male WT KO WT KO Zip11 WT KO WT KO M ale Tubulin Stomach Zip11 Tubulin Cecum Stomach Zip11 Tubulin Zip11 Tubulin Colon M M F M M F WT KO
61 Figure 3 6. Plasma and tissue zinc concentrations of KO mice on dietary restriction. Both plasma and tissue zinc concentrations were measured by AAS. Male and female mice were included in these analyses. (A) Measures of plasma zinc indicating the effe ctiveness of the diet for zinc depletion (n= 9 10) Plasm a was diluted with Milli Q water. (B) Ti ssues were digested with HNO 3 and diluted with Milli Q water. Data are expressed as SD (*=P<0.0 5) (n= 3 4) A B
62 Figure 3 7. Effects of zinc depletion and supplementation on zinc related gene transcripts in the stomach and colon of the KO mice. Tissues were extracted and used for qPCR analysis. Values are normalized to 18S and ZnA levels were set to 1. Data are expressed as SD (*=P<0.05, *=P<0.0 1) (n=3 5) MT, Zip4, and Zip11 transcript expression in the KO stomach (A) and colon (B). A B ** ** **
63 CHAPTER 4 EFFECT OF DIETARY ZINC ON EXPRESSION OF ZIP11 AND OTHER ZINC TRANSPORTERS W ITHIN THE GASTROINTESTINAL TRACT Introductory Remarks The effect of die tary zinc on zinc transporter expression has been studied in rodents for years ( 18 25 43 45 ). Zinc deficiency has been show to affect the regulation of Zip 4 ( 18, 46, 73 ), Zip1 and Zip3 ( 15 ), Zip10 (43), ZnT2 ( 45 ), and several other zinc transporters. Zinc repletion has also been shown to affect the regulation of some zinc transporters, such as Zip10 ( 43 ) and Zip5 ( 16, 71 ). Metallothioneins (MTs) are metal binding proteins that play roles in zinc homeostasis and protection against heavy metal damage MTs can also be indicators for the effectiveness of a zinc deficient diet because in most tissues MT expression will be down regulated during ZnD The response of MT to zinc repletion or supplementation is the opposite of the deficient response. When the re is an excess of zinc in the system, M T will be up regulated by transcription factors that respond to zinc status ( 38, 70 ). With so many zinc transporters and other genes influenced by zinc status, studying the expression of Z ip11 during dietary restric tion i s an important factor when characterizing this novel Zip transporter. No structure, function, or regulatory information is available for the zinc transporter Zip11 (Slc39a11). Zip11 is part of the gufA subfamily of Zip transporters ( 44 ). T hrough a murine Gene Atlas (BioGPS) array analysis Zip11 mRNA was found to be highly expressed in the stomach, large intestine, pancreas, and thioglycollate stimulated peritoneal macrophages in mice. The human Gene Atlas array did not have any GI tract data, bu t revealed high expression in the pancreas and kidney. The Zip11
64 gene is well conserved across several species. The mouse Zip11 gene in found on the antisense strand of chromosome 11, while the human Zip11 gene is found on the antisense strand of chromos ome 17. Since the Zip11 expression pattern revealed it to be highly expressed in the stomach and large intestine, the influence of dietary zinc intake needed to be conducted. The dietary studies in this section involved mice being placed on ZnA, ZnD, and ZnS diets for 2 weeks, or ZnD for 2 weeks, followed by 24 hours of the ZnS diet to replete zinc (ZnR) in some of the ZnD mice With the serum collected from this dietary study, another assay was completed to examine murine serum miRNAs. miRNAs are small non protein coding sequences that target complementary sequences found within the untranslated region (UTR) miRNAs have been found throughout different species, including plants, viruses, and animals (6 ). miRNAs have been shown to strongly influenc e the posttranscriptional regulation of gene expression. One miRNA can target hundreds of mRNA sequences, and vice versa a single mRNA sequence can be the target to several miRNAs. The ability for a single miRNA to regulate several mRNAs can develop a co mplex network of gene expression, which can in turn alter biological functions, such as cell mediated death and proliferation, signaling cascades, and immune response ( 4 ). Different stages of cancer have been shown to have serum miRNAs up or down regulat ed, which has made these small sequences important in biomedical research, especially the fields looking into targeted therapeutic approaches ( 76 ). Since no information can be found on Zip11 the aim of this chapter was to characterize Zip11 expression wit hin gastrointestinal tract, particularly the stomach and
65 colon, and also examine a few select other zinc transporters in these tissues during dietary zinc studies. Results Features of the Zip11 Gene and the Predicted ZIP11 Topology The murine Zip11 gene an d coding sequence (CDS) information was obtained from the NCBI website (Figure 4 1). The gene consists of ten exons, nine of which are translated from the coding sequence. The Zip11 gene has two known isoforms with 21 base pairs removed during alternativ e slicing from the end of exon 5, forming isoform 2. Most ZIP protein structures have eight transmembrane domains, but the predicted protein structure of ZIP11 has six transmembrane domains with the N terminus located in the extracellular surface. A hist idine rich loop is usually found in most Zip transporter proteins, but this area is no t found in the ZIP 11 protein. Isoform differences were analyzed by blasting Slc39a11 variant 1 and 2. The predicted size of the ZIP11 protein is 35 kD. The predicted t opology of ZIP11 was analyzed using several websites that use specific transmembrane prediction sof tware. Combining all of the results, my analysis predicts ZIP11 as having six transmembrane domain s. This topology image was created with a depiction of wh ere the variable region of the isof orms (red box), the in house ZIP 11 antibody sequence (green box), and the ProSci ZIP 11 antibody sequence (blue box) are approximately located. Zip11 RNA and Protein Expression Tissue Distribution Since there are no data a vailable on Zip11 besides the Gene Atlas array, the Zip11 expression in C57BL/6 mice was first examined under n ormal dietary conditions to ana lyze the tissue distribution The Zip11 transcript was found to be highly expressed in murine stomach, cecum and colon (Figure 4 2 A). Figure 4 2A shows
66 Zip11 expression within several tissues, as the relative expression of Zip11 mRNA compared to 18S mRNA. The stomach was separated into three regions ( non glandular body, and pylorus), and Zip11 expression was exam i ned in these regions (Figure 4 2 B). The body and pyloric expression values were compared to the non glandular region values due to th is region having the lowest expression in the stomach. Three to four male samples were used for each gastric region. Th is Zip11 mRNA expression pattern show ed that the highest expression wa s in the glandular regions of the stomach. The in house ZIP11 antibody was used for western blot analysis with ZIP11 visualized at 35 kD the predicted size of ZIP11 (Figure 4 3 A). Th e Ponceau staining of the membrane was used as the loading control. The protein data revealed similar results to the transcript data with the stoma ch and large intestine having the high est expression of ZIP11 protein. In t he small intestine ZIP11 protein expression followed the mRNA with the duodenum presenting the lowest ZIP11 expression and expression increasing through the jejunum to the ileum. qPCR data for kidney tissue revealed much lower Zip11 mRNA expression when compared to the stomach or large intestine, but protein data revealed strong ZIP11 expression The pancreas also showed high expression, while the spleen and brain did not have any ZIP11 protein. A mouse tissue array was purchased to enable visualization of ZIP11 protein in several tis sues (Figure 4 3B). The stomach and colon tissue samples on the slide did not provide a good representation of the tissue, so the figure includes two tissues from C57BL/6 male mice. The ProSci ZIP11 antibody was used for detection of the protein in the s tomach, colon, and the entire tissue array The tissue array images were zoomed to 63 0 X, while the
67 stomach and colon images are at 10 0 X. The immunoperoxidase staining revealed ZIP11 (brown stain) expression in the stomach, small intestine, colon, pancreas liver, and kidney. The nuclei of the different cell types are stained in blue. A small amount of staining was seen in the cerebellum and skeletal muscle, but at levels much lower than the other tissues. The spleen, as with the western blot analysis, s howed no expression of ZIP11. The stomach reveals staining throughout the gastric cells, and the small intestine has staining throughout the villi with a stronger concentration at the apical membrane of the enterocytes. Biomax could not provide a verific ation of which small intestine section was used on the slide, but with the strong staining, the slice was most likely from the distal region. The staining of ZIP11 in the stomach and colon slices on the tissue array showed similar results to my tissue sec tions, but the slices used were not handled carefully while making the array slide and appeared to be torn The colonic epithelial cells appear to have the most ZIP11 staining, while the pancreas and liver seem to have staining throughout the cytoplasm. The kidney section was a slice of the medulla region with staining throughout the cells making up the collecting ducts. The staining in the brain section is in the white matter, with no staining in the granule cell layer Tissue Zinc Concentrations of Stomach, Intestines, and Colon Tissues were collected from mice that had been on a ZnA, ZnD, or ZnS diet for 3 weeks. T issue Zn concentrations were measured by AAS (Figure 4 4 ). A trend of lower tissue zinc during ZnD was recorded in the stomach, duodenu m, colon and liver. Only the colon of the ZnD mice had a significant decrease in colonic zinc concentration compared to those mice fed the ZnA diet The stomach and duodenum both revealed
68 an increase in tissue zinc concentration during zinc supplementat ion w hen compared to the ZnD samples The ZnS increase was not significantly higher than the ZnA, so the se result s show that ZnS does not result in significantly more zinc in the stomach and duodenum when compared to ZnA. Zip11 m RNA Decreases D uring Zinc Deficiency The effectiveness of the ZnD diet to decrease serum zinc or the ZnR diet to effectively return serum zinc levels back to the concentrations of that found in the ZnA samples was examined by measuring the zinc concentration by AAS (Figure 4 5 A). Male mice were placed into the ZnA (n=5) or ZnD (n=10) groups. After 2 weeks on the ZnD diet, five ZnD mice were changed to a high zinc diet for 24 hours (ZnR) In t he ZnR group, the serum zinc increased back to ZnA levels and actually increased signific antly above the ZnA serum Zn concentrations (Figure 4 5A ). MT and Zip11 mRNA expression throughout of the GI tract was analyzed through q PCR of cDNA samples transcribed from the isolated RNA of each tissue Only the stomach and ileum tissues revealed a si gnificant down regulation in MT during ZnD; however, each of the GI tissues analyzed showed a significant up regulation of MT due to ZnR (Figure 4 5 B). In s tomach there was a significant decrease in Zip11 transcript in the ZnD and ZnR group when compare d to the ZnA samples (Figure 4 6 ). With no change between the ZnD and ZnR Zip11 mRNA expression in the stomach, the 24 hour period of ZnR with a diet high in zinc appears to not be a long enough period to recover the Zip11 mRNA levels from the decrease prod uced by ZnD conditions Of the three segments of t he small intestine, the ZnR duodenum samples were the only samples to show a change in Zip11 mRNA expression with a significant decrease in expression.
69 The Zip11 transcript expression in the cecum signifi cantly decreased in both the ZnD and ZnR samples when compared to ZnA samples. The colonic Zip11 mRNA expression was unaffected by ZnD or ZnR. ZIP11 Protein Expres sion Is Variable Among Tissues D uring Zinc Deficiency Using the in house ZIP11 antibody, th e ZIP11 protein was visualized at 35 kD the predicted size of ZIP11 ZIP11 protein expression in the stomach was found to decrease slightly in ZnD samples, and the expression returned to ZnA levels after zinc repletio n (Figure 4 7 A). Relative densitomet ry was used to compare ZIP11 expression data from the blot and ZIP11 was normalized to tubulin (Figure 4 7B) A peptide competition assay provided evidence of the specificity of the in house ZIP 11 antibody. ZIP11 was visualized at 35 kD, but when t he pe ptide, used to make the ZIP 11 antibody, was first incubated with the Z IP 11 antibody, the antibody could no longer b ind to the 35 kD protein, so a band was not visualized (Figure 4 7C). ZIP11 protein expression in the colon was found to have no chan ge duri ng ZnD or ZnR (Figure 4 8A ), which follows the mRNA data that also showed no change at the transcript level. Even though there was n ot a significant response of ZIP 11 to the dietary regimens there was a trend of ZIP11 to increase du ring ZnD and ZnR in th e colon. Relative densitometry was used to compare data provided in the immunoblot, and ZIP11 was normalized to tubulin (Figure 4 8B). A peptide competition assay was also completed with the colonic immunoblot to assess the specificity o f the Z IP 11 in ho use antibody. As w ith the stomach blots, the 35 kD ZIP11 band was not visualized when the Z IP 11 antibody was pre incubated with the peptide (Figure 4 8C). ZIP11 was also analyzed with western blotting for the cecum (Figure 4 9 A) during ZnA, ZnD, and ZnR conditions and the three sections of the small
70 intestine (duodenum, jejunum, and ileum) tissues (Figure 4 9 B) during ZnA and Zn D conditions The cecum ZIP11 expression did not change during zinc restriction or repletion No significant changes in ZIP11 e xpression within the small intestine were seen. The small intestine ZIP11 expression, however, was unusual when comparing this result to the tissue d istribution blot mentioned above. The d uodenum and jejunum should have had lower ZIP11 expression than t he ileum as seen previously in the ZIP11 protein distribution blot but all three tissue sections showed strong ZIP11 bands when comparing ZnA and ZnD samples. ZIP11 expression in the colon and stomach were also visualized using immunoperoxidase staining. The tissue samples were either incubated with the in house ZIP 11 antibody or the antibody plus the peptide to test specificity of the antibody. The results did not show any difference in staining between the ZnA and ZnD stomach samples (Figure 4 10 A), but there seems to be a difference in ZIP11 localization in the gastric mucosa. The darker staining indicating the ZIP11 protein, in the lower regions of the gastric pits seems to partially migrate to the luminal side of the gastric mucosa during ZnD. Gast ric ZIP 11 IHC staining, from another dietary study that included a zin c repletion stage, showed similar patterns to the western blot data (Figure 4 10 B ) The strong ZIP 11 staining s een in the ZnA stomach samples decreases in the ZnD samples and the stain ing appears to return in the ZnR group similar to the staining seen in ZnA samples The similar ZIP 11 migration pattern is also seen among the stomach samples with a darker staining in the lower gastric pits during ZnA, migrating to the luminal gastric s urface during ZnD. The colon revealed distinct staining of ZIP11 within the finger like projections, known as colonic crypts (Figure 4 11 A). The staining appears to
71 increase during ZnD throughout the crypts, while the staining in the ZnR samples appears to increase more than the ZnA samples, but still less than the ZnD samples. The staining at the lumi nal surface suggest s that ZIP 11 is expressed in the epithelial cells of the colon. The cecum ZIP11 staining appears to follow the Zip11 mRNA data presente d previously. The strong staining seen in the ZnA cecum samples decreases dramatically in the ZnD, and the 24 hour ZnR does not seem like enough time to result in up regulation of ZIP11 in the cecum (Figure 4 11 B) Like the stomach ZIP11 staining, the ce cum ZIP11 staining appears throughout the cytoplasm of the epithelial cells, but does not have the distinct staining pattern seen at the surface of the colonic epithelial cells. ZIP11 Localizes to the Nuclei in the Gastric Tissue Since ZIP11 is highly ex pressed in the gastric mucosa, the next step was to see if ZIP11 was expressed on the membrane of the parietal cell, the acid secreting cell of the stomach using immunofluorescence Using H,K ATPase as a membrane marker for parietal cells, I found that i n ZnA tissues, ZIP11 was not co localized with the p arietal cell marker; rather it was co localized with DAPI, the nuclei marker (Figure 4 12 ). The white arrows indicate the co localization of the DAPI and ZIP11 which appear as a pink color (Figure 4 12B) Extensive red fluorescent ZIP11 staining also appears to be located in the cytoplasm of the chief cells, found in the lower portions of the gastric glands (yellow arrows) (Figure 4 12 A ). ZIP11 Localizes to the Colonic Epithelial Cells After seeing the high expression of ZIP11 within the colon, the next step was to see if ZIP11 was localized to the membrane of colonic epithelial cells catenin
72 antibody was used as a marker for the plasma membrane of colonic epithelial cells. Colonic tissue samples reveal ed a co localization of ZIP11 to the colonic epithelial cells with fluorescence seen at the membranes and throughout the epithelial c ells (Figure 4 13 A ). Samples also reveal ed ZIP11 (red) to be localized to epithelial cell nuclei (white arrows) (Figure 4 13B) Co localization of ZIP11 with the nuclei marker (DAPI) was confirmed in other images not shown. ZnD colonic samples also show ed an increase in ZIP11 flu oresc ence (Figure 4 13A) which indicate s an up regulation of ZIP11 during zinc deficiency in the colon as seen in the immunoperoxidase staining (Figure 4 11). ZIP4 Increases in the Colon D uring Zinc Deficiency In a pilot study lo oking at Zip11 expression in the stomach, Zip4 mRNA significantly increased in the stomach during ZnD (data not shown); however, in the latest dietary study Zip4 did not increase significantly during zinc deficiency in the stomach ( Fig ure 4 14 A). This res ult was unusual because during ZnR Zip4 mRNA expression decreased significantly wh en compared to ZnA samples A significant increase in colonic Zip4 transcript s was found in the ZnD samples compared to ZnA samples, and with ZnR Zip4 expression significa ntly decreased to below ZnA levels (Figure 4 14 A ). Zip4 mRNA expression in the small intestine was analyzed to verify the up regulation expected during zinc deficiency. As confirmed by this positive control, t he duodenum, jejunum, and ileum all exhibited a significant up regulation of Zip4 during ZnD, and the 24 hour zinc repletion attenuate d the Zip4 response. The cecum Zip4 mRNA response was similar to that seen in the stomach, with the Zip4 expression only responding in the ZnR samples.
73 ZIP4 protein e xpression in the stomach did not exhibit any changes during zinc deficiency or repletion (Figure 4 14 B). I did not expect to see significant changes in ZIP4 since the transcript levels did not change in ZnD samples and only had a minor decrease in ZnR sam ples. The ZIP4 protein expression in the colon increased significantly due to zinc deficiency (F igure 4 14 C ) Relative densitometry was used to compare immunoblot data (Figure 4 14D). Colonic ZIP4 expression was normalized to tubulin. Zip5 Increase s in the S tomach and Colon D uring Zinc Supplementation The results obtained from this experiment reveal ed that in the stomach a nd colon of mice fed a zinc supplemental diet, Zip5 transcript is significantly up regulated in the ZnS groups when compare d to the Zn A gr oups (Figure 4 15 A ). Zip5 expression throughout the GI tract was also examined during zinc repletion. The 24 hour zinc repletion did not result in any changes in the Zip5 expression, except in the cecum. The Zip5 mRNA response in the cecum was unexp ected but showed a significant down regulation during ZnD and a return to ZnA expression le vels, following Zn R (Figure 4 15 B). ZIP5 protein expression i n the murine stomach revealed a significant increase in both the ZnD and ZnR samples (Figure 4 15 C). T he c olonic ZIP5 expression showed a significant increase only in the ZnR samples (Figure 4 15 D). Both the gastric and colonic immunoblot data is represented graphically by relative densitometry with ZIP5 being normalized to tubulin (Figure 4 15E and F ). S everal microRNAs Increase D uring Zinc Deficiency Serum collected from several mice after two wee ks on a zinc deficient diet was analyzed for expression of zinc responsive miRNAs. The miScript miRNA PCR array
74 protocol developed by Qiagen was used to isolat e and analyze serum miRNA. Among the 84 miRNAs measured, 19 had a fold change of greater than 2 when comparing the ZnD and ZnA samples. A scatter plot provided a graphical representation of the up regulated miRNAs dur ing zinc deficiency (F igure 4 16 A). Among the 19 miRNAs up regulated, 13 were of key interest due to the gastrointestinal cancers each were associated with. These 13 miRNAs were shown graphically as the fold change in the ZnD samples over the ZnA samples (Figure 4 16 B). The miRNAs that wer e of interest to this study were miR 122 5p, miR 17 3p, miR 192 5p, miR 21a 5p, miR 221 3p, miR 29a 3p, miR 34a 5p, miR 31 5p, and miR 148a 39. One of the control genes on the PCR array, SNORD95, was also h ighly up regulated during ZnD. T wo miRNAs in eac h gastrointestinal cancer group had a fold chan ge of greater than or equal to four during ZnD. The gastric cancer miRNAs, miR 21a 5p and miR 34a 5p, had a fold change of 4.6 and 15.1, respectively during ZnD. The colon cancer miRNAs, miR 221 3p and miR 2 9a 3p, each had a fold change around 4.0 during ZnD. Discussion Studying gene expres sion during zinc deficiency, supplementation or repletion can help open avenues for further r esearch into the mechanism of nutritionally regulated gene expression Zinc de ficiency still remains a problem among people in third world countries, so continued research on zinc regulated genes is necessary for the further development of biomarker s and disease related therapies. Continued research of nutritionally regulated zinc transporter genes is also necessary considering t he increase in gastrointestinal related diseases, such as irritable bowel disease ( 3 ).
75 The purpose of this chapter was to characterize Zip11 in the GI tract, and from the results Zip11 does show high express ion in the gastric and large intestinal tissue, along with dietary zinc partially regulating the gene expression. The high expression of Zip11 in the stomach and large intestine was of particular interest because Zip transporters have not been thoroughly examined in the gastric tissues, and the large intestine appear s to play a role in zinc absorption (28). Also zinc has been shown to play a role in gastric acid secretion (54, 56, 77) and influencing colonic health (69). Though Zip11 is unlike Zip4 in th at it is down regulated by zinc deficiency, this Zip transporter may still be involved in zinc processing during normal or high zinc status. The stomach and colon have been shown to not have a strong influence in zinc absorption during normal, healthy st ates; however, the colon has been shown to absorb zinc when the small intestine has been compromised ( 28 ). Other genes that were found to be dysregulated in the stomach and colon during zinc deficiency and/or repletion were MT Zip4 and Zip5 B oth the s tomach and ileum exhibited a significant d own regulation in MT mRNA due to ZnD, which shows that M T mRNA expression in these tissues responds to ZnD, as seen in the liver ( 48 ). The up regulation of Zip4 mRNA in the small intestine confirmed the efficiency of the zinc deficient diet, and the response of Zip4 both mRNA and protein expression, in the colon showed that this tissue respond s similar to small intestine. Hence, the colon could be an additional site of zinc absorption when the system is in a stat e of deficiency or malnutrition. ZIP5 protein expression in the stomach responds with an increase during zinc deficiency, and expression remains high after 24 hours o f repletion, suggesting that this protein does not rapidly adjust to zinc repletion. Thi s response is different than what has been reported previously in that
76 ZIP5 responds to zinc supplementation and not deficiency ( 16 71 ) C olonic ZIP5 protein expression, however, did show similar results to those previously reported in the small intestin e ( 16, 71) in that ZIP5 only responds to additional zinc, as seen in the repletion samples. This finding suggests a possible role for ZIP5 as a monitor of zinc status. Three ZnT transporters related to gastric tissue were also examined in the ZnA and ZnD gastric samples (data not shown). ZnT4, ZnT5, and ZnT6 all showed increases in mRNA expression with ZnT4 and ZnT5 showing significant up regulation in ZnD samples (data not shown). Zinc deficiency does seem to play a regulatory role in zinc transporter expression, which could be important in maintaining proper acid secretion and mucosal integrity during disease states in the stomach. Colonic microbiota composi tion of mice fed ZnA or ZnD diet s were previously analyzed to determine if zinc intake differenc es could be detected Using denaturing gradient gel electrophoresis analysis and 16S rRNA sequencing, Shore et al. (2010) were able to show a significant difference in the microbiota composition when c omparing the ZnA and ZnD mice ( unpublished data ). Zin c absorption in the colon has been shown to be affected by transit time ( 14 ) and diseases resulting in impaired zinc absorption in the upper gut (28 gastroesophageal reflux disease (GERD) Research has also shown the beneficial affects zinc can have on stimulating gut repair and improving gut mucosa by demonstrating the effects zinc carnosine (ZnC, a commercially available health food product) has during in vitro and in vivo studies in cells, rats, mice and humans ( 50 ) ZnC can be found in health stores with claims to support health y gastrointestinal activities. Mahmood et al. were able to show the stimulatory effects ZnC had on cell
77 proliferation and how this compound improved gastrointestinal injury in rats and mice Human volunteers also showed improvement in gut permeabi lity (injury) caused by commercial NSAID s ( 50 ). Continued study of the effects of zinc and the roles of zinc transporters in the maintenance of the colonic mucosa and overall colon health will he lp advance the possibility that zinc could contribute to treatment of distal gut diseases. The localization of ZIP11 to the nucleus of the parietal cell is of interest, considering that the parietal cell is the site of acid secretion in the stomach. Zinc has been shown to be important in the maturation of tubulovesicles within the parietal cell and may play a role in maintaining the integrity of the parietal cell ( 24 ). Zinc deficiency has also been shown to negatively affect the gastric muco sa and the sec retion of acid (77 ). With the localization of ZIP11 to the nucleus of the gastric parietal cells and the decrease of ZIP11 protein expression seen in stomach tissue, ZIP11 could possibly play a role in gene regulation or zinc monitoring within the nucleus With the strong staining of ZIP11 at the base of the gastric glands, where chief cells and Paneth cells are located, another possible role for this zinc transporter could be to influence zymogenic secretions. ZIP11 could also be involved in zinc regula tion within the P aneth cells found in the lower region of the gastric glands. Strong staining in these stem cell granules indicated the ZIP11 protein to be present. Localization of ZIP11 to the colonic epithelial cells indicates that ZIP11 may be involv ed in maintaining mucosal integrity within the colon or may possibly play a role in zinc sequest ration during zinc deficiency. The co localization of ZIP11 and catenin show s that ZIP11 is located throughout the epithelial cell, whether at the plasma membrane or cytoplasm. A strong fluorescence of ZIP11 is also seen within the nuclei
78 of the colonic epithelial cells, and this localization was identified by a co localization of ZIP11 and DAPI (nuclear marker). Colonic pH is also tightly regulated as seen in the stomach which could indicate ZIP11 playing a role in zinc homeostasis to help maintain the pH balance within these tissues. The localization of ZIP11 to the nuclei of gastric and colonic epithelial cells could indicate that this transporter plays a role in gene regulation or influencing zinc concentration within the nuclei. A previous study looked at circulating miRNAs in human serum during acute zinc defi ciency and repletion Most of the miRNAs in the serum that responded to ZnD were dow n regulated when compared to baseline levels (60 ). The study discussed in this dissertation aimed to find circulating miRNAs in mouse serum that were zinc responsive. Th e miRNA analysis provided a large scale PCR array approach that has not been previously used comparing murine ZnD and ZnA serum It was interesting to note that the miRNAs affected by murine zinc deficiency were only up regulated, with no down regulation present. Considering where zinc is absorb ed and the effects zinc can have on the health of the gastric mucosa and intestinal integrity, the miRNA analysis revealed interesting results with 8 out of the 11 gastric cancer and 3 out of the 8 colon cancer r elated miRNAs on the array up re gulated during zinc deficiency. A nother interesting result from this array was that miRNAs previously found to be expressed during liver cancer and injury were up regulated during zinc deficiency. Both miR 21a 5p and miR 31 5p have been studied previously and shown to be up reg ulated in esophageal ZnD tissue ( 4 ). The affect of zinc deficiency on the progression of oral esophageal squamous cell carcinoma (OSCC, ESCC) has been previously discussed ( 58 ). The rat studies com pleted by Alder et al. revealed that the ZnD esophagus had a
79 miRNA signature that was similar to the human ESCC or tongue SCC miRNA profiles, with miR 21 and miR 31 as the strongest up regulated species (4) Something to point out is that the Alder et al. research looked at chronic zinc deficiency in specific tissues (4) while my murine miRNA data showed significant up regulation of these two miRNAs in serum, after only 2 weeks of zinc deficiency, which shows that zinc deficiency can affect regulatory sys tems rather quickly. The p53 tumor suppressor network is critical for regulating cellular responses to DNA damage and activation of cancer related genes. Several miRNAs have been shown to be directly up regulated by p53 after DNA damage. miR 34a is one of these miRNAs that has been shown to be regulated by the p53 tumor suppressor network and it has also been shown to be deleted in several human cancers ( 9 ). miR 34a has also been shown by Liuzzi et al. (2011) to be up regulated in ZnD small intestine a nd thymus (unpublished data). The significant up regulation of miR 34a in the serum of ZnD mice could be an indication of DNA damage caused by zinc deficiency. MiR 122 is a liver specific miRNA that has decreased expression during liver disease and cance r. This miRNA has been shown to be important in regulating hepatic and plasma iron levels When miR 122 was inhibited in mice using an anti miR compound, an up regulation in some genes involved in iron homeostasis ( Hamp, Hfe Hjv and Bmpr1a ) was detecte d ( 8 ). This research by Castoldi, et al showed that hepatic miRNA 122 expression is essential in preventing iron deficiency in the plasma and liver (8). MiR 122 was also significantly up regulated in ZnD murine serum, which indicates that iron homeostasi s is affected during zinc deficiency This miRNA up regulation could be a defensive mechanism of the murine system in trying to prevent liver injury and altered iron
80 homeostasis caused by prolonged zinc deficiency. Further studies into how other dietary components (copper, iron, folate, etc) affect serum miRNAs in the mouse are needed to fully understand the complexity of miRNA expression in response to micronutrients
81 Figure 4 1. Featur es of the murine Zip11 gene. (A) The intron exon organization of murine Zip11 on chromosome 11 is shown, along with the Zip11 coding sequence (CDS) (B) Predicted topology for m ouse ZIP11 showing six transmembrane spanning regions. The red box indicates the variable region of the isoforms. The green box indicate s the region of the in house ZIP 11 antibody sequence. The blue box indicat es the region of the Pro Sci ZIP 11 sequence. A B NM_001166503 CDS Extracellular 113244853 11 3650079
82 Figure 4 2 Tissue distribution of the Zip11 gene transcr ipt in C57BL/6 mice. Tissues were extracted and used for q PCR analysis of Zip11 expression. (A) Zip11 mRNA tissue distribution in C57BL/6 mice. Values were normalized to 18S and pancreatic expression was set to 1. (B) Zip11 mRNA expression distribution in the 3 regions of the mouse stomach. Values were normalized to 18S and fundic expression was set to 1. Data are expressed as SD (*=P<0.0 0 1, **=P<0.00 00001 ) (n=4 5 for each tissue sample). ** A B
83 Figure 4 3 ZIP11 p rotein tissue distribution in the murine model Tissues were extracted and used for western blot analysis and immunoperoxidase staining. (A) ZIP11 was visualized at 35 kD by the in house antibody. The P onceau staining from the blot was used as the loadi ng control. (B) ZIP11 was visualized using the ProSci ZIP 11 antibody and an IHC technique. The stomach and colon tissues were extracted from C57BL/6 male mice, and the other tissues used for the IHC were located on a mouse tissue array slide. The tissue s were visualized with a microscope and the tissue array tissues were visualized with the 63X objective while the stomach and colon are with the 10X. ZIP11 is stained brown and nuclei are stained blue. Stomach Duodenum Jej unum Ileum Cecum Colon Pancreas Liver Spleen Kidney Brain Zip11 Ponceau A Stomach Small Intestine Colon Pancreas Liver S pleen Kidney Cerebellum Muscle B
84 Figure 4 4 Wild type t issue zinc concen trations during ZnA, ZnD, or ZnS di etary conditions Both male and female mice were used in this analysis. Tissues were digested with HNO3 and diluted with 3 volumes of Milli Q water. Tissue zinc concentrations were measured by atomic absorption spectr ophotometry. Data are expressed as SD (*=P<0.05, **=P<0.01) (n= 4 5). ** **
85 Figure 4 5 Effects of dietary zinc depletion and repletion (ZnR) on serum zinc and GI tract MT expression. (A) Measures of serum zinc indicating the effectiveness of the diet for zinc depletion and repletion. Zn conc entration was measured by AAS Tissues were extracted and used for q PCR analysis. (B) Relative MT transcript expression throughout the GI tract during ZnA, ZnD, and Z nR conditions. Values were normalized to 18S and the ZnA data was set to 1. All d ata are expressed as SD (*=P<0. 05, **=P<0.01, ***=P<0.001, =P<0.0001) (n= 3 5) *** *** A B *** ** ** *** ***
86 Figure 4 6. Effects of dietary zinc depletion and repletion on the Zip11 mRNA th roug hout the GI tract. Samples were analyzed using q PCR. Values were normalized to 18S and the ZnA data was set to 1. All data are expressed as SD (**=P<0.01, 5). ** **
87 Figure 4 7 Effects of dietary zinc depletion and repletion on the m urine ZIP11 protein expression in the stomach A r epresentative western blot analysis and relative densitometry of the 35 kD protein normalized to tubulin is provided. (A) Western analysis of t otal stomach ly sate showing the decrease of ZIP 11 duri ng ZnD and the return of the ZIP 11 expression upon ZnR. (B) Immunoblot data represented by densitometric values. (C) Immunoblot images providing evidence of the spec ificity of the in house ZIP 11 antibody by revealing no bands at 35 kD when the antibody and peptide were incubated together (red box). Tubulin is provided as a loading control. A ZnR ZnA ZnD Stomach Tubulin ZIP 11 ZIP 11 + peptide B C ZIP 11 Tubulin
88 Figure 4 8 Effects of dietary zinc depletion and repletion on the m urine ZIP11 protein expression in the colon A representative western blot analysis and relative densitometry of the 35 kD protein, normalized to tubulin is provided. (A) Western analysis of total colon lysate showing no change in ZIP11 during ZnD and ZnR. (B) Immunoblot data represented by densitometric values. (C) Immunoblot images providing evidence of the specificity of the in house ZIP 11 antibody by revealing no bands at 35 kD when the antib ody and peptide were incubated together (red box). Tubulin is provided as a loading control. ZnA ZnR Colon ZIP 11 Tubulin A ZnD Tubulin ZIP 11 ZIP 11 + pepti de B C
89 Figure 4 9 Effects of dietary zinc depletion and repletion on the m urine ZIP11 protein expression in the cecum and small inte stine Tissues were extracted and use d for western blot analysis with the in house ZIP11 antibody. (A) ZIP11 expression in the cecum durin g ZnA and ZnD conditions. ZIP11 and tubulin were visualized at 35 kD and 55 kD, respectively. Tubulin is shown as a loading control. (B) ZIP11 expression in the three sections of the small intestine dur ing ZnA and ZnD conditions. ZIP 11 was visualized at 35 kD and the Ponceau staining was used as the loading control. ZIP 11 Ponceau ZnD ZnA ZnD ZnA ZnA ZnD Duodenum Jeju num Ileum A B ZIP 11 Tubulin Cecum ZnA ZnR ZnD
90 Fig ure 4 10 Visualizing the m urine ZIP11 protein with immunoperoxidase staining in the stomach Tissues were fixed in 10% formalin, and mounted onto slides. The in house ZIP11 antibody was incubated with murine tissues (b, c) and the specificity of the ant ibody was tested by pre incubating with the peptide used to make the Ab (a). Samples we re visualized with a microscope using 10 0 X objective (a,b) or 63 0 X (c) magnification (A) The 63 0 X im ages reveals the m igration of ZIP 11 during ZnD where it is no lo nger located in the base of the gastric glands (B) ZIP 11 staining in stomach tissue of ZnA, ZnD, and ZnR mice. Less ZIP 11 staining is seen in the ZnD samples with the staining returning to ZnA levels in the ZnR samples. ZIP 11 is stained brown, and nucl ei are stained blue. Strong staining in the chief cell region is indicated by blue arrows. A B ZnA ZnD c b a ZnA ZnD ZnR b c
91 Figure 4 11 Visualizing the m urine ZIP11 protein with immunoperoxidase staining in the colon and cecum. The in house ZIP11 antibody was incubated with murine tissues (b,c) and the specificity of the antibody was tested by pre incubating with the peptide used to make the Ab (a). Samples were visualized with a microscope using 100X (a,b) or 630X (c) magnification. (A) The colon re veals a distinct staining of ZIP 11 on the luminal side of the mucosa. ZnD and ZnR samples appear to have more staining than ZnA samples, with the ZnD samples having staining throughout the mucosa. (B) The cecum shows strong ZIP 11 staining in the ZnA samples wit h a decrease of staining appearing in the ZnD samples. The staining in ZnR samples does not appear to recover as seen in the stomach samples. ZIP 11 is stained brown and nuclei are stained blue A c a b ZnA ZnD ZnR ZnA ZnD ZnR b c B
92 Figure 4 12 Immunofluorescence imag ing of ZIP11 in the murine s tomach. A tiled image of DAPI (nuclei marker), H K ATPase (parietal cell marker), ZIP11 (in house antibody) and a m erged image of all 3 fluorescing labels i n the stomach tissue using the 63X objective yellow arrows indicate ZIP11 staining in the chief cell region (A) White arrows indicate co localization o f ZIP11 and DAPI, shown in pink using the 63X objective with 4Z digital zoom (B). I mages were taken using a Laser Scanning Confocal Fluorescent Microscope DAPI H,K ATPase ZIP11 Merged DAPI H,K ATPase ZIP11 Merged A B
93 Figure 4 13. Immunofluorescence imaging of ZIP11 in the murine colon. (A) Images show ZIP11 (in house antibody) in red, cate nin, an epithelial cell marker in green, and a merged image overlapping ZIP11 and catenin. The merged image shows co localization of ZIP11 t o the colonic epithelial cells. (B) The merged imaged is increased in size to visualize the ZIP11 staining (red) in the epithelial cell nuclei (white arrows). Images were obtained using the 20X objective on a Scanning Disk Confocal Fluorescent Microscope. ZIP11 Catenin Merged ZnA ZnD A B
94 Figure 4 14 The effect of ZnD or ZnR on ZIP4 ex pression in murine GI tract (A) Zip4 transcript variation during ZnA, ZnD, or ZnR dietary conditions. Tissues were extracted and used for q PCR analysis of Zip4 mRNA expression The small intestinal expression confirms the efficiency of the ZnD diet. V alues were normalized to 18S and ZnA values were set to 1. Tissues were extracted and used for western blot analysis. ZIP4 expression in the stomach (B) and colon (C) was analyzed with the protein being visualized at ~40 kD. (D) Immunoblot data for colo nic ZIP4 is represented by densitometric values Values are expressed as SD (*=P<0.05, **=P<0.01, =P<0.0001) (n=4 5). ZnA * * A Z IP 4 Tubulin ZnD ZnR ZIP 4 Tubulin Stomach Colon ZnA ZnD B C ZnR D
95 Figure 4 15 The effect of ZnS and ZnR on Zip5 expression in the murine GI tract Tissues were extrac ted and used for q PCR and immunoblot analysis of Zip5 mRNA expression. (A) Zip5 mRNA expression is up regulated in the stomach and colon during ZnS. (B) Zip5 mRNA expression changes only in the cecum tissue with a decrease during ZnD. Values were normalized to 18S and the ZnA values were set to 1. (C) Stomach ZIP5 protein expression reveals an increase in ZnD and ZnR, while the ZIP5 colonic expression is increased in the ZnR (D). Tubulin is shown as a loading control for both immunoblot s. Immunoblot data for stomach (E) and colon (F) is represented by densitometric values Values are expressed as SD (*=P<0.05, **=P<0.001) (n=4 5). ** ** A B Stomach ZnA ZnD ZnR C Colon ZnA ZnD ZnR ZIP5 Tubulin ZIP5 Tubulin D E F
96 Figure 4 16 Identification of serum miR NAs responsive to dietary zinc deficiency in mice using a q PCR based array. Circulating miRNAs were isolated from pooled sera collected from mice after 2 weeks on the zinc deficient diet and were quantified using a pathway focused PCR array, specifically designed to detect known miRNAs in mouse serum. (A) A scatter plot indicating miRNAs in the ZnD sample that had a fold change of above 2. Red = up regulated miRNAs (B) A graphical representation of the specific miRNAs of interest and those that had a f old change of above 2. Values were normalized to cel miR 39 levels Group 1 (ZnD) vs Control Group (ZnA) Log10 (Control Group 2 ^ C t ) Log10 (Group 1 2 ^ C t ) A B
97 CHAPTER 5 CHARACTERIZATION OF ZIP11, ZIP4, AND ZIP5 EXPRESSION IN THE MURINE COLON DURING ACUTE INFLAMMATION Introductory Remarks Bacterial lipopolysaccharide (LPS) challenges have been used in animal models over the years, to imitate an acute inflammatory response ( 2, 10, 48 ) Responses to an infection or inflammation could be fever, increases in several plasma proteins, such as C reactive protein and ceruloplasmin, muscle loss ( result ing in a negative nitrogen balance ) hypozincemia, and many more systematic changes ( 21 ). The changes observed with acute phase proteins during inflammat ory conditions, can be caused by trauma, infection, and other stress inducing situations. An interest ing attribute of acute phase proteins is that they do not all increase at the same time or concentrations in patients with the same disease ( 21 ). Inflammatory cytokines, such as IL 6, IL TNF re produced during early inflammation and wil l activate the production of acute phase proteins The variation in the production of cytokine s by an individual and the responses modulated by these varying cytokines could be one explanation of the differences see n between patients ( 21 ). Plasma zinc le vels decrease significantly during inflammation, and this drop is thought to be caused by the importation of zinc into major organs, mainly the liver ( 26 ). The pro inflammatory cytokine, IL 6, that is the predominant stimulator of acute phase proteins, ha s also been shown to up regulate Zip14 ( 48 ). ZIP 14 is thought to play a major role in the hypozincemia state seen during infection and inflammation. The colon is an interesting organ to research during acute and chronic inflammatory states because of the challenges that can arise in the colon during
98 inflammation and the degree of colon ic damage sustain ed during inflammation. Another key player of colonic health during healthy or disease states is the microbiota l population R esearch is d elving into the complexities of these colonic bacterial populations and the effects this population can have on an individual ( 27, 62 ). Nutrition can play a role in the progression of already developed inflammatory problems within the colon, along with affecting the hea lth of the colonic microbiome ( 23 ) An important relationship exists between the intestinal immune system and the commensal microbiota to ensure that the epithelial cells and the mucosal immune system recognize the difference between non pathogenic and pa thogenic agents ( 27 ). The intestinal immune response involves the recruitment and activation of lymphocytes and macrophages to the site of injury or pathogenic invasion. These activated cells stimulate the production of inflammatory cytokines, such as in terleukin 6 (IL 6 ) tumor interleukin IL ) which will mediate the progression of the inflammatory response ( 67 ). Zinc deficiency has been shown to exacerbate the effects of colitis in a rat model ( 34 ). Suwendi et al were able to sho which in turn affects the inflammatory response and disease progression in the dextran sodium sulfate (DSS) colitis model. The purpose of this chapter was to characterize expression of Zip11 and other zinc transporters in the colon during acute inflammation. To gain more insight into the colon and zinc transporter response to acute inflammation, colon ic tissues were collected from mice fed either ZnA or Zn S for one week and challenged with LPS for 18 hours
99 Results Colonic Tissue Zinc Decreases A fter an LPS Challenge After LPS administration, the LPS/ ZnA samples had significantly lower colonic tissue zinc when compared to the control PBS group (Figure 5 1 A ). The decrease in tissue zinc after LPS injection was abolished when mice were fed a ZnS diet. Z inc transporter s may possibl y be increased in the ZnS group to increase zinc is available for uptake into the colon in the suppleme nted groups. After seeing the difference in the colonic zinc concentration, e xpression of select zinc transporters were analyzed to see if they could possibly explain this diet dependent response to LPS. Colonic Metallothionein m RNA Expressi on Increase s in Response to LPS MT mRNA was analyzed as a positive control due to its respon se to LPS. In the colonic tissue, MT mRNA expression was up regulated after the LPS challenge in the ZnA group (Figure 5 1B ), suggesting there may be more available zinc within the colon Since tissue zinc results did reveal a significant decrease in the LPS/ZnA group, this response must be attributed to a direct response of LPS on MT expression. MT expression responds as expected in the ZnS group with an increase in expression, but was independent of LPS challenge. Zip11 Colonic mRNA Expression Decrease s in Response to LPS T he diets did not play a significant role in the expression of Zip11 mRNA after LPS administration. Zip11 mRNA expression was on ly affected by LPS with a significant decrease in the LPS/ZnA group (Figure 5 2A). ZIP 11 protein did not decrease in response to either diet or LPS (Figure 5 2B) The ZIP 11 protein expression in the LPS treated, ZnA samples appear to increase when compared to the PBS
100 samples, which is the opposite of the results from the RNA but similar to the increase in ZIP11 protein seen in the colon during zinc deficiency. Zip4 Colonic Expression Increases in Response to LPS Zip4 transcript expression was significantly up regulated by LPS in the ZnA group and this response was attenuated with zinc supplementation (Fig ure 5 3A). ZIP4 protein expression did not exhibit the response to LPS or diet as was seen at the transcript level (Figure 5 3B). Zip5 Colonic Expression Decreases in Response to LPS Zip5 transcript expression was signifi cantly down regulated after LPS administration in the ZnA group (Figure 5 4A). Zinc supplementation did not significantly change the response of Zip5 to the LPS. This down regulated response is consistent with the notion suggest ing that the ZIP 5 transporter is being internalized and d egraded, so as not to export zinc out of the blood during an LPS challenge. The ZnA/LPS samples appear to have a decrease in ZIP5 protein expression but not during ZnS conditions (Figure 5 4B). The Zip5 expression pattern at the transcript and protein le vels, suggest that the Zip5 is being down regulated by LPS. Discussion The purpose of this chapter was to characterize Zip11 and a few other zinc transporters in the colon during acute inflammation. Results showed that LPS may regulate Zip11 and Zip4 gene s within the murine colon. The decrease in colonic tissue zinc seen in the LPS/ZnA group was unusual considering the increase of Zip4 and MT mRNA expression in this group With the up regulation of Z ip 4 mRNA one would think that tissue would also have an inc rease of zinc because the ZIP 4 protein should also increase because of the increase in transcription; however, the ZIP4 protein levels only
101 revealed a slight increase in the LPS/ZnA samples. The time point of thi s LPS experiment (18 hours) could be play ing a role in that the ZIP4 protei n up regulation due to LPS from increased Zip4 transcription may require more time Even if there is an up regulation of ZIP4 at a later time point that does not mean that there is zinc available in contents passing through the colon. The liver imports a high amount of z i nc durin g LPS challenges, so Z ip 4 may be up regulated due to the hypozincemia, but zinc may not be available to absorb in the colon. Zip4 mRNA was down regulated by zinc supplementation as seen previously in the small bowel (74) and also similar to resul ts mentioned above in the dietary chapter MT has been previously shown to up regu late in the liver due to LPS ( 48 ) and responded in the murine colon similarly with an increase in the LPS/ZnA samples. MT mRNA expression could also be up regulated due to an increase in another metal, free radicals or ROS (68 ), since the decrease in tissue zinc should not result in an increase in MT mRNA expression. The time point used for this study (18 hours post LPS injection) may play a role in the zinc levels in the c olonic tissue. The decrease in colonic zinc could be indicating that the additional zinc brought into the mucosa had already been exported into the blood stream or that most zinc had already been cleared from the serum and gastrointestinal tract by the ti me it reaches the colon, due to mechanisms involved with the hypozincemia state, resulting from the LPS challenge. The increase in tissue zinc seen in both the LPS and PBS, ZnS groups indicates the potential of other zinc or metal transporters importing z inc or a paracellular transport of residual zinc into the colon. The expression of MT in normal colonic epithelial cells has been found to be significantly higher than expression in diseased colon sections from ulcerative colitis
102 ts (33). These opposing results to inflammation from my results could suggest a difference in regulation of MT during acute and chronic inflammatory states or a difference in MT regulation in humans. The data presented here comes from an acute inflammati on experiment, so the results showing an increase of MT in the ZnA mice after LPS treatment follows the results shown previously by Liuzz i et al (48). As expected, MT expression increased in ZnS samples, regardless of LPS treatment. Further studies are n eeded to analyze the effects zinc supplementation can have on MT, during inflammation, particularly an influence on inflammatory mediators, free radicals, or ROS. Since there appears to be some ZIP11 localization to the colonic epithelial cells, this pat tern could suggest involvement of ZIP11 in maintaining colonic mucosal integrity during inflammation. I n the LPS treated/ ZnA samples Zip11 mRNA e xpression is significantly down regulated at the transcript level but shows an increase at the protein level. There could be a post transcriptional modification i ncreasing the translation of ZIP 11in t he colon in response to the LPS challenge The lower Zip11 m RNA expression could also be suggesting that there is a sufficient supply of ZIP11 protein, as seen by t he increase of ZIP11 in the ZnA/LPS samples, and in turn the system is signaling to lower transcription, until more protein is required. These opposing results with transcription and translation could also mean that the Zi p11 mRNA was up regulated at an earlier tim e point after LPS administration, and at the 18 hour time poi nt, the system is now seeing th at increase in tr anslation of ZIP1 1. This increase in ZIP11 protein in the colon during LPS does follow what was seen in the colon during zinc
103 deficiency. The ZIP11 protein could be more stable during hypozincemia states with in the murine system. The up regulation of Zi p4 in the colon of LPS treated/ ZnA mice shows a response similar to what was also shown in a zinc deficient state. This Zip4 response in the colon further confirms that LPS may create a zinc deficient state throughout several sys tems in a mouse. As expected with supplemented zinc Zip4 expression was reduced in the ZnS samples; however, the protein expression failed to produce this decreas e in expression. The expression of Zip5 mRNA was significantly decreased in the LPS treated groups, regardless of which diet the mice were on. The s e results suggest some influence of LPS on Zip5 tran slocation. During deficiency, ZIP 5 is internalized fro m the basolateral membrane in a model transfection system ( 16 ). Since this study is looking at the response to acute inflammation and ZIP5 is not different in any samples I conclude that Zip5 mRNA is partially down regulated by inflammatory mediators. T he dow n regulation may help to ensure that the colonic epithelial cells are not transporting zinc from the blood back into the cell for excretion in an effort to conserve zinc concentrations during inflammation Colonic tissue responds to LPS as is seen wi th zinc deficiency with an up regulation of Zip4 Since the tissue was collected 18 hours after the LPS challenge, tissue zinc concentrations and ZIP11, ZIP4 and ZIP5 transporters may not have had sufficient time to respond to the proinflammatory conditio ns. Since genes such as MT Zip4 and Zip11 in the colon are responding to LPS, further r esearch needs to elucidate zinc related mechanisms involved in the inflammatory response in the colon.
104 Figure 5 1. The effects o f a ZnA or ZnS diet on murine colonic tissue zinc concentrations and MT mRNA expression after LPS administration. (A) Tissue samples were extracted and digested in HNO3. After the digestion, samples were diluted and measured using AAS. Values are expres sed as SD (n=4). The effect LPS had on colonic zinc concentrations of animals fed a ZnA diet was diminished when mice were fed a ZnS diet. (B) Tissues were extracted and u sed for q PCR analysis. MT expression was significantly up regulated in the ZnA g roup, while zinc supplementation increase MT in the PBS and LPS groups. Values were normalized to TBP and the ZnA/PBS values were set to 1. Values are expressed as SD (n=3 4). P<0.05 P<0.01 P<0.0001 A B
105 Figure 5 2. The effec t of LPS administration on colo nic Zip11 expression when mice were fed either a ZnA or ZnS diet. Tissues were extracted and used for q PCR analysis and western blotting (A) Zip11 mRNA expression was significantly down regulated in the ZnA diet group when administered LPS; however, the high zinc diet did not influence the Zip11 expression pattern when compared to the ZnA group. Values were normalized to TBP and the ZnA/PBS values were set to 1. Values are expressed as SD (n=4 5). (B) ZIP11 expression did not significantly change at the protein level. Tubulin was visualized as a loading control. P<0.001 ZnS ZIP 11 Tubulin PBS LPS + + + + + + + + A B Zn A
106 Figure 5 3. The ef fect ZnA and ZnS diets have on murine Zip4 expression in the colon after LPS administration. Tissues were extracted and used for q PCR analysis or western blotting (A) Zip4 expression was significantly up regulated due to LPS in the ZnA samples. Zip4 expression was significantly down regulated in the ZnS, with no effect of LPS exhibited. Values were normalized to TBP and the ZnA/PBS values we re set to 1. Values are expressed as SD (n=4 5). (B) ZIP4 was visualized at ~40 kD, revealing an increase in expression in the LPS treated, ZnA samples. The dark banding above the 40 kD band is overexposed tubulin that was visualized prior to ZIP4. T ubulin was visualized as a loading control. P<0.0 0 1 P<0.01 Tubulin ZIP4 ZnS + + + + + + + + PBS LPS A B ZnA ZnS
107 Figure 5 4. The e ffect of LPS administration on Zip5 colonic expression when mice were either fed a ZnA or ZnS diet. Tissues were extracted and used for RT analysis or western blotting (A) LPS administration caused a significant down regulation of Zip5 in the ZnA samples. The ZnS samples showed similar responses to LPS but were not significant. ZnS did not affect the regulation of Zip5 whe n LPS was administered. Values were normali zed to TBP and the ZnA/PBS group was set to 1. Values are expressed as SD (n=4 5 ). (B) ZIP5 protein expression appears to be decreased in the ZnA/LPS samples when compared to the ZnA/PBS samples, while the ZnS samples do not appear to change Tubulin was visualized as a loading control. P<0.01 ZnA ZnS Zip5 Tubulin PBS LPS + + + + + + + + A B
108 CHAPTER 6 CONCLUSIONS AND FUTURE DIRECTION The current study was conducted to characterize the novel zinc transporter, Zip11 also known as Slc39a11. Though the random genetrap model did not correctly create a Zip11 knock out species, many informative results were discovered about this zinc transporter. Zip11 is ubiquitously expressed in several tissues of the C57BL/6 mouse, with the stomach, large intestine, pancreas, and kidney having the highest protein expression The main points of this Zip11 research were focused on the stomach a secretory organ and the colon, a n absorptive organ Zip11 mRNA and protein expression in the stomach is down regulated during zinc deficiency and is increased during ZnD at the prot ein level in the colon. Similar to the stomach response (data not shown) the colonic Zip11 mRNA expression is down regulated after LPS treatment. There may be a mechanism involved that down regulates Zip11 mRNA in these tissues when Zip4 expression incr eases. The ex pression patterns of ZIP 11 seen during IHC were also interesting with in the stomach and the colon exhibiting some ZIP 11 localization in the nucleus of the parietal cell throughout the chief cells within the nucleus of the colonic epithelia l cells and partial localization to the membrane of the epithelial cells ZIP 11 could also be an important transporter to study further because of its localization to two pH regulated tissues, the stomach and colon. The regulation of zinc homeostasis in pH regulated tissues will be an important avenue to research further when it comes to understanding more about zinc transporters. Another zinc transporter, Zn T9, has also been mentioned to be expressed in the nucleus during mitosis ( 12 ). Another report m entioned a possible role of zinc transporters in the nucleus when 65 Zn appeared in the nucleus of spleen cells after a 2 hour intragastric
109 65 Zn feeding ( 63 ). Further research using isolated parietal cells could elucidate whether Zip11 is involved in zinc homeostasis in the nucleus The colonic expression of ZIP 11 is visualized on the apical surface of the epith elial cells, suggesting that ZIP 11 may play a role in zinc transport in the colon. The localization of ZIP11 in the colon could also indicate this transporter being important for zinc homeostasis in order to maintain colonic mucosal integrity and function. If a new Zip11 knock out mouse was created, zinc transport in the colon could be analyzed by 65 Zn gavage. The LPS model brought forth some int eresting results with LPS regulating the mRNA of Zip11 (down), Zip4 (up), and Zip5 ( up ) in the colon Further research could look into inflammatory mediated transcription factors responsible for regulating these Zip transporters in the colon. Of interest would also be looking into inflammatory colonic diseases of humans and analyze expression patterns of Zip11 and other zinc transporters. The findings of this dissertation research are important for gaining new insight into a novel zinc transporter, Zip11 that has been previously uncharacterized. By providing the characterization of Zip11 and evidence of its regulation by zinc deficiency, supplementation, repletion and LPS the stomach and colon will nee d to consider Zip11 along with other zinc transporters. Zip11 along with other zinc transporters, particularly Zip4 and Zip5 have the potential to play important roles involving zinc homeostasis when elucidating more information concerning the zinc and a cid output link within the stomach, and when determining zinc and zinc transporter roles in maintaining colonic m ucosal integrity and function.
110 LIST OF REFERENCES 1. Abuin A, Hansen GM, & Zambrowicz B (2007) Gene trap mutagenesis. Handb Exp Pharmacol (178):129 147. 2 Adams JK & Tepperman BL (2001) Colonic production and expression of IL 4, IL 6, and IL 10 in neonatal suckling rats after LPS challenge. Am J Physiol Gastrointest Liver Physiol 280(4):G755 762. 3 Albenberg LG, Lewis JD, & Wu GD (2012) Fo od and the gut microbiota in inflammatory bowel diseases: a critical connection. Curr Opin Gastroenterol 28(4):314 320. 4 Alder H Taccioli C Chen H Jiang Y et al. (2012) Dysregulation of miR 31 and miR 21 induced by zinc deficiency promotes esophageal cancer. Carcinogenesis 33(9):1736 1744. 5 Andrews GK, Wang H, Dey SK, & Palmiter RD (2004) Mouse zinc transporter 1 gene provides an essential function during early embryon ic development. Genesis 40(2):74 81. 6 Bartel DP (2009) MicroRNAs: target recognition and regulatory functions. Cell 136(2):215 233. 7 Beker Aydemir T Chang SM Guthrie GJ Maki AB et al. (2012) Zinc transporter ZIP14 functions in hepatic zinc, iron and glucose homeostasis during the innate immune response (endotoxemia). PLoS One 7(10):e48679. 8 Castoldi M Vujic Spasic M Altamura S Elmn J et al. (2011) The liver specific microRNA miR 122 controls systemic iron homeostasis in mice. J Clin Invest 121(4):1386 1396. 9 Chang TC Wentzel EA Kent OA Ramachandran K et al. (2007) Transactivation of miR 34a by p53 broadly influences gene expression and promotes apoptosis. Mol Cell 26(5):745 752. 10 Chatelais L Jamin A Gras Le Guen C Lalls JP et al. (2011) The level of protein in milk formula modifies ileal sensitivity to LPS later in life in a piglet model. PLoS One 6(5):e19594. 11 Cousins RJ (2010) Gastrointestinal factors influencing zinc absorption and homeostasis. Int J Vitam Nutr Res 80(4 5):243 248. 12 Cousins RJ, Liuzzi JP, & Lichten LA (2006) Mammalian zinc transport, trafficking, and signals. J Biol Chem 281(34):24085 24089.
111 13 C ousins RJ (2006) Zinc (International Life Sciences Institute) Ninth Ed. 14 Davies NT (1980) Studies on the absorption of zinc by rat intestine. Br J Nutr 43(1):189 203. 15 Dufner Beattie J, Huang ZL, Geiser J, Xu W, & Andrews GK (2006) Mouse ZIP1 and ZIP 3 genes together are essential for adaptation to dietary zinc deficiency during pregnancy. Genesis 44(5):239 251. 16 Dufner Beattie J, Kuo YM, Gitschier J, & Andrews GK (2004) The adaptive response to dietary zinc in mice involves the differential cellula r localization and zinc regulation of the zinc transporters ZIP4 and ZIP5. J Biol Chem 279(47):49082 49090. 17 Dufner Beattie J Weaver BP Geiser J Bilgen M et al. (200 7) The mouse acrodermatitis enteropathica gene Slc39a4 (Zip4) is essential for early development and heterozygosity causes hypersensitivity to zinc deficiency. Hum Mol Genet 16(12):1391 1399. 18 Dufner Beattie J Wang F Kuo YM Gitschier J et al. (2003) The acrodermatitis enteropathica gene ZIP4 encodes a tissue specific, zinc regulated zinc transporter in mice. J Biol Chem 278(35):33474 33481. 19 Feeney GP, Zheng D, Kille P, & Hogstrand C (2005) The phylogeny of teleost ZIP and ZnT zinc transporters and their tissue specific expression and response to zinc in zebrafish. Biochim Biophys Acta 1732(1 3):88 95. 20 Fukada T Civic N Furuichi T Shimoda S et al. (2008) The zinc transporter SLC39A13/ZIP13 is required for connective tissue development; its involvement in BMP/TGF beta signaling pathways. PLoS One 3(11):e3642. 21 Gabay C & Kushne r I (1999) Acute phase proteins and other systemic responses to inflammation. N Engl J Med 340(6):448 454. 22 Geibel JP (2005) Secretion and absorption by colonic crypts. Annu Rev Physiol 67:471 490. 23 Gentschew L & Ferguson LR (2012) Role of nutrition and microbiota in susceptibility to inflammatory bowel diseases. Mol Nutr Food Res 56(4):524 535. 24 Gerbino A, Hofer A, McKay B, Lau B, & Soybel D (2004) Divalent cations regulate acidity within the lumen and tubulovesicle compartment of gastric parietal cells. Gastroenterology 126(1):182 195.
112 25 Guo L Lichten LA Ryu MS Liuzzi JP 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. 26 Haase H & Rink L (2009) Functional significance of zinc related signaling pathways in immune cells. Annu Rev Nutr 29:133 152. 27 Hakansson A & Molin G (2011) Gut microbiota and inflammation. Nutrients 3(6):637 682. 28 Hara H, Konishi A, & Kasai T (2000) Contribution of the cecum and colon to zinc absorption in rats. J Nutr 130(1):83 89. 29 Himeno S, Yanagiya T, & Fujishiro H (2009) The role of zinc transporters in cadmium and manganese transport in mammalian cells. Biochimie 91(10):1218 1222. 30 Huang L, Yu YY, Kirschke CP, Gertz ER, & Lloyd KK (2007) Znt7 (Slc30a7) deficient mice display reduced body zinc status and body fat accumulation. J Biol Chem 282(51):37 053 37063. 31 Inoue K Matsuda K Itoh M Kawaguchi H et al. (2002) Osteopenia and male specific sudden cardiac death in mice lacking a zinc transporter gene, Znt5. Hum Mol Genet 11(15):1775 1784. 32 Inoue K Takano H Shimada A Wada E et al. (2006) Role of metallothionein in coagulatory disturbance and systemic inflammation induced by li popolysaccharide in mice. FASEB J 20(3):533 535. 33 Ioachim E, Michael M, Katsanos C, Demou A, & Tsianos EV (2003) The immunohistochemical expression of metallothionein in inflammatory bowel disease. Correlation with HLA DR antigen expression, lymphocyte subpopulations and proliferation associated indices. Histol Histopathol 18(1):75 82. 34 Iwaya H Kashiwaya M Shinoki A Lee JS et al. (2011) Marginal zinc deficiency ex acerbates experimental colitis induced by dextran sulfate sodium in rats. J Nutr 141(6):1077 1082. 35 Johnson L (1985) Functional development of the stomach. Annu Rev Physiol 47:199 215. 36 Johnson L (2012) Physiology of the Gastrointestinal Tract (Elsev ier Inc) Fifth Edition Ed.
113 37 Kararli TT (1995) Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 16(5):351 380. 38 King JC & Cousins RJ (2006) Zinc. Modern Nutri tion in Health and Disease, 10e (Lippincott Williams and Wilkins). 39 Kohler J Dubach JM Naik HB Tai K et al. (2010) Monochloramine induced toxicity and dysregulation of intracellular Zn2+ in parietal cells of rabbit gastric glands. Am J Physiol Gastrointest Liver Physiol 299(1):G170 178. 40 Kopic S, Murek M, & Geibel J (2010) Revisiting the parietal cell. Am J Physiol Cell Physiol 298(1):C1 C10. 41 Kos CH (2004) Cre/lox P system for generating tissue specific knockout mouse models. Nutr Rev 62(6 Pt 1):243 246. 42 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). Methods 50(4):298 301. 43 Lichten LA, Ryu MS, Guo L, Embury J, & Cousins RJ (2011) MTF 1 mediated repression of the zinc transporter Zip10 is alleviated by zinc restriction. PLoS One 6(6):e21526. 44 Lichten L & Cousins R (2009) Mammalian zinc transporters: nutritional and physiologic regulation. Annu Rev Nutr 29:153 176. 45 Liuzzi JP, Blanchard RK, & Cousins RJ (2001) Differential regulation of zinc transporter 1, 2, and 4 mRNA expression by dietary zinc in rats. J Nutr 131(1):46 52. 46 Liuzz i JP, Bobo JA, Lichten LA, Samuelson DA, & Cousins RJ (2004) Responsive transporter genes within the murine intestinal pancreatic axis form a basis of zinc homeostasis. Proceedings of the National Academy of Sciences of the United States of America 101(40) :14355 14360. 47 Liuzzi JP, Guo L, Chang SM, & Cousins RJ (2009) Krppel 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. 48 Liuzzi JP Lichten LA, Rivera S, Blanchard RK, et al. (2005) Interleukin 6 regulates the zinc transporter Zip14 in liver and contributes to the hypozincemia of the acute phase response. Proceedings of the National Academy of Sciences of the United States of America 102(19):6843 6848.
114 49 Liuzzi J, Aydemir F, Nam H, Knutson M, & Cousins R (2006) Zip14 (Slc39a14) mediates non transferrin bound iron uptake into cells. Proc Natl Acad Sci U S A 103(37):13612 13617. 50 Mahmood A FitzGerald AJ Marchbank T Ntatsaki E et al. (2007) Zinc carnosine, a health food supplement that stabilises small bowel integrity and stimulates gut repair processes. Gut 56(2):168 175. 51 Medema JP & Vermeulen L (2011) Microenvironmental regulation of stem cells in intestinal homeostasis and cancer. Nature 474(7351):318 326. 52 Moore JB, Blanchard RK, & Cousins RJ (2003) Dietary zinc modulates gene expression in murine thymus: results from a comprehensive differe ntial display screening. Proc Natl Acad Sci U S A 100(7):3883 3888. 53 Moore JB, Blanchard RK, McCormack WT, & Cousins RJ (2001) cDNA array analysis identifies thymic LCK as upregulated in moderate muri ne zinc deficiency before T lymphocyte population changes. J Nutr 131(12):3189 3196. 54 Naik H, Beshire M, Walsh B, Liu J, & Soybel D (2009) Secretory state regulates Zn2+ transport in gastric parietal cell of the rabbit. Am J Physiol Cell Physiol 297(4): C979 989. 55 Naveh Y, Lee Ambrose LM, Samuelson DA, & Cousins RJ (1993) Malabsorption of zinc in rats with acetic acid induced enteritis and colitis. J Nutr 123(8):1389 1395. 56 Oner G, Bor N, Onuk E, & Oner Z (1981) The role of zinc ion in the developme nt of gastric ulcers in rats. Eur J Pharmacol 70(2):241 243. 57 Poritz LS, Harris LR, Kelly AA, & Koltun WA (2011) Increase in the tight junction protein claudin 1 in intestinal inflammation. Dig Dis Sci 56(10):2802 2809. 58 Prasad AS Beck FW Doerr TD Shamsa FH et al. (1998) Nutritional and zinc status of head and neck cancer patients: an interpretive review. J Am Coll Nutr 17(5):409 418. 59 Richards MP & Cousins RJ (1975 ) Mammalian zinc homeostasis: requirement for RNA and metallothionein synthesis. Biochem Biophys Res Commun 64(4):1215 1223. 60 Ryu MS, Langkamp Henken B, Chang SM, Shankar MN, & Cousins RJ (2011) Genomic analysis, cytokine expression, and microRNA profil ing reveal biomarkers of human dietary zinc depletion and homeostasis. Proc Natl Acad Sci U S A 108(52):20970 20975.
115 61 Samuelson L & Hinkle K (2003) Insights into the regulation of gastric acid secretion through analysis of genetically engineered mice. A nnu Rev Physiol 65:383 400. 62 Sekirov I, Russell SL, Antunes LC, & Finlay BB (2010) Gut microbiota in health and disease. Physiol Rev 90(3):859 904. 63 Shankar AH & Prasad AS (1998) Zinc and immune function: the biological basis of altered resistance to infection. Am J Clin Nutr 68(2 Suppl):447S 463S. 64 Soybel DI & Kohler JE (2011) Zinc and the Gastrointestinal Tract. Zinc in Human Health ed Rink L (IOS Press), pp 448 472. 65 Soybel D (2005) Anatomy and physiology of the stomach. Surg Clin North Am 8 5(5):875 894, v. 66 Stanford WL, Cohn JB, & Cordes SP (2001) Gene trap mutagenesis: past, present and beyond. Nat Rev Genet 2(10):756 768. 67 Suwendi E, Iwaya H, Lee JS, Hara H, & Ishizuka S (2012) Zinc deficiency induces dysregulation of cytokine produc tions in an experimental colitis of rats. Biomed Res 33(6):329 336. 68 Takahashi Y, Orga Y, Ibata K, & Suzuki K (2004) Role of Metallothionein in the Cell Cycle: Protection Against the Retardation of Cell Proliferation of Endogenous Reactive Oxygen Specie s. Journal of Health Science 50(2):154 158. 69 Tran CD, Ball JM, Sundar S, Coyle P, & Howarth GS (2007) The role of zinc and metallothionein in the dextran sulfate sodium induced colitis mouse model. Dig Dis Sci 52(9):2113 2121. 70 Vallee BL & Falchuk KH (1993) The biochemical basis of zinc physiology. Physiol Rev 73(1):79 118. 71 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 Chem 279(49):5 1433 51441. 72 Wang K, Zhou B, Kuo Y, Zemansky J, & Gitschier J (2002) A novel member of a zinc transporter family is defective in acrodermatitis enteropathica. Am J Hum Genet 71(1):66 73. 73 Wang X & Zhou B (2010) Dietary zinc absorption: A play of Zips and ZnTs in the gut. IUBMB Life 62(3):176 182.
116 74 Weaver BP, Dufner Beattie J, Kambe T, & Andrews GK (2007) Novel zinc responsive post transcriptional mechanisms reciprocally regulate expression of the mouse Slc39a4 and Slc39a5 zinc transporters (Zip4 an d Zip5). Biol Chem 388(12):1301 1312. 75 Wilson MP Hugge C Bielinska M Nicholas P et al. (2009) Neural tube defects in mice with reduced levels of inositol 1,3,4 tr isphosphate 5/6 kinase Proc Natl Acad Sci U S A 106(24):9831 9835. 76 Wittmann J & Jck HM (2010) Serum microRNAs as powerful cancer biomarkers. Biochim Biophys Acta 1806(2):200 207. 77 Yamaguchi M, Yoshino T, & Okada S (1980) Effect of zinc on the acid ity of gastric secretion in rats. Toxicol Appl Pharmacol 54(3):526 530. 78 Yao X & Forte J (2003) Cell biology of acid secretion by the parietal cell. Annu Rev Physiol 65:103 131. 79 Yu Y, Kirschke C, & Huang L (2007) Immunohistochemical analysis of ZnT1 4, 5, 6, and 7 in the mouse gastrointestinal tract. J Histochem Cytochem 55(3):223 234.
117 BIOGRAPHICAL SKETCH Alyssa Brooke Maki was born and raised in Hibbing, Minnesota. In 2005, Alyssa graduated with her Bachelor s of Science fro m the University of West Florida, where she majored in molecular biology. In fall 2008, she was accepted into the Food Science and Human Nutrition department at University of Florida, where she began her graduate studies by entering the doctoral program for nutritional sciences working under Dr. Cousins. She received her Ph.D. from the University of Florida in the summer of 2013.