<%BANNER%>

The Colonic (HKalpha2) H+, K+ ATPase and the Effects of Aldosterone in the Kidney

Permanent Link: http://ufdc.ufl.edu/UFE0004245/00001

Material Information

Title: The Colonic (HKalpha2) H+, K+ ATPase and the Effects of Aldosterone in the Kidney
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004245:00001

Permanent Link: http://ufdc.ufl.edu/UFE0004245/00001

Material Information

Title: The Colonic (HKalpha2) H+, K+ ATPase and the Effects of Aldosterone in the Kidney
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0004245:00001


This item has the following downloads:


Full Text

PAGE 1

THE COLONIC (HK2) H+, K+ ATPASE AND THE EFFECTS OF ALDOSTERONE IN THE KIDNEY By MICHELLE L. GUMZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Michelle L. Gumz

PAGE 3

This dissertation is dedicated to my husband Keith and my daughter Madison.

PAGE 4

ACKNOWLEDGMENTS There are many people without whom completing this dissertation would not have been possible. I would first like to thank my thesis advisor, Dr. Brian Cain. Dr. Cain has provided constant support and encouragement over the last five years. He has taught me how to do great science and how to write and speak about my work in an effective manner. I will always be grateful to him for being such a wonderful mentor to me. I would like to thank the members of my committee. Dr. Charles Wingo, as our collaborator, has been very involved in my doctoral work. I would like to extend a special thank you to him for his help and enthusiasm over the years. Dr. Susan Frost, Dr. Philip Laipis, and Dr. Peter McGuire have been very helpful over the last five years. I feel very fortunate to have been able to work with all of my committee members and I appreciate their approachability and insight regarding my project. I am very appreciative of the members of the Cain Laboratory, both past and present, for providing a great work environment. Drs. Tamara Otto and Deepa Bhatt remain great friends and it was a pleasure to share the laboratory with them. I would also like to thank Andrew Hardy, Christina Norris, and Cort Bouldin. I would especially like to thank Dr. Deborah Zies and Tammy Grabar, with whom I spent nearly all of graduate school working side-by-side. I will always be grateful for their friendship and support. Several members of the Biochemistry and Molecular Biology Department have played important roles in my career at the University of Florida. I would like to thank Dr. Robert McKenna for his help with and contribution to the work involving the molecular iv

PAGE 5

model of HK2a, summarized in Chapter 3. I am grateful to Dr. Michael Kilberg and his laboratory as well, for their help with the densitometry and real time PCR described in Chapter 6. Dr. Thomas Yang and his laboratory have always worked closely with the Cain Lab and I appreciate their help in creating a pleasant work environment as well as their assistance with my work. I would specifically like to thank Christine Mione for her help with computer issues and PCR. Dr. Michael Popp of the ICBR was extremely instrumental in completion of the microarray work and I would like to thank him for his help as well. I am very grateful to my family, both the Miller and Gumz sides, for their support and encouragement. I would especially like to thank my parents, Gary and Mary Miller, for instilling in me the work ethic and ambition that it took for me to reach this goal. Finally, and most importantly, I would like to thank my husband Keith for his constant love and understanding. He has been by my side through all of my undergraduate and graduate education and I could never have reached this point without him. Keith has always been supportive of my goals, whether it was by moving across the country so I could attend UF, or by accompanying me on late night trips to the lab. I cannot thank him enough for this or for giving me the greatest gift of all, our daughter Madison. v

PAGE 6

TABLE OF CONTENTS Page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...............................................................................................................x LIST OF FIGURES...........................................................................................................xi ABBREVIATIONS.........................................................................................................xiv ABSTRACT.......................................................................................................................xx CHAPTER 1 BACKGROUND AND SIGNIFICANCE....................................................................1 Physiological Significance............................................................................................1 The Kidney............................................................................................................1 Ion Transport in the Collecting Duct.....................................................................3 Hormonal Regulation of Ion Transport in the Kidney..........................................6 Aldosterone Action in the Kidney.........................................................................7 Ion Transport Disorders.......................................................................................11 P-Type ATPases.........................................................................................................12 Function...............................................................................................................12 Evolution.............................................................................................................12 The X+, K+ ATPases............................................................................................13 Regulation............................................................................................................15 Mechanism of Catalysis......................................................................................17 Structure..............................................................................................................18 The Rabbit Colonic H+, K+ ATPase...........................................................................23 HK Isoforms in the Kidney.................................................................................23 Physiological Significance of HK2...................................................................24 Cation Selectivity................................................................................................27 HK2a and HK2c.............................................................................................27 Expression Systems.............................................................................................31 The Subunit Controversy.................................................................................34 Summary.....................................................................................................................36 vi

PAGE 7

2 MATERIALS AND METHODS...............................................................................38 Molecular Modeling...................................................................................................38 Choice of Templates and Alignment...................................................................38 Modeling..............................................................................................................38 Model Refinement and Validation......................................................................38 Charge Distribution.............................................................................................39 Secondary Structure Prediction...........................................................................39 Recombinant DNA Techniques..................................................................................39 Preparation and Handling of Plasmid DNA........................................................39 Bacterial Transformations...................................................................................39 Plasmid Constructions.........................................................................................40 Site-directed Mutagenesis...................................................................................42 Tissue Culture.............................................................................................................43 Cell Lines.............................................................................................................43 Hormone Treatment of mIMCD3 cells...............................................................43 Transient Transfection.........................................................................................44 Plasma Membrane Preparation............................................................................45 Biochemical Assays....................................................................................................45 BCA Assay..........................................................................................................45 ATPase Assay......................................................................................................46 Luciferase Assay.................................................................................................48 Protein Methods..........................................................................................................49 Western blot Analysis..........................................................................................49 Immunoprecipitation...........................................................................................50 Molecular Biology......................................................................................................53 RNA Isolation......................................................................................................53 RT-PCR...............................................................................................................55 RACE..................................................................................................................56 Affymetrix GeneChip.......................................................................................57 Affymetrix GeneChip Expression Analysis.....................................................58 Northern Blot Analysis........................................................................................59 Southern Blot Analysis........................................................................................60 Real Time RT-PCR.............................................................................................60 Statistical Analysis......................................................................................................61 3 MOLECULAR MODELING OF THE RABBIT COLONIC H+, K+ ATPASE........62 Introduction.................................................................................................................62 Results.........................................................................................................................64 Sequence Alignments..........................................................................................64 Modeling..............................................................................................................66 Charge Distribution.............................................................................................69 Architecture of the HK2a Model......................................................................71 Modeling of the Subunit...................................................................................73 vii

PAGE 8

Discussion...................................................................................................................76 4 CLONING OF THE RABBIT X+, K+ ATPASE SUBUNITS................................79 Introduction.................................................................................................................79 Results.........................................................................................................................80 Degenerate PCR for NaK3................................................................................80 RACE for NaK3................................................................................................83 PCR Amplification of the Full-Length NaK3 cDNA........................................84 Degenerate PCR for NaK2................................................................................87 RACE for NaK................................................................................................88 PCR Amplification of the Full-Length NaK2 cDNA........................................90 Discussion...................................................................................................................91 5 EXPRESSION AND CHARACTERIZATION OF THE RABBIT COLONIC H+, K+ ATPASE.....................................................................................................................98 Introduction.................................................................................................................98 Results.......................................................................................................................100 Correction of HK2 cDNAs.............................................................................100 Expression Constructs.......................................................................................103 NaK1...............................................................................................................105 NaK2...............................................................................................................106 NaK3...............................................................................................................107 Gastric HK.......................................................................................................107 Co-immunoprecipitation of the Rabbit Colonic H+, K+ ATPases:....................108 Analysis of ATPase Activity of the Rabbit Colonic H+, K+ ATPases..............112 Discussion.................................................................................................................115 6 EARLY TRANSCRIPTIONAL EFFECTS OF ALDOSTERONE IN MOUSE IMCD3 CELLS.........................................................................................................118 Introduction...............................................................................................................118 Results.......................................................................................................................119 Characterization of mIMCD3 Cells...................................................................119 Microarray Analysis..........................................................................................121 Analysis of Aldosterone-Responsive mRNAs..................................................122 Induction of Preproendothelin...........................................................................131 Analysis of the Period Homolog Promoter.......................................................132 Discussion.................................................................................................................137 viii

PAGE 9

7 CONCLUSIONS AND FUTURE DIRECTIONS...................................................140 The Rabbit Colonic H+, K+ ATPase.........................................................................140 Future Directions for the Study of the Colonic H+, K+ ATPAse.......................145 Early Transcriptional Effects of Aldosterone in the Mouse Kidney........................147 Future Directions for the Study of Aldosterone Action in the Mouse Kidney..154 LIST OF REFERENCES.................................................................................................156 BIOGRAPHICAL SKETCH...........................................................................................168 ix

PAGE 10

LIST OF TABLES Table page 3-1. Summary of Amino Acid Deletions and Additions in HK2a as Compared to the Ca+2 ATPase.............................................................................................................67 3-2. Stereochemical Statistics...........................................................................................69 4-1. Primers Used to Amplify cDNAs for Rabbit Na+, K+ ATPase 3 and 2.................82 4-2. Protein Sequence Comparison of the Rabbit Na+, K+ ATPase 3 Subunit to Other NaK Subunits.........................................................................................................93 4-3. Protein Sequence Comparison of the Rabbit Na+, K+ ATPase 2 Subunit to Other NaK Subunits.........................................................................................................93 5-1. Summary of HK2 Expression Systems....................................................................99 5-2. Primers Used for Site-directed Mutagenesis of HK2a...........................................102 5-3. Primers Used for PCR Amplification or Mutation of XK Subunits.......................104 5-4. H+, K+ ATPase Expression Constructs and Summary of Co-immunoprecipitation Data........................................................................................................................115 6-1. Comparison of Fold Induction Values......................................................................122 6-2. Transcripts Down-Regulated by Aldosterone..........................................................123 6-3. Transcripts Up-Regulated by Aldosterone...............................................................124 x

PAGE 11

LIST OF FIGURES Figure page 1-1 Structure of the Kidney..................................................................................................2 1-2 Cells of the Collecting Duct..........................................................................................4 1-3. Mechanism of Transcriptional Action of Aldosterone.................................................8 1-4. The H+, K+ ATPase.....................................................................................................14 1-5. Mechanism of Catalysis of the H+, K+ ATPase..........................................................18 1-6. The Ca+2 ATPase in the E1 and E2 Conformations...................................................21 1-7. Rabbit HK2 Isoforms...............................................................................................28 1-8 Model for HK2 Gene Expression..............................................................................30 2-1. Construction of Plasmid pMLG97.............................................................................42 2-2. ATP Hydrolysis Assay...............................................................................................47 2-3. Co-immunoprecipitation of the HK2 and subunits...............................................53 3-1. Primary Structure Alignment of HK2a, Na+, K+ ATPase 1, and Ca+2ATPase......65 3-2. Comparison of the HK2a model with the Ca+2 ATPase..........................................68 3-3: Charge Distribution over Surfaces of HK2a............................................................70 3-4: Possible Architecture of the HK2a Subunit.............................................................72 3-5. Ouabain-binding Residues in HK2a........................................................................74 3-6. Modeling of the Transmembrane Segment of NaK2...............................................75 3-7. A Putative Location for the subunit of the H+, K+ ATPase.....................................77 4-1. Design of Degenerate Primers to Amplify a cDNA Fragment of Rabbit Na+, K+ ATPase 3................................................................................................................81 xi

PAGE 12

4-2. Degenerate PCR for the Rabbit Na+, K+ ATPase 3 Subunit....................................83 4-3 Rapid Amplification of cDNA Ends for the Rabbit Na+, K+ ATPase 3....................84 4-4. PCR Amplification of the Rabbit Na+, K+ ATPase 3 Subunit cDNA......................85 4-5. GenBank Accession Record for Rabbit NaK3.........................................................86 4-6. Design of Degenerate Primers to Amplify a cDNA Fragment of Rabbit Na+, K+ ATPase 2................................................................................................................87 4-7. Degenerate PCR for the Rabbit Na+, K+ ATPase 2 Subunit....................................88 4-8. Rapid Amplification of cDNA Ends for the Rabbit Na+, K+ ATPase 2 Subunit.....89 4-9. PCR Amplification of the Rabbit Na+, K+ ATPase 2 Subunit cDNA......................90 4-10. GenBank Accession Record for Rabbit NaK2.......................................................91 4-11. Alignment of Amino Acid Sequences for the Rabbit X+, K+ ATPase Subunits...94 4-12. Kyte-Doolittle Hydropathy Plots for Rabbit Na+, K+ ATPase 3 and 2................95 4-13. Secondary Structure Prediction for NaK3 and NaK2..........................................96 5-1. Mutated Amino Acids in HK2a.............................................................................101 5-2: pMLG96...................................................................................................................103 5-3. Western Blot Analysis of HK2a and HK2c Subunits..........................................105 5-4. Co-immunoprecipitation of HK2 Subunits and gHK..........................................110 5-5. Co-immunoprecipitation of HK2c Subunits and NaK1-3...................................111 5-6. Analysis of K+-dependent ATPase Activity from Mouse Gastric Mucosa..............112 5-7. ATP Hydrolysis Activity of Transfected COS-1 Cells............................................114 6-1. Characterization of mIMCD3 Cells..........................................................................120 6-2. Up-Regulation of Sgk by Aldosterone......................................................................121 6-3. Dose Response to Aldosterone.................................................................................126 6-4. Time Course of Responses to Aldosterone...............................................................127 6-5. Densitometry Analyses of Time Course Northern Blot Data...................................128 xii

PAGE 13

6-6. Real Time PCR Analysis of Time Course................................................................129 6-7. Effect of Mineralocorticoid and Glucocorticoid Receptor Inhibitors......................130 6-8. Increased Preproendothelin Protein Expression in Aldosterone Treated Cells........132 6-9 Southern Blot Analysis of BAC Clone RP24-277K16..............................................133 6-10 CpG Analysis of the Period Homolog Promoter.....................................................134 6-11 Glucocorticoid Response Elements in the Period Homolog Promoter....................135 6-12 Amplification of the Period Homolog Promoter.....................................................136 6-13. Relative Luciferase Activity of Period Homolog Promoter...................................137 7-1. Model of the Rabbit Colonic H+, K+ ATPase...........................................................143 7-2. Model for Aldosterone Action in mIMCD3 cells.....................................................148 xiii

PAGE 14

ABBREVIATIONS A, alanine ACE, angiotensin converting enzyme ADH, antidiuretic hormone ADP, adenosine diphosphate ANOVA, analysis of variance ANP, atrial natriuretic peptide ATL, ascending thin limb ATP, adenosine triphosphate BCA, bicinchoninic acid bp, base pairs BSA, bovine serum albumin CA, carbonic anhydrase Ca+2, calcium CCD, cortical collecting duct cDNA, complementary deoxyribonucleic acid cGMP, cyclic guaninosine monophosphate CHIF, corticosteroid hormone induced factor Cl-, chloride CMV, cytomegalovirus CTGF, connective tissue growth factor xiv

PAGE 15

D, aspartic acid dCTP, deoxycytosine triphosphate DEPC, diethyl pyrocarbonate DMEM, Delbuccos modified Eagle medium DNA, deoxyribonucleic acid dNTP, deoxynucleotide triphosphate DT, distal tubule DTL, descending thin limn E, glutamic acid E. coli, Escherichia coli ECF, extracellular fluid EDTA, ethylene diamine tetraacetic acid ENaC, epithelial sodium channel ER, endoplasmic reticulum EST, expressed sequence tag f, femtoFBS, fetal bovine serum G, glycine G/MRE, glucocorticoid or mineralocorticoid response element GAPDH, glyceraldehyde-3-phosphate dehydrogenase GFR, glomerular filtration rate GILZ, glucocorticoid-induced leucine zipper protein GR, glucocorticoid receptor xv

PAGE 16

GRE, glucocorticoid response element GTP, guanosine triphosphate H+, hydrogen H2O, water HCO3-, bicarbonate HEK, human embryonic kidney Hr, hour IMCD, inner medullary collecting duct IP, iimunoprecipitation K+, potassium K, lysine Kb, kilobase pair KDa, kilodalton Kir, K+ inward rectifier L, leucine LB, luria broth LDH, lactate dehydrogenase LSD, least significant difference M, methionine micromM, millimolar MAS4, microarray suite 4 MD, macula densa xvi

PAGE 17

mEq, milliequivalent Mg+2, magnesium Mg, milligram Min, minute Mol, mole M, molar MR, mineralocorticoid receptor MRE, mineralocorticoid response element mRNA, messenger ribonucleic acid n, nanoN, asparagine Na+, sodium NADH, nicotinamide adenine dinucleotide NaOH, sodium hydroxide Nedd-4, neural precursor cell expressed developmentally down-regulated 4 NH4, ammonium NP-40, nonidet P-40 OMCD, outer medullary collecting duct P, proline PAGE, polyacrylamide gel electrophoresis PCR, polymerase chain reaction PDB, protein data bank PK, pyruvate kinase xvii

PAGE 18

PKA, protein kinase A PKC, protein kinase C PKG, protein kinase G PMSF, Phenylmethylsulfonylfluoride PT, proximal tubule RACE, rapid amplification of cDNA ends RCCT, rabbit cortical collecting tubule RNA, ribonucleic acid RT-PCR, reverse transcriptase polymerase chain reaction S, serine SAGE, serial analysis of gene expression SA-PMP, streptavidin magnesphere paramagnetic particles SCH28080, 3-(cyanomethyl)-2-methyl-8-(phenylmethoxy) imidazo [1,2a]-pyridine SDS, sodium dodecyl sulfate Sgk, serum and glucocorticoid regulated kinase SOC, salt-optimized borth plus carbon Sp1, stimulating protein 1 SSC, sodium citrate sodium chloride SWR, standard working reagent T, threonine TAE, Tris acetic acid EDTA TAL, thick ascending limb TBS, tris buffered saline xviii

PAGE 19

TGF, tubular glomerular feedback TNP-AMP, 2'(3')-O-(2,4,6-trinitrophenyl)adenosine 5'-diphosphate V, valine Y, tyrosine xix

PAGE 20

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE COLONIC (HK2) H+, K+ ATPASE AND THE EFFECTS OF ALDOSTERONE IN THE KIDNEY By Michelle L. Gumz May 2004 Chair: Brian D. Cain Major Department: Biochemistry and Molecular Biology The kidney is the organ largely responsible for maintenance of ion and acid-base balance. The H+, K+ ATPase is one of the many enzymes the kidney employs to carry out these functions. The H+, K+ ATPase utilizes the energy of ATP hydrolysis to pump K+ into the cells of the renal collecting duct in exchange for H+, transporting both ions against their concentration gradients. Evidence suggests that the colonic H+, K+ ATPase, or HK2 isoform, plays an important role in K+ conservation by the kidney. Two HK2 isoforms, designated HK2a and HK2c, have been identified in the rabbit kidney. The mineralocorticoid aldosterone also acts in the collecting duct of the kidney; it is responsible for maintaining Na+ balance, acid-base balance, and blood pressure. It was the goal of this dissertation to gain a better understanding of renal ion transport and its hormonal regulation. To that end, we have created a molecular model of the HK2a subunit (Specific Aim 1), cloned the four known mammalian X+, K+ ATPase subunits xx

PAGE 21

that may pair with HK2a and HK2c (Specific Aim 2), established a mammalian expression system for the rabbit colonic H+, K+ ATPase (Specific Aim 3), identified the subunit of the gastric H+, K+ ATPase as the putative partner of HK2a and HK2c by co-immunoprecipitation (Specific Aim 4), and characterized the early effects of aldosterone on gene expression in the mouse inner medullary collecting duct (Specific Aim 5). xxi

PAGE 22

CHAPTER 1 BACKGROUND AND SIGNIFICANCE Physiological Significance The Kidney The kidney is both an important regulatory and excretory organ. It plays a vital role in regulation of body fluid osmolality and volume, electrolyte balance, and acid-base balance (66). This organ is also responsible for excreting metabolic and chemical waste from the body as well as for the production and secretion of hormones. The kidney is charged with a difficult task in maintaining the volume and composition of body fluid, and balancing acid-base and electrolytes when the intake by the body of fluid and solutes varies so dramatically on a daily basis. This task is manageable because of the function of the many specialized cells in the kidney that carefully regulate the transport of water and ions into and out of the filtrate passing through this organ (66). The complex anatomy of the kidney underscores its many vital functions (Figure 1-1). The two major regions of a bisected kidney are labeled in Figure 1-1A, an outer region called the cortex and an inner region called the medulla. These regions are made up of nephrons, the functional unit of the kidney. An enlargement of a nephron is shown in panel B of Figure 1-1. The filtrate from the body passes into the nephron through Bowmans capsule via a capillary network called the glomerulus. From there, the filtrate travels through the proximal tubule (PT) and into the Loop of Henle, which is composed of the descending thin limb (DTL), the ascending thin limb (ATL) and the thick 1

PAGE 23

2 Figure 1-1 Structure of the Kidney. (A) A bisected kidney is pictured with the outer region, or cortex, and the inner region, the medulla, labeled. (B) The functional unit of the kidney, the nephron, is shown with each region labeled. PT, proximal tubule; DTL, descending thin limb; ATL, ascending thin limb; TAL, thick ascending limb; MD, macula densa; DT, distal tubule.

PAGE 24

3 ascending limb (TAL). Near the end of the TAL and before the distal tubule (DT) begins, a short segment of the nephron is located, called the macula densa (MD). The cells of the MD are important sensors of the NaCl concentration in the tubular fluid and are therefore a part of an important regulatory mechanism called tubuloglomerular feedback (TGF). Tubuloglomerular feedback is responsible for regulating the rate at which the filtrate passes through the nephron. This regulation occurs by adjusting the glomerular filtration rate (GFR) and renal blood flow accordingly. For example, if GFR is increased and NaCl delivery to the MD rises, then the cells of the MD send a signal that causes vasoconstriction and results in a decrease in GFR and renal blood flow (66). Beyond the MD, the DT extends to the region where two or more nephrons join to form the cortical collecting duct (CCD). The CCD passes into the medulla of the kidney and first becomes the outer medullary collecting duct (OMCD) and then the inner medullary collecting duct (IMCD). The collecting duct is composed of three distinct cell types, principal cells and and intercalated cells (66). Each type is specialized to transport certain ions; these specific functions are discussed in more detail below. Ion Transport in the Collecting Duct Regulation of electrolyte and acid-base balance are two of the kidneys most important functions; this regulation is mediated in part by the ion transporters in the cells of the collecting duct. Failure to maintain the pH of body fluids and the concentrations of ions such as potassium within very narrow margins can have fatal consequences. Nearly all molecular and cellular processes are sensitive to changes in pH. In fact, a pH range from 6.8 to 7.8 is required in order for survival (66). A potassium concentration outside the range of 3.5 to 5.5 mEq/L can result in an ion imbalance, potentially leading to

PAGE 25

4 Figure 1-2 Cells of the Collecting Duct. Apical and basolateral sides of the cells are designated. Solitary arrows represent ion channels while bidirectional arrows represent pumps or exchangers. (A) An -intercalated cell is pictured. These cells secrete H+ and contain H+ ATPases and H+, K+ ATPases on their apical membranes. (B) A -intercalated cell is shown. These cells secrete bicarbonate and have a HCO3-/Clexchanger on their apical membrane. In both types of intercalated cells, it is the action of carbonic anhydrase (CA) that generates H+ and HCO3to be secreted. (C) A principal cell is pictured. These cells are responsible for Na+ and H2O absorption and K+ secretion, as evidenced by the channels depicted on each membrane (66).

PAGE 26

5 cardiac arrest (73). Potassium serves many purposes in the body including maintenance of intracellular osmolarity, establishment of dynamic electrical membrane potentials and neutralization of intracellular negative charges (74). Ion transport occurs all along the renal nephron by transporters and channels contained within specialized cells (66). Three such cell types make up the collecting duct and are pictured in Figure 1-2. Intercalated cells regulate acid base balance by secreting H+, in the case of -intercalated cells, or HCO3-, in the case of -intercalated cells. The H+ and HCO3are generated by the action of carbonic anhydrase (CA) in the intercalated cells. In the acid secreting cells (Figure 1-2A), a basolateral HCO3-/Clantiporter transports HCO3into the bloodstream and Clinto the cell. On the apical membrane of these cells, two different ATPases are responsible for the secretion of H+ into the collecting duct. The H+ ATPase pumps H+ out of the cell while the H+, K+ ATPase pumps H+ out of the cell in exchange for bringing K+ into the cell. The base-secreting -intercalated cells (Figure 1-2B) possess an apical HCO3-/Clantiporter responsible for secretion of bicarbonate and a basolateral H+ ATPase that pumps H+ into the bloodstream. In addition, the intercalated cells contain a basolateral channel for Clexit. The principal cells of the collecting duct (Figure 1-2C) are responsible for the reabsorption of Na+ and H2O and the secretion of K+. These cells contain apical epithelial Na+ channels (ENaC) that bring Na+ into the cell. K+ channels on both sides of the principal cells are responsible for the secretion of K+ into the collecting duct or the bloodstream (66). While K+ is the bodys principal intracellular cation, Na+ is the bodys principal extracellular cation. The balance of Na+ in the body is regulated even more stringently

PAGE 27

6 than that of K+. Specific hormones in the kidney often mediate control of the pumps and channels that transport these ions (80). Hormonal Regulation of Ion Transport in the Kidney Three of the main hormones that regulate ion transport in the collecting duct are vasopressin, atrial natriuretic peptide (ANP), and aldosterone (66). Vasopressin (also called antidiuretic hormone, or ADH) is the most important hormone for regulation of H2O balance. Vasopressin is secreted by the pituitary gland in response to an increase in plasma osmolality or a decrease in the extracellular fluid volume. Action of this hormone causes an increase in the permeability of the collecting duct to H2O, an effect that is mediated by the water channel, aquaporin 2. ANP is a peptide hormone secreted by the atria of the heart in response to high blood pressure and increased extracellular fluid volume. ANP causes increased Na+ and H2O excretion and does so through several levels of downstream effectors. Actions of this hormone result in increased GFR, and inhibition of aldosterone and ADH secretion, which causes a decrease in the reabsorption of Na+ and H2O, respectively. ANP also inhibits Na+ uptake by the cells of the collecting duct through the second messenger, cGMP, which causes inhibition of apical Na+ channels. The mineralocorticoid aldosterone is released in response to low Na+ and/or high K+ concentrations. Aldosterones primary effect is to cause increased Na+ uptake and H2O reabsorption by the collecting duct (66). Release of aldosterone itself is governed at multiple levels; this chain of events is detailed below. In addition to its excretory functions, the kidney also acts as an endocrine organ (80). In response to a decrease in blood pressure, extracellular fluid (ECF) volume, or sodium concentrations, the kidney releases the proteolytic enzyme renin. Renin is released into the blood by the juxtaglomerular cells of the kidney where it acts upon its

PAGE 28

7 substrate, angiotensinogen, a circulating protein produced by the liver. Cleavage of angiotensinogen by renin yields a 10 amino acid peptide, angiotensin I, which is further cleaved to an eight amino acid peptide by angiotensin converting enzyme (ACE). Production of the peptide hormone angiotensin II by ACE occurs on the surface of vascular epithelial cells, primarily in the lung and kidney. Angiotensin II in turn acts on the adrenal zona glomerulosa to stimulate production of aldosterone (80). Aldosterone Action in the Kidney Aldosterone is a mineralocorticoid that plays a vital role in regulating blood pressure as well as potassium and sodium balance. Aldosterone is a steroid hormone and exerts its transcriptional effects through a classical mechanism (Figure 1-3). As a hydrophobic molecule, aldosterone crosses the plasma membrane of a cell and enters the cytoplasm where it binds its receptor. Prior to hormone binding, the receptor is bound by a chaperone in an inactive conformation. After binding, the receptor is released by the chaperone and the hormone/receptor complex translocates into the nucleus where it can bind its recognition sequence and drive transcription of target genes. The canonical receptor for aldosterone is the mineralocorticoid receptor (MR) (5). Aldosterone binds MR with an affinity of 1.3 nM, as determined by Scatchard analysis. The glucocorticoid receptor (GR) can also bind aldosterone with reported KD values ranging from 0.5 to 3.0 nM. MR and GR are capable of forming heteroand homodimers to drive transcription of target genes and their recognition sequences are often similar. They both recognize a fifteen nucleotide, inverted palindrome repeat separated by three random nucleotides (39). Mechanisms exist in aldosterone target cells to ensure that MR is activated specifically by aldosterone and not by other glucocorticoids that may be

PAGE 29

8 Figure 1-3. Mechanism of Transcriptional Action of Aldosterone. Aldosterone crosses the plasma membrane and enters the cell. In the cytoplasm, the mineralocorticoid receptor (MR) and/ or glucocorticoid receptor (GR) binds aldosterone. Binding of hormone releases the receptor from its inactive conformation when bound to a chaperone (C). The hormone/receptor complex then translocates into the nucleus where, in the case of aldosterone, the receptor can homodimerize (MR/MR or GR/GR) or heterodimerize (MR/GR) at a mineralocorticoid response element (MRE) or a glucocorticoid response element (GRE) in the 5 regulatory region of a target gene in order to initiate transcription.

PAGE 30

9 present. Namely, the enzyme 11--hydroxysteroid dehydrogenase inactivates glucocorticoid hormones so that MR activity is induced by aldosterone alone (40). Aldosterone acts directly on the collecting duct of the renal nephron to increase sodium absorption, which in turn increases ECF volume and blood pressure (80). Known aldosterone targets include the basolateral Na+, K+ ATPase (118) as well as the apical epithelial sodium channel (ENaC), both of which function to increase transcellular sodium transport. Lesser-known aldosterone targets are those factors that are active during the time lag between aldosterone release and increased protein levels and activity of the Na+, K+ ATPase and the ENaC. It was recently shown that the serum and glucocorticoid-regulated kinase (sgk) mediates early aldosterone action by regulating ENaC (18). Sgk is upregulated following exposure to aldosterone (77). Its action leads to an increase in ENaC stability due to a reduction in the ubiquitination of the sodium channel. Specifically, sgk phosphorylates the ubiquitin ligase Nedd-4, which purportedly leads to a decrease in the interaction between Nedd-4 and the subunit of ENaC. The subsequent decrease in the degradation of ENaC leads to an increase in the number of active channels at the plasma membrane, and therefore an increase in the transport of Na+ into the cell (48, 119). A connection between the action of aldosterone-induced sgk and the Na+, K+ ATPase has been reported as well (119). Xenopus oocytes expressing the constitutively active form of sgk demonstrated an increase in the number and activity of Na+, K+ ATPase enzymes at the cell surface. Given this observation and the characterization of

PAGE 31

10 sgks action on the apical sodium channel, it would appear that sgk plays a role in regulating the Na+ transport machinery on both sides of epithelial cells. Aldosterone is responsible for regulating K+ levels as well. Recent work investigating the role of sgk in regulation of a K+ channel demonstrates a possible mechanism for the aldosterone-mediated control of K+ levels in the body (128). The primary mode of apical K+ secretion in the kidney is through a family of inward rectifying K+ channels (Kir) encoded by the ROMK gene. One such channel, designated Kir 1.1, may be regulated by the action of sgk. It contains within its N-terminal region a recognition sequence for sgk, RXRXX (S/T); in vitro phosphorylation studies demonstrated that S44, within this sequence, was a substrate for sgk. Introduction of sgk into Kir1.1-expressing cells resulted in an increase in channel density at the plasma membrane. Since this effect was phosphorylation-dependent, it was concluded that phosphorylation of Kir1.1 by sgk leads to increased surface expression of the K+ channel (128). Other than sgk, additional early aldosterone-induced genes include K-Ras2 (103) and CHIF (9). CHIF (corticosteroid hormone induced factor) has been shown to be an accessory subunit of the Na+, K+ ATPase and is discussed in more detail below. K-Ras2 is a member of the Ras superfamily, a group of small GTP binding proteins. In studies using Xenopus oocytes, it was demonstrated that K-Ras2 activity led to an increase in the activity of ENaC channels already present at the membrane (119). Of the approximately 50 mRNAs that were likely to be regulated as defined by differential display experiments (103), only a few have been identified. Therefore, a more complete picture of the early aldosterone-responsive proteins remains to be characterized.

PAGE 32

11 Recent technological innovations such as serial analysis of gene expression (SAGE) and microarray analysis can make this characterization possible (120). SAGE technology was used to identify 34 aldosterone-induced transcripts and 29 aldosterone-repressed transcripts in a mouse kidney cortical collecting duct cell line after four hours of aldosterone treatment (91). These investigators identified several aldosterone responsive transcripts, such as the glucocorticoid-induced leucine zipper protein (GILZ), a transcription factor that could play a role in regulating expression of downstream aldosterone-targets. Although these results are interesting, analyzing cells for aldosterone regulated transcripts four hours after treatment increases the likelihood that early response transcripts have returned to normal levels. Indeed, this study failed to identify sgk as an early aldosterone-responsive transcript, an observation that has been well established by other investigators. Our own work has focused on using microarray analysis to identify aldosterone-regulated transcripts in a mouse inner medullary collecting duct cell line one hour after treatment with hormone ((52) and see Chapter 6). Ion Transport Disorders The importance of aldosterone in control of ion transport and blood pressure is underscored by the fact that every known form of inheritable hypertension results from a defect in the aldosterone signaling pathway (13). In addition to disease resulting from errors in hormone signaling, there are several disorders that stem from mutations in renal transporters. Those mutations affecting the collecting duct are discussed here. Two diseases result from mutations in one or all of the three subunits of ENaC. Pseudohypoaldosteronism type 1a causes increased loss of Na+ and therefore, hypotension results. Liddle syndrome has the opposite effect; a decrease in Na+ excretion causes hypertension. Other examples of CD disorders include a mutation in the water

PAGE 33

12 channel, aquaporin 2, that results in nephrogenic diabetes insipidus, which causes polyuria, polydipsia and plasma hyperosmolality. Finally, distal renal tubular acidosis can be caused by mutations in two different proteins, either the HCO3-/Clexchanger or the H+ ATPase. Symptoms of this disorder include metabolic acidosis, hypokalemia, hypercalciuria and kidney stones (66). P-Type ATPases Function The Na+, K+ ATPase and the H+, K+ ATPase are both members of a large family of membrane bound ATPases and are classified as P-type. Membrane bound, cation-transporting ATPases are categorized into three types: F-type, including the mitochondrial F1F0 ATP synthase; V-type, including the vacuolar H+ ATPase; and P-type (121). The P-type ATPases contain a catalytic subunit that becomes phosphorylated during the catalytic cycle, hence the name P-type. These enzymes utilize the energy of ATP hydrolysis to transport a variety of cations across cell membranes and against their electrochemical gradient. This transport results in an ion gradient that can drive cellular functions such as absorption, secretion, transmembrane signaling, nerve impulse transmission, growth and differentiation (97). Evolution P-type ATPases are further categorized into five branches, based on phylogenetic analysis (84). The Type IA enzymes pump K+, and are only found in bacteria so they may be ancestral to the other P-type ATPases. Type IB pumps transport copper and cadmium and, together with the Type IIA enzymes, are found in eukaryotes, archaebacteria and bacteria; therefore they probably evolved early. The Type III enzymes pump H+ and are found in plants, fungi and archae, but not bacteria; thus they likely evolved later. The

PAGE 34

13 Type IV and V pumps, which transport phospholipids and non-specific substrates, respectively, are found only in eukaryotes so they must have evolved after eukaryotes split from bacteria. We are most concerned with Type II P-type ATPases, or the non-heavy metal ion pumps. These include the single subunit Type IIA and B sarcoplasmic reticulum (SERCA) or plasma membrane (PM) Ca+2 ATPases, respectively, and the type IIC H+, K+ ATPase and the Na+, K+ ATPase. The latter two enzymes are known as the X+, K+ ATPases. The X+, K+ ATPases are oligomeric and contain an and subunit pair as the functional unit. The X+, K+ ATPases The Na+, K+ ATPase and the H+, K+ ATPase comprise a subclass of the P-type ATPases defined by the fact that they are made up of a catalytic subunit and a smaller subunit. Though a matter of some debate, and interaction studies combined with ligand binding experiments suggest that these enzymes likely consist of a dimer of subunits (61). In exchange for K+, these enzymes pump either H+ or Na+ across the membrane. In the kidney, these enzymes are located on opposite sides of the cells in the collecting duct. The Na+, K+ ATPase is localized to the basolateral side and pumps Na+ into the bloodstream in exchange for K+ being brought into the cell. On the apical membrane, the H+, K+ ATPase brings K+ into the cell from the filtrate while pumping H+ into the collecting duct. The subunit of these so-called X+, K+ ATPases contains ten transmembrane helices and is greater than 100 kDa (Figure 1-4). More than a third of this subunit is comprised of a large cytoplasmic loop between transmembrane helices four and five that houses the phosphorylation site and the nucleotide-binding pocket. The subunit also contains the ion transport channels.

PAGE 35

14 The subunit is a single transmembrane protein containing a short N-terminal intracellular domain and a large, glycosylated extracellular domain (Figure 1-4). The extracellular domain contains six highly conserved cysteine residues that comprise three disulfide bridges. Mutational studies have demonstrated that each of the extracellular cysteines is required for the subunit to be able to form an active enzyme complex with Figure 1-4. The H+, K+ ATPase. The enzyme transports H+ out of the cell in exchange for bringing K+ into the cell. The ten transmembrane containing subunit is shown in blue. The subunit houses the ion transport channels and it contains a large cytoplasmic loop between transmembrane domain 4 and 5 that houses the nucleotide binding pocket and the phosphorylation site. The single transmembrane subunit is shown in orange. It has a large extracellular region that contains, in the case of gHK, pictured here, seven glycosylation sites, indicated by blue circles. Three disulfide bridges (blue lines) are found in the extracellular region of all XK subunits.

PAGE 36

15 the subunit (65). The XK subunit has been shown to be necessary for ER exit and proper membrane insertion of the enzyme. The C-terminal transmembrane helices (M5-M10) of the subunit do not insert into the membrane efficiently in the absence of the subunit as indicated by trypsinolysis studies of the enzyme. Association of the subunit with the C-terminal region of has been demonstrated by yeast-two-hybrid and mutagenesis studies; it is thought that this interaction facilitates the correct membrane insertion and packing of the transmembrane helices (see Geering (44) for review). In addition, the subunit is responsible for the structural and functional maturation of the subunit and trafficking of the enzyme to the plasma membrane (11, 45). It has also been shown that the subunit influences the K+ affinity of the X+, K+ ATPases. This effect is likely due to conformational changes brought about by association that would necessarily affect the enzymes ability to bind and occlude cations (27, 44). Regulation The P-type ATPases are regulated on many levels (110). The concentration of the enzymes substrates, including the transported ions and ATP, of course plays a role in activation or inhibition of these ion pumps. Hormonal regulation of these enzymes occurs at the transcriptional and posttranscriptional levels. As discussed above, hormones activated in response to ion concentrations can affect the P-type ATPases by increasing gene expression. Hormone action can also lead to activation of kinases that, in turn, influence the activity of the ion pumps. Co-translational mechanisms of control are also present and include quality control checks in the ER. In addition, regulation can occur posttranslationally in the form of accessory proteins or by phosphorylation of the

PAGE 37

16 subunit. Interaction of such proteins with the Na+, K+ ATPase is the best and most characterized example of such regulation. The effect of phosphorylation on the P-type ATPases has been a controversial issue in the field (110). Several ser/thr kinases are capable of phosphorylating the enzyme, and their recognition sequences normally lie in the N-terminus of the subunit. These kinases include PKA, PKC and PKG. Studies with the Na+, K+ ATPase have suggested that enzyme activity is usually inhibited by phosphorylation, but evidence for the contrary is equally convincing. It would appear that the effect of phosphorylation is tissue and sometimes, even species specific. The presence of a subunit for the Na+, K+ ATPase has been known for some time. However, a more complete understanding of the role that this and other related proteins play in the control of Na+, K+ ATPase activity is just beginning to emerge (43). The NaK subunit is a member of a family of small, single transmembrane proteins termed FXYD for a common motif they possess in their extracellular domain. There are seven members in all, though not all have been studied in depth. The known members include phospholemman (FXYD1), NaK (FXYD2), Mat-8 (FXYD3), CHIF (FXYD4), and RIC (FXYD5). Two more members, FXYD6 and 7, are also known. CHIF, or corticosteroid hormone-induced factor, was identified first as an aldosterone-induced mRNA (9). It is expressed in the collecting duct of the kidney and also the distal colon. NaK is expressed primarily in the renal tubule while phospholemman has been localized to the extraglomerular mesangial cells of the juxtaglomerular apparatus and the afferent arteriole in the kidney. Phospholemman is expressed in the heart and skeletal muscle as well while FXYD7 is specific to the brain (26, 43, 106). The specific distribution of this

PAGE 38

17 family of proteins ensures that activity of the Na+, K+ ATPase is regulated based on the changing needs of individual tissues, as the FXYD proteins have been shown to affect the ion affinity of and transport by the NaK pump (26, 111). Mechanism of Catalysis The P-type ATPases are distinguished from other ATPase families because of the formation of a high-energy phosphorylated intermediate (84). These enzymes become phosphorylated at an aspartic acid residue within the highly conserved sequence DKTG. Phosphorylation drives a conformational change in the enzyme from an E1, or high affinity state, to an E2, or low affinity state. Ion translocation accompanies this conformational change. Following dephosphorylation, the enzyme returns to the E1 state. Figure 1-5 summarizes the catalytic cycle of the H+, K+ ATPase. Following release of K+ into the cell interior, the enzymes intracellular gates are open, allowing intracellular H+ to bind. In the presence of Mg+2, the E1-H+ form binds ATP. Following phosphorylation, the change from E1 to E2 takes place. This occurs as the intracellular gates close and a conformational change takes place to allow the release of H+ into the collecting duct. After the extracellular gates open to release H+, binding of extracellular K+ can take place. Hydrolysis then occurs to release inorganic phosphate from the enzyme. The E2, K+ bound form of the ATPase then undergoes another conformational change back to the E1 state as K+ is released inside the cell. The mechanism of the other Type II P-type ATPases is very similar to that of the H+, K+ ATPase, with some exceptions. The Ca+2 ATPase transports a single cation unidirectionally. The H+, K+ ATPase transports H+ and K+ in opposite directions in a 1:1 ratio while the electrogenic Na+, K+ ATPase transports three Na+ ions for every two K+ ions.

PAGE 39

18 Figure 1-5. Mechanism of Catalysis of the H+, K+ ATPase. The position of the membrane is indicated. After binding of H+, the enzyme becomes phosphorylated and moves from the E1 to the E2 conformation as H+ is occluded. Following release of H+, the enzyme is able to bind K+. Hydrolysis of Pi follows as K+ is brought into the cell (121). Structure A major stride was made in understanding the structure and function of Type II P-type ATPases when a 2.6 resolution crystal structure of the sarcoplasmic reticulum Ca+2 ATPase in the E1 conformation was published (Figure 1-6A)(113). It was determined that the structure contained ten transmembrane helices. This was a very important observation because previous structural data had provided inconclusive evidence concerning the exact number of transmembrane helices in P-type ATPases. It was previously thought that they contained anywhere from seven to ten transmembrane segments (116).

PAGE 40

19 The Ca+2 structure determined by Toyoshima et al. (113)contained three cytoplasmic domains, designated P, N, and A. Both the P and N domains are located in the large cytoplasmic loop between transmembrane helices four and five. The P domain contains the highly conserved phosphorylation site, aspartic acid 351. This residue lies at the C-terminal end of the central -strand in a Rossman fold. A Rossman fold consists of a seven-stranded parallel -sheet with eight short associated helices. This region in the P domain had been predicted to be homologous to the core domain of L-2-haloacid dehalogenase (HAD) based on sequence analysis; the crystal structure of the Ca+2 ATPase revealed that this domain was in fact nearly identical to that of HAD. The P domain consists of two parts that are widely separated by the N domain. The N domain is the largest cytoplasmic domain and it contains the nucleotide-binding site. The ATP binding pocket was examined using Fourier difference maps in the presence and absence of the ATP analog TNP-AMP. A well-characterized and well-conserved lysine at residue 515 was thought to lie at the mouth of the ATP binding site, but analysis of the crystal structure showed that it is actually located in the depth of the nucleotide-binding pocket. The smallest of the cytoplasmic domains, the A domain, so named because it probably functions as an actuator, is thought to undergo substantial movement during active transport. It is comprised of the cytosolic N-terminus of the enzyme and the loop located between transmembrane helices two and three (L23). The N-terminal forty amino acids form two short helices while L23 resembles a distorted jelly roll. The A domain also contains a well-conserved TGES motif (113).

PAGE 41

20 A further breakthrough in understanding the mechanism and structure of P-type ATPases was made upon publication of a 3.1 crystal structure of the Ca+2 ATPase in the E2 state (Figure 1-6B)(114). The differences between the E1 and E2 structures are striking and lend significant insight into the mechanism of this class of enzymes. Dramatic rearrangements of the cytoplasmic and transmembrane domains of the ATPase take place as the enzyme moves from the E1 to the E2 configuration. The open conformation of the E1 state gives way to a more compact structure in the E2 state. The E1 structure (113) had revealed that the phosphorylation site, D351, and the nucleotide-binding pocket were separated by more than 25 so it was known that large domain movements would be necessary in order for phosphorylation of the enzyme to take place. This change in conformation involves a 90 movement of the N domain with respect to the membrane and a 110 horizontal rotation by the A domain. These movements mean that the N and P domains shift by 50 and 23 respectively. Figure 1-6 shows a side-by-side view of the E1 and E2 structures that illustrates the dramatic motion of the cytosolic domains of the enzyme. The smallest of the three cytoplasmic domains, the A domain plays an important role in the E1 to E2 transition as it appears that the E2 structure is stabilized at the interfaces of the A to N and the A to P domains. The transmembrane domain (M domain) also undergoes significant rearrangement during the E1 to E2 transition. Transmembrane helices one and two (M1 and M2) shift upwards toward the cytoplasmic side of the enzyme while M3 and M4 move downward. M5 is the longest of the transmembrane helices and has been described as the central mast of the enzyme (113). The cytoplasmic end of M5 runs into the P domain and, along with M4, is part of the Rossman fold in that region. M4 and M5 move in concert with the

PAGE 42

21 Figure 1-6. The Ca+2 ATPase in the E1 and E2 Conformations. (A) The Ca+2 ATPase in the E1 (Ca+2 bound) state is shown here. (B) The E2 state is pictured. The N and C termini are labeled. The approximate positions of the A (actuator), N (nucleotide-binding), P (phosphorylation) and M (transmembrane) domains are indicated by dashed lines. The loop between transmembrane domains 7 and 8 (L78) is labeled for both the E1 and E2 states for orientation purposes. The Ca+2 ions bound in the E1 state are indicated, as is the molecule of thapsigargin in the E2 state. Figure was created using Protein Explorer (http://molvis.sdsc.edu/protexpl/frntdoor.htm) (last accessed February 1, 2004). P domain during catalysis. A complex hydrogen bond network links the M and P domains and, together with L67, provides a conduit for transmitting a signal for conformational changes upon ion binding by the M domain and phosphorylation of the enzyme.

PAGE 43

22 Of the four transmembrane helices that make up the two cation binding sites, M4-6 and M8, only M8 does not move during catalysis. As the enzyme transitions from the E1-Ca+2 state to the E2 state, in the absence of Ca+2, the ion binding sites are destroyed by the motion of the M domain. As M4 moves downward, M5 bends toward M4 and the unwound portion of the M6 helix rotates by nearly 90. The necessity for such large motions becomes clear when ion binding and phosphorylation are considered together. The enzyme is capable of binding ATP when Ca+2 is not present. However, phosphorylation cannot occur until ion binding takes place. The binding of Ca+2 causes rotation of the A domain and disruption of the A-N interface. Release of the N domain by the A domain allows the enzyme to enter into the closed conformation as the P and N domains come together. This movement allows the transfer of the -phosphate of the bound ATP to D351 (115). Publication of the E1 and E2 structures of the Ca+2 ATPase gave rise to a number of modeling exercises undertaken by groups working on other P-type ATPases, the X+, K+ ATPases, our own work (see Gumz et al. (51) and Chapter 3) included. In particular, in a review article by Sweadner and Donnet (107), the sequences of the Na+, K+ ATPase and the gastric H+, K+ ATPase were mapped onto the E1 Ca+2 ATPase structure. After evaluating extensive biochemical evidence, including cysteine mutagenesis and proteolytic digestion, in combination with the crystal structure data, it was concluded that it was highly likely the X+, K+ ATPases shared the same general structure. Another group undertook a more involved modeling exercise, generating a restraint-based model of L45 of the Na+, K+ ATPase based on the E1 Ca+2 structure (35). The P and N domains contained within this loop were found to share a similar architecture with the Ca+2 pump;

PAGE 44

23 a Rossman fold was formed within the P domain and the phosphorylation residue (D369 in the Na+, K+ ATPase) was found to lie at the C-terminal end of the central -sheet, as predicted by the Ca+2 structure. Importantly, the distance between the nucleotide-binding site and the phosphorylation site was measured to be about 25 a value directly comparable to that observed in the Ca+2 structure. Extensive modeling of the gastric H+, K+ ATPase, based on the Ca+2 ATPase structure, has been performed as well. In fact, a recent publication used gastric H+, K+ ATPase models in both the E1 and E2 conformations to evaluate the docking of K+-competitive inhibitors in the pump (8). Specifically this group looked at the binding of SCH28080, a gastric pump specific antagonist that binds the enzyme reversibly when it is in the E2 form. Using kinetic and mutagenesis studies, they focused on Y801 in M5 and found that mutants at this residue showed 60 to 80 times lower sensitivity to SCH28080 than the wild type enzyme. Y801 was found to face a cavity formed by M1, M4-M6, M8 and L56, L78, and L910 in the E2 structure. It was hypothesized that this cavity forms the binding site for SCH28080 and in fact the compound was docked in the cavity, using computer docking programs. Further evidence for this site binding SCH28080 is that the docking could not occur in the E1 structure, a fact that is consistent with biochemical evidence. The Rabbit Colonic H+, K+ ATPase HK Isoforms in the Kidney The presence of an H+, K+ ATPase in the kidney was first reported when acidification and potassium absorption were measured in the outer medullary collecting duct of rabbits fed a low potassium diet (126). Two H+, K+ ATPase subunit isoforms

PAGE 45

24 have been found in the kidney; the HK1, or gastric, isoform, and the HK2, or colonic, isoform. RT-PCR and in-situ hybridization revealed the presence of HK1 in the rat kidney (2). HK2 was cloned from a rat colon cDNA library and its expression in kidney was shown by Northern blot analysis (28). Although it was first cloned from the distal colon, a large body of evidence suggested that HK2 plays an important role in potassium conservation by the kidney. For example, it has been shown that HK2 message, but not HK1 message, was up regulated in rats fed a low potassium diet (32). Our interest lies in the so-called colonic H+, K+ ATPase, or the HK2 isoform. Physiological Significance of HK2 The best evidence for the importance of the HK2 isoform comes from the role it plays in potassium homeostasis. It has been well established that mRNA levels of the HK2 subunit undergo a substantial increase under low K+ conditions (3, 32, 62). The subsequent increase in the number of active pumps at the apical membrane of collecting duct cells results in increased K+ reabsorption. Therefore, the action of the H+, K+ ATPase is vital in maintaining K+ levels in the 3.5 to 5.5 mEq range. As discussed above, K+ concentrations outside this range can be fatal. When K+ levels rise above 5.5 mEq, hyperkalemia can result. This often occurs in patients on total renal support or those taking certain medications. Hyperkalemia can also result when mechanisms of K+ excretion are inhibited. Because K+ is so important to the maintenance of cellular membrane potential, high levels of K+ perturb voltage levels. Muscle contraction is dependent on membrane potential; thus, hyperkalemia can cause muscle weakness and fatigue. Smooth muscle is also affected and so high K+ levels can lead to respiratory

PAGE 46

25 depression. The worst effect of hyperkalemia is on the heart and can result in potentially fatal cardiac arrhythmias (124). HK2 plays a more important role when K+ levels drop below 3.5 mEq, a condition known as hypokalemia. This condition is most prevalent in patients taking diuretics, but patients with secondary hyperaldosteronism and other renal diseases are at risk as well. The dangers of hypokalemia are evident when its far-reaching implications are considered. Low K+ levels cause a decrease in insulin release, which exacerbates hyperglycemia in diabetic patients. Hypokalemia affects the muscles, causing a decrease in the ability to contract and also decreased blood flow. Other effects of hypokalemia include polyuria, renal cyst formation and hepatic encephalopathy. Low K+ levels perturb acid-base balance as well, which can lead to metabolic alkalosis. This may in part be due to the action of the upregulated H+, K+ ATPase reabsorbing K+ into the cells of the collecting duct; the concomitant movement of H+ out of the cell contributes to intracellular alkalinization. As with hyperkalemia, one of the most dangerous aspects of hypokalemia is its effect on the cardiovascular system. Hypokalemia-related hypertension and ventricular arrhythmias both play a part in the morbidity and mortality associated with hypokalemia (124). While the action of the HK2 isoform may play a role in the disorders discussed above, no specific disorder is associated with the enzyme itself. However, a case study of a novel form of distal renal tubular acidosis suggested that a deficiency in the colonic isoform of the H+, K+ ATPase was responsible for the symptoms (99). A 21-month-old male child suffering from severe hypokalemia, metabolic acidosis, and hypomagnesaemia was evaluated until the age of three. Alkali therapy combined with 10

PAGE 47

26 mmol/kg of potassium citrate per day was required to bring serum K+ levels into the normal range. A defect in the gastric H+, K+ ATPase was ruled out and it was concluded that the novel form of distal renal tubular acidosis was likely due to impaired function of the colonic H+, K+ ATPase. Further evidence for the importance of HK2 comes from studies with knockout mice. Once HK2 null mice were generated, their ability to retain K+ was studied under control or low K+ conditions (75). Under control conditions, no dramatic differences were observed between wild type and the HK2-deficient mice. Once the mice were fed a K+-free diet for 18 days however, the knockout mice lost four times more fecal K+ than the wild type animals. The HK2 null mice also lost twice as much body weight and had lower plasma and muscle K+ levels than the wild type mice. From these observations, it was concluded that the colonic H+, K+ ATPase has a pivotal role in maintaining K+ balance. A subsequent study involving the HK2 knockout animals looked at the effect of dietary Na+ depletion in these mice versus wild type mice (102). The HK2 deficient mice had a higher level of K+ excretion than the wild type animals when fed a Na+ restricted diet. It was also observed that, regardless of diet, the knockout mice exhibited lower levels of ENaC-mediated short circuit current. It was concluded that, not only does HK2 play an important role in K+ conservation, but it also plays an important role in Na+ absorption as part of a coordinated system of transporters in the collecting duct. While these investigators found no evidence to suggest a direct role for the colonic H+, K+ ATPase in Na+ reabsorption, other groups have suggested that this enzyme is capable of transporting Na+ in exchange for K+, as discussed below.

PAGE 48

27 Cation Selectivity As discussed earlier, the subunit of the X+, K+ ATPases influences the ion affinity of the subunit. Because the proper subunit for the colonic H+, K+ ATPase has not been determined (see below), it has been difficult to study this property of the enzyme. However, it is possible that the HK2 enzymes may sometimes transport Na+, rather than H+, in exchange for K+. Several groups have demonstrated that the colonic H+, K+ ATPase is capable of exchanging Na+ for K+, in heterologous expression systems as well as animal tissue. Studies in Xenopus oocytes showed that cells expressing the colonic H+, K+ ATPase had similar intracellular Na+ activity as those cells expressing the Na+, K+ ATPase, and that this effect was K+-dependent and ouabain inhibitable (24). A similar result was seen in HEK 293 cells expressing the human colonic H+, K+ ATPase (50). Experiments with rat colon tissue have also suggested that the colonic H+, K+ ATPase exchanges Na+ for K+. Two groups measured Na+-dependent K+ ATPase activity in the apical membranes of this tissue and used HK2 specific antibodies to show the specificity of the activity (21, 89). A third group studied Na+ transport in rabbit macula densa cells from the kidney, a cell type that does not express detectable levels of the Na+, K+ ATPase enzyme. This study concluded that intracellular Na+ levels in this cell type were regulated by the colonic H+, K+ ATPase (87). HK2a and HK2c Degenerate PCR and Rapid Amplification of cDNA Ends (RACE) were undertaken in our laboratory by Dr. Grady Campbell to clone the HK2 cDNA from rabbit kidney (16). These experiments yielded two rabbit renal cDNAs, designated HK2a and HK2c (Figure 1-7).

PAGE 49

28 Figure 1-7. Rabbit HK2 Isoforms. (A) The HK2 gene contains two transcription start sites, one that gives rise to HK2a (red arrow) and a second that gives rise to HK2c (orange arrow). The splicing for HK2a is indicated by the dark gray lines. Light gray lines indicate intron sequence. Exon 1 is colored blue, exon 2 is green, and the remaining exons are purple. (B). The resulting mRNAs are greater than 4000 bp. Black lines indicate the 5 and 3 UTRs. (C) The HK2a and HK2c transcripts are translated into two different proteins that differ only at their N-termini. HK2c contains a 61 amino acid extension at its N-terminus (green), while HK2a contains two different amino acids at its N-terminus (blue). The C-termini of the two proteins are identical (purple). The putative PKA and casein kinase II phosphorylation sites in the N-terminal Region of HK2c are indicated by the lavendar circles. HK2a and HK2c shared approximately 4 kb of sequence in common and differed only at their 5 ends. The deduced amino acid sequences revealed that HK2c has a 61 amino acid N-terminal extension. The two transcripts give rise to two different proteins that are detectable and separable on a Western blot using membrane protein samples from a rabbit cortical collecting tubule cell line (RCCT-28A). Analysis of the deduced amino acid sequence of HK2c revealed the presence of putative phosphorylation sites in the N-terminus for PKA and casein kinase II.

PAGE 50

29 The existence of the HK2c transcript has been called into question by other investigators (37) but further evidence based on Dr. Campbells work has refuted this claim. Specifically, transcription start site experiments (see below) and immunohistochemical studies have confirmed the existence of HK2c. Work by Dr. Jill Verlander using an HK2c specific antibody demonstrated that the protein was present throughout the nephron (117). Specifically, immunostaining was observed in cells of the connecting segment, the CCD, the outer and inner stripe of the OMCD, the middle third of the IMCD, the cortical TAL, and the MD. Principal cells showed only minimal staining while, as expected, the most intense apical staining occurred in both and intercalated cells. The HK2a and HK2c cDNAs arise from two independent transcription start sites, as evidenced by RNase protection experiments conducted in our laboratory by Dr. Deborah Zies (130). In order to glean information concerning the regulation of these two transcripts, Dr. Zies conducted further experiments using the promoter common to both HK2a and HK2c. An extensive deletion analysis of the HK2 promoter using the luciferase assay system showed that the gene is under negative regulation in RCCT-28A cells. This result, coupled with RT-PCR experiments to detect the HK2 transcript in confluent versus non-confluent RCCT-28A cells, led to a model regarding expression of the HK2 gene (Figure 1-8). It is hypothesized that when RCCT-28A cells are not confluent, the HK2 gene is in a repressed state. Once the cells become confluent, the gene is expressed at a basal level. Increased transcription from the gene would be observed when an activating signal, such as that delivered under low K+ conditions, is received. Further investigation into the

PAGE 51

30 Figure 1-8 Model for HK2 Gene Expression. (A) When cells are not confluent, the gene is repressed, RNA polymerase (RNA pol) cannot bind, and transcription does not occur. (B) When cells become confluent, a basal level of transcription occurs as transcription factors such as Sp1 family members (S) bind. (C)When an activation signal is received (such as low K+ levels), activating transcription factors bind and an increased level of transcription occurs.

PAGE 52

31 mechanisms behind this regulation is needed, but a picture of the complex regulation of the HK2 gene is beginning to emerge. It is one of the aims of this project to functionally express HK2a and HK2c in order to determine their respective pharmacological and activity profiles. The H+, K+ ATPase is probably a tetramer of two and two subunits. In order to express a functional enzyme, the subunits must be co-expressed with a subunit. The subunit that pairs with the HK2 isoform has not been clearly identified but four mammalian X+, K+ ATPases subunits are known. They are Na+, K+ ATPase 1, 2, and 3 and the gastric H+, K+ ATPase These subunits all contain a single transmembrane domain. The extracellular domain of the subunits contains three disulfide bridges and is extensively glycosylated. The subunit functions to traffic the enzyme to the membrane and may also play a role in determining potassium affinities (19). Functional expression systems using the gastric H+, K+ ATPase gHKthe Na+, K+ ATPase 1 (NaK1) and the Na+, K+ ATPase (NaK) subunits co-expressed with an subunit are discussed below. Expression Systems Mammalian cell culture expression systems have been established for human (49), guinea pig (6), and rat (96) colonic H+, K+ ATPase. ATP1AL1, the human H+, K+ ATPase cloned from skin, shares approximately 90% similarity with HK2 isoforms and is thought to represent the human colonic isoform (17). Grishin et al. (49) reported an HEK293 cell line stably transfected with cDNAs for ATP1AL1, the gastric H+, K+ ATPase subunit, and Na+, K+ ATPase 1. These investigators reported measurable activity only for the cells expressing ATP1AL1 and gHK. NaK1 was not an effective

PAGE 53

32 mate for ATP1AL1. Asano et al. (6) reported an HEK293 cell line transiently transfected with cDNAs for guinea pig HK2 and either the gHK subunit or the NaK1 subunit. H+, K+ ATPase activity was reported when HK2 was coexpressed with gHKActivity was also observed when HK2 was coexpressed with NaK1, a result that is inconsistent with the Grishin et al. study. Sangan et al. (96) reported stable transfection of an HEK293 cell line with cDNAs for rat HK2 and either rat NaK1 or NaK3. H+, K+ ATPase activity was observed when HK2 was coexpressed with both NaK and NaK subunits. The Grishin et al. study is at odds with work done by both Sangan et al. and Asano et al. in that the latter two were able to report activity when NaK1 was used. Both Sangan et al. and Asano et al. demonstrated functional expression with NaK1, but the two reports of ouabain resistance differed by three orders of magnitude. It is not clear why different activity and pharmacological profiles have been observed for these HK2 isoforms. Some investigators (49, 96) speculate that the HK2 cDNAs encode different H+, K+ ATPases. However, the human, rat and guinea pig HK2 isoforms all share roughly 87% amino acid sequence identity and 92% similarity. It is difficult to imagine that such high sequence homologies do not represent actual isoforms of the same protein. Indeed, a distance analysis showed that these isoforms share a common ancestor (17). It is much more likely that the contrasting profiles are due to the manner in which these isoforms are expressed and assayed for activity. It is known that the subunit of the X+, K+ ATPases plays an important role in regulation of the enzyme (19). Since the subunit for the colonic H+, K+ ATPase has not been clearly defined; it may be that the contrasting profiles discussed above are due to the different subunits used to express the H+, K+ ATPase (25).

PAGE 54

33 Attempts to establish a functional expression system for the rabbit renal H+, K+ ATPase were made in our laboratory by Dr. Tamaro Otto (83). Dr. Otto first tried to express HK2a and HK2c along with NaK1 using an episomal expression system. HEK293 cells were transfected with two different constructs containing cDNAs for the and subunits. This approach was abandoned because it wasnt possible to control for equal expression between the two episomal constructs. Stable transfection of HEK293 cells was then undertaken. Cells were transfected with a single construct that allowed for simultaneous expression of both and cDNAs. Stable selection and Western blot analysis of clones followed. While several clones expressing NaK1 were isolated, clones expressing HK2a or HK2c were never found. Because Northern blot analysis revealed that full-length transcripts were being made for HK2a and HK2c, it was concluded that the absence of subunit protein might have been due to instability of the protein. Dr. Ottos final approach was to transiently transfect COS-1 cells with the same constructs used for stable transfection in the HEK293 cells. In addition to NaK1, NaK3 was also used in the transient transfections. HK2a and HK2c, as well as NaK1 and NaK3, were detected by Western blot analysis. ATP hydrolysis assays were performed to assess the activity of these H+, K+ ATPases. Although all the necessary proteins were present, no activity above background was detected. It was concluded that functional expression of the rabbit H+, K+ ATPase was not successful in our lab perhaps because the proper subunit was not being used (83). The current controversy surrounding the identity of the subunit in the colonic H+, K+ ATPase is discussed below.

PAGE 55

34 The Subunit Controversy Pairing of the and subunits of the X+, K+ ATPases was originally thought to be promiscuous (72). Mounting evidence concerning the colonic H+, K+ ATPase however, suggests that the functionality of this ion pump depends on pairing with the proper subunit. Although the identity of the subunit partner for HK2 has not been definitively demonstrated, reports link HK2 to any one of the four known mammalian X+, K+ ATPase subunits. The expression systems discussed above suggested that HK2 was able to form functional, stable complexes with gHK (6, 49), NaK1 (6, 96), and also NaK3 (96). Subsequent studies were directed to ascertain the identity of the colonic H+, K+ ATPase subunit in vivo. Two independent laboratories performed co-immunoprecipitation experiments in order to identify the subunit that associates with HK2 in vivo (20, 67). Both studies used a polyclonal antibody directed against amino acids 686 to 698 of the published rat HK2 sequence. Both groups performed immunoprecipitation of HK2 from the kidneys of potassium-depleted rats; Codina (20) used distal colon tissue as well. Kraut (67) analyzed the immunoprecipitate by Western blot using antibodies against gHK and NaK1 while Codina only looked for the presence of NaK1. In both cases, the presence of NaK1 was confirmed, suggesting that this subunit associates with HK2 in the kidney. However, the data from both studies are unconvincing. Western blots from both publications appear to be of poor quality. In addition, important negative controls were missing from both studies.

PAGE 56

35 Adding further confusion to the field was a report of the identification of an actual colonic subunit. Sangan et al. (95) isolated a subunit cDNA from a rat colon cDNA library that was termed the colonic H+, K+ ATPase subunit. This cDNA was in fact 100% homologous to the known rat NaK3 subunit (72). It was termed the H+, K+ ATPase colonic subunit based on the following observations: the colonic subunit was co-immunoprecipitated with HK2 from distal colon, its message was up-regulated in the distal colon of potassium depleted rats, and colonic protein levels were increased in these rats in the apical membrane only, an observation consistent with association of the subunit with HK2 and not with the basolaterally located Na+, K+ ATPase subunit. Additionally, an antibody against NaK1 did not react with the HK2 immunoprecipitate, a result in direct conflict with those discussed above (20, 67). In order to clarify the confusion concerning the identity of the colonic subunit, Geering et al. (46) used a Xenopus oocyte expression system to assess the stability of the H+, K+ ATPases formed by expression of the four known mammalian X+, K+ ATPase subunits with the HK2 subunit. Because co-expression with a subunit is required for the structural maturation of the X+, K+ ATPase subunits, a proteolysis assay was employed to observe the effects of the various subunits on the folding of HK2. In addition, the stability of association of the / complexes was assayed by treatment with detergent. These investigators showed that, of the mammalian X+, K+ ATPase subunits, only NaK2 and gHK were able to form stable complexes with HK2. Although NaK2 and gHK formed the most stable complexes with HK2, it was noted that these complexes were still highly sensitive to trypsin, indicating that while / association was

PAGE 57

36 occurring, correct folding of the enzyme was not. Thus, these investigators concluded that the real colonic subunit had not yet been identified. Based on investigation into the human genome project, it would appear that no other X+, K+ ATPase -like open reading frames exist, other than those already identified. Therefore, it appears as if one of the four known mammalian X+, K+ ATPase subunits is likely to be the proper partner for HK2. The Xenopus system used by Geering et al. was the first effort to examine all four subunits in parallel. This was a much-needed investigation into this area. However, the differences between the oocyte system and mammalian cell culture systems must be noted. Xenopus oocytes are grown at 30C, while mammalian cells are grown at 37C, which is physiological temperature for the colonic H+, K+ ATPase. ER quality control mechanisms that would otherwise prevent exit of improperly folded proteins are inhibited in Xenopus oocytes when proteins are being overexpressed. In addition, Xenopus oocytes employ a mechanism to regulate the number of exogenous cell-surface ion pumps with respect to endogenous ion pumps. This could lead to cell surface expression of even incorrectly assembled pumps (46). Thus, a mammalian cell culture system is a better candidate for studying the colonic H+, K+ ATPase. Our own investigation into this area is described in Chapter 5. Summary The important role of the colonic H+, K+ ATPase in maintaining potassium homeostasis and acid-base balance in the body has been well established. However, the controversy surrounding the identity of the subunit of the colonic H+, K+ ATPase remains unresolved. It has become necessary to undertake a parallel study of all four known X+, K+ ATPase subunits similar to that performed by Geering et al., but in a

PAGE 58

37 mammalian cell culture system. The goal of this work was to establish a mammalian cell culture expression system in order to perform structure/function studies of the colonic H+, K+ ATPase. At the same time, in order to gain a better understanding of aldosterone action in the kidney, we performed a microarray analysis to determine the early transcription effects of aldosterone on the inner medullary collecting duct. Taken together, the data presented in this dissertation provide a more complete picture of ion transport and its hormonal regulation in the kidney.

PAGE 59

CHAPTER 2 MATERIALS AND METHODS Molecular Modeling Choice of Templates and Alignment The template selected for the modeling exercise was the rabbit sarcoplasmic reticulum Ca+2 ATPase (113). The primary amino acid sequence of the HK2a subunit (Genbank AAB80941.1) of the rabbit colonic H+, K+ ATPase was aligned with that of the Ca+2 ATPase (Genbank P04191) and the NaK1 subunit (Genbank NP_036636) of the Na+, K+ ATPase using Clustal W (112) with default settings. The gap penalties used were a gap opening penalty of 10 and a gap extension penalty of 0.05. Modeling The PDB coordinates for the Ca+2 ATPase (1EUL) were downloaded from the Brookhaven protein database (http:// www.rcsb.org ) (last accessed Febuary 1, 2004) and loaded into the program O (63). Individual amino acids of the Ca+2 ATPase were mutated to match that of the primary sequence of HK2a according to the pairwise sequence alignment. Insertions and deletions with respect to the Ca+2 ATPase sequence were made interactively and energy minimized locally using the graphics program O, version 7 (63). The transmembrane domain predictor program TMHMM (68) and the program O (63) were used to build an helix representing the single transmembrane domain of NaK2. Model Refinement and Validation The entire model was globally energy minimized using the CNS software package (14). This minimization was also performed on the model of NaK2. The program 38

PAGE 60

39 PROCHECK (69) was used to investigate the stereochemistry of the model. The coordinate files for the HK2a model and the Ca+2 ATPase were loaded into the program and used to generate secondary structure predictions and stereochemical statistics. Charge Distribution Surface potential energy of the HK2a model and the Ca+2 ATPase were generated using the program GRASP (79). Parameters were set to .566 and 8.501 for HK2a and .965 and 8.493 for the Ca+2 ATPase. Secondary Structure Prediction Secondary structure elements of the HK2a transmembrane domain were visualized using Bobscript (34). Data from the Ca+2 ATPase crystal structure were used in conjunction with the amino acid sequence alignment to assign the locations of the transmembrane helices in HK2a. Recombinant DNA Techniques Preparation and Handling of Plasmid DNA Plasmid DNA was purifed using the Qiagen Maxi Prep kit or Mini Prep kit. Restriction endonucleases were purchased from New England Biolabs. Electrophoresis was performed using 1% agarose (Fisher) gels in 1X TAE buffer (40 mM Tris, 1 mM EDTA, plus acetic acid to a pH of 7.9). All nucleotide sequencing reactions were carried out by the Interdisciplinary Center for Biotechnology Research at the University of Florida. Synthesis of oligonucleotides was performed by Invitrogen. Bacterial Transformations Three different strains of E. coli bacteria were used for this dissertation project, DH5 (Invitrogen), XL1 Blue (Stratagene) and TOP 10 (Invitrogen). The majority of

PAGE 61

40 plasmid transformations were performed using DH5 cells. Ten microliters of a ligation reaction or 1 l of plasmid was incubated with 25 l of competent DH5 (Invitrogen) on ice for at least ten minutes. The cells were then heat shocked for 30 sec at 37C and allowed to recover on ice for 2 min. For plasmid transformations, cells were plated immediately following the heat shock on LB agar plates containing the proper antibiotic. For ligation transformations, 250 l of LB media without antibiotic was added to the cells and they were allowed to grow at 37C for 1 hr on a roller drum. Transformations into XL1 Blue cells were performed only after QuikChange mutagenesis (see below). Competent XL1 Blue cells were incubated with 2 l of -mercaptoethanol (-ME) on ice for 10 min. Plasmid was added and allowed to incubate for at least 10 min on ice. Heat shocking was performed at 42C for 30 sec, followed by a 2 min recovery on ice. Cells were then grown at 37 C for 1 hr on a roller drum before being plated on an LB agar plate containing the appropriate antibiotic. TOP 10 cells were used for transformations following ligations using the TOPO cloning kit (see below). Four microliters of the ligation reaction were added to 25 l aliquots of competent cells and allowed to incubate for 10 min on ice. Cells were heat shocked at 42C for 30 sec and then allowed to recover on ice for 2 min. Growth at 37C for 1 hr on a roller drum was carried out in 250 l of SOC medium that was provided by Invitrogen. Plasmid Constructions Two types of ligations were carried out for this project. Either a highly concentrated preparation of DNA ligase (HC DNA ligase, Invitrogen) was used or PCR products were cloned into pCR2.1 using the topoisomerase (Invitrogen) attached to the

PAGE 62

41 vector. Ligations using HC DNA ligase were performed in a total volume of 20 l. The reaction was carried out overnight at 16C in a 1X DNA ligase buffer plus 10 fmol of vector. Insert was added at 3X and 10X the molar vector concentration. Vector and insert were always gel purified from 1% agarose gels using the Qiagen gel extraction kit prior to the ligation reaction. An example of this type of ligation is the construction of plasmid pMLG97, described in Figure 2-1. PCR amplification of inserts for the TOPO cloning vector is described in subsequent chapters. For the majority of pCR2.1 ligations, PCR was performed using the Taq polymerase, which adds A overhangs to its products. The plasmid pCR2.1 contains T overhangs and is connected to a topoisomerase to facilitate the ligation reaction. The TOPO cloning reaction was carried out in a total of 6 l, in a 1X buffer supplied by Invitrogen plus 1 l of salt solution, 4 l of insert (not to exceed 25 ng total), and 1 l TOPO. The insert was always gel purified from a 1% agarose gel using the Qiagen gel extraction kit. When using a higher fidelity polymerase was necessary, PCR of the pCR2.1 inserts was performed using Pfu Turbo (Stratagene). Following gel purification the Pfu-amplified product was incubated with dATP, 1X PCR buffer and 1 l of Taq polymerase to add on A overhangs, as recommended by Invitrogen. All TOPO ligation reactions were allowed to proceed at room temperature for 15 min. The transformation reactions were carried out as described above.

PAGE 63

42 Figure 2-1. Construction of Plasmid pMLG97. Plasmids pMLG84 and pMLG23 were digested sequentially with NotI and BstBI. The HK2a insert and the pMLG23 vector bands were gel purified and combined in a ligation reaction to generate pMLG97. Site-directed Mutagenesis All mutagenesis performed in this dissertation was carried out using the QuikChange XL kit from Stratagene, designed to mutagenize plasmids greater than 8 kb. The primers designed to perform QuikChange are described in subsequent chapters. QuikChange PCR reactions were set up as described by the manufacturer. Each reaction consisted of 1 l Pfu Turbo, 2 l dNTP (provided by Stratagene), 5 l of 10X buffer, 1-5 l template and H2O to a 50 l total volume. Varying amounts of template were used; 5 ng, 10 ng, 20 ng or 50 ng of the plasmid in question served as template. The cycling

PAGE 64

43 parameters for QuikChange PCR were as follows: 95C x 5 min; 18 cycles of 95C x 30 sec, 60C x 30 sec, 68C x 16 min; 68C x 7 min final extension and a final hold at 4C. Tissue Culture Cell Lines Cells from the green monkey kidney cell line, COS-1, were used for the expression studies described in Chapter 5. Cells between passages 20 and 40 were used for all experiments. Cells were grown in a T-75 flask (Corning) in DMEM media plus 10% FBS (Invitrogen) until they reached 90-100% confluency, whereupon they were plated in 100 mm dishes for transfection (see below). Mouse inner medullary collecting duct (mIMCD3) cells between passages 15 and 25 were used for the aldosterone experiments described in Chapter 6. Cells were grown in a T-75 flask (Corning) in DMEM-F12 media plus 10% FBS (Invitrogen) before being plated on collagen-coated Costar Transwell-COL inserts (Fisher) in DMEM-F12 plus 10% FBS for hormone studies. Hormone Treatment of mIMCD3 cells Mouse IMCD3 cells were grown one day past confluency on collagen-coated Costar Transwell-COL inserts (Fisher). Twenty-four hours before treatment with aldosterone, media was changed to phenol-red free DMEM-F12 (Invitrogen) plus 10% charcoal/dextran stripped FBS (Biosource International). Cells grown for this length of time in Transwell-COL inserts typically exhibit a transepithelial resistance of 703.3 +/24.7 Ohm/cm2 (Shen-Ling Xia, personal communication). Cells were treated for varying times with vehicle (ethanol) or aldosterone (Sigma) at concentrations ranging from 10-10 to 10-6 M. Aldosterone was dissolved in phenol red-free DMEM plus 10% charcoal stripped FBS plus 10% ethanol to a final concentration of 2.77 x 10-4 M. Vehicle

PAGE 65

44 consisted of the same media minus the aldosterone. Inhibitor studies were performed using 1 M mifepristone (RU-486) and/or 1 M spironolactone (Sigma). For the microarray experiment, cells were treated for 1 h with vehicle or 10-6 M aldosterone. Transient Transfection Plasmids were transiently transfected into the green monkey kidney cell line, COS-1, using the Fugene (Roche) transfection reagent according to the manufacturers directions. Cells were grown in DMEM 10 media and plated in 100 mm dishes. When cells reached 50 to 60 % confluency, transfection was performed. For transfection of COS-1 cells in 100 mm dishes, 10 l of Fugene reagent were combined with 200 l DMEM 10 in a 1.5 mL polypropylene tube. Three micrograms of plasmid DNA were added to the tube. A 20 min incubation at room temperature followed. When the incubation was complete, the DNA-Fugene mixture was added dropwise to the cells. Two days after transfection, plasma membranes were isolated. Transfection of mIMCD3 cells for the period homolog promoter studies described in Chapter 6 was performed using Fugene reagent as well. Cells were plated in 24 well dishes (Fisher) and transfected when they reached 50-60% confluency. DNA-Fugene mixtures were prepared in 3.3X batches in order to accommodate transfection of triplicate wells. Four microliters of Fugene reagent were combined with 83 l DMEM 10 in a 1.5 ml polypropylene tube. One microgram of experimental plasmid DNA was added along with 0.66 g of pRL-TK (see below for description of luciferase assay) was added to the tube. A 20 min incubation at room temperature followed. The DNA-Fugene mixture was then added dropwise to each well.

PAGE 66

45 Plasma Membrane Preparation To isolate plasma membranes, transfected COS-1 cells were washed twice in ice cold PBS and then scraped three times into 2 mL PBS (for a total of 6 ml per transfection). Cells were spun for 10 min at 1000 x g and then resuspended in 2 ml 10 mM Tris, 1 mM PMSF, pH 7.5. Resuspended cells were kept on ice for 10 min prior to homogenization. After twenty-five strokes, 2 mL of 10 mM Tris, 0.5 M sucrose, 1mM PMSF, pH 7.5 were added. Twenty-five more strokes were then performed. Cells were spun again for 10 min at 1000 x g to pellet nuclei. The supernatant was spun for 90 min at 100,000 x g to pellet the final membrane fraction. This membrane pellet was resuspended in 100 l of 5 mM Tris, 250 mM sucrose, pH 7.4, unless the membrane fraction was to be used for ATPase assays. In that case, the membrane pellets were resuspended in 100 l of a hypotonic solution (see below). Biochemical Assays BCA Assay To determine the protein concentration of the final membrane fraction, the bicinchoninic acid (BCA) assay was used. The standard working reagent (SWR) consisted of 50 ml Solution A (1% 4,4-dicarboxy-2,2-biquinoline, 2% Na2CO3-H2O, 0.16% sodium tartrate, 0.4% NaOH, and 0.95% NaHCO3) combined with 1 mL Solution B (4% CuSO4-5H2O) and 5 mL 10% ultra pure SDS. Samples were brought to a final volume of 100 l before addition of 2 ml of SWR and then incubated first for 30 min in a 37C water bath and then for 10 min at room temperature. An OD562 was recorded for each sample in duplicate. A standard curve was generated using bovine serum albumin with a concentration range from 0 to 80 g. Protein concentrations were determined

PAGE 67

46 using linear regression of the standard curve. A 100 mm dish of transfected cells typically yielded about 50 g of total membrane protein. ATPase Assay K+-dependent ATP hydrolysis activity was measured using the method of Garg and Narang (42). A coupled assay approach was employed based on the hydrolysis of ATP to ADP, which is coupled to NADH oxidation in the reactions described in Figure 2-2A. Following hydrolysis of ATP to ADP by the ATPase enzyme, ADP serves as substrate in the conversion of phosphoenolpyruvate to pyruvate by pyruvate kinase (PK). In the final reaction, pyruvate is converted to lactate by the action of lactate dehydrogenase (LDH) with NADH being oxidized to NAD+ in the process. The decrease in fluorescence of NADH following catalysis was measured in a TD-700 Laboratory Fluorometer (Turner Designs). The assay reactions were carried out in 300 l total volume in an imidazole buffer containing 25 mM imidazole, 10 mM MgCl2, 0.6 mM phosphoenolpyruvate, 1.1 mM Na2ATP, 0.017 mM NADH, 0.33 mM EDTA, 3.2 U/ml PK, 4.1 U/ml LDH, 2.5 mM sodium azide to inhibit mitochondrial and other endogenous ATPases, and varying concentrations of ouabain and SCH28080 (in DMSO) to specifically inhibit endogenous Na+, K+ ATPase and gastric H+, K+ ATPase. Assays were performed in the presence and absence of 2.5 mM KCl in order to determine the K+-dependent ATPase activity. A standard curve was generated using known amounts of ADP; a typical standard curve is depicted in Figure 2-2B. The assay reactions were set up in 1.5 ml microfuge tubes using three separate solutions. Membrane fractions were resuspended or diluted in 100 l of a hypotonic solution (H-solution) containing 1 mM imidazole, 1 mM MgCl2, 1 mM EDTA, and 0.1%

PAGE 68

47 Figure 2-2. ATP Hydrolysis Assay. (A) The three coupled reactions used to measure ATP hydrolysis activity of the membrane fraction of transfected COS-1 cells. In reaction 1, the expressed H+, K+ ATPase should be able to hydrolyze ATP to ADP. ADP then serves as the substrate in reaction 2 for the pyruvate kinase (PK)-catalyzed conversion of phosphoenolpyruvate (PEP) to pyruvate. In reaction 3 pyruvate is converted to lactate by the action of lactate dehydrogenase (LDH). LDH uses NADH as a cofactor and its oxidation into NAD causes a decrease in fluorescence that can be measured as an indicator of ATPase activity. (B) A typical standard curve is shown with known amounts of ADP on the x-axis and the decreasing level of fluorescence on the y axis. The equation of the line is used to extrapolate the amount of ADP generated in a given reaction from the measured fluorescence value. BSA. The enzyme-containing H-solution was added to 100 l of an incubation solution (I-solution) containing 25 mM imidazole, 30 mM MgCl2, 7.5 mM sodium azide, and a varying concentration of ouabain and SCH28080. The reaction was started by adding 100

PAGE 69

48 l of a reaction solution (R-solution) containing 50 mM imidazole, 3.3 mM Na2ATP, 1.8 mM PEP, 0.05 mM NADH, 9 M NH4Cl, 9.6 U/ml PK, and 4.1 U/ml LDH. The microfuge tubes containing the final reaction mix were incubated for the time points indicated (see Chapter 5). Fluorescence was measured using an excitation wavelength of 345 nm and an emission wavelength of 465 nm. When measuring ATPase activity of mouse stomach tissue, ATPase assays were performed as described above in the absence of 2.5 mM KCl, in the presence of 2.5 mM KCl, and in the presence of 2.5 mM KCl plus 0.1 mM SCH28080. Assays using mouse stomach tissue were conducted in a constant concentration of 1 mM ouabain to inhibit any endogenous Na+, K+ ATPase. When ATPase activity of the plasma membrane fraction of transfected COS-1 cells was measured, the assays were performed as described above using plus or minus 2.5 mM KCl in the presence of constant 0.1 mM SCH28080 to inhibit endogenous gastric H+, K+ ATPase activity. A constant concentration of 1 M ouabain was used to inhibit any endogenous Na+, K+ ATPase activity; this level of ouabain is high enough to inhibit the Na+, K+ ATPase which has a Ki for ouabain in the submicromolar range, but not high enough to inhibit the colonic H+, K+ ATPase, which has been reported to be either resistant to ouabain, or sensitive to concentrations in the low millimolar range. Luciferase Assay Following transfection and hormone treatment, mIMCD3 cells were lysed directly in 24 well plates using 1X Passive Lysis Buffer (Promega) on an orbital shaker for 15 min. Each well was then scraped with a 200 l pipet tip to facilitate lysis. Twenty microliters of the cell lysate were added to 100 l of Luciferase Assay Reagent II

PAGE 70

49 (Promega). The reaction was mixed by pipetting and measurement of firefly luciferase activity in relative light units was performed in a Berthold Detection Systems Luminometer (Sirius). Immediately following the initial reading, 100 l of Stop & Glo Reagent (Promega) was added to the firefly luciferase reaction. A second reading for Renilla luciferase activity was then recorded. Protein Methods Western blot Analysis Western blot analysis of the membrane fractions of transfected cells was performed using antibodies against the epitope tags or a monoclonal antibody for NaK1. One-hundred micrograms of membrane protein were run on a 10% Tris-glycine gel for SDS-PAGE. Proteins were electrically transferred onto nitrocellulose using transfer buffer (25 mM Tris-Base, 192 mM glycine, 20% (v/v) methanol, pH 8.3). Blots were blocked in 5% nonfat milk in Tris-buffered saline (TBS) at 4C overnight. Primary antibody incubations were done in 2.5% nonfat milk in TBS for 1 hr at room temperature. Three types of primary antibodies were used for this dissertation. To detect the HK2a or HK2c subunits, an anti-V5 antibody (Invitrogen) was used at a 1:2500 dilution to detect the C-terminal V5 epitopes on the proteins. NaK2, NaK3, and gHK were detected using an anti-c-myc antibody (Oncogene Research Products) at a concentration of 2.5 g/ml. NaK1 was detected using an anti-NaK1 antibody (Upstate Biotechnology Corporation) at a concentration of 0.5 g/ml. Secondary antibody incubations were performed using a 1:10,000 dilution of horseradish peroxidase-linked sheep anti-mouse immunoglobulin (Amersham) for 1 hr in 2.5 % nonfat milk in TBS. After both antibody incubations, blots

PAGE 71

50 were washed three times for 5 min each in TBS plus 1% Tween. Antibody binding was detected by chemiluminescence using the ECL system (Amersham Pharmacia). Western blot analysis was also performed on total cell lysates from aldosterone treated mIMCD3 cells. Cells were trypsinized and collected in PBS. Following 5 min of centrifugation at 150 x g, cells were resuspended in 1 mL cell lysis buffer (50 mM Tris pH 7.8, 150 mM NaCl, 1% NP-40). One-hundred fifty micrograms of total cellular protein were run on an 18% Tris-glycine SDS-PAGE gel. Proteins were electrically blotted onto nitrocellulose in transfer buffer (25 mM Tris-base, 192 mM glycine, 20% (v/v) methanol, pH 8.3). Blocking was performed in 5% nonfat milk (Biorad) in Tris-buffered saline (TBS) plus 0.1% Tween at room temperature for 1 h. To detect endothelin, either a rabbit polyclonal antibody (Chemicon, AB3280) or a mouse monoclonal antibody was used (Sigma E0771), both at a dilution of 1:1000. The primary antibody incubation was done in 5% nonfat milk in TBS plus 0.1% Tween overnight at 4C. Secondary antibody incubation was performed using a 1:5000 dilution of horseradish peroxidase-linked donkey anti-rabbit or sheep anti-mouse immunoglobulin for 1 h at room temperature in 5% nonfat milk in TBS plus 0.1% Tween. After the secondary antibody incubation, blots were washed three times for 15 min in TBS plus 0.1% Tween (Sigma). Chemiluminescence was used to detect antibody binding via the ECL system (Amersham). Immunoprecipitation For co-immunoprecipitation experiments, fourteen 100 mm dishes were transfected with the plasmid in question. Twelve to fourteen dishes were typically enough to yield 1 mg of membrane protein. The membrane protein fraction was solubilized according to the method of Asano et al. (7). A 2X lysis buffer was prepared containing 2% Nonidet P

PAGE 72

51 40, 300 mM NaCl, 1.0 mM EDTA, and 100 mM Tris-HCl, pH 7.4. An equal volume of the lysis buffer was added to the membrane protein fraction and incubation at 4C on a nutator proceeded for 30 min. The membranes were pelleted by centrifugation in a microcentrifuge for 20 min at 16000 x g. The protein-containing supernatant was then immunoprecipitated using an anti-V5 antibody overnight at 4C on a nutator. Preparation of the V5 antibody for immunoprecipitation is discussed below. The Seize Primary Immunoprecipitation Kit was purchased from Pierce. Fifty micrograms of a monoclonal anti-V5 antibody was purchased from Invitrogen. Two-hundred microliters of AminoLink Plus 50% Gel Slurry was aliquoted into a HandeeTM Spin Cup Column, provided with the kit. The column was spun in a microcentrifuge for 30 sec at 14000 rpm and the flow-through discarded. The gel was washed with 400 l of Coupling Buffer (0.1 M NaPO4, 0.15 M NaCl, pH 7.2) by adding the coupling buffer to the gel in the column, inverting the column several times, and then centrifuging the column as described above. This step was repeated for a total of two washes. Fifty microliters of the anti-V5 antibody were diluted into a total volume of 200 l Coupling Buffer and added to the gel in the spin column. Two microliters of 5 M sodium cyanoborohydride were added and the column was inverted five times. The antibody AminoLink mixture was incubated overnight at 4C on a nutator. The following day, the spin column was centrifuged and the flow-through discarded. The antibody AminoLink mixture was then washed with 400 l Quenching Buffer (1 M Tris-HCl, pH 7.4, supplied with the kit). After discarding the flow through from the previous wash, 400 l of Quenching Buffer were added to the gel, followed by addition of 4 l of 5 M sodium cyanoborohydride. The spin column was the incubated at room temperature on a

PAGE 73

52 nutator for 30 min. After centrifugation and disposal of the flow through, the antibody coupled gel was washed six times with 400 l Wash Buffer (1 M NaCl, supplied with the kit) and then three times with 400 l Binding Buffer (0.14 M NaCl, 0.008 M sodium phosphate, 0.002 M potassium phosphate, 0.01 M KCl, pH 7.4). The antibody-coupled gel was stored in 400 l Binding Buffer at 4C. In order to immunoprecipitate the H+, K+ ATPase, the antibody-coupled gel was incubated with the membrane-protein containing supernatant described above. A graphical representation of the preparation of the antibody coupled gel and the immunoprecipitation is shown in Figure 2-3A and B. The volume limit of the spin column is 500 l so the incubation of protein with the gel was normally performed in a 1.5 ml microfuge tube. The immunoprecipitation proceeded overnight at 4C on a nutator. The following day, the mixture was centrifuged through the spin column in 500 l aliquots and the flow through from the immunoprecipitation saved for further analysis. The gel was then washed three times with 400 l IP buffer (0.025 M Tris, 0.15 M NaCl, pH 7.2). Elution of the immunoprecipitated protein was performed a total of four times using 100 l of ImmunoPure IgG Elution Buffer (pH 2.8, provided with the kit). The wash and elution fractions were collected and saved for future analysis as well; 100 l of each of the IP flow-through, wash and elution fractions were loaded on an SDS-PAGE gel and transferred to nitrocellulose for Western blot analysis (described above).

PAGE 74

53 Figure 2-3. Co-immunoprecipitation of the HK2 and subunits. (A) Preparation of the AminoLink Plus coupled antibody is facilitated by a covalent linkage between the activated support and the anti-V5 antibody. (B) The colonic H+, K+ ATPase is immunoprecipitated by the immobilized anti-V5 antibody via the V5 epitope at the C-terminus of the HK2 subunit. If there is a subunit bound to the HK2 subunit, it should be co-immunoprecipitated with HK2 and be detectable by Western blot analysis with an anti-c-myc antibody directed against the C-terminal epitope tag on the XK subunit or with a subunit specific antibody. Molecular Biology RNA Isolation Total RNA was isolated using Trizol reagent (Invitrogen) essentially by the manufacturers instructions with the following modifications. Cells were lysed directly in the Transwell-COL inserts for five minutes using 1 mL of Trizol reagent per insert. The organic and aqueous phases were separated by centrifugation at 3200 x g for 20 min.

PAGE 75

54 RNA was precipitated from the aqueous layer by addition of 3 mL of isopropanol. The solution was allowed to sit at room temperature for 15 min and then the RNA was pelleted by centrifugation at 3200 x g for 15 min. The RNA pellet was washed in 6 mL of 75 % ethanol and resuspended in 100 L DEPC-treated water. Before resuspension of the RNA, the pellet was allowed to dry at room temperature for 10-20 min. The DEPC-treated water was heated to 50C before being added to the RNA pellet. After resuspension of the RNA by pipetting, it was incubated at 50C for 5 min to aid in complete resuspension. Isolation of mRNA from rabbit kidney medulla was performed using the PolyATract System 1000 (Promega). One gram of rabbit kidney medulla was kindly provided by the Wingo Laboratory. The tissue was homogenized in 4 ml of the GTC Extraction Buffer provided with the kit (containing guanidine thiocyanate) plus 164 l -ME in a 10 ml homogenization tube for 1 hr using a Con-Torque power unit (Eberbach Corp.) The homogenate was diluted into 8 ml of Dilution Buffer (provided with the kit) that had been preheated to 70C. One-hundred sixty four microliters of -ME was added as well. One thousand picomoles (20 l) of a biotinylated oligo dT probe (provided with the kit) was added and the sample was heated at 70C for 5 min; centrifugation at 12000 x g followed for 15 min in order to pellet cellular debris. The mRNA-oligo dT-containing supernatant was then mixed for 2 min at room temperature with Streptavidin MagneSphere Paramagnetic Particles (SA-PMPs) that had been prewashed in 0.5 X SSC (1.5 M NaCl, 150 mM sodium citrate, pH 7.0). The SA-PMPs were then pulled to the side of the tube using a PolyATract System 1000 Magnetic Separation Stand. The cleared supernatant was discarded and the SA-PMPs resuspended in 2 ml 0.5X SSC and

PAGE 76

55 moved to a 2 ml Eppendorf tube. The SA-PMPs were again pulled aside using the magnetic stand and then resuspended in 2 ml 0.5X SSC for a total of three washes. The mRNA was eluted from the SA-PMPs in 2 ml of DEPC-treated H2O and moved to a 50 ml centrifuge tube containing an additional 4 ml of DEPC-H2O. The mRNA was then precipitated by addition of 600 l 3 M sodium acetate, pH 5.2 plus 6 ml isopropanol. An overnight incubation at C was carried out. The following day the tube was spun at 16000 x g for 20 min. The mRNA pellet was washed with 2 ml of 75% ethanol and then air dried for 15 min. The mRNA was resuspended in 100 l DEPC-H2O and incubated at 55C for 10 min. RT-PCR Two micrograms of total RNA from rabbit kidney medulla, cortex or colon were used in the reverse transcriptase reactions. The RNA had been isolated by a previous graduate student in the Cain Laboratory, Dr. Grady Campbell (15). The reactions took place in a total volume of 20 l. One microliter of Superscript II (Invitrogen) was added to an RNase free tube containing 2 l of 10x Superscript II buffer (Invitrogen), 2 l of dNTP (10 mM each, Qiagen), 1 l RNase inhibitor (Promega), 1 l random decamers (Ambion), 2 g of total RNA, and a volume of DEPC-treated H20 to 20 l total. Twenty-five percent of the RT reaction was used as template in subsequent PCR reactions. The PCR amplification of cDNA fragments for creating Northern blot probes is described below. The Primers MG24 and MG25 (GCAGATCAGCCTTCAGTTCGTGCGG/ CATGCATGGAGTCTAGAAGCTT) were designed based on the published mouse mineralocorticoid receptor (MR) (AW910225) to amplify a 238 base pair product.

PAGE 77

56 Primers MG81 and MG82 (GTGCTCTCCCACACCCCCTGCAGC/ CTGGCCCCCGTCACTGTGGACAGC) were designed to the published sequence of the subunit of the epithelial sodium channel (ENaC (AF112185) in order to amplify a 505 bp product. Primers MG83 and MG84 (CTGGAAATCACCAAGGCCCACACG/ CAGGAACTGCCCGTGCACGTGCTC) were designed to the published sequence of hydroxysteroid 11 dehydrogenase 2 (11HSD2) (NM_008289) in order to amplify a 480 bp product. PCR was performed using Taq PCR Master Mix plus Q-solution (Qiagen) and the following cycling parameters: 94C x 5 min presoak; 25 cycles of 94C x 30 sec, 62.5C (ENaC) or 61.8C (11HSD2) or 56.6C (MR) x 30 sec, 72C x 1 min; 72C x 10 min final extension. Products were purified from a 1% agarose gel and either cloned into the TA cloning vector (pCR2.1, Invitrogen) or sent directly for sequence analysis. Probes for CTGF, period homolog, sgk and endothelin were prepared in the manner described above (Table 1). Primer sets were as follows: sgk MG 34/35 CCTCCAACCCTCACGCCAAAC/CTTCCAGGAGGTGCCTTGCCG; CTGF KF3/4 CCCCTGTCCGAATCCAGGCTC/GCGCACGTCCATGCTGCATAG; period homolog MG52/53CACGCGTCCGGCGGAGCTTCTGGG/GGCAATGGAGCTGCTGGGTGGGGA; preproendothelin MG54/54 GTTCGTGACTTTCCAAGGAGCTCC/CTCAGCTTTCAACTTTGCAACACG. RACE Rapid amplification of cDNA ends (RACE) was performed using the MarathonTM cDNA Amplification Kit from Clontech according to the manufacturers instructions. One microgram of rabbit kidney medulla mRNA was used to create a double stranded cDNA (ds cDNA) library to serve as template for the RACE PCR reactions. First and

PAGE 78

57 second strand synthesis were performed exactly as described by the manufacturer. Radioactive nucleotides were not used in these reactions. The MarathonTM cDNA Adaptor was ligated onto both ends of the ds cDNA library in a ligation reaction using 5 l of ds cDNA as described by the manufacturer, with the following exception: the ligation reaction was allowed to proceed overnight at 37C. While duplicate ligation reactions were performed at 16C and at room temperature, the adaptor-ligated library created by the 37C ligation was the only one that yielded usable results. The 5 and 3 RACE reactions performed to amplify cDNA fragments for NaK2 and NaK3 are described in Chapter 4. For all of those reactions, the 37C adaptor-ligated ds cDNA library was used at a 1/50 dilution as template. Affymetrix GeneChip The murine genome array, U74Av2, was purchased from Affymetrix through the Interdisciplinary Center for Biotechnology Research (ICBR) at the University of Florida. Using total RNA from aldosterone or vehicle treated cells, first and second strand synthesis were performed using Invitrogen reagents according to the protocol provided in the Affymetrix GeneChip Expression Analysis Manual. In vitro transcription reactions were then carried out using the BioArray High Yield RNA Transcript Labeling Kit (ENZO). Biotin-labeled CTP and UTP were included in the reaction cocktail. Fragmentation of the cRNA, hybridization, staining and scanning of the microarray were performed according to the GeneChip Expression Analysis Manual provided by Affymetrix. Three independent microarray experiments were performed.

PAGE 79

58 Affymetrix GeneChip Expression Analysis Analysis of intensity data was performed using Microarray Suite Version 4 (MAS4) (Affymetrix). Global scaling was applied to all arrays such that the mean intensity of each array was equivalent. In global scaling the raw signal value of each probe cell is multiplied by a scaling factor. The scaling factor is determined by first calculating the mean intensity of each array that is equivalent to the mean raw signal value, minus background, of probe cells, excluding the highest and lowest 2% of values. The mean intensity is multiplied by the scaling factor in order to equal the target intensity. The target value of all chips was 2500. The scaled signal form each probe cell was used to generate a quantitative hybridization signal for each gene with MAS4. MAS4 was also used to perform comparison expression analyses to examine the intensity data between two different arrays. Initially, a comparison expression analysis was performed for two samples, untreated and treated where the untreated cells served as the baseline. MAS4 algorithms, which use both qualitative and quantitative metrics, produced a list of differentially expressed genes. This list was further filtered to include only genes whose expression was altered by at least 2 fold. The experiment was repeated two additional times on independent RNA samples isolated from untreated and treated cells. The scaled signal values from all six hybridization experiments have been deposited in the Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/geo) (last accessed March 21, 2004). The accession numbers are GSM6603, GSM6604, GSM6605, GSM6606, GSM6607, and GSM6623. The data series is represented by the accession number GSE434.

PAGE 80

59 Northern Blot Analysis Northern blot analysis was done according to the method of Davis et al. (30). RNA was isolated as described above. Twenty micrograms of total RNA was electrophoresed on a 1% agarose, 0.22 M formaldehyde denaturing gel. Depurination of the RNA was performed in 50 mM NaOH for 25 min followed by neutralization 25 min in 100 mM Tris, pH 7.0. The gel was then equilibrated in 10X SSC for 30 min. Capillary transfer was allowed to proceed overnight in 10X SSC. The following day the RNA was crosslinked onto the nylon membrane (Amersham-N) in a XL1000 UV crosslinker (Spectronics Corporation) under the optimal crosslink. Generation of 32P labeled probes was performed using the Redi Prime II random labeling kit (Amersham) and 25 ng of template DNA. All reaction components were provided in dry form in an Eppendorf tube, except for 32P--dCTP and the template DNA. The template DNA was diluted into 45 l of TE (10 mM Tris HCl, 1 mM EDTA, pH 8.0) and heated to 95C for 5 min. The DNA was snap cooled on ice for 5 min and then added to the reaction tube. Five microliters of Redivue 32P--dCTP (Amersham) was added and mixed into the tube by pipetting. The reaction was incubated at 37C for 10 min and then stopped by addition of 5 l of 0.2 M EDTA. Nylon membranes containing the RNA samples to be probed were prehybridised in hybridisation solution (250 mM Na2HPO4, 10 mg/ml BSA, 1 mM EDTA, 7% SDS) at 65C for at least 30 min. The labeled probe DNA was denatured at 95C for 5 min and then added to the hybridization tube. Hybridization continued overnight at 65C. The following day, blots were washed at 65C three times for 15 min in wash solution (20 mM Na2HPO4, 1% SDS, pH 7.2) in a shaking water bath. Following the washes, blots were exposed to Kodak Biomax MS film for an appropriate length of time. A probe for

PAGE 81

60 the mRNA of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used to control for loading. Southern Blot Analysis Southern blot analysis was performed exactly as described for Northern blot analysis, above, with the following exceptions. DNA samples were electrophoresed on 0.8% agarose gels prepared with 1X TAE. Depurination of DNA samples in the gel was performed in a denaturing solution (0.5 M NaOH, 1.5 M NaCl) for 30 min at room temp. Neutralization of the membrane was carried out in neutralization buffer (1 M Tris, 3 M NaCl) for 30 min at room tem. Following equilibration in 10X SSC, the samples were transferred to a nylon membrane (Amersham-N+) via capillary action. Probe hybridization was performed as described above. Real Time RT-PCR Reactions were performed using TaqMan One-Step RT-PCR Master Mix Reagents (Applied Biosystems) according to the manufacturers instructions. Cycling parameters were as follows: 48C x 30 min, 95C x 10 min presoak; 45 cycles of 95C x 15 sec, 60C x 1 min. Primers and TaqMan probes were generated via Assays-by-Design (Applied Biosystems). The sequence of the primers and probes were as follows: sgk forward CGCCAAGTCCCTCTCAACAA, sgk reverse TTGGCGTGAGGGTTGGA, sgk Probe 6FAMTCAACCTGGGTCCGTC MGBNFQ; preproendothelin forward TGCCACCTGGACATCATCTG, preproendothelin reverse CTCCCAGTCCATACGGTACGA, preproendothelin probe 6FAM AACACTCCCGAGCGC MGBFNQ; period homolog forward CCAGGTGTCGTGATTAAATTAGTCA, period homolog reverse

PAGE 82

61 GGGCTTTTGAGGTCTGGATAAA, period homolog probe 6FAM TCAGAGACAGGCGTCCT MGBFNQ. As a negative control, TaqMan Rodent GAPDH Control Reagents were used to perform real time RT-PCR for GAPDH. Reactions were carried out in a DNA Engine Opticon 2 Continuous Fluorescence Detector and data was analyzed using Opticon Monitor 2 software (MJ Research Inc.). Statistical Analysis Means are represented as plus or minus standard error. A 2-way ANOVA was performed on the real time PCR data to compare fluorescent intensity at each time point to control. The Dunnett test for error protection was used with a confidence interval of 95%. A 1-way ANOVA was performed on the fold change data for the inhibitor study to compare each condition to control. The LSD test for error protection was used with a confidence interval of 90%. A 1-way ANOVA with the Dunnett test for error protection and a confidence interval of 95% was used to evaluate the Western blot fold change data.

PAGE 83

CHAPTER 3 MOLECULAR MODELING OF THE RABBIT COLONIC H+, K+ ATPASE Introduction The H+, K+ ATPase is a member of a large family of integral membrane proteins classified as P-type ATPases. Typically, these enzymes consist of a large subunit of about 100 kDa. The P-type ATPase family includes the Ca+2 ATPase and the X+, K+ ATPases, the H+, K+ ATPase and the Na+, K+ ATPase. Unlike the single subunit Ca+2 ATPase, the Type II P-type ATPases, such as Na+, K+ ATPases and the H+, K+ ATPases, also contain smaller subunits. The subunits are approximately 30 kDa proteins that are glycosylated to varying degrees. P-type ATPases use the energy of ATP hydrolysis to pump ions against their concentration gradients. Phosphorylation of the subunit leads to a conformational change from a high affinity E1 state to a low affinity E2 state. Changes in ion affinity that accompany the conformation change promote ion translocation (84). Determination of a high resolution structure for the colonic H+, K+ ATPase has remained impractical because the enzyme is present in relatively low abundance in animal tissues and no high yield expression system has been established. However, a 2.6 crystal structure was reported for the rabbit sarcoplasmic reticulum Ca+2 ATPase in the E1 conformation (113) and provided the basis for generation of a molecular model of the HK2 subunit. The Ca+2 ATPase crystal structure showed the existence of ten transmembrane helices and three cytoplasmic domains, designated A, N and P. The A domain is made up of the N-terminus and the loop between transmembrane helices two and three (L23). The P and N domains lie within the large loop between transmembrane 62

PAGE 84

63 helices four and five (L45). The N domain, which contains the nucleotide-binding site, is the largest of the three cytoplasmic domains. The P domain contains the aspartic acid residue that becomes phosphorylated during the enzymes catalytic cycle. This aspartic acid residue lies at the C-terminal end of the central sheet in a Rossman fold. Following publication of the Ca+2 ATPase structure, an 11 structure of the duck Na+, K+ ATPase was deduced using cryo-electron microscopy (90). The Na+, K+ ATPase structure in the E2 conformation appeared to be similar to that observed for the 8 Ca+2 ATPase structure in the E2 conformation (129). In both structures, the three cytoplasmic domains were easily recognizable. Conservation of the structure of this catalytic core provided very strong support for the idea that these enzymes, as well as other P-type ATPases, probably undergo similar E1 to E2 conformation changes that accompany ion translocation. One of the important observations made by Rice et al. (90) was that of an extracellular mass component of the Na+, K+ ATPase. This likely corresponds to the location of the subunit, which was shown to lie near transmembrane helices seven and ten (M7 and M10) of the NaK1 subunit. Additional evidence for the three-dimensional structure of the Na+, K+ ATPase was presented by Hebert et al. (54) in a 9.5 structure also deduced from electron micrographs. Densities from within this structure suggested a location for the subunit near M8 and M10 of the Na+, K+ ATPase. This density was taken from a plane closer to the membrane than that used to assign subunit density in the structure of Rice et al. It was the purpose of the work described in this chapter to generate a molecular model of the rabbit colonic H+, K+ ATPase 2 subunit based on the atomic coordinates of the Ca+2 ATPase (113). The model indicates possible points of entry for the transported

PAGE 85

64 ions, likely secondary structure elements of the transmembrane domain, and a putative ouabain binding site. Incorporating data from the duck Na+, K+ ATPase structure (90) and a recent model of the NaK1 subunit (53), speculation on the position of the subunit was also made. Results Sequence Alignments Sequence alignments were generated in order to determine if the primary structures of the rabbit HK2a subunit and the rabbit sarcoplasmic reticulum Ca+2 ATPase were sufficiently conserved to consider molecular modeling of the colonic H+, K+ ATPase. These alignments revealed that the two proteins share 29% identity and 47% similarity. Other researchers have found similar levels of primary sequence similarity adequate for modeling (35). HK2a shares 65% identity with the Na+, K+ ATPase compared to 29% with the Ca+2 ATPase. The amino acid sequence of rat NaK1 was included as an added feature of the alignment for this reason and because of the growing body of structural data on the Na+, K+ ATPase. Clustal W (112) was used to produce the alignment of the deduced amino acid sequences of the three proteins (Figure 3-1). Amino acid conservation was observed within several regions of all three proteins. These included the well-characterized phosphorylation site (position 385 in HK2a) and surrounding sequence (ICSDKTGTLT) and the region near the nucleotide-binding pocket (KGAPE). The latter contains the conserved lysine (position 517 in HK2a) that has been shown to lie within the binding pocket of P-type ATPases. Many of the amino acids within the ten transmembrane helices were also conserved. In all three proteins, the large cytoplasmic

PAGE 86

65 Figure 3-1. Primary Structure Alignment of HK2a, Na+, K+ ATPase 1, and Ca+2 ATPase. Amino acid sequences were aligned using Clustal W with gap penalties. The numbering system at the right corresponds to each individual amino acid sequence. Identical amino acids between HK2a (top) and Ca+2 ATPase (bottom) are shown in green. Similar amino acids are shown in pink. When applicable, similar and identical amino acids of the Na+, K+ ATPase (middle) were highlighted as well. Underlined regions correspond to the ten transmembrane domains of the Ca+2 ATPase. Aspartic acid 385 and lysine 517 are denoted by *.

PAGE 87

66 loop that houses the P and the N domains was located between transmembrane helices four and five (L45). In summary, the similarities between the HK2a subunit and the Ca+2 ATPase appeared sufficient to justify molecular modeling of the colonic H+, K+ ATPase. Modeling The model of HK2a was built using the program O (63) and was based on the published atomic coordinates of the Ca+2 ATPase (113). Using the amino acid sequence alignment as a guide (Figure 3-1), amino acids of the Ca+2 ATPase were changed to match the primary sequence of HK2a. A total of 739 individual amino acid substitutions were made; of these, 204 changes were to similar amino acids. HK2a is forty-eight amino acids longer than the Ca+2 ATPase at its N-terminus. Secondary structure prediction using Gor II (47) suggested that these amino acids would predominantly form a helix. Since there was no counterpart for these in the Ca+2 ATPase structure they were necessarily omitted from the model. As might be expected, all insertions and deletions with respect to the Ca+2 ATPase occurred in both HK2a and NaK1. In total, thirty-three amino acids were added and fifty amino acids removed with respect to the Ca+2 sequence. The deletions and insertions are summarized in Table 3-1. The insertions and deletions in HK2a as compared to the Ca+2 ATPase are highlighted in Figure 3-2. In Panel A, amino acids that were added in HK2a are highlighted in green; in Panel B, amino acids that were deleted from the Ca+2 ATPase are highlighted in red. Areas of insertion or deletion occurred only in extra-membraneous loop regions such as the extracellular loops and the P domain (Figure 3-2A). Examples of insertions contributing to changes in the extracellular domain include a three amino acid

PAGE 88

67 Table 3-1. Summary of Amino Acid Deletions and Additions in HK2a as Compared to the Ca+2 ATPase Additions Sizea Locationb 3 L12 20 L45 2 L45 1 L45 7 L78 Deletions Sizea Locationb 2 L23 2 L23 3 L23 10 L34 14 L45 18 L45 1 L45 aSize indicates the number of amino acids. bLocation indicates in which transmembrane loop the addition or deletion occurred. addition in L12 and a seven amino acid addition in L78. A more substantial area of addition is a twenty amino acid insertion in L45 that occurs in the P domain. Since this insertion is also present in the Na+, K+ ATPase, it may represent the protuberance from the P domain discussed by Rice et al. (90). This mass appeared as a slight protrusion in the P domain that was distinctly absent in the Ca+2 ATPase. Both of the major insertions into HK2a appeared to fill space vacated by deletions of amino acids present in the Ca+2 ATPase (Figure 3-2). For example, ten Ca+2 ATPase amino acids were absent from L34 in HK2a. This space in the extracellular domain was occupied by the insertions in L12 and L78 in the model. Similarly, the space vacated by the deletion of material from L45 of the Ca+2 ATPase was filled by the large twenty amino acid insertion. Smaller areas of deletion were not obviously replaced by additions in HK2a but these areas were located on the surfaces of the N and A domains (Figure 3-2B).

PAGE 89

68 Figure 3-2. Comparison of the HK2a model with the Ca+2 ATPase (A) The C trace of the HK2a subunit model is shown with insertions labeled in green. For emphasis, the side chains of the added amino acids are highlighted as well. (B) The C trace of the Ca+2 ATPase is seen with amino acids that are absent in HK2a displayed in red. Figure was prepared using O, Version 7. Following completion of the primary structure changes, CNS (14) was used to perform a global energy minimization on the model. PROCHECK (69) was then used to evaluate the stereochemical properties of the model. The CNS minimization and PROCHECK evaluation were also performed on the structure of the Ca+2 ATPase using the published coordinates. Table 3-2 lists pertinent stereochemical parameters of the model as compared to the Ca+2 ATPase crystal structure. Importantly, the root mean square deviations (r.m.s.d.) listed for the HK2a model were strikingly similar to those observed for the Ca+2 ATPase for the great majority of the parameters. The bond lengths and angles values both fell within accepted limits. The omega torsion angle, representing

PAGE 90

69 Table 3-2. Stereochemical Statistics Statistic aHK2a aCa+2 Bond Length 0.0569 0.0481 Bond Angles 1.9344 1.7945 Omega Angle 1.3 1.2 Zeta Angle 1.8 1.9 H Bond Energy 0.7 0.8 Bad Contacts 2.6 1.3 Chi-1 gauche minus 19.5 17.6 Chi-1 trans 18.1 17.6 Chi-1 gauche plus 18.5 17 Chi-1 pooled 19.4 17.8 Chi-2 trans 14.9 15.6 Overall G-factor -0.28 -0.2 aValues displayed are root mean square deviations (r.m.s.d.) between the actual and most likely values. rotation around the peptide bond, and the zeta torsion angle, giving a measure of C tetrahedral distortion, were similar to that observed in the Ca+2 ATPase structure. As a measure of stability of side chains, the chi angles were also very similar between the two structures. An important parameter for judging the overall quality of a structure is the G factor, used to estimate the normalcy of the structure. PROCHECK calculates on average overall G factor by evaluating phi-psi and chi1-chi2 values for each residue. Accepted G values are typically greater than .5 and any value that falls below .0 requires further investigation, according to the program. Notably, the overall G factor for the HK2a model was .28, compared to a value of .2 for the Ca+2 ATPase structure. Charge Distribution The program GRASP was used to visualize the charge distribution over the surface of the HK2a model (Figure 3-3). The area surrounding the phosphorylation site is rich

PAGE 91

70 Figure 3-3: Charge Distribution over Surfaces of HK2a. Areas of negative charge are designated in red and areas of positive charge are indicated in blue. (A) A view of the charge distribution over the entire model is seen here. (B) The figure in panel A was rotated 180 degrees in the x direction. (C). The figure in panel B was rotated 90 degrees in the y direction. (D) The figure in panel B was rotated -90 degrees in the y direction. Figures were prepared with GRASP.

PAGE 92

71 in positively charged amino acids that probably interact with the phosphate group (Figure 3-3A). A concentration of basic amino acids was found in the region linking the P domain to the transmembrane domain (arrow in Figure 3-3B). Acidic amino acids clustered at the extracellular and intracellular faces of the molecule (Figures 3-3C and D, respectively). Substitutions in the extracellular loops of HK2a actually resulted in a net increase of eight positive charges with respect to the comparable region in the Ca+2 ATPase. In fact, the overall charge of the Ca+2 ATPase is -28 compared to -11 for HK2a. This disparity in charge may represent the difference between the two pumps for association-dissociation of a proton as opposed to a divalent calcium ion, or it may reflect the requirement of the H+, K+ ATPase to transport cations in both directions. Nevertheless, the negatively charged regions at the extracellular face of HK2a probably represent the location of the entrance to the ion channel. Architecture of the HK2a Model The relative positions of the phosphorylation site and the nucleotide-binding site were maintained in the HK2a model. The two amino acids used to indicate the locations of the P and N domains of the enzyme were the phosphorylation site aspartic acid at position 385 and the lysine at position 517 in the nucleotide-binding pocket (Figure 3-4). The distance between these two residues was measured to be 29 This value was very close to that determined for the same two amino acids using the reported coordinates for the Ca+2 ATPase (113) and in a model of L45 of the Na+, K+ ATPase (35). The amino acid sequence alignment, together with data from the crystal structure of the Ca+2 ATPase, was used to assign putative secondary structure elements to the model of HK2a (Figure 3-4). While there is little direct evidence about the secondary structure

PAGE 93

72 Figure 3-4: Possible Architecture of the HK2a Subunit. The residue in green is the lysine (position 517, Figure 1) that lies in the nucleotide binding pocket and the residue in red is the aspartic acid at the phosphorylation site (position 385). Based on an alignment with the Ca+2 ATPase, ten transmembrane helices were modeled into HK2a. The ten transmembrane helices are drawn as rods and shown in different colors for clarity. From left to right, the transmembrane helix designations are M10 (blue), M7 (gray), M8 (purple), M9 (orange), M5 (pink), M6 (green), M3 (brown), M4 (cyan), M2 (teal), and M1 (almond).

PAGE 94

73 of the H+, K+ ATPase, it has been determined that the Ca+2 ATPase, the N. crassa H+ ATPase, and the duck and pig kidney Na+, K+ ATPases all contain ten transmembrane helices (113). Ten transmembrane domains were readily apparent in the HK2a model. Each occupied a position comparable to transmembrane domains in other P-type ATPases for which structural information is available. It has been reported that the colonic H+, K+ ATPase has partial sensitivity to the cardiac glycoside ouabain (49), while the Na+, K+ ATPase has been long recognized as highly sensitive to the drug. Extensive mutagenesis studies with the Na+, K+ ATPase have revealed the participation of several key residues in conferring ouabain sensitivity to the pump (for review see (107)). In an attempt to glean information from the model about the ouabain sensitivity of our enzyme, we used the alignment in Figure 3-1 and our model of HK2a to look at the conservation of these key residues as well as their placement in the H+, K+ ATPase. Of the eighteen residues that have been defined as contributing to ouabain sensitivity in the Na+, K+ ATPase (107), twelve were conserved in the colonic H+, K+ ATPase. Of the twelve conserved amino acids, nine were located at the extracellular face of the molecule where ouabain binding occurs (Figure 3-5). The remaining residues appeared to be too far from the extracellular binding site to participate in an interaction with the drug and might be involved with functions outside the ouabain-binding site. Modeling of the Subunit The H+, K+ ATPase and the Na+, K+ ATPase both require a subunit to function. Structural as well as biochemical data have revealed several important interactions between the and subunits of Type II P-type ATPases. Together with our model of

PAGE 95

74 Figure 3-5. Ouabain-binding Residues in HK2a. Ouabain binding residues from the Na+, K+ ATPase that are conserved in HK2a are highlighted in red. The model from Figure 4 was rotated to give a view of the extracellular face of the enzyme and the clustering of the putative ouabain binding residues in this area. Helix colors and designations are the same as described for Figure 3-4. HK2a, the data concerning and interactions were used to consider a possible location for the subunit. Although the identity of the subunit that pairs with the HK2a subunit has not yet been determined, four possible partners are known. These are Na+, K+ ATPase 1, 2, 3 and gastric H+, K+ ATPase all share similar molecular architectures. The probable topology consists of a short intracellular region, a single -helical transmembrane domain, and a large glycosylated extracellular domain that contains six well-conserved cysteines thought to form three disulfide bridges. Studies in Xenopus oocytes have demonstrated that, of the four possible subunit partners, NaK2 is an effective partner for HK2 (46). We have recently cloned and expressed rabbit NaK2 (Genbank AY069937). We have built an helix representing the single

PAGE 96

75 Figure 3-6. Modeling of the Transmembrane Segment of NaK2. (A) The program TMHMM was used to predict the transmembrane segment of NaK2, spanning amino acids 37 through 62. (B) Helical Wheel Custom Images and Interactive Java Applet (http://marqusee9.Berkeley.edu/kael/helical.htm) (last accessed June 30, 2003) was used to draw a helical wheel plot for the 27 amino acids that span the membrane domain of the NaK subunit. (C) The program O was used to build a model of the transmembrane helix of NaK2. In this view, the helix is seen from the side, with the intracellular end on the left and the extracellular end on the right. (D) Clustal W alignment of NaK2, NaK3, NaK1, and gHK. Conserved residues are highlighted in green.

PAGE 97

76 transmembrane domain of NaK2 (Figure 3-6). Figure 3-6C shows the model of the transmembrane helix of NaK2. The positions of several phenylalanines were evident along one face of the helix, consistent with the helical wheel prediction. Two of these are conserved across all four subunits, as illustrated in the alignment in Figure 3-6D. Discussion Generation of a three-dimensional model of HK2a allows consideration of a location for the subunit that accounts for all known / interactions in Type II P-type ATPases. The 11 structure of the duck Na+, K+ ATPase deduced from electron micrographs showed a large extracellular mass in the vicinity of M7 and M10 (90). Placement of the subunit in this area would allow for possible contact of the extracellular region of with L78 of the subunit, an interaction for which there is direct biochemical evidence (23). A yeast-two-hybrid system coupled with alanine scanning mutagenesis was used to identify four particular amino acids, SYGQ, in L78 that were necessary for and interaction in the Na+, K+ ATPase (23). The corresponding amino acids in the rabbit colonic H+, K+ ATPase are NYGQ as determined by our laboratory (16) and independently by Naray-Fejes-Toth et al. (37) (Figure 3-7A). In contrast, the 9.5 resolution structure of the pig Na+, K+ ATPase by Hebert et al. (54) revealed density attributable to near M8 and M10. Interaction of the transmembrane domain of the subunit with M8 in the Na+, K+ ATPase was supported by cross-linking studies (82). A model of NaK1 was recently built and used to place the subunit at either of two placesnear M7 and M10 or M8 and M10 (53). Importantly, the cross-section of electron density used to suggest proximity of to M7 and M10 in the duck Na+, K+ ATPase structure was taken closer to the extracellular side of the membrane than

PAGE 98

77 Figure 3-7. A Putative Location for the subunit of the H+, K+ ATPase. (A) Clustal W alignment of the L78 amino acid sequence of eleven Type II P-type ATPase subunits. The conserved four amino acids are highlighted. (B) The four amino acids that have been shown to be necessary for interaction with the subunit are highlighted in L78. The model of HK2a pictured in Figure 4 is shown here with a possible location for the subunit indicated with dashed lines. Helix colors and designations are the same as described for Figure 3-4.

PAGE 99

78 the cross-section used to show near M8 and M10 in the pig structure. We propose that the subunit lies near M7 and M10 at the extracellular face and near M8 and M10 closer to the cytoplasmic surface in the HK2a model (Figure 3-7B). In summary, we have generated a structural model for the HK2a subunit of the rabbit colonic H+, K+ ATPase. The gross structural features and specific locations of functional features of the enzyme strongly resemble those of the P-type ATPases for which structures have been determined. The model of the colonic H+, K+ ATPase provides a basis for an understanding of the position of features such as the subunit, the likely ouabain binding site, and the major intracellular domains, including the nucleotide-binding site and phosphorylation site.

PAGE 100

CHAPTER 4 CLONING OF THE RABBIT X+, K+ ATPASE SUBUNITS Introduction The subunit of the X+, K+ ATPases has been shown to be necessary for ER exit and proper membrane insertion of the enzyme. The subunit is also responsible for the structural and functional maturation of the subunit and trafficking of the enzyme to the plasma membrane (11, 45). Pairing of the and subunits of the X+, K+ ATPases was originally thought to be promiscuous (72). Mounting evidence concerning the colonic H+, K+ ATPase however, suggests that the functionality of this ion pump depends on pairing with an appropriate subunit. Although the identity of the subunit partner for HK2 has not been definitively demonstrated, various reports have linked HK2 to any one of the four known mammalian X+, K+ ATPase subunits. Mammalian expression systems have suggested that HK2 was able to form functional, stable complexes with gHK (49) (6), NaK1 (6) (96), and also NaK3 (96). Co-immunoprecipitation experiments using animal tissue suggested that NaK1 (20, 67) and NaK3 (96) associate with HK2 in vivo. In order to clarify the confusion concerning the identity of the colonic subunit, Geering et al. (46) used a Xenopus oocyte expression system to assess the stability of the H+, K+ ATPases formed by expression of the four known mammalian XK subunits with the HK2 subunit. These investigators showed that, of the mammalian X+, K+ ATPase subunits, only NaK2 and gHK were able to form stable complexes with HK2. 79

PAGE 101

80 However, it was noted that these complexes were still highly sensitive to trypsin. Thus, these investigators concluded that the true colonic subunit had not yet been identified. Based on investigation into the human genome project, it would appear that no other apparent XK-like open reading frames exist, other than those already identified. Therefore, it appears as if one of the four known mammalian XKsubunits is likely to be the proper partner for HK2. The four members of the mammalian XK family have never been tested in parallel in a mammalian cell culture expression system. The Xenopus system used by Geering et al. was the first effort to examine all four subunits in parallel. However, the contrasts between a Xenopus oocyte system and a mammalian cell culture system are dramatic, including differences in growth temperature and ER quality control mechanisms. Thus, a mammalian cell culture system appeared to be a better candidate for studying a mammalian enzyme. In order to determine the identity of the XK subunit that pairs with HK2a and HK2c, it was necessary to test the ability of each of the four subunits to form a functional complex with the subunit. We therefore needed the cDNAs of NaK1, NaK2, NaK3 and gHK. The cDNAs for NaK1 and gHK were cloned and expressed by previous members of the laboratory. It was the purpose of the work described in this chapter to obtain the cDNAs for the remaining members of the rabbit XK subunit family, NaK3 and NaK2. Results Degenerate PCR for NaK3 Primers were designed in conserved regions of the published sequences for human, mouse and rat NaK3 subunit cDNAs (Figure 4-1). The sequences of all the primers used

PAGE 102

81 for the experiments described in this chapter are listed in Table 4-1. For NaK3, a two-fold degenerate forward primer (MG8) and a four-fold degenerate reverse primer (MG9) Figure 4-1. Design of Degenerate Primers to Amplify a cDNA fragment of Rabbit Na+, K+ ATPase 3. Known sequences from human, mouse and rat subunits were used to design degenerate primers for NaK3 were used to amplify a fragment of the cDNA sequence. cDNA libraries derived from rabbit kidney cortex and medulla and rabbit distal colon were used as templates, as described in the RT-PCR section of Chapter 2. PCR was performed using Qiagen master mix plus Q solution and the following cycling parameters: 94C x 5 min presoak; 25 cycles of 94C x 30 sec, 56.2C x 30 sec, 72C x 1 min; 72C x 10 min final extension. The results of this experiment indicated that NaK3 was present in rabbit kidney medulla and cortex as well as the distal colon (Figure 4-2). The 766 base pair PCR product from medulla was gel purified and cloned into pCR2.1 TOPO for sequence analysis. A BLAST search of the nucleotide sequence results from this original PCR

PAGE 103

82 Table 4-1. Primers Used to Amplify cDNAs for Rabbit Na+, K+ ATPase 3 and 2 Primer subunit Reaction Sequencea MG8 (Fd) NaK3 Degenerate PCR TGGAAGCTSTTCATCTACAACCCG MG9 (Rv) NaK3 Degenerate PCR AACWCGTCCCAARAACTTGTCACG MG10 NaK3 3 RACE GTGCGTCTGATCCAAGCTCCTATAGAG MG11 NaK3 5 RACE CCTTGGCGTTCCTTCAGGCTTTAA MG17 (Fd) NaK3 Full-length cDNA GTCCGCACGCACACCATGACG MG18 (Rv) NaK3 Full-length cDNA CCACTTTTATTTCCTTTTCAGGGCTCAC MG28 (Fd) NaK2 Degenerate PCR CCAGTTYATGGGSCGCACCGG MG29 (Rv) NaK2 Degenerate PCR CRTTGGGGGTCACATTCAGGA MG30 NaK2 5 RACE CGCCCCAAGACTGAGAACCTTGACG MG31 NaK2 3 RACE GAGCAGTTGCCCAGCTGAGTTGCGG nMG30 NaK2 5 RACE nested CAGTGACACTGAAAGCTGGGACCAG nMG31 NaK2 3 RACE, nested AGCTGAGTGCGGTTGAATTGGCAAG MG36 (Fd) NaK2 Full-length cDNA ATGGTCATCCAGAAAGAGAAGAAGAGCTG MG37 (Rv) NaK2 Full-length cDNA GAGGCTCAGACAAGTCATAGCTCTG .aS=C or G, W=A or T, R=A or G, Y=C or T. product revealed 87% identity to the published human sequence. Such strong homology supported the interpretation that this sequence corresponded to the rabbit NaK3 isoform. This sequence was used to design primers for R apid A mplification of c DNA E nds (RACE).

PAGE 104

83 Figure 4-2. Degenerate PCR for the Rabbit Na+, K+ ATPase 3 Subunit. PCR was performed using degenerate primers designed using sequences from human, mouse, and rat NaK3 cDNAs. Lane 1: 1 kb ladder, Lane 2: cortex plus reverse transcriptase (RT), Lane 3: cortex minus RT, Lane 4: distal colon plus RT, Lane 5: distal colon minus RT, Lane 6: medulla plus RT, Lane 7: medulla minus RT. RACE for NaK3 5 and 3 RACE reactions were performed using the Marathon cDNA Amplification Kit (Clontech) adaptor primer 1 (AP1) and gene specific primers. PCR was performed using the Advantage 2 Polymerase (Clontech) and the following cycling parameters: 94C x 1 min presoak; 5 cycles of 94C x 30 sec, 72C x 3 min; 5 cycles of 94C x 30sec, 70C x 3 min; 25 cycles of 94C x 20 sec, 68C x 3 min, 68C x 3 min final extension. Products were amplified from an adaptor primer-ligated double-stranded cDNA library from rabbit kidney medulla. The library was created using mRNA from rabbit kidney medulla as described in the RACE section of Chapter 2. 5 RACE for NaK3 was performed using the primer AP1 and the gene specific primer MG10; this reaction generated an approximately 600 bp band (lanes 2 and 3 in Figure 4-3). The 3 end of the NaK3 cDNA was amplified using the primers AP1 and MG11; this reaction generated an approximately 1100 bp band (lane 4 in Figure 4-3).

PAGE 105

84 PCR was performed under the conditions described above. PCR reactions were performed in triplicate and products were gel purified and cloned into pCR2.1 TOPO for sequence analysis. The sequence derived from these RACE reactions was used to design gene specific primers to amplify the full-length cDNA for NaK3. Figure 4-3 Rapid Amplification of cDNA Ends for the Rabbit Na+, K+ ATPase 3 Subunit. RACE products are boxed. Lane 4 was run on the same gel; intervening lanes have been cropped out. Lane 1: 1 kb ladder; lane 2: 5 RACE reaction 1; lane 3: 5 RACE reaction 2; lane 4: 3 RACE reaction. PCR Amplification of the Full-Length NaK3 cDNA Based on the sequence data obtained from the 5 and 3 RACE reactions, gene specific primers were designed for NaK3 to amplify the full-length cDNA. PCR was performed using Qiagen master mix plus Q solution, the primers MG17 and MG18 and the following cycling parameters: 94C x 5 min presoak; 25 cycles of 94C x 30 sec, 57.5C x 30 sec, 72C x 2 min; 72C x 10 min final extension. The product from a single

PAGE 106

85 NaK3 full-length PCR reaction is shown in Figure 4-4. Sequence data was collected from triplicate reactions. The sequence of the rabbit NaK3 cDNA was derived from Figure 4-4. PCR Amplification of the Rabbit Na+, K+ ATPase 3 Subunit cDNA. Primers were designed to amplify the full-length cDNA of NaK3. Lane 1: 1 kb ladder, Lane 2: full-length PCR product, Lane 3: H2O as template. both the 5 and 3 RACE reactions and the PCR amplification of the full-length cDNA, meaning that six independent sequences were used to generate the final cDNA sequence. The rabbit NaK3 cDNA sequence was deposited into Genbank and assigned the accession number AF302929 (Figure 4-5).

PAGE 107

86 Figure 4-5. GenBank Accession Record for Rabbit NaK3.

PAGE 108

87 Degenerate PCR for NaK2 A similar approach was used to amplify a 751 bp fragment of the NaK2 cDNA. A four-fold degenerate forward primer (MG28) and a two-fold degenerate reverse primer Figure 4-6. Design of Degenerate Primers to Amplify a cDNA fragment of Rabbit Na+, K+ ATPase 2. Known sequences from human, mouse and rat subunits were used to design degenerate primers for NaK2. (MG29) were designed, as described in Figure 4-6. Degenerate PCR was performed as described above for NaK3, but with an annealing temperature of 59.2C. As with NaK3, the results of the degenerate PCR indicated that NaK2 is present in rabbit kidney medulla and cortex as well as the distal colon (Figure 4-7). A BLAST search of the nucleotide sequence results from the NaK2 degenerate PCR product revealed 97% identity to the published human sequence. Consequently, the sequence data was used to design gene specific primers for RACE.

PAGE 109

88 Figure 4-7. Degenerate PCR for the Rabbit Na+, K+ ATPase 2 Subunit. PCR was performed using degenerate primers designed using sequences from human, mouse, and rat NaK cDNAs. The band in lane 6 was excised and gel purified for cloning purposes. Lane 1: 1 kb ladder, Lane 2: cortex plus reverse transcriptase (RT), Lane 3: cortex minus RT, Lane 4: distal colon plus RT, Lane 5: distal colon minus RT, Lane 6: medulla plus RT, Lane 7: medulla minus RT. RACE for NaK The RACE reactions for NaK2 proved to be more difficult to perform than those for NaK3. When the 5 and 3 RACE reactions were performed with AP1 and the gene specific primers MG30 and MG31, respectively, the PCR reactions yielded only smears and no discrete bands. To circumvent this problem, nested PCR was performed. This involved the design of primers that lay inside the area of cDNA amplified in the original PCR reaction (Figure 4-8A). In order to perform nested PCR, a dilution of the original PCR reaction was used as template. A nested 5 RACE reaction was performed using the adaptor primer 2 (AP2), which lies inside AP1, and the gene specific primer nMG30, which lies inside MG30. The template for this nested 5 RACE reaction was a 1/250 dilution of the original 5 RACE reaction performed with the primers AP1 and MG30.

PAGE 110

89 This reaction was performed a total of three times and the products were cloned and sequenced. Figure 4-8B, lanes 2 and 3, shows the products of two of these reactions. A nested 3 RACE reaction was also performed using AP2 and nMG31. The template for this reaction was a 1/250 dilution of the original 3 RACE reaction performed with AP2 and MG31. The nested 3 RACE reaction was also performed three times and the products were cloned and sequenced. Figure 4-8B, lanes 4 and 5, shows the Figure 4-8. Rapid Amplification of cDNA ends for the Rabbit Na+, K+ ATPase 2 Subunit. (A) Diagram of the adaptor ligated cDNA library from rabbit medulla. Location of the primers used to amplify the 5 and 3 ends of the rabbit NaK2 cDNA are shown. Primers used for initial RACE reactions are shown in black; primers used for nested PCR reactions are shown in purple. (B) Nested PCR reactions were performed to amplify the 5 and 3 ends of the NaK2 cDNA using dilutions of the initial RACE reactions as template. Lane 1: 1 kb ladder, Lanes 2 and 3: 5 nested RACE, Lanes 4 and 5: 3 nested RACE. Bands were excised and gel purified for cloning purposes.

PAGE 111

90 products of two such reactions. As with NaK3, the sequenced derived from the 5 and 3 RACE reactions was used to design gene specific primers in order to amplify the full-length cDNA. PCR Amplification of the Full-Length NaK2 cDNA The full-length PCR for NaK2 was performed using the primers MG36 and MG37 and the following parameters: 94C x 5 min presoak; 30 cycles of 94C x 30 sec, 59C x 30 sec, 72C x 1 min; 72C x 10 min final extension. Again, PCR reactions were performed three times and gel-purified products were cloned into pCR2.1 TOPO for sequence analysis. Figure 4-9 illustrates a gel with three bands that have been excised, corresponding to the triplicate PCR reactions for the full-length NaK2 cDNA. Sequence Figure 4-9. PCR Amplification of the Rabbit Na+, K+ ATPase 2 Subunit cDNA. Primers were designed to amplify the full-length cDNA of NaK3. Bands were excised and gel purified for cloning purposes. Lane 1: 1 kb ladder, Lanes 2-4: full-length PCR product, Lane 5: H2O as template. data from the 5 and 3 nested RACE reactions, together with that from the full-length cDNA PCR, was used to derive the complete cDNA sequence for NaK2. As with NaK3, six independently derived sequences were used to generate the final cDNA sequence. The rabbit NaK2 cDNA sequence was deposited into Genbank and assigned the accession number AY069937 (Figure 4-10).

PAGE 112

91 Figure 4-10. GenBank Accession Record for Rabbit NaK2.

PAGE 113

92 Discussion Cloning of the full-length cDNAs for rabbit NaK3 and NaK2 was completed using a combination of degenerate PCR and RACE. It was first important to compare these sequences to the known X+, K+ ATPase sequences from other species to ensure the specificity of each. The completed cDNA sequences were fed into a translation program at Molecular Toolkit ( http://arbl.cvmbs.colostate.edu/molkit/ ) (last accessed October 31, 2003) to generate the deduced amino acid sequences. The open reading frame was chosen based on alignments with known NaK3 or NaK2 nucleotide sequences from other species. The deduced amino acid sequence for NaK3 consisted of 279 amino acids while that of NaK2 was 290 amino acids. Comparison of the deduced amino acid sequences for rabbit NaK3 and NaK2 to known sequences revealed very high homology to previously published Na+, K+ ATPase 3 and 2 sequences, respectively (Tables 4-2 and 4-3). Notably, homology to other members of the XK family was very low for both NaK3 and NaK2, suggesting that the cloned sequences were indeed specific to the intended subunit. These amino acid sequences were then examined for the well-known markers of the subunit family. The probable topology of the XK subunits is shown in Figure 1-4. It consists of a short intracellular domain, a single transmembrane domain, and a large extracellular region that contains three disulfide bridges and a variable number of N-linked glycosylation sites. These features of the XK subunits are highlighted in the amino acid sequence alignment of the rabbit X+, K+ ATPase subunits, NaK2, NaK3, NaK1 and gHK in Figure 4-11.

PAGE 114

93 Table 4-2. Protein Sequence Comparison of the Rabbit Na+, K+ ATPase 3 Subunit to Other NaK SubunitsXK Subunit % Similarity % Identity Rat NaK3 88 76 Mouse NaK3 84 73 Human NaK3 92 85 Rabbit NaK1 53 36 Rabbit NaK2 61 47 Rabbit gHK 53 34 Table 4-3. Protein Sequence Comparison of the Rabbit Na+, K+ ATPase 2 Subunit to Other NaK SubunitsXK subunit % Similarity % Identity Rat NaK2 98 97 Mouse NaK2 99 97 Human NaK2 98 98 Rabbit NaK1 55 40 Rabbit NaK3 61 47 Rabbit gHK 56 39 First, the deduced amino acid sequences of rabbit NaK3 and NaK2 were examined for probable transmembrane domains. A single transmembrane domain is one of the defining characteristics of the X+, K+ ATPase subunits. Importantly, the deduced amino acid sequences of NaK3 and NaK2 both contain an apparent single transmembrane domain, as evidenced by the Kyte-Doolittle hydropathy plots shown in Figure 4-12. A single broad peak starting around amino acid 35 and extending to about amino acid 62 indicates the presence of a likely transmembrane domain for both the NaK3 (Figure 4-12A) and NaK2 (Figure 4-12B) sequences.

PAGE 115

94 Figure 4-11. Alignment of Amino Acid Sequences for the Rabbit X+, K+ ATPase subunits. Deduced amino acid sequences of NaK1, NaK2, NaK3, and gHK were aligned using Clustal W. Highlighted in red are the conserved cysteine residues that comprise three disulfide bridges. N-linked glycosylation sites are denoted in green. The predicted transmembrane segment is underlined and highlighted in purple. Secondary structure predictions were also generated for NaK3 and NaK2 (Figure 4-13A and B). As expected, these predictions reveal that the amino acids purported to lie within the transmembrane regions of NaK3 and NaK2 are also most likely to form an alpha helix, as evidenced by the cluster of blue lines surrounding position 50 in the amino acid sequences of NaK2 and NaK3. The largest domain of the XK subunits lies in the extracellular region. The secondary structure predictions reveal a mixture of helix, sheet and coil conformations in this area, an observation that is consistent with the fact that not much is known about the structure of this region.

PAGE 116

95 Figure 4-12. Kyte-Doolittle Hydropathy Plots for Rabbit Na+, K+ ATPase 3 and 2 Subunits. Hydropathy plots for NaK3 (Panel A) and NaK2 (Panel B) were generated using their deduced amino acid sequences and a web program available at http://fasta.bioch.virginia.edu/o_fasta/grease.htm (last accessed March 21, 2004). Another important characteristic of the X+, K+ ATPase subunits is the presence of three highly conserved disulfide bridges in the extracellular domain. The six conserved cysteines that give rise to these disulfide bridges are highlighted in red in the alignment pictured in Figure 4-11. The positions of the cysteine residues are conserved between

PAGE 117

96 Figure 4-13. Secondary Structure Prediction for NaK3 and NaK2. The deduced amino acid sequences of NaK3 (Panel A) and NaK2 (Panel B) were fed into a secondary structure prediction program (http://pbil.ibcp.fr/cgi-bin/secpred_gib.pl via GorII at www.expasy.ch ) (last accessed October 31, 2003). Numbering from left to right represents each amino acid, progressing from the N-terminus to the C-terminus. Residues in blue are scored most likely to lie in a helical formation, residues in red are most likely to be found in a-sheet and residues in purple are likely to be found in a coil formation. NaK2 and NaK3 and provide further evidence that these sequences represent the remaining members of the rabbit XK family. Indeed, the conservation of these residues is consistent with evidence demonstrating that the disulfide bridges are required for enzyme assembly (12). The number of N-linked glycosylation sites, highlighted in green in Figure 4-11, in the XK subunits is variable. NaK2 contains nine putative sites while NaK3 has only two. The remaining members of the rabbit XK family, NaK1 and gHK, contain three and seven sites, respectively. This variability among the XK subunits is consistent with

PAGE 118

97 the observation that, while glycosylation is necessary for enzyme assembly and activity, the number and position of the glycosylation sites does not seem to have an effect (12). In conclusion, with the cloning of NaK3 and NaK2, the complete cDNAs for all four members of the rabbit XK subunit family are now in hand. Obtaining the four subunits is vital to the establishment of a functional expression system for the rabbit colonic H+, K+ ATPase because the identity of the colonic subunit is not known. Expression of the NaK3 and NaK2 subunits, which will be discussed in the following chapter, will allow for all four XK subunits to be tested in a mammalian tissue culture system for their ability to pair with and form a functional complex with HK2a and HK2c.

PAGE 119

CHAPTER 5 EXPRESSION AND CHARACTERIZATION OF THE RABBIT COLONIC H+, K+ ATPASE Introduction Degenerate PCR and Rapid Amplification of cDNA Ends (RACE) were undertaken in our laboratory to clone the HK2 cDNA from rabbit kidney. These experiments yielded two rabbit renal cDNAs, designated HK2a and HK2c (16). They share approximately 4 kb of sequence in common and differ only at their 5 ends. The deduced amino acid sequences revealed that HK2c has a 61 amino acid N-terminal extension (Figure 1-7). These two different cDNAs are under the control of the same promoter and are products of alternative splicing (130). It was one of the aims of this project to functionally express HK2a and HK2c in order to determine their respective pharmacological and activity profiles. The H+, K+ ATPase is a tetramer of two and two subunits. In order to express a functional enzyme, the subunit must be co-expressed with a subunit. As discussed in the previous chapter, the subunit that pairs with the HK2 isoform has not been clearly identified. The four possible mammalian X+, K+ ATPases subunits are NaK1, NaK2, NaK3 and gHK. Mammalian cell culture expression systems have been established for the human (49), guinea pig (6), and rat (96) colonic H+, K+ ATPases. Results from these expression systems are summarized in Table 5-1. These expression systems are at odds with each other concerning subunit pairing and ouabain resistance. For example, while 98

PAGE 120

99 Table 5-1. Summary of HK2 Expression SystemsSpecies / Composition Fold Activity over Control Resistance to [Ouabain] Humana ATP1AL1/ gHK 2 1 M ATP1AL1/ NaK1 0 N/A Guinea Pigb HK2/gHK 4 1 M HK2/ NaK1 5 1 M Ratc HK2/ NaK1 20 1 mM HK2/NaK3 25 1 mM aBased on 86Rb uptake experiments in Grishin et al.(49). bBased on ATP hydrolysis experiments in Asano et al.(6). cBased on 86Rb uptake experiments in Sangan et al.(96). Grishin et al. reported no detectable activity using NaK1 (49), the Asano (6)and Sangan (96) studies showed levels of activity using this subunit that differed by four-fold. Furthermore, the latter two groups reported levels of ouabain resistance that differed by three orders of magnitude. It is not clear why different activity and pharmacological profiles have been observed for these HK2 isoforms. It is likely that the contrasting profiles are due to the manner in which these isoforms are expressed and assayed for activity. It is known that the subunit of the X+, K+ ATPases plays an important role in regulation of the enzyme (19). Since the subunit for the colonic H+, K+ ATPase has not been clearly defined, it may be that the contrasting profiles discussed above are due to the different subunits used to express the H+, K+ ATPase (25). Attempts to establish a functional expression system for the rabbit renal H+, K+ ATPase have previously been made in our laboratory (83). Following unsuccessful attempts using an episomal expression system as well as a stable transfection system, transient transfection was undertaken. Expression of NaK1 or NaK3 subunits and HK2a and HK2c was confirmed in the transient system. ATP hydrolysis assays were performed to assess the activity of these H+, K+ ATPases, but no activity above

PAGE 121

100 background was detected. It was concluded that functional expression of the rabbit H+, K+ ATPase was not successful in our lab perhaps because the proper subunit was not being used. It became the goal of this project to establish a functional expression system for the rabbit colonic H+, K+ ATPases, not only to look at the differences between HK2a and HK2c, but also to resolve the controversy surrounding the identity of the proper subunit for this enzyme. In order to resolve the subunit issue, the rabbit H+, K+ ATPase expression system would be the first mammalian system to use all four possible subunit partners in combination with the HK2 subunits. Results Correction of HK2 cDNAs Before further attempts at a functional expression system were made, it was deemed necessary to sequence the HK2 cDNAs. Given the fact that extensive cloning had been done using the cDNAs, it was possible that spontaneous mutations had occurred. The HK2a cDNA was sequenced in its entirety in both the forward and reverse directions. Since HK2a and HK2c differ only at the N-terminus, only the most 5 end of the HK2c cDNA was sequenced. The nucleotide sequence results were aligned with the published cDNA sequences (16). The amino acid changes that were found are summarized in the alignment pictured in Figure 5-1. Not shown are the 61 N-terminal amino acids of HK2c; no amino acid changes were found in this region. Of the nine amino acids that were changed, three of these (starred in Figure 5-1) were determined to be the most potentially deleterious based on conservation and/or charge effect. These three amino acids, a proline at position 235, a glutamic acid at position 597,

PAGE 122

101 Figure 5-1. Mutated Amino Acids in HK2a. Figure 3-1 is shown here to highlight mutations found in the HK2a cDNA. Those amino acids that differed from the published sequence are shown in red above their respective positions. Starred amino acids were deemed to be potentially destructive changes and were mutated back to the original sequence.

PAGE 123

102 and a threonine at position 1006, were changed back to the original amino acids using site-directed mutagenesis. The primers used to perform QuikChange (Stratagene) mutagenesis on the HK2a cDNA are shown in Table 5-2. Plasmid pTLC46 was used as template in a PCR reaction to change the threonine at position 1006 back to an isoleucine. Primers MG 48 and MG 49 were used and the amino acid change was confirmed by sequence analysis. The resulting plasmid, pMLG59, was used as template in a PCR reaction to change the Table 5-2. Primers Used for Site-directed Mutagenesis of HK2a. Primer Sequencea Amino Acid Change MG48 (Fd) gtggccgtaccccatgccatc ctgatctgggtatacgatg TI1006 MG49 (Rv) catcgtatacccagatcaggat ggcatggggtacggccac TI1006 MG71 (Fd) cctccaacttatgttttgtgggc ctcttatcaatgattgatcctcc EG597 MG72 (Rv) ggaggatcaatcattgataagaggcc cacaaaacataagttggagg EG597 MG73 (Fd) ctggagagtctgagccgcagt cccgctcaagtgggttc PS235 MG74 (Rv) gaacccacttgagcgggact gcggctcagactctccag PS235 aMutated codons are underlined. glutamic acid at position 597 back to a glycine with the primers MG 71 and MG 72. The resulting plasmid, pMLG83, was used as template in a final PCR reaction to change the proline at position 235 back to a serine using the primers MG 73 and MG 74. The PCR parameters were the same for all of the QuikChange reactions, as described in Chapter 2. The HK2a cDNA in the resulting plasmid, pMLG84, was sequenced in its entirety in the forward and reverse directions to confirm the amino acid changes and to ensure that no undesired mutations had occurred. The 5 end of the HK2c cDNA in plasmid pTLC35 was sequenced in the forward and reverse directions and no amino acid changes were found. It was necessary to generate a corrected HK2c cDNA that would share 3 sequence with the corrected

PAGE 124

103 HK2a cDNA in pMLG84. In order to do this, the unique N-terminus of the HK2c subunit was ligated into pMLG84. Plasmid pTLC35 and pMLG84 were digested with the restriction enzymes NotI and EcoRV. The 300 bp fragment from pTLC35 and the 7700 bp fragment from pMLG84 were gel purified and combined in a ligation reaction. The resulting plasmid, pMLG94, was confirmed by restriction analysis. Expression Constructs The corrected HK2a and HK2c cDNAs were used to generate the subsequent plasmids needed to establish a functional expression system for the rabbit colonic H+, K+ ATPase. The mammalian expression plasmid pBUDCE4 (Figure 5-2) contains two multiple cloning sites driven by two different promoters. Dr. Tamara Otto (83) had Figure 5-2: pMLG96. The plasmid pMLG96 was created using the backbone vector pBUDCE4, purchased from Invitrogen. It contains two multiple cloning sites driven by two promoters. The HK2a cDNA was cloned in front of the EF1 promoter using the restriction sites indicated; the gHK cDNA was cloned in front of the CMV promoter using the restriction sites indicated. The plasmid contains an SV40 promoter for replication in mammalian cells and a pUC origin for replication in E. coli. A zeocin resistance gene is also located on the plasmid and can be used for selection in either E.coli or mammalian cells.

PAGE 125

104 previously engineered the HK2 cDNAs with a NotI site at the 5 end and a BstBI site at the 3 end. These sites enabled the cDNAs to be cloned in front of the EF1 promoter of pBUDCE4 and in frame with the C-terminal V5 epitope. The subunits were cloned in front of the CMV promoter using various restriction enzymes sites. The primers used to amplify the coding regions and facilitate the cloning of the subunits into pBUDCE4 are summarized in Table 5-3. Eight plasmids had to be constructed containing either HK2a or HK2c in combination with each of the four possible subunit partners. Figure 5-3 shows the expression of HK2a and HK2c from two of these plasmids. Evidence of subunit protein expression is shown in the first lane of each of the Western blots pictured in Figures 5-4 and 5-5. Each of the plasmid constructions is discussed below. Table 5-3. Primers Used for PCR Amplification or Mutation of XK SubunitsPrimer Sequence cDNA Purpose MG42 (Fd) aagctt ccaagatggtcatcca NaK2 Amplify coding region MG43 (Rv) gaggaagggggatcc ggttttgttb NaK2 Amplify coding region MG15 (Fd) aagctt gtccgcacgcacaccatgacga NaK3 Amplify coding region MG16 (Rv) cctactcataggatcc acgtgctatgb NaK3 Amplify coding region MG46 (Fd) gtcgacaccatggccgccttgcaggagc gHK Amplify coding region MG47 (Rv) cccacgaccgcgggatcc cttctggatcb gHK Amplify coding region MG56 (Fd) ccgaccccagcttcggc ttcaaggaaggaaagd gHK DG171 MG57 (Rv) ctttccttccttgaagcc gaagctggggtcggd gHK DG171 a5 HindIII site is underlined. bBamHI site is underlined. c5 AccI site is underlined. dMutated codon is underlined.

PAGE 126

105 Figure 5-3. Western Blot Analysis of HK2a and HK2c Subunits. Plasma membranes were prepared from COS-1 cells that were transfected with pMLG92 (HK2a) and pMLG95 (HK2c). Western blot analysis was performed using an antibody directed against the V5 epitope tags located at the C-termini of HK2a and HK2c. Lane 1: untransfected control; lane 2: HK2a; lane 3: HK2c. NaK1 Dr. Tamara Otto cloned the cDNA for NaK1 into pBUDCE4 using the restriction enzyme sites XbaI and BamHI to create plasmid pTLC33 (83). The cDNA for NaK1 is not in frame with the C-terminal epitope tag in pBUDCE4, but a monoclonal antibody for the protein is available. In order to move the corrected HK2 cDNAs into pBUDCE4 with NaK1, pTLC33 was completely digested with BstBI and partially digested with NotI due to the presence of an additional NotI site within the NaK1 cDNA. Plasmids pMLG84 and pMLG94, containing the corrected HK2a and HK2c cDNAs, respectively, were digested completely with BstBI and NotI. The HK2a and HK2c fragments were gel purified along with the pBUDCE4 backbone containing the NaK1 cDNA and combined in a ligation reaction. The resulting plasmids, pMLG99

PAGE 127

106 (HK2a/NaK1) and pMLG100 (HK2c/NaK1) were confirmed by restriction analysis. NaK2 Using the sequence data derived from PCR amplification of the full-length cDNA of NaK2 (see Chapter 4), primers were designed to amplify the coding region of the cDNA. The forward primer MG 42 was designed with a HindIII site at the very 5 end. The reverse primer MG 43 was designed to mutate the stop codon and contained a BamHI site to facilitate in-frame cloning with the C-terminal c-myc epitope. RT-PCR was performed using rabbit kidney medulla RNA as a template and the following cycling parameters: 94C x 5 min presoak; 5 cycles of 94C x 30 sec, 55C x 30 sec, 72C x 1 min; 25 cycles of 94C x 30 sec, 58C x 30 sec, 72C x 1 min; 72C x 10 min final extension. The resulting 884 bp product was gel purified and cloned into pCR2.1 to make pMLG47.5. Before any subsequent cloning was done, the cDNA was sequenced in its entirety in the forward and reverse directions to ensure that no PCR errors had occurred. In order to move the NaK2 cDNA into pBUDCE4, the plasmids pMLG47.5 and pBUDCE4 were digested with HindIII and BamHI. The vector and insert bands were gel purified and combined in a ligation reaction. The resulting plasmid, pMLG49, was confirmed by restriction analysis. To generate pBUDCE4 constructs containing both the HK2 and XK cDNAs in comibination, the plasmids pMLG49, pMLG84 and pMLG94 were digested with BstBI and NotI. Again, the NaK2-containing vector and the HK2 insert bands were gel purified and combined in a ligation reaction. The resulting plasmids, pMLG92 (HK2a/NaK2) and pMLG95 (HK2c/NaK2), were confirmed by restriction analysis.

PAGE 128

107 NaK3 The coding region for NaK3 was amplified using the primers MG 15 and MG 16 in the manner described above for NaK2. The pBUDCE4 plasmids containing NaK3 alone (pMLG23), HK2a/NaK3 (pMLG97) and HK2c/NaK3 (pMLG98) were constructed in exactly the same way as that described above for NaK2. Gastric HK The forward primer MG 46 and the reverse primer MG 47 were designed to amplify the coding region of gHK. The 5 primer contained an AccI site while the 3 primer contained a BamHI site and changed the stop codon in order to clone the cDNA in frame with the C-terminal c-myc epitope. The following PCR parameters were used to amplify the coding region of the gHK cDNA: 94C x 5 min, 25 cycles of 94C x 30 sec, 64.9C x 30 sec, 72C x 1 min; 72C x 10 min final extension. The template used was the plasmid pTLC50, which contains a gHK cDNA lacking convenient restriction enzyme sites (83). The resulting 901 bp fragment was cloned into pCR2.1 to generate pMLG54 and sequenced. A mutation was found in the extracellular region, which resulted in a change at amino acid 171 from a glycine to an aspartic acid. This mutation was carried over from pTLC50 and was not due to a PCR error. In order to change this amino acid back to the original glycine, QuikChange mutagenesis was performed using the primers MG56 and MG57 and the QuikChange cycling parameters described in Chapter 2. The resulting plasmid, pMLG65, was sequenced in its entirety to confirm the change and to ensure that no undesired mutations occurred. The gHK cDNA was then moved from pMLG65 into pBUDCE4 using the restriction enzyme sites AccI and BamHI to generate pMLG66. The pBUDCE4 plasmids containing HK2a/gHK (pMLG93) and

PAGE 129

108 HK2c/gHK (pMLG96) were constructed in the same manner described above for NaK2 and NaK3. Co-immunoprecipitation of the Rabbit Colonic H+, K+ ATPases: It was thought that previous attempts in our laboratory to establish a functional expression system had been unsuccessful perhaps because the proper subunit was not being used. The subunit of the X+, K+ ATPases is known to play a role in assembly and trafficking of the enzyme (11, 45). It was therefore necessary to make sure the enzyme was being assembled and trafficked to the plasma membrane before the various H+, K+ ATPases were assayed for activity. Co-immunoprecipitation of the and subunits from the plasma membranes of transfected cells was used as a means to evaluate which / pairs were interacting and were located at the cell membrane. COS-1 cells were transfected with each of the eight pBUDCE4 plasmids containing all HK2/XK combinations in turn. As a negative control, pMLG66, containing only the gHK cDNA and no subunit cDNA, was transfected into COS-1 cells as well. Plasma membranes were prepared two days post-transfection. One milligram of membrane protein was solubilized by a 30 min incubation in a NP-40 containing lysis buffer and membranes pelleted by a 20 min spin at 16000 x g. The protein-containing supernatant was immunoprecipitated overnight using an anti-V5 antibody that was covalently linked to a solid support, Amino Link Plus (Promega). The following day, the antibodyAmino Link Plus-antigen complex was washed and a low-pH IgG elution buffer used to elute the antigen complex. Fifty micrograms of membrane protein, and the flow-through, wash and eluate fractions were run on a SDS-PAGE gel and transferred to a nitrocellulose membrane. Western blot analysis was performed using an anti-NaK1 or an anti-c-myc

PAGE 130

109 antibody to detect each of the four XK subunits. Presence of a subunit in any of the eluate fractions indicated that the subunit in question was likely interacting with the HK2 subunit and that the complex was likely being properly trafficked to the plasma membrane. Following co-immunoprecipitation, it appeared that only the gHK subunit was interacting with HK2. This result is evident in lanes 7-9 of Figure 5-4A and Figure 5-4B where the presence of the gHK subunit shows that it co-immunoprecipitated with both HK2a and HK2c, respectively. In order to demonstrate that there was not a non-specific interaction between gHK and the anti-V5 antibody, plasma membranes from COS-1 cells transfected with only the gHK cDNA were immunoprecipitated with the anti-V5 antibody (Figure 5-4C). The absence of any subunit in the eluate lanes (lanes 6-9) was evidence that the interaction seen in Figures 5-4A and B was specific. Co-immunoprecipitations were also performed using plasma membranes from COS-1 cells transfected to express the HK2 subunits with each of the NaK subunits. None of the NaK subunits were co-immunoprecipitated with HK2c (Figure 5-5) or HK2a. This negative result was observed consistently, in at least two independent experiments. It was therefore concluded that the combinations of HK2a/gHK and HK2c/gHKwere the most likely to exhibit ATPase activity.

PAGE 131

110 Figure 5-4. Co-immunoprecipitation of HK2 Subunits and gHK. COS-1 cells were transfected to express HK2a/gHK (A), HK2c/gHK (B), or gHK alone as a negative control (C). Plasma membranes were immunoprecipitated with an anti-V5 antibody directed against the epitope tags of HK2a and HK2c. Western blot analysis was performed with an anti-myc antibody directed against the epitope tag of gHK. Lane designations are as follows: Lane 1: 50g membrane protein, lane 2: flow-through, lanes 3-5: washes 1-3, lanes 6-9: eluates 1-4.

PAGE 132

111 Figure 5-5. Co-immunoprecipitation of HK2c Subunits and NaK1-3. COS-1 cells were transfected to express HK2c/NaK1 (A), HK2c/NaK2 (B), and HK2c/NaK3 (C). Plasma membranes were immunoprecipitated with an anti-V5 antibody directed against the epitope tag of HK2c. Western blot analysis was performed with an anti-NaK1 antibody or an anti-myc antibody directed against the epitope tags of NaK2 and NaK3. Lane designations are as follows: Lane 1: 50 g membrane protein, lane 2: flow-through, lanes 3-5: washes 1-3, lanes 6-9: eluates 1-4.

PAGE 133

112 Analysis of ATPase Activity of the Rabbit Colonic H+, K+ ATPases The method chosen for measuring ATPase activity was a coupled assay approach to measure K+-dependent ADP production by the H+, K+ ATPase. ATPase activity was indicated by a decrease in fluorescence as the ADP-dependent reduction of NADH by lactate dehydrogenase occurred (42) (see Chapter 2). A standard curve was generated using known amounts of ADP and was used to extrapolate the mole amounts of ADP released following ATP hydrolysis by the H+, K+ ATPase. As a positive control, ATP hydrolysis activity of the gastric H+, K+ ATPase was measured using mouse gastric mucosa (Figure 5-6). Blue bars represent activity in the absence of K+, the presence of K+ and the presence of K+ plus the gastric H+, K+ ATPase inhibitor SCH28080. The level of specific activity of the membrane fraction from mouse stomach Figure 5-6. Analysis of K+-dependent ATPase Activity from Mouse Gastric Mucosa. A coupled ATP hydrolysis assay was performed as described in Chapter 2. Two g of membrane protein from mouse stomach was used in the assay. Fluorescent measurements were recorded 30 min following addition of reaction buffer and were taken in the absence of K+, the presence of K+ and the presence of K+ plus the gastric H+, K+ ATPase specific inhibitor, SCH28080 (SCH). Activity is presented as moles of ADP hydrolyzed per mg of protein per hour.

PAGE 134

113 tissue was approximately 2 mol ADP/mg/hr, a level that was comparable to that measured using the same assay and protein sample in the Wingo laboratory (personal communication Jiahong Shao). Importantly, the activity was SCH28080-inhibited, demonstrating that it is specific to the gastric H+, K+ ATPase. This result was important not only because it demonstrated that the assay worked in our hands, but also because it was the first demonstration of ATPase activity of a mammalian enzyme in our laboratory. The next step was to measure ATPase activity from a recombinant mammalian enzyme. Since expression levels of HK2c seemed to be higher than those of HK2a (see Figure 5-3), the HK2c/gHKconstruct was chosen to first evaluate ATPase activity in the plasma membranes of transfected COS-1 cells. Mock transfection as a negative control and transfection with the plasmid pMLG96 were performed. Two days later, plasma membranes were isolated and rendered permeable by one freeze/thaw cycle. Each assay was carried out in duplicate using 50 g of membrane protein. SCH28080 and ouabain were included in the assay to inhibit any endogenous gastric H+, K+ ATPase or Na+, K+ ATPase activity. Fluorescent readings were taken at 3, 6, 9, 10, 20 and 30 minutes following addition of the reaction buffer to the assay tube containing the membrane protein fraction of transfected COS-1 cells. The results are summarized in Figure 5-6 and are presented as K+-dependent ATPase activity. Blue bars represent activity in the membrane fraction from mock-transfected cells while magenta bars represent activity in the membrane fraction from pMLG96-transfected cells. This activity is defined as moles of ADP hydrolyzed per mg of protein per hour in the presence of K+, minus that hydrolyzed in the absence of K+ (K+ activity (no K+ activity)). The highest activity was seen at 3 minutes following addition of membrane protein and

PAGE 135

114 Figure 5-7. ATP Hydrolysis Activity of Transfected COS-1 Cells. K+-dependent ATPase activity is defined as mol of ADP hydrolyzed per mg of membrane protein per hour in the presence of K+ minus that hydrolyzed in the absence of K+. Blue bars represent activity levels in mock transfected cells and magenta bars represent activity in pMLG96 (HK2c/gHK)-transfected cells. (A) K+-dependent ATPase activity was measured at 3, 6, and 9 minutes following addition of reaction buffer. (B) K+-dependent ATPase activity was measured at10, 20, and 30 minutes following addition of reaction buffer (see Chapter 2). measured nearly 0.8 moles of ADP hydrolyzed per mg of protein per hour. Unfortunately, the level of activity was the same between plasma membrane fractions from mock-transfected and pMLG96-transfected cells. That result stayed the same at all subsequent time points and in multiple independent experiments. Since Western blot analysis and co-immunoprecipitation experiments showed that the HK2c/gHK combination had high expression levels and was interacting and being properly trafficked, it was concluded that this construct had the best possible chance of

PAGE 136

115 demonstrating activity. Because there were no detectable levels of activity with the HK2c/gHK combination, this line of research was set aside. Discussion The HK2a, HK2c, and gHK cDNAs were sequenced in their entirety to make sure their sequence matched the published nucleotide sequence. The mutations that were found were likely due to repeated engineering in E. coli over the course of several years and multiple graduate students. The mutations that were considered the most potentially destructive were corrected and the new cDNAs cloned into a series of pBUDCE4 plasmids. In all, eight plasmids were generated in order to test the interaction between all possible HK2/XK combinations (see column 1 in Table 5-4). Table 5-4. H+, K+ ATPase Expression Constructs and Summary of Co-immunoprecipitation Data. Plasmid cDNAs interaction pMLG93 HK2a + gHK Yes pMLG96 HK2c + gHK Yes pMLG92 2aNaK No pMLG95 HK2c +NaK No pMLG97 HK2a + NaK3 No pMLG98 HK2c + NaK3 No pMLG99 HK2a + NaK1 No pMLG100 HK2c + NaK1 No The use of co-immunoprecipitation to evaluate the interaction between the HK2 subunits and all four possible XKsubunits represents the first such look into / interaction in a mammalian expression system, to our knowledge. The systematic testing of the eight HK2/XK combinations by co-immunoprecipitation was deemed the best possible approach in determining which pair(s) would be likely to exhibit activity. Surprisingly, only one of the XK subunits appeared to interact with either HK2

PAGE 137

116 subunit. It was consistently observed that gHK was able to co-immunoprecipitate with HK2a and HK2c while no interaction was observed with any of the NaK subunits. The results of the co-immunoprecipitation experiments are presented in the third column of Table 5-4. While the goal of determining the activity profiles of the HK2a and HK2c subunits was not realized, the co-immunoprecipitation data enables an interpretation concerning the identity of the subunit for rabbit HK2. Because gHK was the only one of the four possible XK partners to pair with and traffic HK2a and HK2c to the plasma membrane, it would appear that it was the best possible subunit for the rabbit enzyme. However, confirmation of this hypothesis requires establishment of a functional expression system for the rabbit enzyme. Given the consistent observation that gHK was able to interact with and properly traffic both HK2a and HK2c, it was disappointing that there was no detectable ATP hydrolysis activity in the plasma membrane fractions of the colonic H+, K+ ATPase-expressing cells. The lack of detectable activity in transfected COS-1 cells was perplexing, especially considering the fact that activity has been reported for the colonic H+, K+ ATPase in heterologous expression systems for the rat (96), guinea pig (6) and human (49) enzymes. However, the inconsistencies among these expression systems (see Table 5-1) and the absence of subsequent publications from these laboratories lead one to question the reliability of the respective expression systems. Nevertheless, the question remains as to why COS-1 cells that are expressing an assembled colonic H+, K+ ATPase at the plasma membrane do not exhibit ATP hydrolysis activity in our laboratory. Recent work in our laboratory regarding the rabbit HK2 gene showed that transcription of the

PAGE 138

117 HK2 gene appeared to be repressed unless rabbit kidney cells were grown to 100% confluency in tissue culture dishes (130). This suggests that specific cellular signals are required in order for transcriptional activation of the HK2 gene. While the identity of this signal(s) is still under investigation, a similar event may be required in order to activate the protein itself. It is possible that further investigation into the regulation of the HK2 gene could lend insight into why the expressed H+, K+ ATPase enzyme is not active in transfected COS-1 cells. Elucidation of the mechanism of activation of the colonic H+, K+ ATPase in rabbit kidney cells could perhaps enable the successful establishment of a functional mammalian expression system for this enzyme.

PAGE 139

CHAPTER 6 EARLY TRANSCRIPTIONAL EFFECTS OF ALDOSTERONE IN MOUSE IMCD3 CELLS Introduction Known molecular targets of aldosterone action include the basolateral Na+, K+ ATPase and the apical epithelial sodium channel (ENaC), both of which are critical for Na+ absorption. Aldosterone has both short and long term effects on these ion transporters. The transcriptional effects of aldosterone on ENaC and the Na+, K+ ATPase occur after four hours; these late effects of aldosterone on ion transport are well characterized (120). The more immediate transcriptional effects of aldosterone have not been fully investigated, but certain candidate genes have been identified. For example, aldosterone induces the expression of the serum and glucocorticoid-regulated kinase (sgk) as soon as 30 min after hormone treatment (77, 98). Sgk has been linked to increases in both the number and the activity of ENaC (18). Expression of sgk in Xenopus oocytes results a three-fold increase in the number of functional ENaC at the cell surface (4). Sgk-mediated phosphorylation of the ubiquitin ligase Nedd-4 leads to inhibition of ENaC subunit degradation and therefore an increase in ENaC activity (31, 98). Aside from sgk, there is limited information on early aldosterone-responsive genes. SAGE technology has recently been used to identify 34 aldosterone-induced transcripts and 29 aldosterone-repressed transcripts in a mouse kidney cortical collecting duct cell line after 4 h of aldosterone treatment (91). However, the acute transcriptional effects of aldosterone should occur in a much shorter time frame. 118

PAGE 140

119 Studies with primary cultures derived from the inner medullary collecting duct (IMCD) have suggested that this is a target epithelium for the action of aldosterone and may be an important terminal site of Na+ absorption and acid secretion in the collecting duct (100, 101, 122). There is a dramatic increase in Na+ transport in cultured IMCD cells in response to aldosterone (57, 58). However, the factors that mediate the acute effects of aldosterone largely remain to be defined. Previous studies that have examined the early aldosterone responsive genes used non-mammalian cell lines or considered responses several hours after exposure to hormone (91, 103). It was the purpose of this project to use microarray technology to analyze a mouse kidney inner medullary collecting duct cell line for acute aldosterone regulation following 1 h of exposure to hormone. Numerous, previously unreported, aldosterone-regulated transcripts have been identified using this method and the regulation of selected genes has been verified using traditional biochemical approaches. These findings represent a novel list of early aldosterone-regulated transcripts. Results Characterization of mIMCD3 Cells In order to demonstrate that the mIMCD3 cell line had retained properties reflecting collecting duct cells, the presence of the alpha subunit of the epithelial sodium channel (ENaC was investigated as a representative collecting duct specific transcript (70). RT-PCR was performed to amplify a 505 base pair fragment of ENaC (Figure 6-1). The resulting product, evident in lane 2 of Figure 6-1, was gel purified and sequenced;

PAGE 141

120 Figure 6-1. Characterization of mIMCD3 Cells. Primers were designed according to published sequences to amplify fragments of ENaC, 11HSD2, and MR from mIMCD3 cells. RT-PCR products were separated on a 1% agarose gel and visualized with ethidium bromide. Presence or absence of reverse transcriptase and template is indicated by +/-; in the absence of an RNA template, an equal volume of H2O was used. the product was identical to the published ENaC sequence. The presence of this transcript indicated that mIMCD3 cells were indeed representative of the collecting duct. Next, RT-PCR was performed to amplify 480 and 238 base pair fragments of hydroxysteroid 11 dehydrogenase 2 (11HSD2) and the mineralocorticoid receptor (MR), respectively (lanes 5 and 8 of Figure 6-1). 11HSD2 and the MR represent transcripts specific to aldosterone-responsive cells. The resulting products were gel purified and their sequences determined. Both sequences were identical to the published sequence of the transcript in question. Another marker of aldosterone-responsiveness is induction of a known aldosterone-responsive transcript, sgk. As a preliminary indication of whether aldosterone regulates sgk in this cell line, mIMCD3 cells were treated for 1 h with 10-6 M aldosterone or vehicle (ethanol), total RNA was isolated and Northern blot

PAGE 142

121 analysis was performed (Figure 6-2). Densitometry analysis of the results revealed an approximate five-fold increase in sgk expression over control cells (Table 6-1), a value that is comparable to levels observed in other studies (77, 98). In summary, mIMCD3 cells were found to be representative of the collecting duct and also to be aldosterone-responsive. These cells were therefore selected to study the acute transcriptional effects of aldosterone. Figure 6-2. Up-Regulation of Sgk by Aldosterone. Total RNA was isolated from mIMCD3 cells treated with vehicle or 10-6 M aldosterone for 1 h. Top panel is an ethidium bromide stained 1 % denaturing agarose gel containing the RNA samples. The 28S and 18S ribosomal RNA bands are indicated. Northern analysis was performed using cDNA probes for sgk and GAPDH. Microarray Analysis To prepare RNA samples for microarray analysis, mIMCD3 cells were treated for 1 h with vehicle or 10-6 M aldosterone. Northern blot analysis for sgk was conducted to

PAGE 143

122 Table 6-1. Comparison of Fold Induction Values Transcript GenBank Microarray Fold Changea Northern Fold-Changeb Real Time PCR Fold Change Sgkc AF205855d 3.1 +/0.12 5.1 +/0.7e 5.2 +/0.7h CTGF M70642 2.1 +/0.10f 3.0 +/0.6g No data Period Homolog AF022992 4.1 +/0.20 7.9 +/3.0g 7.2 +/1.5h Preproendothelin U35233 2.5 +/0.23 6.2 +/1.7g 3.0 +/0.3h aFold induction values were calculated by Affymetrix Suite Version 4, and are listed as +/standard error, n=3.bFold induction values were calculated using densitometry values from Northern blots and are listed as +/standard error, n=3.cFor sgk, n=6.dA BLAST search indicated that mouse sgk is most similar to human sgk-1. ep<0.001. fFor CTGF, n=2.gp<0.1. hp<0.01. confirm the hormone response before the samples were used in the microarray experiments. Three independently prepared sets of control and treated RNA samples were used to generate targets for hybridization of the Affymetrix murine U74Av2 array, an oligonucleotide array that contains probe sets representing 12,000 mouse genes. The expression of numerous genes was affected by aldosterone treatment in mIMCD3 cells. Those transcripts that increased or decreased in the same manner in response to aldosterone in at least two of the three hybridization experiments are listed in Tables 6-2 and 6-3, respectively. Numerous ESTs appeared in the lists of both upand down-regulated transcripts, shown in Tables 6-2 and 6-3, respectively. Subsequently, the online Affymetrix analysis tool NetAffx ( www.Affymetrix.com ) (last accessed May 1, 2003) has now identified many of these ESTs; sgk was among the recently annotated ESTs. Analysis of Aldosterone-Responsive mRNAs As expected, sgk was up-regulated in all three expression analysis experiments. Several other transcripts were consistently up-regulated in the expression analysis experiments. These include connective tissue growth factor (CTGF), period homolog and

PAGE 144

123 Table 6-2. Transcripts Down-Regulated by Aldosterone GenBank Fold Changea Transcript AV332798 -2.5 +/1.2 EST AI843571 -2.1 +/0.8 Similar to hypothetical protein MGC12966b AF020771 -2 +/0.7 Importin alpha Q1 mRNA AV277198 -1.9 +/0.7 Weakly similar to T28931 hypothetical protein C52B9.3b U48721 -1.9 +/0.6 Zinc finger protein 60 X53177 -1.8 +/0.3 Integrin alpha 4 (Cd49d) W91600 -1.7 +/0.5 EST aFold change values were calculated by Affymetrix Suite Version 4, n=2. bEST designations were amended using NetAffx. preproendothelin. These three transcripts, along with sgk, were selected as a representative group of up-regulated transcripts. Northern blot analyses were performed to confirm the regulation effects indicated by the initial data analysis of the microarray experiments. cDNA probes were constructed by RT-PCR for each of the four transcripts (see Chapter 2). Sgk, CTGF, period homolog and preproendothelin all showed clear evidence of increased mRNA in response to acute exposure to aldosterone. Northern blot analyses and real time PCR data suggested greater induction than those calculated from the microarray data (Table 6-1). No significant changes in expression of the GAPDH mRNA, used as a loading control throughout this study, were observed. Of the four transcripts tested, period homolog underwent the greatest induction according to the microarray data, the Northern blot data and the real time PCR data. The period homolog transcript was induced greater than seven-fold in mIMCD3 cells exposed to aldosterone for 1 h, as measured by Northern blot and real time PCR. To examine the effects of a range of aldosterone concentrations, mIMCD3 cells were treated with increasing amounts of hormone for 1 h (Figure 6-3). The most

PAGE 145

124 Table 6-3. Transcripts Up-Regulated by Aldosterone GenBank Fold Changea Transcript AF022992 4.1 +/0.3c Period homolog (Drosophila) AF019385 3.4 +/0.7 Heparan sulfate glucosaminyl 3-O-sulfotransferase AW046181 3.1 +/0.2c Sgkb AF032459 3.0 +/1.3 BimEL AF058799 2.8 +/1.6 14-3-3 protein gamma U35233 2.6 +/0.8c Endothelin 1 AI839289 2.6 +/0.8 SMT3 (supressor of mif 2, 3) homolog 2, pseudogene 4 D50032 2.5 +/1.3 TGN38B X05546 2.4 +/0.8 DBA/2 RNA fragment for gag related peptide U07982 2.3 +/0.2 Endothelin 1 AF077660 2.3 +/0.8 Homeodomain-interacting protein kinase 3 AW046627 2.3 +/0.5 Serine threonine kinase pim 3, clone MGC:27707b AA434661 2.2 +/1.0 EST D13458 2.2 +/0.3 Prostaglandin E receptor EP4 subtype U36340 2.2 +/1.0 CACCC-box binding protein BKLF AI845584 2.2 +/0.1 Dual specificity phosphatase 6b U50413 2.2 +/0.8 Phosphatidylinositol 3-kinase, regulatory subunit, polypeptide 1 (p85 alpha) AF072127 2.2 +/0.8 Claudin-1 U50384 2.2 +/0.5 Retinoic acid-responsive (NN8-4AG) AF077658 2.1 +/0.3 Homeodomain-interacting protein kinase 1 M70642 2.1 +/0.1 Connective Tissue Growth Factord X61800 2.1 +/0.4 CCAAT/enhancer binding protein (C/EBP), delta AA517845 2.1 +/0.3 EST AF036893 2.0 +/0.5 Circadian clock protein (Per2)

PAGE 146

125 Table 6-3. Continued GenBank Fold Changea Transcript Y12713 2.0 +/0.1 Endogenous retroviral sequence MuERV-L gag, pol and dUTPase X66084 2.0 +/0.9 CD44 antigen AV380793 2.0 +/0.9 EST AA759910 1.9 +/0.8 EST AW121876 1.9 +/0.7 Similar to hypothetical protein FLJ20505b AV315224 1.9 +/0.4 Nuclear protein 220b AI507266 1.9 +/0.4 Lymphoid nuclear protein related to AF4-likeb M29009 1.9 +/0.4 Complement factor H-related protein AF037437 1.8 +/0.6 Prosaposin AW047616 1.8 +/0.5 HLA-B-associated transcript 3b AA420397 1.8 +/0.6 EST AB017202 1.8 +/0.5 Entactin-2 M26005 1.8 +/0.3 Endogenous retrovirus truncated gag protein AA797843 1.7 +/0.5 EST AI841387 1.7 +/0.6 Clone MGC:7674 IMAGE:3496398b AW049716 1.7 +/0.4 Epidermal growth factor receptorb aFold change values were calculated by Affymetrix Suite Version 4, n=2 unless stated otherwise. bEST designations were amended using NetAffx. cFold change values from three hybridization experiments were averaged. Means are represented plus or minus standard error, n=3. dThis transcript was previously named Fibroblast Inducible Secreted Protein. dramatic increase in expression for all transcripts tested was observed after treatment with 10-6 M aldosterone. Although the concentration of hormone in these in vitro experiments was greater than the physiological concentration of aldosterone, the actual effective intracellular concentration was probably much lower. Recent studies have indicated that steroid hormones may not diffuse across cell membranes as freely as was once thought (88).

PAGE 147

126 Figure 6-3. Dose Response to Aldosterone. mIMCD3 cells were treated for 1 h with vehicle or increasing concentrations of aldosterone from 10-10 M to 10-6 M. The 28S and 18S ribosomal RNA bands are indicated. Northern blot analysis of RNA samples was performed using probes for the transcripts indicated. The next experiments examined the levels of expression of sgk, CTGF, period homolog, and preproendothelin as a function of time. Northern blot results revealed that all four messages showed a clear increase in expression after 1 h (Figure 6-4). Sgk mRNA decreased over the next 12 h and then showed a sharp increase beyond 24 h. A similar pattern was observed for CTGF message. In contrast, period homolog expression slightly decreased at 6 h and then remained relatively low for the duration of the experiment. Preproendothelin mRNA showed a biphasic response similar to sgk and CTGF, but with

PAGE 148

127 different timing. The recovery of preproendothelin mRNA was evident at 12 h and continued to increase. In order to control for a feeding effect, cells were treated with Figure 6-4. Time Course of Responses to Aldosterone. mIMCD3 cells were treated with 10-6 M aldosterone for the time periods indicated. The 28S and 18S ribosomal RNA bands are indicated. Northern blot analysis of RNA samples from vehicle or aldosterone treated cells was conducted using probes for the transcripts indicated. vehicle alone over these same time points. Densitometry analysis of Northern data from vehicle and aldosterone treated cells was performed and is summarized graphically in Figure 6-5. Triangles represent vehicle treatment; squares represent aldosterone treatment. Importantly, no significant changes in expression were observed for any of the four transcripts tested in cells treated with vehicle alone.

PAGE 149

128 Figure 6-5. Densitometry Analyses of Time Course Northern Blot Data. Densitometry was performed using Northern blot data from vehicle (triangles) or 10-6 aldosterone (squares)-treated mIMCD3 cells. Data is presented as fold change versus time of vehicle or aldosterone treatment. Transcript color designations are as follows: sgk, cyan; preproendothelin, maroon; CTGF, teal; period homolog, blue. To further validate, in a quantitative manner, the time course Northern results, real time PCR was performed using the aldosterone-treated RNA samples as template (Figure 6-6). Expression patterns similar to those seen in the Northern blot analysis were observed for all transcripts tested. Sgk showed a sharp five-fold increase at 1 h, which decreased to three-fold at 6 h. The level of sgk transcript continued to climb from 12 to

PAGE 150

129 48 h, hitting a peak of approximately seven-fold at the final time point. Period homolog peaked at 1 h with an increase of greater than seven-fold. Period homolog levels then dropped and stayed between threeand four-fold over control for the remainder of the experiment. Preproendothelin showed a gradual increase from three-fold over control at 1 h to a peak of greater than six-fold over control at 12 h. Preproendothelin levels then remained high for the duration of the experiment. GAPDH showed no significant changes in expression over time. Figure 6-6. Real Time PCR Analysis of Time Course. RNA samples from duplicate time course experiments were used as template in duplicate real time RT-PCR reactions. Time of aldosterone treatment is indicated. Error bars indicate plus or minus standard error, n=4. Indicates significant difference from control, p<0.05.

PAGE 151

130 The effects of aldosterone can be mediated through the MR as well as the glucocorticoid receptor (GR). The effects of GR and MR often overlap and the two receptors can heterodimerize to drive transcription of some genes (5, 36). It has previously been shown in primary cultures of IMCD cells that either MR or GR can activate electrogenic Na+ transport (57). Specific inhibitors of both receptors were used alone or in combination in order to determine the contribution of each to the aldosterone-mediated changes in gene expression (Figure 6-7). Densitometry analysis of repeated Figure 6-7. Effect of Mineralocorticoid and Glucocorticoid Receptor Inhibitors. mIMCD3 cells were treated for 1 h with vehicle (V), 10-6 M aldosterone (A), A plus 10-6 M mifepristone (A+M), A plus 10-6 M spironolactone (A+S), or A plus mifepristone and spironolactone (A+M+S). Densitometry was performed on repeated Northern blot data. All values were normalized against GAPDH mRNA levels. Error bars indicate plus or minus standard error, n=3. Indicates significant difference from control, p<0.1. Northern blot data demonstrated similar patterns of expression for all messages tested. MR and GR inhibitors, used alone or in combination, dropped expression to levels that

PAGE 152

131 were not significantly different from control. Use of GR or MR specific inhibitors suggested then that both receptors contribute to the aldosterone-mediated effects on gene expression in mIMCD3 cells. This is not a surprising result, given that MR and GR are known to heterodimerize with each other. Induction of Preproendothelin In order to correlate protein expression with the changes in mRNA levels, a cell lysate was prepared from aldosterone or vehicle treated cells following 6 h of hormone exposure. One-hundred fifty micrograms of total cellular protein were run on an SDS-PAGE gel and electrically transferred to a nitrocellulose membrane. Western blot analysis was conducted using a monoclonal antibody raised against endothelin-1. A representative blot is pictured in Figure 6-8A, and shows a definite increase in preproendothelin levels in aldosterone-treated cells as compared to vehicle-treated cells. In addition, an anti-endothelin-1 polyclonal antibody was used. Both antibodies recognized the same band at approximately 23 kDa, which is the expected size of unprocessed preproendothelin. Equal loading of samples was visualized on a duplicate gel stained with Coomassie brilliant blue (Figure 6-8B). As determined by densitometry analysis of repeated Western blot data, preproendothelin protein levels were increased greater than five-fold in aldosterone-treated cells compared to control cells in four independent experiments (Figure 6-8C).

PAGE 153

132 Figure 6-8. Increased Preproendothelin Protein Expression in Aldosterone Treated Cells. (A) mIMCD3 cells were treated for 6 h with vehicle or 10-6 M aldosterone. Western blot analysis was performed using a monoclonal antibody raised against endothelin. (B) Equal loading was visualized on a duplicate gel stained with Coomassie brilliant blue. (C) Densitometry analysis was performed on repeated Western blot data. Error bars indicate plus or minus standard error, n=4. *Indicates significant difference from control, p<0.05. Analysis of the Period Homolog Promoter In an effort to further characterize the effect of aldosterone on the period homolog gene, an analysis of the period homolog promoter was initiated. A bacterial artificial chromosome (BAC) clone was purchased from BACPAC CHORI. The clone, RP24

PAGE 154

133 277K16, contained 150 kbp of mouse chromosome 11, including the period homolog gene. To confirm that clone RP24-277K16 did in fact contain the period homolog gene Southern blot analysis was performed. The BAC clone was digested with EcoRI, yielding multiple fragments (Figure 6-9A). Based on sequence analysis of mouse chromosome 11 (obtained from the National Center for Biotechnology Information, http://www.ncbi.nlm.nih.gov/) (last accessed March 24, 2004), a 20 kb EcoRI fragment should contain the period homolog gene. In fact, a probe to exon 1 of the period homolog cDNA did hybridize to the 20 kb band, as indicated by the signal in lane 2 of Figure 6-9B. The probe hybridized to the uncut BAC DNA as well (lane 3 in Figure 6-9B). These results clearly demonstrated that the BAC clone contained the gene in question. Figure 6-9 Southern Blot Analysis of BAC Clone RP24-277K16. (A) BAC clone RP24-277K16 was digested with EcoRI. Samples were analyzed on a 0.8% agarose gel and visualized with ethidium bromide. Lane 1: DNA ladder, lane 2: BAC clone digested with EcoRI, lane 3: uncut BAC clone. (B) The DNA samples in panel A were transferred to a nylon membrane via capillary action. Southern blot analysis was performed using a probe to period homolog exon 1. Lane designations are the same as in Panel A. The band in lane 2 corresponds to the expected 20 kb EcoRI fragment containing the period homolog gene. Next, a CpG analysis was performed on the sequence from mouse chromosome 11 thought to contain the period homolog promoter (Figure 6-10). The nucleotide C,

PAGE 155

134 followed by G, is underrepresented in the mammalian genome. Clusters of these so-called CpGs, called CpG islands, are often indicative of promoter regions (130). The sequence analyzed extended from position 50 to +151 (relative to the period homolog transcription start site); this amount of sequence should include the majority of the period homolog 5 regulatory elements. A cluster of CpG dinucleotides was observed starting around position and extending past the transcription start site. The analysis suggests then that this sequence is indicative of a promoter. Figure 6-10 CpG Analysis of the Period Homolog Promoter. The number of CpG dinucleotides within 50 bp windows was counted in the sequence extending from to +150, relative to the approximate transcription start site of the period homolog cDNA. The same sequence was next analyzed for transcription factor binding sites, using the transcription factor binding site prediction program TESS (http://www.cbil.upenn.edu/tess) (last accessed March 23, 2004). Binding sites for many transcription factors were found, including general transcription factors such as Sp1. Due to the induction of period homolog expression by aldosterone, we looked specifically for the presence of putative GREs or MREs. Only GREs were found and the positions of those seven possible response elements are illustrated in Figure 6-11. Blue bars indicate

PAGE 156

135 Figure 6-11 Glucocorticoid Response Elements in the Period Homolog Promoter. The sequence from the PCR product was analyzed for putative GREs. Height of blue lines indicates the strength of match to a consensus GRE; numbering indicates the position of the putative GRE relative to the period homolog transcription start site. the presence of the putative GREs and the height of the bars corresponds to the strength of match to a consensus GRE sequence. The contribution of these sites to the aldosterone response can be tested using the luciferase assay system. In order to begin a biochemical analysis of the promoter, a PCR approach was used to amplify a portion of the period homolog 5 regulatory region. Using sequence from mouse chromosome 11, the primers MG 92 (CCTGTGGCCCAGGTATCCTCCCTGAAAAGG) and MG 93 (GTCTGGGCCATACAGTGGAGGACGAAACAG) were designed to amplify a 2002 base pair fragment of the period homolog gene that extended to the base immediately 5 of the translation start site (up to, but not including the ATG). The PCR product was amplified using the BAC clone DNA as template and the following parameters: 94C x 5 min; 30 cycles of 94C x 30 sec, 63.2C x 30 sec, 72C x 3 min; 72C x 10 min final extension. The 2002 bp band, shown in lane 2 of Figure 6-12, was excised, gel purified, and cloned into pCR2.1 TOPO for sequence analysis. The sequence of the PCR product exactly matched that from the period homolog gene in mouse chromosome 11.

PAGE 157

136 Figure 6-12 Amplification of the Period Homolog Promoter. PCR was used to amplify 2002 bp of sequence corresponding to the 5 regulatory region of the period homolog gene. Lane 1: 1 kb ladder, lane 2: PCR reaction 1 with BAC clone DNA as template (band was excised for cloning purposes), lane 3: 10% of PCR reaction 1, lane 4: PCR reaction 2 with H2O only as template as a negative control. Upon confirmation that the PCR product sequence matched that of the intended region of the period homolog gene, the putative period homolog promoter was moved into the plasmid pGL3 (Promega) to determine if it exhibited promoter activity. A KpnI/XhoI fragment containing the 2002 base pair PCR product was moved from pCR2.1 TOPO to pGL3 and in front of the firefly luciferase reporter gene. The resulting plasmid, pMLG107, was transfected into mIMCD3 cells for analysis of luciferase activity as an indicator of promoter activity. As a negative control, a second batch of cells were transfected with empty pGL3 vector. As a positive control, a third group of cells were transfected with a pGL3 plasmid containing the SV40 promoter. In addition, all of the cells were co-transfected with the plasmid pRL-TK in order to control for transfection efficiency. Twenty-four hours post transfection, a dual luciferase assay was performed (see Chapter 2). The results are summarized graphically in Figure 6-13 and are presented as relative luciferase activity. The highest activity was observed from pMLG107, containing the 5 regulatory region of the period homolog gene; this value was set to

PAGE 158

137 Figure 6-13. Relative Luciferase Activity of Period Homolog Promoter. mIMCD3 cells were transfected with plasmid constructs containing 2002 bp of the putative period homolog promoter (pMLG107), the SV40 promoter (pGL3-SV40) or empty vector (pGL3). Luciferase activity was normalized to a background control plasmid and is presented relative to the highest measured value. 100%. The SV40 promoter also exhibited a high level of luciferase activity while the empty pGL3 vector showed less than 5% activity, as expected. These results suggest that the plasmid pMLG107 contained promoter elements. Future work in the Cain laboratory will involve a deletion analysis of the putative period homolog 5 regulatory region contained in pMLG107, using the location of the putative GREs (pictured in Figure 6-11) as a guide. Identification of the aldosterone response element(s) will be the immediate goal. Discussion The late effects of aldosterone on ion channels and transporters such as ENaC and the Na+, K+ ATPase have been well characterized, but the genes that are immediately affected by aldosterone are not as well known. We have used oligonucleotide array technology to examine the effects of aldosterone on gene expression after 1 h in a mouse inner medullary collecting duct cell line. Not surprisingly, the results of three hybridization experiments suggested that the expression of numerous genes changed in

PAGE 159

138 aldosterone treated cells compared to control cells. Northern blot analysis directly confirmed the up-regulation of sgk, CTGF, period homolog and preproendothelin transcripts. Time course studies showed a biphasic response of sgk, CTGF and preproendothelin mRNAs to aldosterone treatment, indicating that aldosterone signaling is complex in mIMCD3 cells with a long-term component in addition to the acute response. The use of MR and GR specific inhibitors indicated that both receptors contribute to the aldosterone-mediated regulation of the transcripts tested. Western blot analysis demonstrated that the increases in preproendothelin mRNA levels also resulted in increased levels of protein expression. Finally, cloning of the period homolog promoter will enable a deletion analysis aimed at identifying the hormone response element(s) responsible for the aldosterone-mediated regulation of that gene. These findings represent a novel list of aldosterone-regulated transcripts. Three independent methods, microarray analysis, Northern blot, and real-time PCR, were used to validate the aldosterone-responsiveness of sgk, period homolog and preproendothelin. Induction of CTGF was verified by two of these methods. The consistent aldosterone-mediated response of these transcripts lends validity to the remaining transcripts on the lists of aldosterone-regulated transcripts in Tables 6-2 and 6-3. The aldosterone-induced and repressed transcripts contain several kinases, transcription factors, and signaling proteins. It is becoming apparent that these molecules must play a vital role in mediating aldosterone action by regulating existing transporters or causing changes in expression of transporters. To our knowledge, these results represent the first investigation into aldosterone regulation as early as 1 h after treatment using microarray technology. Previous studies

PAGE 160

139 (91) examined the effect of aldosterone at 4 h, at which time the most immediate effects of aldosterone have likely already occurred. Indeed, the time course data presented here showed that by 6 h, the expression of all transcripts tested had substantially decreased relative to the levels observed at 1 h. The expression of several additional transcripts was affected by aldosterone as described by the microarray data and these effects remain to be confirmed and studied. The identity of these genes indicates that the aldosterone signaling pathways are more complex than previously thought. Subsequent investigation into the functions of these early response genes, together with the results presented here, will provide greater insight into the signaling pathways initiated by aldosterone and will further clarify our knowledge of the critical role aldosterone plays in regulating ion homeostasis and cardiovascular disease.

PAGE 161

CHAPTER 7 CONCLUSIONS AND FUTURE DIRECTIONS The Rabbit Colonic H+, K+ ATPase The colonic H+, K+ ATPase plays a vital role in maintenance of K+ homeostasis and acid-base balance by the kidney. The enzyme belongs to a large family of membrane bound enzymes, the P-type ATPases. Much less is known about the colonic H+, K+ ATPase than the closely related Na+, K+ ATPase and the gastric H+, K+ ATPase. These three enzymes, collectively known as the X+, K+ ATPases, consist of an and a subunit. The identity of the subunit of the colonic H+, K+ ATPase has remained controversial and it was one of the aims of this dissertation to identify the proper subunit for the rabbit enzyme. To that end, cDNAs for the four possible subunits, NaK1, NaK2, NaK3 and gHK were obtained and expressed in combination with the two rabbit HK2 isoforms, HK2a and HK2c. Co-immunoprecipitation of the expressed HK2 and XK subunits was performed in order to determine which subunit(s) was able to assemble with and traffic the HK2 subunit to the plasma membrane. The results suggested that only gHK was able to form a complex with HK2a and HK2c that was found at the plasma membrane. Subsequent ATP hydrolysis assays did not show this complex to be active, however. Although this line of research was abandoned because of the lack of functionality of the expressed enzyme, three important benefits have emerged from this work. First, if any experimentation with this system were to be performed in the future, it would be vital to 140

PAGE 162

141 have all the necessary cDNAs in hand. As described in Chapter 4, the cDNAs for NaK2 and NaK3 were cloned and expressed, thereby completing the family of rabbit renal XK subunits for this laboratory. There are eight possible combinations of / pairs to form the rabbit colonic H+, K+ ATPase. Eight expression vectors containing these combinations were generated and confirmed, as discussed in Chapter 5. Second, although functionality of the expressed enzymes was not demonstrated, the co-immunoprecipitation experiments yielded an important result. It was repeatedly observed that gHK was able to assemble with and traffic both HK2a and HK2c to the plasma membrane, while this result was never observed with any of the NaK isoforms. It would appear then that gHK is the likeliest mate for the rabbit HK2 subunits in our system. While this result remains to be confirmed in vivo, it is in agreement with data from expression systems using the human and guinea pig colonic H+, K+ ATPases (49) (6). Third, the generation of a molecular model for the HK2a subunit represented a significant contribution to the body of knowledge about the colonic H+, K+ ATPase (51). The fact that the model was energy minimized to a level comparable to the 2.6 resolution structure of the Ca+2 ATPase supports its validity. Ten transmembrane helices were constructed in the M domain of the model, a characteristic shared with other recent structures proposed for the P-type ATPases. Importantly, the distance measured between the lysine residue in the nucleotide-binding pocket and the aspartic acid residue at the phosphorylation site was measured to be approximately 29 a value similar to that observed in both the Ca+2 ATPase crystal structure and a molecular model of L45 of the Na+, K+ ATPase. These observations from the HK2a model support the conservation of

PAGE 163

142 the structure of this family of enzymes. One of the most intriguing aspects of the HK2a model was the placement of the subunit at the C-terminal region of the subunit. Figure 7-1 shows a cartoon of the gHK subunit placed within the HK2a model. The gHK subunit contains three disulfide bridges and seven glycosylation sites in the extracellular domain, as pictured. As described in Chapter 3, the subunit likely passes through the membrane at such an angle as to allow for interaction of its N-terminal region with M8 and M10, and also L67. The C-terminal region of gHK is then located near M7 and M10 where it can interact with L78. Placement of the subunit of the colonic H+, K+ ATPase in this position allows for visualization of previously demonstrated biochemical data (23; 53; 82). Although there is no direct biochemical evidence for the rabbit enzyme, it is another validation of the HK2a model that it supports biochemical evidence from other related P-type ATPases. Of course, the molecular model of HK2a would be the most useful if a functional expression system existed to test it. Although the effort to establish such a system, described here, as well as a previous attempt (83), have both been unsuccessful, future work built upon the two previous studies could lead to establishment of a useful system. The work of Dr. Otto and the efforts described in this dissertation show that use of the proper subunit is paramount to formation of the colonic H+, K+ ATPase, and that the proper subunit is likely to be gHK. Although the co-immunoprecipitation studies described in Chapter 5 suggest that assembly and trafficking are taking place, no ATP hydrolysis activity was detected in the plasma membranes of transfected cells. Since the enzyme is assembled and should be properly localized within the cell, it is possible that

PAGE 164

143 Figure 7-1. Model of the Rabbit Colonic H+, K+ ATPase. The HK2a model is shown with dashed orange lines representing a putative location for the gHK subunit, near M7 and M10. Three extracellular disulfide bridges are indicated by white bars and seven possible glycosylation sites by lavendar circles. an unknown signal is needed to turn the pump on. It has been well established that mRNA and protein levels for the HK2 subunit are induced under conditions when K+ levels are low (22; 62). Under normal K+ conditions, the mRNA for HK2 is often undetectable, indicating that protein expression is probably very low as well. It is possible then, that a signal generated under low K+ conditions is required not only to induce expression of the protein, but also to activate it once it reaches the plasma

PAGE 165

144 membrane. The identity of such a signal may be revealed by further studies with the HK2 gene promoter, as discussed in Chapter 5. Such an activation signal has not been required to generate functional expression systems involving the Na+, K+ ATPase or the gastric H+, K+ ATPase. However, the subunit mRNAs for these enzymes are expressed at readily detectable levels in several cell types under normal conditions, which is not the case for the colonic H+, K+ ATPase. The colonic H+, K+ ATPase, specifically the HK2c subunit, was expressed at detectable levels in several very specialized cell types in the rabbit kidney, as demonstrated by immunohistochemical localization (117). Colocalization of HK2c with carbonic anhydrase II and the Cl-/HCO3exchanger demonstrated that the protein was specifically located in and intercalated cells in the CCD and OMCD. If the colonic H+, K+ ATPase requires an accessory protein(s) to function, such as one of the FXYD family members that interacts with the Na+, K+ ATPase (see Chapter 1), expression of such an accessory protein would be expected in an intercalated cell in the collecting duct. The cell types so far used to establish a functional expression system for the colonic H+, K+ ATPase, COS-1 (see Chapter 5 and Otto (83)) and HEK 293 (83) do not exhibit properties of collecting duct intercalated cells. Therefore, if the colonic H+, K+ ATPase requires an as-yet-unidentified accessory protein in order to function properly, the enzyme would need to be expressed in an intercalated collecting duct cell line. In summary, the work described in this dissertation concerning the rabbit colonic H+, K+ ATPase has led to several important benefits. Generation of the NaK2 and NaK3 cDNAs enabled construction of expression plasmids for all eight possible 2/XK pairs. The molecular model of the 2a model led to a hypothesis on the

PAGE 166

145 position of the subunit within the enzyme and revealed several conserved characteristics shared among the Type II P-type ATPases. The model of 2a will become extremely vital if and when a functional expression system is established for the rabbit colonic H+, K+ ATPase. Generation of a such a system may necessitate further experimentation with the 2 gene promoter to determine what signals may be necessary to activate the protein. In addition, successful functional expression of the colonic H+, K+ ATPase may require the use of a specialized mammalian cell line that shares the characteristics of an intercalated collecting duct cell. In fact, a recent search of the American Type Culture Collection revealed that a rabbit cortical intercalated cell line, Clone C, is available (http://www.atcc.org) (last accessed March 23, 2004). Future Directions for the Study of the Colonic H+, K+ ATPAse Future directions of this line of research could involve use of the newly deposited Clone C cell line, mentioned above. This cell line would be a good candidate for heterologous expression of the H+, K+ ATPase, but analysis of the endogenous gene and enzyme should be carried out first. This study would be similar to the work of Dr. Grady Campbell with another rabbit kidney cell line, RCCT-28A (15). Northern and Western blot analysis could be conducted to determine the levels of HK2a and HK2c expression in this cell line. K+-dependent ATP hydrolysis activity could be measured as well. Given the known induction of HK2 message and protein under low K+ conditions, it would be interesting to look at mRNA, protein and activity levels of HK2a and HK2c following K+ deprivation in Clone C cells. Study of the HK2 gene in this cell line should also be carried out. Previous work in our laboratory by Dr. Deborah Zies indicated that the HK2 gene was repressed in the RCCT-28A cell line, which is also a

PAGE 167

146 rabbit intercalated cell line (130). It would be interesting to know if the same type of regulation is observed in Clone C cells. Dr. Zies generated several deletion constructs of the HK2 promoter for use in the dual luciferase assay system as part of her analysis of the gene. These constructs would be very useful in characterizing the state of the HK2 gene in Clone C cells. Following characterization of endogenous HK2 in Clone C cells, cells could be transfected with pMLG96 to express HK2c and gHK. Transfection could also be performed in RCCT-28A cells, for that matter. It may be more advantageous to perform stable transfections with pMLG96 to enable long-term study of the enzyme. As discussed in Chapter 1 (see Figure 1-8), the confluency of kidney cells grown in culture does have an effect on their cellular properties. Stably transfected cells could be grown to or past confluency to evaluate the effect of cell differentiation on the activity of the expressed enzyme; these growth conditions were not possible in the present study due to the nature of transient transfection. The co-immunoprecipitation technique described in Chapter 5 could be used with the stably transfected cell line to evaluate whether HK2c and/or gHK are interacting with any endogenous accessory proteins. If a functional expression system can be established for the colonic H+, K+ ATPase, then a second model of HK2a, based on the E2 Ca+2 ATPase structure, should be generated. Such a modeling exercise would be identical to that described in Chapter 3. The same amino acid sequence alignment pictured in Figure 3-1 could be used in conjunction with the E2 Ca+2 ATPase coordinates to create a model of HK2a in the E2 conformation. Having molecular models of HK2a in the E1 and E2 conformations, would make it possible to study the catalytic mechanism of this enzyme

PAGE 168

147 In the short term, the future experiments described here could help characterize the HK2 gene and enzyme in Clone C cells. It could also be determined whether or not confluency has an effect on colonic H+, K+ ATPase activity in transfected cells, and whether or not an accessory subunit is required for enzyme activity. In the long term, having molecular models of the colonic H+, K+ ATPase in both its high and low affinity states would make it possible to study the catalytic mechanism of this enzyme and would facilitate an in-depth structure-function analysis. Also, a stable expression system, in conjunction with our hypothesis concerning the location of the subunit (Figure 7-1), would make it possible to study HK2c/gHK interaction as a model for / interaction in the P-type ATPase family of enzymes. Early Transcriptional Effects of Aldosterone in the Mouse Kidney Whereas the late effects of aldosterone on ion channels and transporters such as ENaC and the Na+, K+ ATPase have been well characterized, the genes that are immediately affected by aldosterone are not as well known. Our use of oligonucleotide array technology to examine the effects of aldosterone on gene expression in a mouse IMCD cell line suggested that the expression of numerous genes changed in aldosterone treated cells compared to control cells. Northern blot analysis directly confirmed the up-regulation of sgk, CTGF, period homolog and preproendothelin transcripts. Time course studies showed a biphasic response of sgk, CTGF, and preproendothelin mRNAs to aldosterone treatment, indicating that aldosterone signaling was complex in mIMCD3 cells with a long-term component in addition to the acute response. The use of MR and GR specific inhibitors indicated that both receptors contribute to the aldosterone-mediated regulation of the transcripts tested. Western blot analysis demonstrated that the

PAGE 169

148 increases in preproendothelin mRNA levels also resulted in increased levels of protein expression. Our results, together with what is known about the functions of the proteins encoded by the four transcripts, have led to a picture of aldosterone action in the mouse kidney (Figure 7-2). Figure 7-2. Model for Aldosterone Action in mIMCD3 cells. Treatment with aldosterone leads to heterodimerization of MR and GR, resulting in increased transcription from the genes for sgk, period homolog, CTGF, and preproendothelin. Shapes representing each of these are indicated. Possible downstream actions are described as well. Phosphorylation of Nedd-4 by sgk prevents degradation of ENaC. Period homolog is a likely transcription factor affecting expression of as-yet-unidentified genes. CTGF plays a role in several renal diseases and may function in aldosterone-mediated cardiac and/or renal fibrosis. Preproendothelin is cleaved twice to yield the peptide hormone endothelin-1, an extremely potent vasoconstrictor.

PAGE 170

149 The up-regulation of sgk by aldosterone within 1 h is a well-established effect (77; 98). For this reason, we used sgk as an internal control for the effectiveness of aldosterone treatment. Recently, a role for sgk in Na+ absorption has been proposed. Several investigators have linked activity of sgk to the number and activity of epithelial Na+ channels at the apical membrane. When coexpressed with ENaC, sgk was able to stimulate Na+ current (85). In addition, sgk activity has been shown to lead to translocation of ENaC subunits to the apical membrane (71). The effect of sgk on ENaC activity is likely due to its interaction with the ubiquitin ligase Nedd-4. Through inhibition of Nedd-4, sgk causes a decrease in ENaC degradation (48; 119). Although sgk was first identified as a glucocorticoid induced kinase, studies with the human sgk gene failed to show the presence of a GRE in the promoter (78). However, our own search of the human sgk promoter did reveal a putative GRE. It may have been previously overlooked due to more stringent search parameters. Nevertheless, Fejes-Toth et al., using a GR antagonist, demonstrated that the induction of sgk expression by aldosterone in cortical collecting cells was specific to MR (77). CTGF stimulates extracellular matrix formation, migration, fibroblast proliferation, and adhesion. We have confirmed an increase in CTGF message upon aldosterone treatment. This result is in accord with previous work concerning the role of CTGF in the kidney. Stenson et al. (104) and Taal et al. (108) demonstrated that upregulation of CTGF in 5/6 nephrectomy was ablated in the presence of an inhibitor of angiotensin converting enzyme (ACE). In addition, Dammeier et al. (29) demonstrated the upregulation of CTGF in the kidney, as well as the heart and skin, after systemic treatment with the glucocorticoid analog dexamethasone. Increased expression of CTGF

PAGE 171

150 was observed in several renal diseases, including diabetic nephropathy, IgA nephropathy, focal and segmental glomerulonephritis, and mesangial proliferative lesions of crescentic glomerulonephritis (60). Whether CTGF is instrumental in the development of the cardiac and renal fibrosis, which is mediated by aldosterone, will require further investigation. Understanding the relationship between aldosterone and CTGF may lead to insight concerning treatment for renal and cardiovascular disease and may also lead to the development of novel therapies. Period homolog is the mammalian homolog of the Drosophila period gene, a regulator of circadian rhythms. In Drosophila, period is a transcription factor lacking inherent DNA binding ability. Period pairs with the DNA binding protein timeless to affect transcription of target genes. It is therefore likely that period homolog may exert downstream transcriptional effects. Sun et al. termed the mouse period homolog m-rigui and found a circadian pattern of expression in the suprachiasmatic nucleus (SCN) (105). They also demonstrated expression of period homolog in several tissues, including kidney (105). Regulation of period homolog by aldosterone was not surprising given the demonstration of Per1 upregulation by another corticosteroid, the glucocorticoid analog dexamethasone. Balsalobre et al. also demonstrated that Per1 expression increased in mouse kidney after 5 h of dexamethasone treatment (10), a result that correlates with the continued increased expression we observed after 6 h of aldosterone treatment. Regulation of a circadian rhythm protein by aldosterone is not surprising given the circadian release pattern of aldosterone itself (64). In fact, it has been known for some time that plasma cortisol concentrations fluctuate in a circadian pattern (76).

PAGE 172

151 Of particular note is the up-regulation of preproendothelin by aldosterone. Endothelin-1 is a peptide hormone secreted by vascular endothelial cells and is the most potent vasoconstrictor known. Preproendothelin undergoes significant posttranslational processing with enzymatic hydrolysis by two enzymes prior to the production of the active form, a 21 amino acid peptide (1). Interestingly, it has been demonstrated that endothelin-1 itself leads to increased release of aldosterone by acting on the renin-angiotensin-aldosterone pathway (86; 93). Given the present evidence for upregulation of preproendothelin by aldosterone, it is clear that additional regulatory mechanisms must exist to prevent a positive feedback system. In fact, the effects of endothelin-1 are mediated through two different receptors, ET-A and ET-B, with different results. Both receptor subtypes are expressed in the adrenal zona glomerulosa where aldosterone is produced (86). In the endothelium, ET-A receptors mediate vasoconstriction whereas ET-B receptors result in vasodilation (33). The vasodilation effects of endothelin-1 occur mostly at lower concentrations of the peptide hormone and are mediated by ET-B receptors. Higher concentrations of endothelin-1 result in vasoconstriction and are mediated by ET-A receptors. The ET-A and ET-B receptors are both able to transcriptionally upregulate the aldosterone synthase gene in humans (93). In the mature animal, endothelin-1 has other effects aside from vasoconstriction. Most notably, endothelin-1 has been shown to affect both Na+ transport (41) and H+ secretion (125) in the collecting duct. In addition, evidence supports the role of endothelin-1 in the pathogenesis of renal interstitial fibrosis, potassium depletion and diabetic nephropathy (56). Aldosterone is a known stimulator of H+ secretion (81). The finding that preproendothelin is one of the early genes stimulated by aldosterone suggests

PAGE 173

152 an important role for this hormone as a mediator of aldosterone action. Given aldosterones known effect on cardiac fibrosis (92) and in view of our observation that aldosterone also stimulates CTGF, a more coordinated action of aldosterone may emerge from these studies. Further investigation into the relationship between aldosterone, CTGF and endothelin-1 must be made to elucidate the mechanisms governing each of their roles in cardiovascular disease. These findings represent a novel list of aldosterone-regulated transcripts. The physiological significance of these studies must be viewed at two levels. The first level is comprised of the immediate mediators of aldosterone action. Endothelin and CTGF have emerged from the present study as two candidates that connect the known physiological actions of aldosterone; these actions are, respectively, the enhancement of renal net acid excretion with the attendant metabolic alkalosis and the effects of aldosterone on cardiac and renal fibrosis. The second level is only made possible with the use of oligonucleotide array technology, which enables us to examine the global, genomic effects of aldosterone. Although caution must be exercised since these effects reflect changes at the mRNA level, several of the transcripts listed in Tables 6-2 and 6-3 have well-known functions and, in many cases, are part of well-established signaling pathways. For example, p85, the regulatory subunit of the phosphatidylinositol 3-kinase, was shown to be upregulated by aldosterone. Of all the up-regulated transcripts in Table 6-3, five were designated as kinases using Net Affx (http://www.Affymetrix.com) (last accessed May 1, 2003). These include the aforementioned PI3K and sgk, and also homeodomain interacting protein kinases 1 and 3, and the epidermal growth factor receptor. Although not designated in the

PAGE 174

153 NetAffx kinase search, the serine threonine kinase pim 3 is also listed in Table 6-3. Another aldosterone induced transcript that may play a role in making connections within and between signaling pathways is 14-3-3 gamma. The 14-3-3 family contains adaptor proteins that mediate interactions between components of signal transduction pathways (38). In addition to transcripts that may be involved in signal transduction, there are several aldosterone-regulated transcripts that play a role in regulating transcription. From Table 6-2, zinc finger protein 60 and from Table 6-3, the CCAAT/enhancer binding protein delta were both designated as a transcription regulatory proteins. However, NetAffx showed the presence of several other nuclear-located transcripts that likely play a role in regulation of transcription. These were period homolog (Per 1), Per2, and the CACCC-box binding protein BKLF. No transcripts were found when searching Tables 6-2 and 6-3 for transport-related transcripts. This was surprising given the role aldosterone plays in regulating ion transport. However, the kinases and transcription factors contained within these lists must play a vital role in mediating aldosterone action by regulating existing transporters or causing changes in expression of transporters. Our results represent the first investigation into aldosterone regulation in the IMCD as early as 1 h after treatment using microarray technology. This study has yielded novel results, including the upregulation of CTGF, preproendothelin and period homolog by aldosterone. There are far-reaching implications of these genes being regulated by a mineralocorticoid. Sgk plays a role in activation of ENaC and it likely has other, as-yet-unidentified cellular targets. CTGF functions in several renal diseases (127). Period homolog regulates circadian rhythms and at least one other circadian rhythm gene, Per2, also appeared in the list of up-regulated transcripts. Finally, endothelin-1 participates in

PAGE 175

154 regulation of Na+ and acid-base homeostasis, blood pressure control, and cardiovascular function. The expression of several additional transcripts was affected by aldosterone as described by the Affymetrix data and these effects remain to be confirmed and studied. The identity of these genes indicates that the aldosterone signaling pathways are more complex than previously thought. Subsequent investigation into the functions of these early response genes will provide greater insight into the signaling pathways initiated by aldosterone and will further clarify our knowledge of the critical role aldosterone plays in regulating ion homeostasis and cardiovascular disease. Future Directions for the Study of Aldosterone Action in the Mouse Kidney Future directions for this line of research will include a deletion analysis of the period homolog promoter in order to identify the aldosterone response element. Based on the inhibitor study described in Figure 6-7, the mineralocorticoid receptor and the glucocorticoid receptor are both contributing to the aldosterone response of the period homolog gene. Sequence analysis of the 2002 base pairs of period homolog 5 regulatory sequence did not reveal the presence of any MREs but several GREs were found (see Figure 6-11). It is likely that one or more of these sites play a role in the aldosterone regulation of period homolog. An as-yet-unidentified MRE could be involved as well. Preliminary experiments have already been performed in order to begin a deletion analysis based on progressively removing the putative GREs in order to determine which site or sites are contributing to the aldosterone response. Aldosterone responsiveness of the period homolog promoter needs to be confirmed. If the 5 regulatory elements are not responsive, then other areas of the gene should be analyzed for possible hormone response elements. Using the luciferase assay system in combination with mutagenesis of the promoter constructs will allow the aldosterone response element to be mapped.

PAGE 176

155 Other experiments could involve elucidating the downstream action of period homolog. This may involve a yeast-two-hybrid study to determine the identity of other proteins that interact with period homolog. Furthermore, chromatin immunoprecipitation could be performed to investigate the interaction of period homolog with its putative downstream target genes. In the short term, these studies should result in identification of the response element(s) in the period homolog gene that mediate aldosterone regulation. In the long term, the role of period homolog in mediating aldosterone action could be clarified. These same approaches could be used to study both the regulation and the downstream action of CTGF and preproendothelin as well in order to determine their respective functions as part of the aldosterone-signaling cascade. In summary, while functional expression of the colonic H+, K+ ATPase was not established, generation of the HK2a model and the identification of gHK as the likely partner for the HK2 subunits constitute a significant contribution to the understanding of the enzyme. Further investigation, perhaps with different cell types or under activating conditions, could result in functional expression of the colonic H+, K+ ATPase. Analysis of the early transcriptional effects of aldosterone showed that the hormone induces expression of several transcripts; these results have far-reaching implications. Further investigation into regulation of these genes by the hormone will yield a more complete understanding of aldosterone action. Together these data help to better characterize renal ion transport and its hormonal regulation.

PAGE 177

LIST OF REFERENCES 1. Agapitov AV and Haynes WG. Role of endothelin in cardiovascular disease. J Renin Angiotensin Aldosterone Syst 3: 1-15, 2002. 2. Ahn KY and Kone BC. Expression and cellular localization of mRNA encoding the "gastric" isoform of H(+)-K(+)-ATPase alpha-subunit in rat kidney. Am J Physiol 268: F99-109, 1995. 3. Ahn KY, Park KY, Kim KK, and Kone BC. Chronic hypokalemia enhances expression of the H(+)-K(+)-ATPase alpha 2-subunit gene in renal medulla. Am J Physiol 271: F314-321, 1996. 4. Alvarez de la Rosa D, Zhang P, Naray-Fejes-Toth A, Fejes-Toth G, and Canessa CM. The serum and glucocorticoid kinase sgk increases the abundance of epithelial sodium channels in the plasma membrane of Xenopus oocytes. J Biol Chem 274: 37834-37839, 1999. 5. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Housman DE, and Evans RM. Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237: 268-275, 1987. 6. Asano S, Hoshina S, Nakaie Y, Watanabe T, Sato M, Suzuki Y, and Takeguchi N. Functional expression of putative H+-K+-ATPase from guinea pig distal colon. Am J Physiol 275: C669-674, 1998. 7. Asano S, Kimura T, Ueno S, Kawamura M, and Takeguchi N. Chimeric domain analysis of the compatibility between H(+), K(+)-ATPase and Na(+),K(+)-ATPase beta-subunits for the functional expression of gastric H(+),K(+)-ATPase. J Biol Chem 274: 22257-22265, 1999. 8. Asano S, Yoshida A, Yashiro H, Kobayashi Y, Morisato A, Ogawa H, Takeguchi N, and Morii M. The cavity structure for docking the K+-competitive inhibitors in the gastric proton pump. J Biol Chem 279: 13968-13975, 2003. 9. Attali B, Latter H, Rachamim N, and Garty H. A corticosteroid-induced gene expressing an "IsK-like" K+ channel activity in Xenopus oocytes. Proc Natl Acad Sci U S A 92: 6092-6096, 1995. 156

PAGE 178

157 10. Balsalobre A, Brown SA, Marcacci L, Tronche F, Kellendonk C, Reichardt HM, Schutz G, and Schibler U. Resetting of circadian time in peripheral tissues by glucocorticoid signaling. Science 289: 2344-2347, 2000. 11. Beggah AT, Beguin P, Bamberg K, Sachs G, and Geering K. Beta-subunit assembly is essential for the correct packing and the stable membrane insertion of the H,K-ATPase alpha-subunit. J Biol Chem 274: 8217-8223, 1999. 12. Beggah AT, Jaunin P, and Geering K. Role of glycosylation and disulfide bond formation in the beta subunit in the folding and functional expression of Na,K-ATPase. J Biol Chem 272: 10318-10326, 1997. 13. Booth RE, Johnson JP, and Stockand JD. Aldosterone. Adv Physiol Educ 26: 8-20, 2002. 14. Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, Read RJ, Rice LM, Simonson T, and Warren GL. Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta Crystallogr D Biol Crystallogr 54: 905-921, 1998. 15. Campbell W. Characterization of the rabbit renal H,K-ATPases (Doctoral Dissertation). Gainesville: University of Florida, 1998. 16. Campbell WG, Weiner ID, Wingo CS, and Cain BD. H-K-ATPase in the RCCT-28A rabbit cortical collecting duct cell line. Am J Physiol 276: F237-245, 1999. 17. Caviston TL, Campbell WG, Wingo CS, and Cain BD. Molecular identification of the renal H+,K+-ATPases. Semin Nephrol 19: 431-437, 1999. 18. Chen SYea. Epithelial sodium channel regulated by aldosterone-induced protein sgk. Proc Natl Acad Sci U S A 96: 2514-2519, 1999. 19. Chow DC and Forte JG. Functional significance of the beta-subunit for heterodimeric P-type ATPases. J Exp Biol 198: 1-17, 1995. 20. Codina J, Delmas-Mata JT, and DuBose TD, Jr. The -subunit of the colonic H+,K+-ATPase assembles with 1Na+,K+-ATPase in kidney and distal colon. J Biol Chem 273: 7894-7899, 1998. 21. Codina J, Pressley TA, and DuBose TD, Jr. The colonic H+,K+-ATPase functions as a Na+-dependent K+(NH4+)-ATPase in apical membranes from rat distal colon. J Biol Chem 274: 19693-19698, 1999. 22. Codina J, Wall SM, and DuBose TD, Jr. Contrasting functional and regulatory profiles of the renal H+,K+-ATPases. Semin Nephrol 19: 399-404, 1999.

PAGE 179

158 23. Colonna TE, Huynh L, and Fambrough DM. Subunit interactions in the Na,K-ATPase explored with the yeast twohybrid system. J Biol Chem 272: 12366-12372, 1997. 24. Cougnon M, Bouyer P, Planelles G, and Jaisser F. Does the colonic H,K-ATPase also act as an Na,K-ATPase? Proc Natl Acad Sci U S A 95: 6516-6520, 1998. 25. Cougnon M, Planelles G, Crowson MS, Shull GE, Rossier BC, and Jaisser F. The rat distal colon P-ATPase alpha subunit encodes a ouabain-sensitive H+, K+-ATPase. J Biol Chem 271: 7277-7280, 1996. 26. Crambert G and Geering K. FXYD proteins: new tissue-specific regulators of the ubiquitous Na,K-ATPase. Sci STKE 2003: RE1, 2003. 27. Crambert G, Hasler U, Beggah AT, Yu C, Modyanov NN, Horisberger JD, Lelievre L, and Geering K. Transport and pharmacological properties of nine different human Na, K-ATPase isozymes. J Biol Chem 275: 1976-1986, 2000. 28. Crowson MS and Shull GE. Isolation and characterization of a cDNA encoding the putative distal colon H+,K(+)-ATPase. Similarity of deduced amino acid sequence to gastric H+,K(+)-ATPase and Na+,K(+)-ATPase and mRNA expression in distal colon, kidney, and uterus. J Biol Chem 267: 13740-13748, 1992. 29. Dammeier J, Beer HD, Brauchle M, and Werner S. Dexamethasone is a novel potent inducer of connective tissue growth factor expression. Implications for glucocorticoid therapy. J Biol Chem 273: 18185-18190, 1998. 30. Davis LG, Kuehl WM, and Battey JF. Basic Methods in Molecular Biology, Second Edition. Norwalk: Appleton and Lange, 1994. 31. Debonneville C, Flores SY, Kamynina E, Plant PJ, Tauxe C, Thomas MA, Munster C, Chraibi A, Pratt JH, Horisberger JD, Pearce D, Loffing J, and Staub O. Phosphorylation of Nedd4-2 by Sgk1 regulates epithelial Na(+) channel cell surface expression. Embo J 20: 7052-7059, 2001. 32. DuBose TD, Jr., Codina J, Burges A, and Pressley TA. Regulation of H(+)-K(+)-ATPase expression in kidney. Am J Physiol 269: F500-507, 1995. 33. Dupuis J. Endothelin receptor antagonists and their developing role in cardiovascular therapeutics. Can J Cardiol 16: 903-910, 2000. 34. Esnouf RM. An extensively modified version of molscript that includes generally enhanced colouring capabilities. Journal of Molecular Graphics Modelling 15: 132-134, 1997.

PAGE 180

159 35. Ettrich R, Melichercik, M., Teisinger, J., Ettrichova, O., Krumscheid, R., Hofbauerova, K., Kvasnicka, P., Schoner, W., and Amler, E. Three-dimensional structure of the large cytoplasmic H4-H5 loop of Na+/K+ ATPase deduced by restraint-based comparative modeling shows only one ATP binding site. Journal of Molecular Modeling 7: 184-192, 2001. 36. Farman N and Rafestin-Oblin ME. Multiple aspects of mineralocorticoid selectivity. Am J Physiol Renal Physiol 280: F181-192, 2001. 37. Fejes-Toth G, Naray-Fejes-Toth A, and Velazquez H. Intrarenal distribution of the colonic H,K-ATPase mRNA in rabbit. Kidney Int 56: 1029-1036, 1999. 38. Fountoulakis M, Cairns N, and Lubec G. Increased levels of 14-3-3 gamma and epsilon proteins in brain of patients with Alzheimer's disease and Down syndrome. J Neural Transm Suppl 57: 323-335, 1999. 39. Funder JW. Glucocorticoid and mineralocorticoid receptors: biology and clinical relevance. Annu Rev Med 48: 231-240, 1997. 40. Funder JW, Pearce PT, Smith R, and Smith AI. Mineralocorticoid action: target tissue specificity is enzyme, not receptor, mediated. Science 242: 583-585, 1988. 41. Gallego MS and Ling BN. Regulation of amiloride-sensitive Na+ channels by endothelin-1 in distal nephron cells. Am J Physiol 271: F451-460, 1996. 42. Garg LC and Narang N. Ouabain-insensitive K-adenosine triphosphatase in distal nephron segments of the rabbit. J Clin Invest 81: 1204-1208, 1988. 43. Garty H, Lindzen M, Fuzesi M, Aizman R, Goldshleger R, Asher C, and Karlish SJ. A specific functional interaction between CHIF and Na,K-ATPase: role of FXYD proteins in the cellular regulation of the pump. Ann N Y Acad Sci 986: 395-400, 2003. 44. Geering K. The functional role of beta subunits in oligomeric P-type ATPases. J Bioenerg Biomembr 33: 425-438, 2001. 45. Geering K. The functional role of the beta-subunit in the maturation and intracellular transport of Na,K-ATPase. FEBS Lett 285: 189-193, 1991. 46. Geering K, Crambert G, Yu C, Korneenko TV, Pestov NB, and Modyanov NN. Intersubunit interactions in human X,K-ATPases: role of membrane domains M9 and M10 in the assembly process and association efficiency of human nongastric H,K-ATPase subunits (ATP1AL1) with known subunits. Biochemistry 39: 12688-12698, 2000.

PAGE 181

160 47. Gibrat JF, Garnier J, and Robson B. Further developments of protein secondary structure prediction using information theory. New parameters and consideration of residue pairs. J Mol Biol 198: 425-443, 1987. 48. Gormley K, Dong Y, and Sagnella GA. Regulation of the epithelial sodium channel by accessory proteins. Biochem J 371: 1-14, 2003. 49. Grishin AV, Bevensee MO, Modyanov NN, Rajendran V, Boron WF, and Caplan MJ. Functional expression of the cDNA encoded by the human ATP1AL1 gene. Am J Physiol 271: F539-551, 1996. 50. Grishin AV and Caplan MJ. ATP1AL1, a member of the non-gastric H,K-ATPase family, functions as a sodium pump. J Biol Chem 273: 27772-27778, 1998. 51. Gumz ML, Duda D, McKenna R, Wingo CS, and Cain BD. Molecular modeling of the rabbit colonic (HKalpha2a) H(+), K(+) ATPase. J Mol Model (Online) 9: 283-289, 2003. 52. Gumz ML, Popp MP, Wingo CS, and Cain BD. Early transcriptional effects of aldosterone in a mouse inner medullary collecting duct cell line. Am J Physiol Renal Physiol 285: F664-673, 2003. 53. Hasler U, Crambert G, Horisberger JD, and Geering K. Structural and functional features of the transmembrane domain of the Na,K-ATPase beta subunit revealed by tryptophan scanning. J Biol Chem 276: 16356-16364, 2001. 54. Hebert H, Purhonen P, Vorum H, Thomsen K, and Maunsbach AB. Three-dimensional structure of renal Na,K-ATPase from cryo-electron microscopy of two-dimensional crystals. J Mol Biol 314: 479-494, 2001. 55. Heinemeyer T, Wingender E, Reuter I, Hermjakob H, Kel AE, Kel OV, Ignatieva EV, Ananko EA, Podkolodnaya OA, Kolpakov FA, Podkolodny NL, and Kolchanov NA. Databases on transcriptional regulation: TRANSFAC, TRRD and COMPEL. Nucleic Acids Res 26: 362-367, 1998. 56. Hocher B, Lun A, Priem F, Neumayer HH, and Raschack M. Renal endothelin system in diabetes: comparison of angiotensinconverting enzyme inhibition and endothelin-A antagonism. J Cardiovasc Pharmacol 31: S492-495, 1998. 57. Husted RF, Laplace JR, and Stokes JB. Enhancement of electrogenic Na+ transport across rat inner medullary collecting duct by glucocorticoid and by mineralocorticoid hormones. J Clin Invest 86: 498-506, 1990. 58. Husted RF and Stokes JB. Separate regulation of Na+ and anion transport by IMCD: location, aldosterone, hypertonicity, TGF-beta 1, and cAMP. Am J Physiol 271: F433-439, 1996.

PAGE 182

161 59. Inoue A, Yanagisawa M, Takuwa Y, Mitsui Y, Kobayashi M, and Masaki T. The human preproendothelin-1 gene. Complete nucleotide sequence and regulation of expression. J Biol Chem 264: 14954-14959, 1989. 60. Ito Y, Aten J, Bende RJ, Oemar BS, Rabelink TJ, Weening JJ, and Goldschmeding R. Expression of connective tissue growth factor in human renal fibrosis. Kidney Int 53: 853-861, 1998. 61. Ivanov AV, Modyanov NN, and Askari A. Role of the self-association of beta subunits in the oligomeric structure of Na+/K+-ATPase. Biochem J 364: 293-299, 2002. 62. Jaisser F, Escoubet B, Coutry N, Eugene E, Bonvalet JP, and Farman N. Differential regulation of putative K(+)-ATPase by low-K+ diet and corticosteroids in rat distal colon and kidney. Am J Physiol 270: C679-687, 1996. 63. Jones TA, Zou JY, Cowan SW, and Kjeldgaard. Improved methods for binding protein models in electron density maps and the location of errors in these models. Acta Crystallogr A 47: 110-119, 1991. 64. Khaldoun M, Khaldoun T, Mellado M, Cambar J, and Brudieux R. Circadian rhythm in plasma aldosterone concentration and its seasonal modulation in the camel (Camelus dromedarius) living in the Algerian Sahara desert. Chronobiol Int 19: 683-693, 2002. 65. Kimura T, Tabuchi Y, Takeguchi N, and Asano S. Mutational study on the roles of disulfide bonds in the beta-subunit of gastric H+,K+-ATPase. J Biol Chem 277: 20671-20677, 2002. 66. Koeppen BM and Stanton BA. Renal Physiology, Third Edition. St. Louis: Mosby, Inc., 2001. 67. Kraut JA, Hiura J, Shin JM, Smolka A, Sachs G, and Scott D. The Na(+)-K(+)-ATPase beta 1 subunit is associated with the HK alpha 2 protein in the rat kidney. Kidney Int 53: 958-962, 1998. 68. Krogh A, Larsson B, von Heijne G, and Sonnhammer EL. Predicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J Mol Biol 305: 567-580, 2001. 69. Laskowski RA, MacArthur, M.W., Moss, D.S., and Thornton, J.M. PROCHECK a program to check the stereochemical quality of protein structures. J Appl Cryst 26: 283-291, 1993.

PAGE 183

162 70. Loffing J, Pietri L, Aregger F, Bloch-Faure M, Ziegler U, Meneton P, Rossier BC, and Kaissling B. Differential subcellular localization of ENaC subunits in mouse kidney in response to highand low-Na diets. Am J Physiol Renal Physiol 279: F252-258, 2000. 71. Loffing J, Zecevic M, Feraille E, Kaissling B, Asher C, Rossier BC, Firestone GL, Pearce D, and Verrey F. Aldosterone induces rapid apical translocation of ENaC in early portion of renal collecting system: possible role of SGK. Am J Physiol Renal Physiol 280: F675-682, 2001. 72. Malik N, Canfield VA, Beckers MC, Gros P, and Levenson R. Identification of the mammalian Na,K-ATPase 3 subunit. J Biol Chem 271: 22754-22758, 1996. 73. Marieb EN. Essentials of Human Anatomy and Physiology, Sixth Edition. San Francisco: Benjamin Cummings, 2000. 74. Meneton P, Lesage, F., and Barhanin, J. Potassium ATPases, Channels, amd Transporters: An Overview. Seminars in Nephrology 19: 438-457, 1999. 75. Meneton P, Schultheis PJ, Greeb J, Nieman ML, Liu LH, Clarke LL, Duffy JJ, Doetschman T, Lorenz JN, and Shull GE. Increased sensitivity to K+ deprivation in colonic H,K-ATPase-deficient mice. J Clin Invest 101: 536-542, 1998. 76. Moore-ede MC, Schmelzer WS, Kass DA, and Herd JA. Cortisol-mediated synchrinization of circadian rhythm in urinary potassium excretion. Am J Physiol 233: R230-238, 1977. 77. Naray-Fejes-Toth A, Canessa C, Cleaveland ES, Aldrich G, and Fejes-Toth G. sgk is an aldosterone-induced kinase in the renal collecting duct. Effects on epithelial na+ channels. J Biol Chem 274: 16973-16978, 1999. 78. Naray-Fejes-Toth A, Fejes-Toth G, Volk KA, and Stokes JB. SGK is a primary glucocorticoid-induced gene in the human. J Steroid Biochem Mol Biol 75: 51-56, 2000. 79. Nicholls A, Sharp KA, and Honig B. Protein folding and association: insights from the interfacial and thermodynamic properties of hydrocarbons. Proteins 11: 281-296, 1991. 80. Norman A and Litwack G. Hormones, Second Edition. San Diego: Academic Press, 1997. 81. Oberleithner H, Steigner W, Silbernagl S, Vogel U, Gstraunthaler G, and Pfaller W. Madin-Darby canine kidney cells. III. Aldosterone stimulates an apical H+/K+ pump. Pflugers Arch 416: 540-547, 1990.

PAGE 184

163 82. Or E, Goldshleger R, and Karlish SJ. Characterization of disulfide cross-links between fragments of proteolyzed Na,K-ATPase. Implications for spatial organization of trans-membrane helices. J Biol Chem 274: 2802-2809, 1999. 83. Otto TC. Molecular Analysis of F1F0 Adenosine Triphosphate Synthase and Renal H+/K+-Adenosine Triphosphatases (Doctoral Dissertation). Gainesville: University of Florida, 2001. 84. Palmgren MGaA, K.B. Evolution of P-Type ATPases. Biochim Biophys Acta 1365: 37-45, 1998. 85. Pearce D, Verrey F, Chen SY, Mastroberardino L, Meijer OC, Wang J, and Bhargava A. Role of SGK in mineralocorticoid-regulated sodium transport. Kidney Int 57: 1283-1289, 2000. 86. Pecci A, Cozza EN, Devlin M, Gomez-Sanchez CE, and Gomez-Sanchez EP. Endothelin-1 stimulation of aldosterone and zona glomerulosa ouabainsensitive sodium/potassium-ATPase. J Steroid Biochem Mol Biol 50: 49-53, 1994. 87. Peti-Peterdi J, Bebok Z, Lapointe JY, and Bell PD. Novel regulation of cell [Na(+)] in macula densa cells: apical Na(+) recycling by H-K-ATPase. Am J Physiol Renal Physiol 282: F324-329, 2002. 88. Pietras RJ, Nemere I, and Szego CM. Steroid hormone receptors in target cell membranes. Endocrine 14: 417-427, 2001. 89. Rajendran VM, Sangan P, Geibel J, and Binder HJ. Ouabain-sensitive H,K-ATPase functions as Na,K-ATPase in apical membranes of rat distal colon. J Biol Chem 275: 13035-13040, 2000. 90. Rice WJ, Young HS, Martin DW, Sachs JR, and Stokes DL. Structure of Na(+),K(+)-ATPase at 11-A Resolution: Comparison with Ca(2+)-ATPase in E(1) and E(2) States. Biophys J 80: 2187-2197, 2001. 91. Robert-Nicoud M, Flahaut M, Elalouf JM, Nicod M, Salinas M, Bens M, Doucet A, Wincker P, Artiguenave F, Horisberger JD, Vandewalle A, Rossier BC, and Firsov D. Transcriptome of a mouse kidney cortical collecting duct cell line: effects of aldosterone and vasopressin. Proc Natl Acad Sci U S A 98: 2712-2716, 2001. 92. Rossi GP, Cavallin M, Nussdorfer GG, and Pessina AC. The endothelin-aldosterone axis and cardiovascular diseases. J Cardiovasc Pharmacol 38 Suppl 2: S49-52, 2001. 93. Rossi GP, Sacchetto A, Cesari M, and Pessina AC. Interactions between endothelin-1 and the renin-angiotensin-aldosterone system. Cardiovasc Res 43: 300-307, 1999.

PAGE 185

164 94. Ryseck RP, Macdonald-Bravo H, Mattei MG, and Bravo R. Structure, mapping, and expression of fisp-12, a growth factorinducible gene encoding a secreted cysteine-rich protein. Cell Growth Differ 2: 225-233, 1991. 95. Sangan P, Kolla SS, Rajendran VM, Kashgarian M, and Binder HJ. Colonic H-K-ATPase beta-subunit: identification in apical membranes and regulation by dietary K depletion. Am J Physiol 276: C350-360, 1999. 96. Sangan P, Thevananther S, Sangan S, Rajendran VM, and Binder HJ. Colonic H-K-ATPase and -subunits express ouabain-insensitive H-K-ATPase. Am J Physiol Cell Physiol 278: C182-189, 2000. 97. Scarborough GA. Crystallization, Structure and Dynamics of the Proton-Translocating P-Type ATPases. The Journal of Experimental Biology 203: 147-154, 2000. 98. Shigaev A, Asher C, Latter H, Garty H, and Reuveny E. Regulation of sgk by aldosterone and its effects on the epithelial Na(+) channel. Am J Physiol Renal Physiol 278: F613-619, 2000. 99. Simpson AM and Schwartz GJ. Distal renal tubular acidosis with severe hypokalaemia, probably caused by colonic H(+)-K(+)-ATPase deficiency. Arch Dis Child 84: 504-507, 2001. 100. Sonnenberg H, Cheema-Dhadli S, Goldstein MB, Stinebaugh BJ, Wilson DR, and Halperin ML. Ammonia addition into the medullary collecting duct of the rat. Kidney Int 19: 281-287, 1981. 101. Sonnenberg H and Chong C. Comparison of micropuncture of microcatheterization in papillary collecting duct. Am J Physiol 239: F95-95, 1980. 102. Spicer Z, Clarke LL, Gawenis LR, and Shull GE. Colonic H(+)-K(+)-ATPase in K(+) conservation and electrogenic Na(+) absorption during Na(+) restriction. Am J Physiol Gastrointest Liver Physiol 281: G1369-1377, 2001. 103. Spindler B, Mastroberardino L, Custer M, and Verrey F. Characterization of early aldosterone-induced RNAs identified in A6 kidney epithelia. Pflugers Arch 434: 323-331, 1997. 104. Stenson C, B. McMahon, M. Clarkson. CTGF expression in the glomeruli following 5/6 nephrectomy: regulation by cytokines and abrogation by ACE-inhibitors. J Am Soc Nephrol 10: 580, 1999. 105. Sun ZS, Albrecht U, Zhuchenko O, Bailey J, Eichele G, and Lee CC. RIGUI, a putative mammalian ortholog of the Drosophila period gene. Cell 90: 1003-1011, 1997.

PAGE 186

165 106. Sweadner KJ, Arystarkhova E, Donnet C, and Wetzel RK. FXYD proteins as regulators of the Na,K-ATPase in the kidney. Ann N Y Acad Sci 986: 382-387, 2003. 107. Sweadner KJ and Donnet C. Structural similarities of Na,K-ATPase and SERCA, the Ca(2+)-ATPase of the sarcoplasmic reticulum. Biochem J 356: 685-704, 2001. 108. Taal MW, T Jiang, H.G. Rennke. Increases in CTGF and TGF-beta 1 expression, predictors of fibrosis in remnant kidneys, are prevented by chronic treatment with candesartan. J Am Soc Nephrol 10: 671, 1999. 109. Taruscio D, Zoraqi GK, Falchi M, Iosi F, Paradisi S, Di Fiore B, Lavia P, and Falbo V. The human per1 gene: genomic organization and promoter analysis of the first human orthologue of the Drosophila period gene. Gene 253: 161-170, 2000. 110. Therien AG and Blostein R. Mechanisms of sodium pump regulation. Am J Physiol Cell Physiol 279: C541-566, 2000. 111. Therien AG, Pu HX, Karlish SJ, and Blostein R. Molecular and functional studies of the gamma subunit of the sodium pump. J Bioenerg Biomembr 33: 407-414, 2001. 112. Thompson J, Higgins D, and Gibson T. CLUSTALW: improving the sensitivity of progressive multiple sequence alignment through sequence weighing, position-specific gap penalties and weight matrix choice. Nucl Acid Res 22: 4673-4680, 1994. 113. Toyoshima C, Nakasako M, Nomura H, and Ogawa H. Crystal structure of the calcium pump of sarcoplasmic reticulum at 2.6 A resolution [see comments]. Nature 405: 647-655, 2000. 114. Toyoshima C and Nomura H. Structural changes in the calcium pump accompanying the dissociation of calcium. Nature 418: 605-611, 2002. 115. Toyoshima C, Nomura H, and Sugita Y. Structural basis of ion pumping by Ca(2+)-ATPase of sarcoplasmic reticulum. FEBS Lett 555: 106-110, 2003. 116. Toyoshima C, Sasabe H, and Stokes DL. Three-dimensional cryo-electron microscopy of the calcium ion pump in the sarcoplasmic reticulum membrane. Nature 362: 467-471, 1993. 117. Verlander JW, Moudy RM, Campbell WG, Cain BD, and Wingo CS. Immunohistochemical localization of H-K-ATPase alpha(2c)-subunit in rabbit kidney. Am J Physiol Renal Physiol 281: F357-365, 2001.

PAGE 187

166 118. Verrey F, Kraehenbuhl JP, and Rossier BC. Aldosterone induces a rapid increase in the rate of Na,K-ATPase gene transcription in cultured kidney cells. Mol Endocrinol 3: 1369-1376, 1989. 119. Verrey F, Loffing J, Zecevic M, Heitzmann D, and Staub O. SGK1: aldosterone-induced relay of Na+ transport regulation in distal kidney nephron cells. Cell Physiol Biochem 13: 21-28, 2003. 120. Verrey F, Pearce D, Pfeiffer R, Spindler B, Mastroberardino L, Summa V, and Zecevic M. Pleiotropic action of aldosterone in epithelia mediated by transcription and post-transcription mechanisms. Kidney International 57: 1277-1282, 2000. 121. Voet D and Voet J. Biochemistry. New York City: John Wiley & Sons, Inc., 1995. 122. Volk KA, Sigmund RD, Snyder PM, McDonald FJ, Welsh MJ, and Stokes JB. rENaC is the predominant Na+ channel in the apical membrane of the rat renal inner medullary collecting duct. J Clin Invest 96: 2748-2757, 1995. 123. Waldegger S, Erdel M, Nagl UO, Barth P, Raber G, Steuer S, Utermann G, Paulmichl M, and Lang F. Genomic organization and chromosomal localization of the human SGK protein kinase gene. Genomics 51: 299-302, 1998. 124. Weiner ID and Wingo CS. Hyperkalemia: a potential silent killer. J Am Soc Nephrol 9: 1535-1543, 1998. 125. Wesson DE and Dolson GM. Endothelin-1 increases rat distal tubule acidification in vivo. Am J Physiol 273: F586-594, 1997. 126. Wingo CS. Active proton secretion and potassium absorption in the rabbit outer medullary collecting duct. Functional evidence for proton-potassiumactivated adenosine triphosphatase. J Clin Invest 84: 361-365, 1989. 127. Yokoi H, Mukoyama M, Sugawara A, Mori K, Nagae T, Makino H, Suganami T, Yahata K, Fujinaga Y, Tanaka I, and Nakao K. Role of connective tissue growth factor in fibronectin expression and tubulointerstitial fibrosis. Am J Physiol Renal Physiol 282: F933-942, 2002. 128. Yoo D, Kim BY, Campo C, Nance L, King A, Maouyo D, and Welling PA. Cell surface expression of the ROMK (Kir 1.1) channel is regulated by the aldosterone-induced kinase, SGK-1, and protein kinase A. J Biol Chem 278: 23066-23075, 2003. 129. Zhang P, Toyoshima C, Yonekura K, Green NM, and Stokes DL. Structure of the calcium pump from sarcoplasmic reticulum at 8-A resolution. Nature 392: 835-839, 1998.

PAGE 188

167 130. Zies DM. Molecular Characterization of the Rabbit HK2 Gene (Doctoral Dissertation). Gainesville: University of Florida, 2003.

PAGE 189

BIOGRAPHICAL SKETCH Michelle Louise Gumz was born in Phoenix, Arizona, on December 14, 1976. The next year, she moved to California where her family eventually settled in Hinkley. In 1994 she graduated from Barstow High School as Salutatorian and class president. Michelle married Keith Gumz in 1996. She graduated from the University of California, Riverside, in 1999 with a Bachelor of Science Degree in biochemistry, and a minor in history. Keith and Michelle moved to Florida in the spring of 1999 and Michelle started graduate school at the University of Florida that fall. In the spring of 2000, Michelle joined the laboratory of Dr. Brian Cain where she began the work described in this dissertation. In August 2003, Michelle and Keiths daughter, Madison, was born. After graduating from the University of Florida, Michelle plans to move with her husband and daughter to either Arizona or California, where she is currently seeking a postdoctoral position. 168