<%BANNER%>

Functional Studies of Purified Recombinant Bk Channels and the Mechanosensitive Channel of High Conductance (mscl) Recon...

Permanent Link: http://ufdc.ufl.edu/UFE0041236/00001

Material Information

Title: Functional Studies of Purified Recombinant Bk Channels and the Mechanosensitive Channel of High Conductance (mscl) Reconstituted in Bilayer Lipid Membranes Tethered to a Microelectrode Array Device
Physical Description: 1 online resource (210 p.)
Language: english
Creator: Okeyo, George
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The measurement of single-channel activity of the high conductance calcium-activated potassium (BK) channels incorporated in tethered bilayer lipid membranes is a major step towards the development of biosensors based on ion channels. Recombinant BK channels were expressed with hexa-histidine fusion tags and isolated from Xenopus laevis oocyte membranes by detergent solubilization. The histidine tag was included to facilitate purification of the desired channels by immobilized metal-ion affinity chromatography, which provided a convenient purification protocol that yielded high purity BK channels which in turn were successfully reconstituted in vesicles consisting of phytanoyl phospholipids. Biophysical characterization of vesicular and solid-supported bilayer lipid membrane assemblies of two different phytanoyl lipids, 1,2-diphytanoyl-sn-glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoyl-sn-glycero-3-phosphoethanolamine (DPhPE) were performed. These studies revealed the appropriate physical conditions and optimal lipid concentrations required to form stable membranes suitable for both reconstitution of proteins and for tethering on solid substrates to form electrically stable membranes. Results showed that the most stable biomimetic model membranes were formed using a combination of DPhPC and DPhPE lipids in a 7:3 ratio; these were then interfaced to microelectrode array devices for incorporation of channels for functional studies. Formation of the tethered bilayer lipid membrane on a gold substrate and the successful incorporation of BK channels in this membrane system were verified with surface plasmon resonance-enhanced ellipsometry. Functional studies of incorporated channels were performed using a modified patch clamp electrophysiology technique and single-channel events were recorded. Results indicate that the IMAC purified BKCa channels were functional and that their electrical properties in the tBLM system had similarities with channels investigated under conventional patch-clamp conditions. However, conductance levels were lower in the tBLM and channels in this membrane system exhibited slower gating kinetics. Pharmacological studies of tBLM incorporated BK channels showed that these channels were sensitive to tetraethylammonium compounds at micromolar concentrations, just like wild type channels investigated by patch clamp techniques. The mechanosensitive channel of large conductance (MscL), a stretch-activated ion channel isolated from E. coli was also studied in this experimental set-up. This study represents the first documented report on the investigation of MscL in a supported bilayer membrane. Results obtained here demonstrate that the MscL channel can be activated by voltage and that the channel is gated in response to stress in the lipid membrane as opposed to pressure across it. Furthermore, these findings show the possibility of using MscL as a release valve for engineered membrane devices; one step closer to mimicking the true function of the living cell.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by George Okeyo.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Fanucci, Gail E.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041236:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041236/00001

Material Information

Title: Functional Studies of Purified Recombinant Bk Channels and the Mechanosensitive Channel of High Conductance (mscl) Reconstituted in Bilayer Lipid Membranes Tethered to a Microelectrode Array Device
Physical Description: 1 online resource (210 p.)
Language: english
Creator: Okeyo, George
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The measurement of single-channel activity of the high conductance calcium-activated potassium (BK) channels incorporated in tethered bilayer lipid membranes is a major step towards the development of biosensors based on ion channels. Recombinant BK channels were expressed with hexa-histidine fusion tags and isolated from Xenopus laevis oocyte membranes by detergent solubilization. The histidine tag was included to facilitate purification of the desired channels by immobilized metal-ion affinity chromatography, which provided a convenient purification protocol that yielded high purity BK channels which in turn were successfully reconstituted in vesicles consisting of phytanoyl phospholipids. Biophysical characterization of vesicular and solid-supported bilayer lipid membrane assemblies of two different phytanoyl lipids, 1,2-diphytanoyl-sn-glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoyl-sn-glycero-3-phosphoethanolamine (DPhPE) were performed. These studies revealed the appropriate physical conditions and optimal lipid concentrations required to form stable membranes suitable for both reconstitution of proteins and for tethering on solid substrates to form electrically stable membranes. Results showed that the most stable biomimetic model membranes were formed using a combination of DPhPC and DPhPE lipids in a 7:3 ratio; these were then interfaced to microelectrode array devices for incorporation of channels for functional studies. Formation of the tethered bilayer lipid membrane on a gold substrate and the successful incorporation of BK channels in this membrane system were verified with surface plasmon resonance-enhanced ellipsometry. Functional studies of incorporated channels were performed using a modified patch clamp electrophysiology technique and single-channel events were recorded. Results indicate that the IMAC purified BKCa channels were functional and that their electrical properties in the tBLM system had similarities with channels investigated under conventional patch-clamp conditions. However, conductance levels were lower in the tBLM and channels in this membrane system exhibited slower gating kinetics. Pharmacological studies of tBLM incorporated BK channels showed that these channels were sensitive to tetraethylammonium compounds at micromolar concentrations, just like wild type channels investigated by patch clamp techniques. The mechanosensitive channel of large conductance (MscL), a stretch-activated ion channel isolated from E. coli was also studied in this experimental set-up. This study represents the first documented report on the investigation of MscL in a supported bilayer membrane. Results obtained here demonstrate that the MscL channel can be activated by voltage and that the channel is gated in response to stress in the lipid membrane as opposed to pressure across it. Furthermore, these findings show the possibility of using MscL as a release valve for engineered membrane devices; one step closer to mimicking the true function of the living cell.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by George Okeyo.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Fanucci, Gail E.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041236:00001


This item has the following downloads:


Full Text

PAGE 1

1 FUNCTIONAL STUDIES O F PURIFI ED RECOMBINANT BKCA CHANNELS AND THE MECHANOSENSITIVE CHANNEL OF HIGH CONDUCT ANCE (MSCL) RECONSTITUTED IN BIL AYER LIPID MEMBRANES TETHERED TO A MICROELECTRODE ARRAY DEVICE By GEORGE ODHIAMBO OKEYO A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 George Odhiambo Okeyo

PAGE 3

3 To m y Parents: Zephaniah Jura Okeyo and Peres Atieno Jura

PAGE 4

4 ACKNOWLEDGMENTS I would first like to extend my utmost appreciation to Dr Gail Fanucci for taking over the responsibility of being my advisor after the departure of Dr Randol ph Duran from the University of Florida Additionally, her support, encouragement and technical advice over the years cannot go unnoticed, and for that I will always be grateful. I also thank Dr Duran for taking me into his group when I joined the Univers ity of Florida and for his support while there. Special thanks also go to Dr Peter Anderson for his commitment, advice and the special interest he took to make sure that I succeeded in all my work. I thank him too for facilitating all those trips to the Whitney Laboratory for Marine Bioscience and his technical advice which I always asked for in short notice, but received promptly, courtesy of his sacrificed time I thank Dr Joanna Long for her encouragement and for letting me in her laboratory to learn from Dr Chris Williams whom I also thank I appreciate the support of the rest of my committee members ; Dr Nicole Horenstein for her encouragement and Dr Ben jamin Smith for making all challenges seem easily manageable. My experiences in the Chemistry d epartment over the years were enriched by past members of the Duran group, Dr Martin Andersson, Dr Maria Stjerndahl, Dr Henk Keizer, Dr Firouzeh Sabri, Jorge Chavez, Eric Greeley, Brian Dorvel and all the others that I interacted with, whom I thank for their friendship and support. Id like to thank those at the Whitney Laboratory who made my life easier, especially Becky Price and Dr Christelle Brouchard. I thank Mandy Blackburn and all members of the Fanucci research group who offered constructive cr iticism and support.

PAGE 5

5 Finally, I would like to thank my family for their love and support. I thank my wife Janetricks Chebukati for her unconditional love, the scientific discussions and for putting up with me. I thank my daughters Lavender and Subi for bri ghtening my life. I also thank my parents and siblings who helped to shape me into the man I am today. Above all else, I thank God through whom I have all that I have and in whom all things are possible. Glory be to God!

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ...................................................................................................... 4 LIST OF TABLES .............................................................................................................. 10 LIST OF FIGURES ............................................................................................................ 11 LIST OF ABBREVIATIONS .............................................................................................. 15 ABSTRACT ........................................................................................................................ 17 CHAPTER 1 INTRODUCTION TO ION CHANNELS AND BIOMIMETIC MODEL MEMBRANE SYST EMS ............................................................................................. 19 Potassium Ion Channels ............................................................................................. 20 Calcium Activated Potassium Channels ............................................................. 22 High conductance calcium activated potassium channels .......................... 25 BKCa channel modulation .............................................................................. 25 Pharmacology ................................................................................................ 28 Voltage -Gated Potassium Channels ................................................................... 29 Membrane Lipids ........................................................................................................ 30 Phosphoglycerides ............................................................................................... 31 Cholesterol ........................................................................................................... 33 Biomimetic Membrane Systems ................................................................................. 35 Vesicles ................................................................................................................ 35 Supported Lipid Bilayers ...................................................................................... 36 Tethered Bilayer Lipid Membranes ..................................................................... 39 Electrical Pr operties of Membranes ........................................................................... 42 Stochastic Sensing ..................................................................................................... 46 2 EXPERIMENTAL PROCEDURES AND TECHNIQUES ........................................... 48 Introduction ................................................................................................................. 48 Techniques and Methods ........................................................................................... 49 Immobilized Metal Ion Affinity Chromatography (IMAC) .................................... 49 Theory ............................................................................................................ 49 Experimental settings .................................................................................... 50 Patch Clamp Electrophysiology ........................................................................... 50 Theory ............................................................................................................ 50 Experimental Settings ................................................................................... 54 Dynamic Light Scattering ..................................................................................... 61 Theory ............................................................................................................ 61 Experimental settings .................................................................................... 66

PAGE 7

7 Negative Staining Transmission Electron Microscopy ........................................ 67 Theory ............................................................................................................ 67 Experimental settings .................................................................................... 68 Atom ic Force Microscopy .................................................................................... 69 Theory ............................................................................................................ 69 Experimental settings .................................................................................... 71 Experi mental Procedures ........................................................................................... 73 Molecular Biology ................................................................................................. 73 DNA manipulation ......................................................................................... 73 RNA synthesis and injection into oocytes .................................................... 75 Sodium dodecyl sulfate polyacrylamide gel electrophoresis ....................... 76 Western blotting ............................................................................................. 77 Vesicle Formation Using DPhPC and DPhPE Lipids ......................................... 78 Reconstitution of Recombinant Proteins in Liposomes ...................................... 78 3 BKCa CHANNEL EXPRESSION, PURIFICATION AND FUNCTIONAL RECONSTITUTION IN LIPID VESICLES .................................................................. 80 Introduction ................................................................................................................. 80 Materials and Methods ............................................................................................... 85 Lipids and Chemicals ........................................................................................... 85 Oocytes ................................................................................................................ 86 Plasmids ............................................................................................................... 86 DNA Preparation and Manipulation ..................................................................... 87 Expression in Oocytes ......................................................................................... 88 Analysis of Expression: Electrophysiological Recordings .................................. 89 M embrane Extraction of Expressed Channels ................................................... 90 Immo bilized Metal Ion Affinity Chromatography ................................................. 91 Western Blot Analysis .......................................................................................... 91 Lipid Vesicle Formation........................................................................................ 92 Reconstitution of Recombinant Proteins in Artificial Liposomes ........................ 93 Negative -Staining Transmission Electron Microscopy ....................................... 94 Results and Discussion .............................................................................................. 94 Expression of Recombinant BKCa Channels ....................................................... 94 Analysis of Expression by Two -Elec trode Voltage Clamping ............................. 95 Isolation and Solubilization of Expressed mRFP1-BKCa Channels .................. 101 Immobilized Metal Ion Affinity Chrom atography ............................................... 102 Reconstitution of BKCa Channels into Liposomes ............................................. 111 Conclusions .............................................................................................................. 113 4 DIPHYTANOYLPHOSPHATIDYLCHOLINE AND -ETHANOLAMINE LIPID MIXT URE CHARACTERIZATION OF VESICLES AND PLANAR BILAYER FORMATION ............................................................................................................ 115 Introduction ............................................................................................................... 115 Materials and Methods ............................................................................................. 118 Lipids and Chemicals ......................................................................................... 118

PAGE 8

8 Gold Surfaces ..................................................................................................... 118 Vesicle Preparation ............................................................................................ 119 Cryogenic Transmission Electron Microscopy (Cryo-TEM) .............................. 119 Dynamic Light Scattering (DLS) ........................................................................ 120 NMR Diffusion .................................................................................................... 120 Quartz Crystal Microbalance with Dissipation Monitoring (QCM -D) ................ 121 Atomic Force Microscopy (AFM) ....................................................................... 122 Results ...................................................................................................................... 123 Cryogenic Transmission Electron M icroscopy (Cryo -TEM) .............................. 123 DLS and NMR Diffusion ..................................................................................... 125 Quartz Crystal Microbalance with Dissipation Monitoring (QCM -D) ................ 127 Atomic Force Microscopy (AFM) ....................................................................... 130 Discussion ................................................................................................................. 132 Conclusions .............................................................................................................. 136 5 INCORPORATION OF RECOMBINANT BKCa CHANNELS IN A TETHERED LIPID MEMBRANE AND FUNCTIONAL ANALYSIS .............................................. 138 Introduction ............................................................................................................... 138 Materials and Methods ............................................................................................. 142 The Microelectrode Array (MEA) Device ........................................................... 142 Lipids .................................................................................................................. 143 BKCa Channel ..................................................................................................... 143 Electrophysiology ............................................................................................... 144 Pharmacology .................................................................................................... 144 Lipid Vesicle Formation...................................................................................... 145 Preparation of the Tethered Bilayer .................................................................. 145 Charac terization of tBLM Formation and BKCa Membrane Insertion ............... 146 Results ...................................................................................................................... 146 Membrane Insertion of BKCa Channels ............................................................. 146 Electrophysiology ............................................................................................... 149 Single -Channel Analysis of the BKCa Channel .................................................. 151 BKCa Channel Open Probability ......................................................................... 154 BKCa Conductance ............................................................................................. 156 Pharmacology .................................................................................................... 157 D iscussion ................................................................................................................. 161 Conclusions .............................................................................................................. 166 6 VOLTAGE-INDUCED GATING OF THE MECHANOSENSITIVE CHANNEL OF LARGE CONDUCTANCE (MscL) IN TETHERED B ILAYER LIPID MEMBRANES ........................................................................................................... 168 Introduction ............................................................................................................... 168 The E. coli MscL Structure and Function .......................................................... 168 Gating in the MscL ............................................................................................. 169 Materials and Methods ............................................................................................. 171 Lipids and Chemicals ......................................................................................... 171

PAGE 9

9 The Microelectrode Array Device ...................................................................... 171 MscL Isolation and Purification .......................................................................... 172 Preparation of the Tet hered Bilayer .................................................................. 173 Electrophysiology ............................................................................................... 174 Results and Discussion ............................................................................................ 174 Conclusions .............................................................................................................. 179 7 CONCLUSION AND FUTURE DIRECTIONS ......................................................... 180 APPENDIX A DYNAMIC LIGHT SCATTERING ANALYSIS OF TEMPERATURE EFFECTS ON SIZES OF PHYTANOYL LIPID VESICLES OF DIFFERENT MIXTURES ...... 186 B X -RAY PHOTOELECTRON SPECTROSCOPY USED FOR SURFACE ANALYSIS OF PLASMA-TREATED GOLD ............................................................ 188 C DNA AND AMINO ACID SEQUENCES OF PROTEINS ......................................... 189 D PLASMID MAP FOR THE pCDNAOX AND MSLO GENE INSERTS .................. 191 LI ST OF REFERENCES ................................................................................................. 192 BIOGRAPHICAL SKETCH .............................................................................................. 210

PAGE 10

10 LIST OF TABLES Table page 1 -1 Summary of functional characteristics defining vertebrate calcium activated potassium channels ............................................................................................... 23 1 -2 Summary of BKCa Channel modulatory molecules ............................................... 29 1 -3 Major Phosphoglycerides and their head groups. ................................................. 33 1 -4 Classification of liposomes by size. ....................................................................... 35 5 -1 Comparison of the different ion channels incorporated in the tBLM and their molecular weights ................................................................................................. 164 C -1 Full Length mslo Amino Acid Sequences (BKCa Channel) ................................. 189 C -2 Truncated BKCa Channel at position 335 with 12 additional residues coded for at the C -terminus to introduce a stop codon .................................................. 189 C -3 Red fluorescent protein (mRFP1) DNA and protein sequences ......................... 190 C -5 E. coli Mechanosensitive Channel of Large Conductance (MscL) DNA sequences ............................................................................................................ 190 C -4 E. coli Mech anosensitive Channel of Large Conductance (MscL) amino acid sequences ............................................................................................................ 190

PAGE 11

11 LIST OF FIGURES Figure page 1 -1 Sequence alignment showing the conserved region known as the selectivity filter in members of the potassium channel family. ............................................... 21 1 -2 The structure of the K+ selectivity filter and single letter amino acid code of the signature sequence. ......................................................................................... 22 1 -3 Topology maps of calcium activated potassium channels showing the position of helices and locations of the N and C -termini ..................................... 24 1 -4 BKCa channel modulation by voltage and calcium. ............................................... 26 1 -5 Sequence alignment for Charybdotoxin (ChTX) and Iberiotoxin (IbTX) with non-homologous residues highlighted. .................................................................. 28 1 -6 Structure of phosphatidic acid showing its polar and hydrophilic domains. ......... 31 1 -7 The molecular structure of cholesterol. ................................................................. 34 1 -8 Vesicular m odel membranes ................................................................................. 36 1 -9 Immobilization methods for supported membrane s .............................................. 38 1 -10 The tethered bil ayer lipid membrane (tBLM) ......................................................... 39 1 -11 Structure of the 2,3 -di -O -phytanyl -sn glycerol 1 -tetraethylene glycol D,L lipoic acid ester lipid (DPTL) .................................................................................. 41 1 -12 Structures of 1,2-diphytanoyl -sn -glycero 3 -phosphoethanolamine (DPhPE) and 1,2diphytanoyl -sn glycero -3 phosphocholine (DPhPC) lipids ...................... 42 1 -13 Equivalent circuit diagram of a membrane ............................................................ 45 1 -14 Schematic of stochastic sensing by an engineered pore in a planar bilayer. ...... 46 2 -1 Principal scheme of a typical TEVC re cording arrangement. ............................... 52 2 -2 Event detection occurri ng at level zero and level one. ......................................... 57 2 -3 Histogram fitted by the maximum likelihood with a continuous Gaussian curve representing open and closed event distributions. ..................................... 59 2 -4 Possibilities for the interaction of a laser b eam with a liquid sample ................... 62 2 -5 A simple representation of the basic concept of a transmission electron microscope operating in the bright field mode. ..................................................... 68

PAGE 12

12 2 -6 A simple representation of the basic setup of the atomic force microscope. ....... 71 2 -7 Data analysis of a Z axis calibration grid performed in contact mod e ................. 72 3 -1 Summary of the approaches used to introduce genetic material into oocytes. ... 81 3 -2 Illustration of the full length BKCa channel showing the point of truncation and the truncated product. ............................................................................................ 82 3 -3 Illustration depicting the chimeric C -terminally deleted BKCa channel with positions of fusion proteins and number s of amino acid residues. ....................... 83 3 -4 Bright field microscopy images of oocytes fo r analysis of expression ................. 95 3 -5 TEVC of uninjecte d oocytes as a control for monitoring expression levels of BKCa channels. ....................................................................................................... 96 3 -6 TEVC currents showing expression of mRFP1-tagged BKCa chan nels ............... 97 3 -7 Currents recorded from mRFP1-tagged BKCa channels to test for voltage dependence of expressed channels ...................................................................... 98 3 -8 The current voltage curve for the expressed mRFP1-tagged BKCa channel showing the voltage -dependence of this channel in oocytes. ............................ 100 3 -9 Purification scheme for the histidine -tagged mRFP1BKCa channel. ................. 102 3 -10 UV spectra of fractions from manual mRFP1-BKCa channel purification. .......... 103 3 -11 Image of a 420% gradient Tris -HCl gel stained with Coomassie Blue showing lanes wit h purified fractions of mRFP1-BKCa channels from the Ni2+ affinity column. ...................................................................................................... 104 3 -12 Chromatogram showing the purification stages of mRFP1-BKCa channels on the AKTA Prime automated protein purification system ..................................... 105 3 -13 Electrophoresis and immunoblotting of purified BKCa channels ........................ 106 3 -14 Absorbance spectra for bovin e serum albumin standards in Coomassie blue G -25 0 .................................................................................................................... 109 3 -15 Standard curve plotted from A595 nm values from BSA for the estimation of protein concentration ............................................................................................ 110 3 -16 Negative -staining of 100 nm extruded DPhPC: DPhPE lipid vesicles at a 7:3 molar ratio ............................................................................................................. 111 3 -17 Negative -staining TEM images after dialysis of octyl gluco side for p roteoliposome formation .................................................................................... 112

PAGE 13

13 4 -1 TEM micrographs of hydrated dispersions composed of extruded vesicles ...... 123 4 -2 TEM mi crographs of samples of varying lipid composition. ................................ 124 4 -3 Average effective diameter as a function of lip id ratio ........................................ 126 4 -4 Cha nges in resonant frequency and dissipation versus time for adsorption of pure DPhPC vesicles onto sil ica substrates ........................................................ 127 4 -5 Changes in resonant frequency and dissipation versus time for the a dsorption of pure DPhPC vesicles onto oxidized gold ...................................... 129 4 -6 AFM images showing vesicle fusion on gold substrates .................................... 131 4 -7 AFM i mages of pure DPhPC vesicles deposited on an ultra flat gold sur face .. 131 5 -1 The tethered bilayer membrane on a m icroelectrode array device .................... 141 5 -2 Fitted Kinetic data for incorporation of the BKCa Channel in the tBLM ............... 147 5 -3 Schematic showing tBLM formation and incorporation of BKCa channels ......... 148 5 -4 Recording from an electrically stable membrane formed by fusion of 7: 3 DPhPC: DPhPE vesicles ..................................................................................... 150 5 -5 Single -channel activity of the BKCa channel in a tB LM at +120 mV ................... 151 5 -6 Single -channel activity of BKCa channel in tBLM at 80 mV a pplied voltage ....... 152 5 -7 Single -channel activity of the BKCa channel in tBLM at 50 mV applied potential ................................................................................................................ 153 5 -8 Open probability versus applied voltage fitted to a Boltzmann distribution ........ 154 5 -9 Logarithm histogram of square root ordinate that bins dwell -times of all intervals and fitted lifetimes from three exponentials. ......................................... 155 5 -10 Current voltage rel ationship for the BKCa channel .............................................. 156 5 -11 Single channel traces of channels at 100 mV applied voltage under different concentrations of TEA solutions .......................................................................... 158 5 -12 Dose-response curve of the BKCa channel mut ant ............................................ 159 5 -13 Voltage dependence of blockade by 500 M t etraethylammonium (TEA). ....... 160 6 -1 Crystal structure of MscL from M. tuberculosis adopted from the protein data bank (PDB) ........................................................................................................... 169

PAGE 14

14 6 -2 Current plotted against the applied voltage of a giga -seal ................................ 175 6 -3 A stable trace recorded from a 7:3 DPhPC/DPhPE bilayer membrane over a duration of 10 minutes. ......................................................................................... 175 6 -4 Single channe l behavior of MscL presented as measured current at an applied transmembrane potenti al of + 300 mV ................................................... 176 6 -5 A trace for one MscL showing only one conductivity st ate ................................ 177 A-1 Diameters of 100 % DPhPC vesicles at 25 intensity of scattered light. ................................................................................... 186 A-2 Diameters of vesicles of DPhPC: DPhPE lipid mixtures at a 7:3 molar ratio at 25 ...................... 186 A-3 Diameters of vesicles of DPhPC: DPhPE lipid mixtures at a 5:5 molar rat io at 25 ...................... 187 B-1 XPS results showing gold, oxygen and carbon as the only elements present on plasma-treated gold. ....................................................................................... 188 D -1 Map for the pCDNAOX plasmid containing the mslo BKCa channel gene insert and the restriction sites. ....................................................................................... 191

PAGE 15

15 LIST OF ABBREVIATION S AFM Atomic force microscopy BKCa H igh -conductance, calcium activated potassium (maxi-K) channel BLM Bilayer lipid membranes ChTX Charybdotoxin DLS Dynamic light scattering DPhPE 1, 2-diphytanoyl -sn -glycero 3 -phosphoethanolamine DPhPC 1, 2-diphytanoyl -sn -glycero 3 -phosphocholine DPTL 2,3d i -O phytanyl-sn glycero-1 -tetraethylene glycol DL -lipoic acid ester IbTX Iberiotoxin Kcps Kilocounts per second pS Picosiemens PE Phosphatidylethanolamine PC Phosphatidylcholine mRFP Monomeric red fluorescence protein PIPES 1,4Piperazinebisethanesulfonic acid OG Octyl glucoside M L V Large multilamellar vesicles LUV Large unilamellar vesicles MEA Microelectrode array MOPS 3 -morpholinopropanesulfonic acid RCK Regulator of potassium conductance RT Room temperature Rpm Revolut ions per minute SAM Self assembled monolayers

PAGE 16

16 SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SUV Small unilamellar vesicles sBLM Supported bilayer lipid membranes tBLM Tethered lipid bilayer membrane TEA Tetraethyammonium TEM Transmiss ion electron microscopy TEVC Two electrode voltage clamp

PAGE 17

17 Abstract of Dissertation Pr esented to the Graduate School of the University of Florida in Partial Fulf illment of the Requirements for t he Degree of Doctor of Philosophy FUNCTIONAL STUDIES OF PURIFIED RECOMBINANT BKCA CHANNELS AND THE MECHANOSENSITIVE CHANNEL OF HIGH CONDUCTANCE (MSCL) RECONSTITUTED IN BILAYER LIPI D MEMBRANES TETHERED TO A MICROELECTRODE ARRAY DEVICE By George Odhiambo Okeyo December 2009 Chair: Gail E. Fanucci Major: Chemistry The measurement of single-channel acti vity of the high conductance calciumactivated potassium (BKCa) channels incorporated in tether ed bilayer lipid membranes is a major step towards the development of biosensors based on ion channels. Recombinant BKCa channels were expressed with hexa-histidine fusion tags and isolated from Xenopus laevis oocyte membranes by detergent solubilization. The histidine tag was included to facilitate purification of the desired channels by immobilized metal-ion affinity chromatography which provided a convenient purification protocol that yielded high purity BKCa channels which in turn were successfully reconstituted in vesicles consisting of phytanoyl phospholipids. Biophysical characterization of vesicu lar and solid-supported bilayer lipid membrane assemblies of two different phytanoyl lipids, 1,2-diphytanoylsn -glycero-3phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE) were performed. These studies re vealed the appropriate physical conditions and optimal lipid concentrations required to form stable membranes suitable for both reconstitution of proteins and for tethering on so lid substrates to form electrically stable

PAGE 18

18 membranes. Results showed that the most stable biomimetic model membranes were formed using a combination of DPhPC and DPhP E lipids in a 7:3 rati o; these were then interfaced to microelectrode array devices fo r incorporation of channels for functional studies. Formation of the tethered bilayer lip id membrane on a gold substrate and the successful incorporation of BKCa channels in this membrane system were verified with surface plasmon resonanceenhanced ellipsometry. Functional studies of incorporated c hannels were performed using a modified patch clamp electrophysiology technique and single-channel events were recorded. Results indicate that the IMAC purified BKCa channels were functional and that their electrical properties in the tBLM system had similarities with channels investigated under conventional patch-clamp conditions. However, conductance levels were lower in the tBLM and channels in this membrane system exhibited slower gating kinetics. Pharmacological studies of tBLM incorporated BKCa channels showed that these channels were sensitive to tetraethy lammonium compounds at micromolar concentrations, just like wild type channels investigated by patch clamp techniques. The mechanosensitive channel of large cond uctance (MscL), a stretch-activated ion channel isolated from E. coli was also studied in this experimental set-up. This study represents the first document ed report on the investigation of MscL in a supported bilayer membrane. Results obtained here dem onstrate that the MscL channel can be activated by voltage and that the channel is gated in response to stress in the lipid membrane as opposed to pressure across it Furthermore, these findings show the possibility of using MscL as a release va lve for engineered membrane devices; one step closer to mimicking the true function of the living cell.

PAGE 19

19 CHAPTER 1 INTRODUCTION TO ION CHANNELS AND BIOMIMETIC MODEL MEMBRANE SYSTEMS Ion channels are integral membrane prot eins that form aqueous pores in cell membranes through which sele cted ions and other small molecules can translocate from one side to the other. The permeation of ions across ion channels usually occurs very rapidly at nearly diffusion limited rates ca 108 ions s-1 per channel.1 According to electron microscopy and X-ray crystallography studies, typical aqueous pores within ion channels have lengths of approximately 3 nm and radii which vary along the pore length but which may be as low as 0.2 nm in places. Typically, channel pores have comparable dimensions to the sizes of transported molecules or ions. Many types of cells from diverse organism s including bacteria, viruses, plants and animals have ion channels responsible for the regulation of numerous cellular and physiological functions. Ion channels regulate their physiological roles by switching between open and closed states to allow ion flow in a proce ss of gating which can occur on a wide range of timescales through confo rmational changes of the channel interior. The gating action can be driven by allosteric modulation of conformation due to ligand binding, depolarizing membr ane potentials and mechanical stimuli such as membrane tension, as happens with mechanosensitive c hannels. Mutations in ion channel genes result in alteration of gating properties, re sulting in either excessive or diminished activity thereby causing a number of diseases referred to as channelopathies. Ion channels are therefore drug targets and ar e being studied for the development of therapeutic options for the correction of these disorders which include epilepsy, myotonia, arrhythmias and migraines.2-4

PAGE 20

20 Most ion channels exhibit great selectivity for different ions and show a high level of recognition of specific compounds. Fo r that reason, ion channels make ideal candidates as sensor elements in the rapidly growing field of biosensor development for chemical and biological analytes.5-7 Nature has produced subfamilies of channels selective for each of the prin cipal species of ions in living tissues: potassium, sodium, calcium and chloride ions. In each of these subfamilies, there are many differences in functional properties associated with the types of ions allowed through channels. These functional properties are voltage-dependenc e, gating kinetics and modulation by intracellular factors, for in stance calmodulin and calcium.8 Conductance is a signature functional property that varies between ion channels by at least two orders of magnitude, and it gives a measure of the current that can pass through a single open channel at a given membrane potential.9 Potassium Ion Channels Potassium channels are responsible for a diversity of physiological processes in cells such as hormone secretion, cell volu me regulation and electric impulse formation in electrically excitable cells. Because pot assium channels are t he only ion-selective cation channels that have an equilibrium pot ential near the typical cellular resting potential, they play a vital role in dete rmining the resting potential of most cells.10 The classification of potassium channels is broadly defined by transmembrane topology, as reflected in primary sequence: the si x-transmembrane-helix voltage-gated (Kv) and the two-transmembrane-helix inward-rectifier (Kir) subtypes.11, 12 Fully assembled K+ channels of the two transm embrane and six transmembrane architectures are formed by homoor heterotetramers of principal subunits, often supplemented by auxiliary subunits. BKCa channels have an extra transmembrane domain, the S0 which leads to

PAGE 21

21 an external NH2 terminus, therefore they possess the seven transmembrane architecture. A highly conserved segment is found between the two most carboxyterminal transmembrane helices in all potassium channels, bearing the unique sequence of amino acids TM xTVGYG commonly referred to as the K+ signature sequence.13 A few variations in the signature sequence are known to exist as can be observed in Figure 1-1, which shows seque nce alignments of several different ligandgated and voltage-gated K+ channels with the characteristic selectivity filter portion of the signature sequence highlighted. hBK: YLLMVTMSTVGYGDVYAKTTLGRLFMVFFILGGLAMFASYVPEIIELIGN Shaker:WWAVVTMTTVGYGDMTPVGFWGKIVGSLCVVAGVLTIALPVPVIVSNFNY KcsA: WWSVETATTVGYGDLYPVTLWGRLVAVVVMVAGITSFGLVTAALATWFVG MthK: YWTFVTIATVGYGDYSPSTPLGMYFTVTLIVLGIGTFAVAVERLLEFLIN Dradio:YWAVVTVTTVGYGDISPKTGLGKFIATLAMLSGYAIIAVPTGIVTVGLQQ Ecoli: YFSIETMSTVGYGDIVPVSESARLFTISVIISGITVFATSMTSIFGPLIR hDRK1: WWATITMTTVGYGDIYPKTLLGKIVGGLCCIAGVLVIALPIPIIVNNFSE hGIRK2:LFSIETETTIGYGYRVITDKCPEGIILLLIQSVLGSIVNAFMVGCMFVKI hIRK1: LFSIETQTTIGYGFRCVTDECPIAVFMVVFQSIVGCIIDAFIIGAVMAKM hSK3: WLISITFLSIGYGDMVPHTYCGKGVCLLTGIMGAGCTALVVAVVARKLEL hERG2: YFTFSSLTSVGFGNVSPNTNSEKIFSICVMLIGSLMYASIFGNVSAIIQR Figure 1-1. Sequence alignment showing the conserved region known as the selectivity filter in members of th e potassium channel family. Numerous positively charged amino acid re sidues are found in a conserved region on the fourth transmembrane segment of Kv type channels.14 This region undergoes an outward conformational movement which is energetically favored by depolarizing voltage, and constitutes the central event which initiates channel opening. Potassium channels open to allow the rapid permeation of K+ into cells while rejecting the biologically abundant potent ial competitors Ca2+, Na+ and Mg2+. The exceptionally high ion selectivity and the hi gh throughput rate of K+ channels is made possible by a precise coordination of dehydrated K+ by the protein and multiple ion occupancy within the

PAGE 22

22 permeation pathway of the channel.15 The mechanism of ion selectivity depends on the arrangement of backbone car bonyl groups of the K+ signature sequence of the channel in which a dehydrated K+ exactly fits (Figure 1-2). The smaller Na+ ion would fit so loosely in this region; therefore it energetically favors being in aqueous solution.16 Figure 1-2. The structure of the K+ selectivity filter and single letter amino acid code of the signature sequence. Calcium-Activated Potassium Channels Calcium-activated potassium channels are a large family of potassium channels responsible for translocation of K+ across membranes in response to the binding of Ca2+ at intracellular receptor sites. They are gated by increase in cytoplasmic calcium concentrations which occurs largely in resp onse to calcium influx via voltage-gated calcium channels that open during action potentials.17 Calcium-activated potassium channels are classified into three broad families based on their biophysical and pharmacological properties. These families ar e the large or high conductance calciumactivated potassium (BKCa) channels,18 small conductance calcium-activated potassium (SK) channels19, 20 and the intermediate conductance calcium-activated potassium (IK) channels.21, 22 Table 1-1 summarizes the properties of calcium-activated channels.

PAGE 23

23 Table 1-1. Summary of functional characteri stics defining vertebrate calcium-activated potassium channels Property Channel Type BKCa IKCaSKCa Intracellular Ca 2 + needed for activation 1-10 nM (at -50 mV) 50-900 nM 50-900 nM Voltage dependence e-fold/ 9 to 15 mV None None Single-channel conductance 90-250 pS 20-80 pS 4-20 pS Blockers Charybdotoxin (nM) Iberiotoxin (nM) TEA (nM) -KTx (nM) Clotrimazole (nM) Apamin (nM) Scyllatoxin (nM) Mammalian clones slo1 (slo2,3) IK1/SK4 SK1,2,3 Functional properties of the three subtypes of calcium-activated ion channels differ markedly. BKCa channels are highly selective for potassium ions, have single-channel conductances of 90 pS to 300 pS and require both calcium and membrane depolarization to initiate gating. SK channe ls on the other hand have a single-channel conductance of 2 to 20 pS and unlike BKCa channels, are voltage insensitive. The activity of SK channels following action potentia ls leads to much longer lasting currents than those mediated by BKCa channels, owing to their higher affinity for calcium at hyperpolarized membrane potentials. IK channels show conductances of 20 to 100 pS and are also voltage insensitive.22 Other differences in functional properties between calcium-activated potassium channels are in their response to modulatory agents. Both IK and SK channels are unaffected by BKCa channel blockers isolated from scorpion toxins.19 However, SK channels are potently blocked by the bee venom apamin, amongst other compounds.20 IK channels are distinct ly separable from BK and SK channels since they show no response to sco rpion toxins or apamin. They are poorly

PAGE 24

24 studied due to their sparse distribution in different cell types. Figure 1-3 illustrates the topology maps of BK and SK channels. Figure 1-3. Topology maps of calcium-a ctivated potassium channels showing the position of helices and locations of the Nand C-termini. A) High conductance calcium-activated potassium (BK) c hannel. B) Small conductance calciumactivated potassium (SK) channel. A B

PAGE 25

25 High conductance calcium-activated potassium channels High conductance calcium-activated potassium (BKCa) or maxi-K channels which are the main subjects of this study were the first calcium-activated channels to be identified. They are also called SLO fa mily channels, a name derived from the conserved slowpoke gene that encodes this channel, which was first cloned in Drosophila melanogaster.23, 24 Structurally, BKCa channels comprise a pore-forming tetrameric domain made of seven putative transmembrane segments and a domain comprising two -helical transmembrane domains connected by a large glycosylated extracellular loop, with intracellular amino and carboxy termini.25, 26 The -and subunits are associated with each other noncov alently in the form of an octameric complex in a 1:1 stoichiometry.27-29 BKCa channels have an extrace llular N-terminus on the -subunit, which distinguishe s them from other potassi um channels, and this comes as a result of the extra transmembrane segm ent they possess, relative to the other potassium channels.26 They also have a large cytoplasmic C-terminal domain which controls different functional properties of t he channel. In the C-terminal domain region, two regulators of potassium conductance ( RCK) domains have been identified on each subunit and these connect to the pore region by a short linker.30, 31 The RCK domains have been proposed to form a gating ring where calcium binding causes a conformational change that promotes c hannel opening by pulling the linker connecting the gating ring to the pore region.32, 33 BKCa channel modulation BKCa channels are activated by both depolariz ing membrane potentia ls as well as elevated concentrations of cytosolic calciu m. The calcium dependence for the gating of these channels relies on membrane potential,34 the Kd for calcium being in the

PAGE 26

26 micromolar range at resting membrane potent ials (~ -60 mV) but in the nanomolar range at depolarized potentials (+20 to +40 mV).35 Interestingly, some BKCa channels can open in the absence of calcium and it appears that the effe cts of calcium and membrane potential are independent processes.36 Depolarization causes a conformational change in the voltage sensor regions S0 through S4 perhaps involving a twisting outward motion of the positively ch arged S4 helix. The volt age sensitivity of the channel is conferred by acidic residues in the S2 and S3 segments, along with basic residues at every third pos ition in the S4 segment.37 Each of four voltage sensor movements is coupled to the opening of the conduction pathway, lined by the loop connecting the S5 and S6 regions, driving t he concerted gating transition toward the open state. Likewise, Ca2+ binding causes conformational changes at the cytoplasmic Ca2+ sensors and these changes are coupled to t he opening transition that is optimally driven forward when all voltage sensors and Ca2+ sensors are activated (Figure 1-4).38 Figure 1-4. BKCa channel modulation by voltage and calcium. (Adopted from Ref. 38)

PAGE 27

27 An understanding of the structur al implications of the BKCa channel to modulation by voltage and bound calcium ions would ex plain the observations made during functional studies of C-terminally deleted recombinant channels in this study. The Cterminal region of the BKCa channel contains multiple regulatory sites which include the bulky region in the transmembrane domain referred to as the Ca2+ bowl located just before the S10 segment, implicated in Ca2+ regulation.39 Changes in Ca2+ concentration that regulate channel opening span a range of 0.5 mM to 50 mM, the wide range explained by the existence of multiple ind ependent calcium binding sites. Advances in understanding the regulation of BKCa channels reveal that bot h the calcium bowl located within the S9-S10 peptide, and the RCK domain around the S7-S8 segments play a role in the regulation of channel opening by Ca2+.40, 41 A study comparing the Ca2+ sensitivity of wild-type BKCa currents arising from -subunits to that arising from -subunits with a mutation abolishing the function of the Ca2+ bowl, show current s from the mutant channel still exhibiting subs tantial regulation by Ca2+, albeit with less channel activation at low [Ca2+] relative to wild-type currents.42 Piskorowski and Aldrich performed studies documenting Ca2+ dependent gating of BKCa channels with completely truncated Ctermini, suggesting the existence of an additional undefined Ca2+-activation site different from that located in the RCK domain and the one within the calcium bowl.43 Besides the conventional modulatory mechanisms of depolarizing membrane potentials and elevated concentrati ons of intracellular calcium, BKCa channels can be modulated by post-translational modifications. Glycosylation is known to affect both cellsurface expression and channel activity of BKCa channels.28, 44, 45 N-linked glycosylation sites on the -subunits of these channels provide a means by which modifications can

PAGE 28

28 be done and these influence modulation of c hannel activity as a result of the -subunit regulatory effect or an alterati on of the interaction between the and -subunits.45 Other known post-translational modification s that affect channel modulations include phosphorylation and oxidative reduction.46-48 The effects of post-translational modifications on channel function arising from using oocytes as an expression system were observed during the study docu mented in this dissertation. Pharmacology The well understood signature responses of BKCa channels to agonists (activators) and antagonists (blockers) can provide useful information that makes these channels ideal for development of ion channel based bios ensors, which is the long term goal of this project. Located between t he S5 and S6 segments of the -subunit is the pore loop (P-loop) which is part of the pore-forming motif of BKCa channels and bears the receptor site for the binding of the pore blockers: iberiotoxin (IbT X) and charybdotoxin (ChTX).49 IbTX possesses several acidic residues and has an overall lower net positive charge than ChTX which may explain its selective inhibition of the BKCa channel. IbTX is therefore a better f unctional probe for BKCa channel pharmacology unlike ChTX which is less selective and blocks other potassium channels as well. The sequence alignment comparing the amino acid residu es of these peptidyl blockers is shown in Figure 1-5. ChTX: FTNVSCTTSKECWSVCQRLHNTSRGKCMNKKCRCYS IbTX: FTDVDCSVSKECWSVCKDLFGVDRGKCMGKKCRCYQ Figure 1-5. Sequence alignment for Charybdotoxin (ChTX) and Iberiotoxin (IbTX) with non-homologous residues highlighted. The mechanism for IbTX inhibition of the BKCa channel is the same as that of ChTX, but the dissociation rate of IbTX is much slower.50, 51 These inhibitors are useful not only as agents for investigations of the pharmacology of recombinant BKCa channels

PAGE 29

29 in this study, but can also give an under standing of the channels orientation in the tethered bilayer membrane. Bo th IbTX and ChTX work by binding to residues in the external pore of the channel th rough a freely reversible bi molecular reaction driven by electrostatic and hydrophobic interactions. In the tethered bilayer membrane, modulation of current through channels in response to ChTX and IbTX binding should only be observed if channel incorporation into the membrane occurs with the N-terminus facing outwards. Other structural types of small molecule compounds such as glycotriterpenes possess BKCa channel agonist activity. Dehydrosoyasaponin 1 (DHS-1) inhibits ChTX binding to BKCa channels through an allosteric mechanism and it reversibly increases the open probability of channels at concentrations as low as 10 mM.52 Quaternary ammonium compounds such as tetraethylammonium (TEA) are known pharmacological blockers of the BKCa channel used in this study.53 Table 1-2. Summary of BKCa Channel modulatory molecules Blockers Peptide Blockers: Iberiotoxin, Charybdotoxin, Slototoxin Non-peptide Blockers: Paxill ine, Penitrem, Tetrandrine Non-specific Blockers: Tetraethylammonium, Tetrabutylammonium, Clotrimazole Openers Maxikdiol, Dihydrosoyasaponin-1, Estradiol, Primaric acid, NS004, NS619, NS88 Mefenamic acid, Niflumic acid, Flufenamic acid Voltage-Gated Potassium Channels The simplest member of the voltage-gated potassi um channel family is homotetrameric in structure, with each subunit surrounding the water-filled conduction pathway and bearing a voltage sensor. Unlike BKCa channels, voltage-gated K+ channels have intracellular amino and carboxy termini and six transmembrane regions.54 Each subunit has these six subunits (S1-S6) with the voltage sensor located

PAGE 30

30 on the S4 subunit and the pore located between S5 and S6. Permeability of ions is regulated by channel opening and closing wh ich involves two main mechanisms, a conformational constriction of the permeati on pathway, or a conditional plugging of the pore by an auto-inhibitory part of the channel protein.55 The closing of the channel occurs when the intracellular entrance pinc hes shut and the S6 segment obstructs the entrance from the cytoplasmic surface to the water-filled core of the conduction pathway.56 The second mechanism responsible for the closing of voltage-gated potassium channels is the N-terminal pepti de block which involves a tethered peptide blocker attached to the N terminus (N-type i nactivation) that gave this mechanism the name ball and chain mechanism.57, 58 A third mechanism is postulated to occur via a pinching shut of the pore at the narrowest part of the selectivity filter.59 This inactivation mechanism is thought to regulat e repetitive electrical activity and may also determine a physiological response to accumulation of extracellular potassium ions. Membrane Lipids Biological membranes are complex bilaye r structures that form boundaries between different cell compartm ents and are composed of a di versity of lipids, sterols and membrane proteins. This general membrane composition is found in most organisms, however, a few exceptions are known, for instance, prokaryotes which lack sterols in their membranes.60 Lipids can be broadly classified into two groups; saponifiable lipids, which upon hydrolysis cont ain salts of fatty acids, and the nonsaponifiable which do not contai n fatty acids. Saponifiable lipids are categorized based on their backbone structure to which fatty acid s are covalently attached. Fatty acids are categorized based on their hydrocarbon chain lengths (C4-C36) and degree of saturation; and these determine the physical properties of membranes formed by these

PAGE 31

31 lipids. Membrane lipids play numerous des ignated roles such as storage of energy reserves, provision of a matrix to help main tain correct protein folding and stability, and therefore maintain opt imal cellular function,61, 62 and signal transduction.63, 64 The most abundant membrane lipids are phospholipids, which are der ived from glycerol or sphingosine, though phosphogly cerides are the most co mmonly found phospholipids. Sterols are also a major component of biologi cal membranes where they play important functional and structural roles. Cholestero l is the best studied sterol, and is known to influence phase transitions of phospholipid bilayers, cause a space effect within membrane phospholipids, and affe ct lateral diffusion of lipids in membranes among other roles.65 Phosphoglycerides Phosphoglycerides constitute the major component of structural lipids and are characterized by a terminal hydroxyl group of glycerol esterified to phosphoric acid, with the other two hydroxyl groups esterified to fatty acids, resulting in phosphatidic acid illustrated in Figure 1-6. Phosphoglycerides are amphiphilic and are classified into groups based upon the identity of the moiety a ttached to the phosphate. This moiety is referred to as the headgroup. Hydrophobic Hydrophilic Figure 1-6. Structure of phos phatidic acid showing its polar and hydrophilic domains.

PAGE 32

32 Phosphoglycerides can also be categorized by the number of carbon atoms present in the acyl chains whose carbonyl groups are es terified to the hydroxyl groups on the first and second carbon groups on the glycerol backbone. Glyceride nomenclature is often in terms of the stereos pecific numbering ( sn ) system with most phosphoglycerides having the phosphate lo cated at the sn -3 position of the glycerol. The long-chain hydrocarbons attach to the sn -1 and sn -2 positions of the glycerol through ester linkages.66 Biological membranes differ in terms of their lipid compositions based on the different environmental conditions within which they are found. A variety of ratios of individual 1, 2-diacylphos phoglycerides or phospholipids are found among different strains of organisms based on ecological factor s. Phosphatidylcholine (PC) is a bilayerforming lipid found in a majority of eukar yotic and prokaryotic membranes, but not archaebacteria.67 Biological membranes have both bilayer and nonbilayer-forming lipid components combined together in different ratios which ensure optimal function of cells in which they are localized. Under physiol ogical conditions, the nonbilayer-forming lipid components are represented by phosphatidylethanolamine (PE) and/or monogalactosyl or monoglucosyl diacylglycerol. The la tter neutral lipids are usually found in chloroplasts. The anionic lipids phosphatidylserine (PS), phosphatidylglycerol (PG), cardiolipin (CL) and phosphatidic acids (PA) confer negative charge to membranes. CL and PA can form nonbilayer structures in the presence of special divalent cations. The strength of interactions of glycerophosp holipids with cholesterol increases in the following order: phosphatidylet hanolamine, phosphatidylseri ne and phosphatidylcholine respectively.68

PAGE 33

33 Table 1-3. Major Phosphoglycerides and their head groups. Phosphoglyceride Formula Headgroup R R Cholesterol Cholesterol is a type of membrane lipid that belongs to the class of compounds known as steroids, which includes saponins and bile acids. T he common feature of steroids is the cyclopentanophenanthr ene fundamental ring system or the Choline Ethanolamine Inositol Serine Glycerol Phosphatidylglycerol Diphosphatidylglycerol (DPG) Phosphatidylglycerol (PG) Phosphatidylserine (PS) Phosphatidylinositol (PI) Phosphatidylet hanolamine (PE) Phosphatidylcholine (PC)

PAGE 34

34 perhydrogenated form of this ring structure.69 Cholesterol is the most common sterol and is found in mammalian plasma membranes comprising about 30% of the total lipid mass. It is an unsaturated alcohol with the cholestane structur e consisting of an acyclic 4-ring system with three fused six-member ed rings and one five-membered ring. The ring fusions in cholesterol are all trans conferring a compact and relatively flat molecular arrangement with the stereochemistry of each of it s atoms or groups defined with respect to the plane of the ring syst em as illustrated in Figure 1-7. Figure 1-7. The molecular structure of cholesterol. Cholesterol modulates membrane struct ure, dynamics and function in cells through sterol-protein interactions and also plays the role of altering the lateral distribution of components in the cell memb rane. Cholesterol is quite rigid in membranes, exhibiting only axial diffusion about an axis per pendicular to the membrane surface therefore causing a condensing effect resulting in ordering of membrane lipids. Another major role played by cholesterol is the inhibition of permeation of small molecules through membranes. Normally, a disordering of the lipid bilayer causes defects in the bilayer structure which would allow small molecules to occupy small voids in the membrane and permeate across the bila yer. Bilayer ordering by cholesterol reduces incidences of these defects ther efore reduces the membrane permeability.

PAGE 35

35 Biomimetic Membrane Systems Lipids adopt a bilayer assembly struct ure in biological membranes and this structure provides a barrier that defines the structural components within the cell and also separates the cell interior from the exte rnal environment. The complexity of natural membrane systems as regards the delicate ba lance of lipids and different types of integral and peripheral membrane proteins ma kes the in situ study of any single cell membrane component a major challenge. He nce there is need to develop simpler membrane mimics that can be used to study cellular structures in isolation. Vesicles The simplest model membrane system is that of lipid bi layers which form when amphiphilic lipid molecules dissolved in aqueous solution spontaneously aggregate to form multilamellar vesicles (MLVs) in a pr ocess partly driven by the hydrophobic effect.70 The closure of bilayers to form vesicles is related to curvature induced by the constituent lipids, and is dependent on the size of the hydrophilic head groups relative to the hydrocarbon backbone.66 Multilamellar vesicles can be mechanically disrupted through sonication, or forced through polycarbonate membrane pores in a process of extrusion to form unilamellar vesicles which vary in size ranging from few nanometers to tens of microns as summarized in Table 1-4. Table 1-4. Classification of liposomes by size. Liposomal structure Small Unilamellar Vesicles (SUV) Large Unilamellar Vesicles (LUV) Very Large Unilamellar Vesicles (VLUV) Giant Unilamellar Vesicles (GUV) Size Range 4 50 nm 50 500 nm 500 5000 nm > 5000 nm Small unilamellar vesicles (SUVs) have le ss symmetry in lipid distribution between the inner and outer bilayer leaflets, and have high surface curvatur e therefore are less

PAGE 36

36 stable thermodynamically and tend to fuse to form larger vesicles.71 As a result, large unilamellar vesicles are more favored as mo del membranes to avoid these lipid packing problems associated with SUVs and because of the significant LUV interior volume. The traditional method involving use of vesicular lipids has been as free suspensions in solution, but in recent years, several groups have developed systems whereby vesicles are tethered to supports using a number of recognition elements such as biotin-streptavidin coupling, or tet hering by DNA hybridiz ation (Figure 1-8).72 Figure 1-8. Vesicular model membranes. A) Vesicles tethered to solid supports. B) Free vesicle suspensions in solution. Supported Lipid Bilayers The formation of lipid bilayers supported on substrates was physically characterized for the first time by Ta mm and McConnell when they deposited lipids on glass, quartz and oxidized single-crystal silicon wafers via Langmuir-Blodgett transfer.73 The McConnell group demonstrated the lateral diffusion of lipid molecules in such supported membranes as being identical to diffusion in conventional multilayer systems as facilitated by the 1 2 nm thick wa ter-filled space between the bilayer and the hydrophilic substrate. Later it was shown that SUVs deposited on glass cover slips A B

PAGE 37

37 could spontaneously fuse to form supported bilayer membranes.74 Subsequent work on supported lipid bilayers explai ned that fusion of phospholipid vesicles to substrates is dependent upon buffer pH, ionic strength as well as hydration and steric forces.75 Electrostatic attractions and van der Waals forces had for a long time also been known to play a significant role in substrate bilayer interactions.76 Since the initial studies, planar lipid syst ems on solid supports proved valuable as suitable mimics to the natural lipid environment for immobi lization of proteins for functional studies.77 A variety of substrates are curr ently used with supported BLMs and include glass or silica substrates,78-81 and unfunctionalized metal surfaces.82-84 Surface sensitive characterization techniques such as ellipsometry, surface plasmon resonance spectroscopy (SPR), quartz crystal micr obalance with dissipation (QCM-D), atomic force microscopy (AFM) and ma ny others, are often used to characterize SLBs as model membranes.85-87 Studies done by the combinat ion of QCM-D and AFM have revealed mechanisms through which the self-o rganization of lipids occurs during the formation of the SLB, a process which has been determined to involve adhesion and rupture of vesicles on the support followed by the evolution of the supported bilayer patches which aggregate to form a continuous bilayer.88 The rupture of vesicles that have adhered to the surface for the formation of SLBs depends on a threshold surface density of vesi cles that has to be overcome, and is further enhanced by proximit y to other adhered vesicles.89 Upon vesicle rupture, the resultant bilayer patch exposes an edge which due to the hydrophobicity of the hydrocarbon acyl chains is energetically unfavorable, therefore interacts with neighboring adhered or vesicles in solution depending on overall dens ity of vesicles.

PAGE 38

38 Eventually, a cascade of rupture events leads to extended bilayer patches that are continuous over the whole surface.90 A characteristic of supported lipid bilayers is the proximity between the membrane and substrate. This close contact is associat ed with loss of function and lateral mobility of transmembrane proteins inco rporated in SLBs; therefore modifications were made to increase the space beneath the membrane by in tegration of polymer supports into the SLB.91 Other modifications to the original SL B include efforts to mimic the biological membrane as closely as possible by the Tamm group through use of Langmuir transfers and vesicle fusion to create hybrid polyme r-cushioned bilayers with an asymmetry of lipid compositions between the two leaflets.92 The supported lipid bilayer membranes that can currently be assembled are illustrated in Figure 1-9. Figure 1-9. Immobilization methods for supported membranes. A) Membrane adsorbed on support with a separating thin water film. B) Formed on monolayer of alkanethiols or alkanesilanes on Au or Si /SiO2 substrates. C) Supported on a polymer matrix containing hydrophobic chains. D) Tethered through anchoring molecules. Despite the advances made in the developm ent of supported bilayers suitable for the integration of membrane proteins for f unctional studies, SLBs still lack a well-defined ionic reservoir on both sides of the membrane.93 The limited volume below the SLB is a A B C D

PAGE 39

39 deterrent to reconstitution of proteins with bulky extramembranous domains due to concerns about protein denaturat ion and sub-optimal function as a result of hydrophobic mismatch in the proteins t herefore limiting the possible appl ications for such systems. Additionally, the exposed regions of tr ansmembrane proteins may interact with substrates upon which the bilayer is support ed with a pinning effect thereby inhibiting lateral mobility and proper function of proteins.94 Attempts to address the problem of the limited volume below the membrane using po lymer supports yielded patchy bilayers with defects which were unsuitable for reco nstitution of integral membrane proteins.77 Tethered Bilayer Lipid Membranes Generally, the tBLM is characterized by a pr oximal layer divided into three parts to include the bilayer forming lipids, a s pacer group and an anchorin g group (Figure 1-10). Figure 1-10. The tethered bilayer lipid membrane (tBLM). Shown are the spacer, linking unit, bilayer forming lipids and the anchor group by which the membrane is tethered onto a solid substrate. Anchor Spacer Linking Unit Inner Leaflet Outer Leaflet

PAGE 40

40 Tethered bilayer lipid membranes (tBLMs) ar e emerging to be useful structural and functional mimics of biologic al membranes offering greater chemical, electrical and mechanical stability than ot her model membrane systems.93 To date, gold and silica are the most commonly used substrates for the fo rmation of tBLMs each of which is used depending on the desired applications.93, 95 Applications for tBLMs so far is in drug discovery, basic research towards the better understanding of membrane dynamics and protein-lipid interactions or biosensing.96 Gold substrates can be used as electrodes and therefore allow for electrical characterizati on of the system as well as for electrical measurements of ion channels incorporated within tBLMs .97-99 The planar configuration of tBLMs allows for application of a number of powerful surface or intersurface-sensitive techniques similar to those used in SLB studies for the characteriza tion of structural and dynamic organization, as well as order in these systems.100 The spacer group on the tBLM creates an ionic reservoir below the membrane therefore allows the lateral m obility critical for optimal function of incorporated integral proteins.95 The quality of the ionic reservoir coul d be improved by increasing the volume of the reservoir by increasing the length of the hydrophilic regi on of the tethering molecules or by laterally spac ing the reservoir forming lipids.93 Anchor groups form covalent attachments by gold-sulf ur interactions on the substrate7, 101, 102 or by the bonds formed by lipid-polymer conjugates on silica substrates.95 The first step of formation of the tBLM ty pically involves self-assembly of thiols, disulfides, thiophenols and other organosulfur monolayers which serve as the tethering moiety onto the substrates.103 After the disposition of the self-assembled monolayer (SAM), different lipids can be applied to comp lete the formation of the tBLM and this is

PAGE 41

41 done either by vesicle fusion of SUVs to form tethered bilayers on the SAM or by a layer by layer transfer of lipid monolayers through Langmuir-Blodgett or Langmuir Schfer techniques.95 For the work documented in this dissertation, tBLMs were formed on a gold substrate using a SAM covalently attac hed by gold-sulfur interactions. The SAM is based on the archaea analogue 2, 3-di-O-phytanyl-sn-glyce rol-1-tetraethylene glycolD,L-lipoic acid ester lipid (DPTL) which bears two phytanoyl chains bound to a spacer through a chiral glycerol unit, illu strated in Figure 1-11. Figure 1-11. Structure of t he 2,3-di-O-phytanyl-sn-glycerol-1-tetraethylene glycol-D,Llipoic acid ester lipid (DPTL) Phytanoyl lipids associated with archaea bacteria and other extremophiles are then used to prepare liposomes for fusion to the SAMs for the fo rmation of tethered bilayer membranes. These lipids are characteri zed by their acyl chains composed of the phytanoyl functionality which consis ts of a highly branched 3,7,11,15tetramethylhexadecyl group that confers th ermal stability and fluidity to membranes.104 Phytanoyl lipids have also been demonstrated to have electrical stability that makes them more suitable for use in the incor poration of ion channels and other pore-forming proteins during recordings of electrophysiol ogical measurements. Two phytanoyl lipids were used for the preparation of vesi cles for this study and these are 1,2-diphytanoylsn-glycero-3-phosphoethanolamine (DPhPE ) and 1,2-diphytanoyl-sn-glycero-3phosphocholine (DPhPC) illustrated in Figure 112 and used in various ratios that would provide optimal electrical and mechanical stability

PAGE 42

42 O O O O O P O N+ H O O H H H O O O O O P O N+ H O O Figure 1-12. Structures of 1,2-diphytanoyl-sn-glycero -3-phosphoethanolam ine (DPhPE) and 1,2-diphytanoyl-sn-glycero-3 -phosphocholine (DPhPC) lipids Electrical Properties of Membranes Electrochemical forces on cell membranes are involved in facilitation of the regulation of chemical exchanges within cells and the surrounding environment. Membranes are semipermeable and therefore allow ionic interchange across the cell. The predominant ions withi n and around cells include Na+, K+, Ca2+ and Cland are found in markedly different extracellular and intracellular concentrations. Voltage and concentration gradients drive ions across memb ranes with a net flux from regions of higher concentration to lower concentrations.1 The potential difference (volt age) is a measure of how mu ch electrical work is required to move a charge in a frictionless manner from one poi nt to another. The equilibrium potential is the memb rane potential at which there is no net flux of ions from one side of the membrane to t he other. The equilibrium potentia l for a single ion species across a membrane separating two regions of unequal concentration is described by the Nernst potential (Eeq): Eeq = (RT/zF)ln{C(out)/C(in)} = 2.303 (RT/zF)log10{C(out)/C(in)} in volts (1-1) DPhPE DPhPC

PAGE 43

43 Where R is the gas constant (8.314 V C K-1 mol-1), T is the absolute te mperature, z the charge of the ion, F is Faradays constant (9.648 104 C mol-1), and C(out) and C(in) are the ionic concentrations outside and inside the cell respectively. Assuming room temperature is 20 C, 2.303(RT/zF) = 58 mV for a univalent ion.1 For a system with more than one permeable i onic species, the equilibrium voltage will depend on the concentration and relative pe rmeability of the individual ions. The Goldman-Hodgkin-Katz (GHK) equations define absolute ion permeabilities in terms of flux measurements, and ion per meability ratios in terms of zero-current potential measurements without the cont ributions of ion channels.105 The GHK equation therefore can define resting potentials and in a cell with K+, Na+ and Clas the permeant ions, the equation would be: e CliKi N i CleKe NClPKPNaaP ClPKPNaaP F RT V ][][][ ][][][ lnrest (1-2) Where Erev is the resting potential, [K+]e the outside K+ concentration, [Na+]e the outside Na+ concentration, [Cl-]e the outside Clconcentration. In the case of only one permeant ion, Erev becomes the Nernst potential for that ion. Conductance (G) and resistance (R) are used to quantify the ease or di fficulty of current (I) flow between two points and they are defined by the Ohms law: I = GV or I = (1/R) V (1-3) Conductance varies with the salt ionic concentration and is affected by the viscosity of solutions on either side of the membrane. T he bilayer lipid membrane acts as a barrier to the flow of ions and ther efore increases the resistance of the system. Two conducting electrodes can be used to measure the conduc tance of a system if each is positioned on either side of the membr ane. Because the membrane is a dielectric insulator

PAGE 44

44 between two conductors, charge can build up at either interface. Capacitance is a measure of how much charge separation is r equired to create a given voltage difference and it is measured in farads and can be defined by the following equation: C=Q/E (1-4) Where Q is the charge transferred, E is the potential difference and C is the capacitance. The rate of change of potentia l due to current flow, Im, at the membrane is given by: dE/dT = Im/C (1-5) Capacitance depends on the dist ance between the two charged surfaces (d) and the dielectric constant ( ) of the membrane in between. The di electric constant is a measure of the polarizability of the material and t he degree to which any permanent electric dipoles which may be present in the materi al respond to the voltage difference. The capacitance for a bilayer is given by the equation: C = 0A/d (1-6) Where is the dielectric constant and 0 is a constant of the permittivity of free space with the value 8.85 x 10-12 coulomb V-1m-1. The specific capacitance is the capacitance per unit area which depends on the charge s eparation per unit ar ea of the membrane and for phospholipids and biomembranes has similar values of around 1 farad/cm2, which corresponds to a dielectric cons tant of 2 and distance of about 25 .106-108 The general effect of membrane capacitance is to slow down the response to any current by a time, that depends on the resistance and capacitance of the membrane. The discharge rate of the membrane capacit or can be expressed by the equation: dE/dt = Im/C = -E/RC = -E/ (1-7)

PAGE 45

45 When this equation is solved, it gives an expression for the potential across the membrane as a function of time, t. If the initial potentia l is given as E0, the relationship is expressed as: E = E0 exp (-t/RC) = E0 exp (-t/ ) (1-8) Due to the great variation in time const ants for different biological membranes the resting membrane resistances can vary broadly from 10 to 106 ohms.109 The equivalent circuit in Figure 1-13 can be used to represen t the electrical properties of the bilayer membrane and how they are related to each other. EiCmgi Figure 1-13. Equivalent circuit diagram of a membrane. The electrical properties of a membrane can be modeled using the conductance, gi of ions, crossing the membrane with a membrane potential, Ei in series. These are in parallel with the capacitance, Cm. Subsequent to the formation of a solid supported lipid bila yer or a tethered bilayer lipid, membrane capacity does not vary signifi cantly since the bilayer determines its value. On the other hand, membrane resist ance may vary over several orders of magnitudes in response to the lipid compos ition, preparative fact ors and experimental

PAGE 46

46 parameters indicating whether a memb rane is electrically dense or not.66 Electrostatic interactions between the lipid bilayer and the surface of the supporti ng substrate as well as the membrane compressibility are key parameters determini ng formation of electrically dense supported or tethered bilayer lipid membranes.110 Stochastic Sensing Stochastic sensing is a technique which re lies on single-molecule detection and is usually based on transmembrane pores devel oped as sensing elements with binding sites for analytes. The first known exam ple of a single-molecule experiment on a functional biomolecule was the observation of current flow through a single ionconducting channel formed by the peptide gr amicidin in a planar lipid bilayer.111 Singlechannel currents passing through a transmemb rane pore can be modulated by channel blockers or antagonists reversibly bound within the pore as illustrated in Figure 1-14. Figure 1-14. Schematic of stochastic sensin g by an engineered pore in a planar bilayer. Stochastic sensing monitors individual binding events and can reveal the concentration of analytes and based on struct ure specific current signatures, for instance the mean duration and am plitude of events, the ident ity of the analyte can be

PAGE 47

47 deduced.112 Quantitative kinetic information can be obtained about the interaction of an analyte to its binding site. In a simple case, the sensor elem ent has two states; the state where the binding site is occupied by t he analyte and another where the binding site is unoccupied, each giving a different output.113 In a simple equilibrium, off = 1/ Koff where off is the mean dwell time of the analyte and Koff is the dissociation rate constant and on = 1/Kon[A] where on is the mean time between binding events Kon is the association rate constant and [A] is the analyte concentration.113 If the support onto which the bila yer membrane is tethered is a conducting substrate, for ex ample, the work documented in Chapter 5 in this dissertation, where the bilayer lipid membra ne is tethered to a gold substrate, an alternating potential can be applied to allow impedance meas urements. Advances have been made to engineer channel-forming peptides and proteins to detect a myriad of chemical analytes with pot ential applications in high throughput pharmaceutical screening, detection of heavy-metal cont aminants in water and many others.114 Chapter 5 of this dissertation documents single-channel recordings representing functional activity of the BKCa channel in a bilayer membrane tethered on gold, as well as pharmacological aspects of bound tetraethylammonium compounds.

PAGE 48

48 CHAPTER 2 EXPERIMENTAL PROCEDURES AND TECHNIQUES Introduction Chimeric BKCa channels with a red fluorescent protein (mRFP1) tag were heterologously expressed in Xenopus laevis oocyte membranes. The approach used for expression involved in vitro transcription followed by micr oinjection of complementary RNA into oocytes.115 Translation of RNA occurred followed by trafficking of functional channels to the oocyte membrane. The su ccess of expression was quantified by twoelectrode voltage clamping of oocytes. Membrane extracts containing channels were detergent solubilized and these were then purified by immobilized metal ion affinity chromatography. Purification was determined by SDS-PAGE and the identity of the protein verified by immunoblotting. Protein reconstitution into liposomes was mediated by dialysis and the followed negative-staining transmission electron microscopy imaging to visualize the formed proteoliposomes. It was important to demonstr ate functional activity of reconstituted channels, and this was done in a tethered bilayer lipid membrane (tBLM) by use of a modified patch clamp electrophysiology tec hnique. The insertion process of channels into the membrane was analyzed by surface plasmon resonance-enhanced ellipsometry. Quartz crystal microbalance with dissipation (QCM-D) and atomic force microscopy (AFM) were used to characteri ze the formation of model membranes by phytanoyl lipids on solid substrates. Cryo-tra nsmission electron microscopy imaging of lipid dispersions was performed at the Cent er of Chemistry and C hemical Engineering, Lund University, Lund, Sweden. Cryo-TEM prov ided us with images which allowed for the determination of the structural morphol ogy of vesicle mesophases and analysis of

PAGE 49

49 vesicle sizes. For a further characterization of the sizes of vesicles we used dynamic light scattering (DLS) and NMR diffusion. Briefl y described below is the methodology of the experimental work as well as the basic principles of the techniques used for the characterization studies. Techniques and Methods Immobilized Metal Ion Affini ty Chromatography (IMAC) Theory IMAC is a protein purification technolog y that makes use of specific binding relationships between the side chains of certai n amino acids, in most cases, histidine and Lewis metal ions such as Cu2+, Ni2+ and Zn2+. A chelating agent attached to a stationary support combines with metal ions to form a metal ion complex referred to as the immobilized metal chelate complex (IMCC ) thereby capturing of the metal ion. The earliest applications of IMAC made use of surface histidines that occur naturally in the target protein, however, advances to t he technique allowed t he inclusion of a hexahistidine tail or tag on the target protein thereby allo wing for more specific and stronger binding affinities for use in more selective purification.116-118 Because IMAC relies on affinity interactions between amino acids and metal ions rather than biological func tion, it can function under denaturing conditions and has become a popular method for the purification of proteins ex pressed in inclusion bodies. Systems have been developed which bear capabilit y to specifically detect histidinetagged proteins by relying on anti-hexahist idine antibody for secondary antibody reporter enzyme conjugate detection, or a reporter enzyme (horseradish peroxidase and alkaline phosphatase) linked to a chelat or for direct metal chelate complex

PAGE 50

50 detection. Samples can then be queried by either method for the presence of the histidine tag in an immunoblotting assay format. Experimental settings The instrument used for automated purif ication was the KTA prime plus (GE Healthcare Life Sciences) using Unicor n PrimeView software and equipped with a UV and conductivity monitor. Supernatants obt ained after solubilization of membrane extracts are loaded onto the Histrap FF (GE Healthcare Life Sciences, PA, USA) columns pre-packed with Ni2+ Sepharose media made from highly cross-linked agarose matrix. Elution is performed through colu mns equilibrated with binding buffer (10 mM octyl glucoside, 20 mM Tris buffer, 500 mM KCl and 20 mM imidazole, adjusted to pH 7.5) at a flow rate of 1 mL min 1 and washed with the same buffer. Unbound material is washed off using 5 column volumes of bindi ng buffer, and then the samples eluted with an elution buffer containing 500 mM imidazol e and all other components of the binding buffer at pH 7.4 and a fl ow rate of 1 mL min 1. Patch Clamp Electrophysiology Theory The patch clamp electrophysiological technique allows the recording of macroscopic whole-cell or microscopic si ngle-channel currents flow ing across biological membranes through ion channels. The technique allows the experimental control and manipulation of the voltage of membrane patches or the whole cell, thus allowing the study of voltage dependence of ion channels and recordings of very low electrical currents amounting to a picoampere (10 -12 A).119 The development of the gigas eal led to the establishment of various recording configurations such as the cell attached, inside-out, outside-out or whole-cell and these

PAGE 51

51 have allowed the possibility of making patch re cordings from the cell surface or cell-free membrane patches as well as intracellular re cordings. However, the basis of all patch clamp recordings began with the cell atta ched method. In this method, a glass micropipette tip containing the referenc e electrode and electrolyte solution is micromanipulated into the cells surface, isolat ing a patch of the membrane. Suction is next applied to create a seal which indicates that a significant amount of electrical resistance was obtained between the patch c ontaining the ion channels and the tip. This electrical resistance is needed to measur e ion channel currents free of background noise. The current flow is reflective of all the ion channels within this patch and measured with a specific patch clamp amplifier. The amplif ied current transitions are then recorded and analyzed with a data acquisition program. Each channel exhibits a unique amplitude threshold, lifetime profile, and conductance parameters which relates to the single channel activity of the ion channel. In recent years, variations of the patch-clamp technique have been developed making this a powerful technique capable of providing answers to questions that previously could not be addre ssed. Information that can be obtained from patch clamp recordings include the conductance, vo ltage dependence, selectiv ity, open probability, dwell times, resistance of the membrane, and pharmacolo gical profile of the ion channels.120 Two-electrode voltage clamping : Two-electrode voltage clamping is a high throughput measurement configuration wh ich allows measurement of current through channels in cells using a pair of electrodes while cont rolling the membrane potential. The voltage electrode measures membrane voltage ( Vm) while clamping the membrane potential at

PAGE 52

52 a desired level ( Vh) by injecting current via a second electrode. Any change in membrane current caused by opening or clos ure of ion channels is immediately followed by a change in the current equal in amplitude but opposite in sign at the output of the amplifier connected to the current electrode. The exper imental set-up is illustrated in Figure 2-1. Experimental chamber oocyte Virtual Ground Circuit negative feedback amplifier voltage current electrodes ND96 agarose/KCl bridge 1 agarose/KCl bridge 2 3 M KCl 3 M KCl Vcommand Voltage monitoring amplifier Figure 2-1. Principal schem e of a typical TEVC recording arrangement. Bold lines denote Ag/AgCl electrodes. The oocyte is placed in the middle of an experimental chamber filled with ND96 recording solution. The amplifier output is monitored on a computer. Because it makes use of two electrodes, TEVC is especially useful for th e recording of measurements in cells with large dimensions such as oocytes. We used TEVC for the analysis of the success of expression of the BKCa channels. Oocytes that were in jected with complementary RNA encoding for BKCa channels were incubated for three days after which the oocytes were analyzed for the activity of ion channels. Oo cytes were placed in a 1 ml plexiglass experimental chamber, impaled with two 0.3 to 2 megohm electrodes and the

PAGE 53

53 membrane potential clamped at -90 mV using a Warner OC725C oocyte clamp. The clamping was done after a brief duration of 0.5 min to 2 minutes to allow for a partial recovery of the oocyte from the injury produced by the electrodes. Data was collected and analyzed using pClamp 8.0 software. Curr ent-voltage relationships were obtained by stepping to voltage clamp to -70 mV and then increasing in 10 mV steps of 100ms duration to +120 mV. The maximum curr ent at each voltage was recorded. Specialized patch clamping on tethered bilayers : In recent years biomimetic membrane systems have been developed ba sed on supported bilayers on solid substrates. Chapter 1 of this dissertation discusses the diffe rent types of systems that are currently in use. The studies documented here involve use of the tethered bilayer lipid membrane (tBLM), a system composed of a gold substrate with two interconnected pads; the sensor pad within which is the tBLM with incorporated ion channels, and a probe pad which serves as the ground elec trode. In this modified patch clamp technique, a thiol-modified monolayer is coval ently attached to a gold substrate by self assembly, forming the leaflet of the membrane proximal to the substrate, and the outer leaflet is deposited via vesicle fusion on top of the monolayer to form a complete hybrid bilayer. Alternatively, instead of fusing vesicles of different lipid mixtures, the bilayer can be formed by direct fusion of proteoliposomes containing ion channels or other poreforming peptides of interest, thereby achievin g both bilayer membrane formation as well as ion channel incorporation in one process. A buffer or water droplet is added onto the sensor pad in order to maintain hydration of the tethered bilayer and ion channel. A conventional patch microelectrode filled with t he same buffer solution is then dipped into the bulk solution on the sensor pad by mi cromanipulation and a potential of 3 mV

PAGE 54

54 pulsed to measure the electrical resistance of the tBLM. It is imperat ive that resistances in the giga-ohm range be observed before any electrical measurements can be recorded using this system to ensure that any currents passing across the membrane be those traversing through the ion channel duri ng ion translocation rather than through membrane defects. Experimental Settings The experimental setup for the specialized patch clamping of tethered bilayers consisted of a patch-clamp amplifier (A xopatch 200B, Molecular Devices), data acquisition and digitizer (Digidata 1322A, Mole cular Devices), a mi cromanipulator, two electrodes, a data recording system, an anti-vibration table, and a Faraday cage with a dust cover. A noise and dust free envir onment for the recording systems was established with an anti-vibrat ion or vibration isolation table surrounded by a Faraday cage. A piezoelectric anti-vibr ation table provided the stage for patch clamp recording. To reduce the amount of electrical noise a copper mesh was placed around the stage thereby shielding the set-up from all external electrical si gnals and electrical noise. The seal resistance between both the reference and ground electrode must be at least 1 G in order to observe single ion channel acti vity and ideally create an electrical stable bilayer.69 A high resistance enables the complete electrical isolation of the membrane patch, and reduces the current noise. Experiments were not done if the resistance was not in the giga-ohm range. T he discrepancy of the current noise is related to Johnson voltage noise Si: Si 2 = 4kTfc/R (2-1) Where k is the Boltzmanns constant, T is temperature (in Kelvins), fc is the low pass filter (Hz), and R is the resi stance. The Johnson noise for a 1 G resistance at 298

PAGE 55

55 K, with a 1 kHz filter is 0.1 pA, and a 10 G resistance results in current noise of 0.04 pA. Now if the resistance is 100 M under the same conditions as the first seal, the noise becomes 0.4 pA. Therefore the higher the seal resistance, the less noise observed. This resistance in electroph ysiology is termed the gigaseal. A strong gigaseal depends on the pipette mate rial used, the size of the tip, and a sample area free of microscopic debris. Both soft (thin wall CEE BEE glass) and hard glass pipettes may be used for patch clamp measurements. However, hard glass like borosilicate can establish greater resistance be cause it has thicker walls. The diameter of the tip depends on the pulling tec hnique and can range form 0.5 to 5 m.70 In the tethered bilayer system, the resistance meas ured depends on the complete coverage of the sensor pad by the membr ane thereby creating a barrier between the two electrodes. Software used during experiments included Ax on Instruments Clam pex 8.0 or 9.0 for data collection and Clampfit 9. 0 for data analysis. Episodi c and gap-free were the two main software protocols involved in characterizing the bilayer and the single channel activity. The Episodic protocol was used to determine the conductance of the bilayer. During these experiments autom ated potentials from negative 200 mV to positive 200 mV were applied to the system for several seconds. From these studies, we determined the stability of the bilaye r system. The gap free protocol gave us a chance to observe the ion channels activity over a long period of time with different potentials. With the gap free protocol, we controlled the change in potential manually. Therefore a number of experiments were conducted using the gap-free protocol. The recording filter was 5 kHz low pass 8-pole Bessel.

PAGE 56

56 Data analysis was done using the event detection for single channel statistics within the Clampfit program. From the event detection, we obtai ned statistics on the channels average current during open and clos ed states, dwell time (average time spent at one level during the duration), and current transitions. We also determined the probability to which the channel was in an open and closed state. In order to conduct this analysis, signals were first filtered at 1 kHz by an 8-pole lowpass Bessel filter to reduce the high frequency noise during record ing. Once the signal was filtered, the baseline was manually adjusted to zero, fo r the closed state and event detection was performed. Event detection : Event detection for single channel analysis goes through a single trace and counts the number of events at di fferent amplitude levels. An event is a continuous section of the trac e at one level. In the event viewer, we manually set the different levels that are detected. Level ze ro accounts for when the channel is closed or not conducting and level one, or two correspond to the open states of the channel. This opening is called a threshold. Each change in amplitude to a different level must meet a maximum threshold in order to be count ed as an event. The maximum amplitude Amax/A0, threshold depends on the error function erf, the frequency of th e filter recording system, and the duration of the event w. Amax/A0 = erf (2.668 fcw) (2-2) Because this value varies, we set a 10% level contribution within the software in which the changes in amplitude contributing to a new event have to be within 10% of that levels amplitude. For instance, if the existing open level is 5 pA and a new event begins at 5.5 pA, the 10% contribution setti ng will reduce the transition to 5.05 pA.

PAGE 57

57 (0.9 x 5.0 pA) + (0.1 x 5.5) = 5.05pA (2-3) Rapid deviations of the trace that are not within the range of the current level were ignored. Due to the possibility of the channel being open or closed upon entering the event detection, the first and last events were also excluded from the statistics. Once a count of all the different ev ents at both the open and closed levels was obtained, single channel analysis of the recombinant BKCa channel was performed. Figure 2-2. Event detection occurring at level zero and level one. Level zero indicates the closed state of the channel or t he non-conducting level and it is adjusted to fit the baseline. Level one is roughly at -3.0-3.3 pA. Events are highlighted in red and accounts for the open state of the channel. Single ion channel activity : Single ion channels exhi bit stochastic activity. Theoretically the durations of events and t he order in which they occur are random variables. The channel has no memory system of opening or closing at a given time, and behaves independently of specific poten tials, and lipid environment. For this reason, parameters represent ative of single channel activity, and the information contained in each event are meas ured from statistical distri butions. There are two main objectives for analyzing single ion channel activity of channels: First, amplitude studies

PAGE 58

58 to understand ion permeation through open channels at differ ent ionic compositions, and different membrane potentials, and secondly, the duration of open and shut times or the dwell time for lifetime and kinetic studies. Point amplitude histograms : The conductance of biological ion channels is restricted by the channels topology including its maximu m pore expansion, leng th and structure in the lipid bilayer membrane. Becaus e of this restriction, there is a specific range in which channels will conduct. For instance, the co nductance of the wild type, full-length BKCa channel is between 90 pS to 200 pS. The la rge range is also due in part to the calculation of conductance and the method used for recording single channel activity. Conductance can be calculated by stati onary fluctuation using voltage clamp measurements on large ensembles of channel s or measurements of unitary current analysis using patch clamp. Variance calculati ons are used for stationary fluctuations to calculate N, the numbe r of the independent and i dentical channels open and contributing to macroscopic current I from single chann el conductance i: I 2 = iI-I2/N (2-4) Where I 2 is the variance from each channel. Patch clamp techniques require a gigaseal which allows the isolation of single ion channels resulting in direct measurement s of unitary current. The conductance is calculated from the slope of current-volt age curves derived from Ohms law. Unitary conductance is calculated from point amp litude histograms at given potentials. All digitized current can be plotted as a histogram with a peak for each closed and open level. The area under the peak is proportional to the time spent at that level as shown in Figure 2-3. Point-amplitude hi stograms for each experiment at a known

PAGE 59

59 potential were plotted from informati on obtained from the single channel event detection. A statistical fr equency count with a minimum bin width of 0.05 mm was performed and fitted by the maximum likelih ood with a continuous Gaussian curve using Origin software as in Figure 2-3. -10-8-6-4-202 0 1000 2000 3000 4000 5000 6000 Count (N)Current (pA) Open= 0.53 Closed=0.48 -10-8-6-4-202 0 1000 2000 3000 4000 5000 6000 Count (N)Current (pA) Open= 0.53 Closed=0.48 Figure 2-3. Histogram fitted by the maximum likelihood with a continuous Gaussian curve representing open and clos ed event distributions. The open Popen /close Pclose probability and the unitary conductance are calculated from this distribution. Both probabi lities were estimated as a fraction of the time the channel remained in the respective state, divided by the overall recording time. Unitary conductance is calculated by dividi ng the average amplitude from the amplitude histograms by the applied potential. Pore form ing proteins must r eproducibly conduct in a specific range to be considered as ion c hannels. Kinetic calculat ions of lifetime are determined by using a probabilit y density function (pdf). Probability density function : As mentioned throughout this chapter, single channel activity exists in at least two states: t he open and closed state. Cons idering the simplest kinetic model of just two states, the transiti on mechanism from one st ate to the other is denoted by:

PAGE 60

60 k12 k 21closed Open State: 1 2 k12 k 21closed Open k12 k 21closed Open State: 1 2 Where k21 is the transition from cl osed to the open state, and k12 is the transition from open to the closed state. This unimo lecular transition reaction involves conformational changes of the ion channel from one state to the other.121 Random thermal movements allow the bonds of the prot ein to vibrate, bend, and stretch in the correct rapid transition on the picosecond scal e. This unsystematic motion leads to the randomness in the lifetime of the open and closed state.122 The probability p that the channel will overcome the energy barrier hol ding the channel open and fail to close is (1-p) p. The probability that it will succ eed in the next attempt r is independent of the first attempt: P(r) = (1-p)r-1p, r = 1, 2, 3, (2-5) The lifetime is therefore a probability distribution and is expressed by a probability density function (pdf) where the area under the curve represent s the probability that the lifetime is less than the specified time t.123 The exponential density of a time interval with mean of = 1/ and equals: f (t) = i 1e-t/ t>0 (2-6) The area under the curve for any pdf must equal unity. Theref ore an ensemble of ion channels will lead to a corresponding number of exponential distributions a and the total area of the i th component, will also have to equal unity when i is mean.123 f (t) = aii -1e-t/ (2-7)

PAGE 61

61 Data obtained from pdf are displayed in multi-component histograms where they are usually characterized by three differ ent time values. These time values are representative of the different open and closed levels exhibited by the ion channel. Dynamic Light Scattering Theory The principle of the dynamic light scattering technique is that t he scattering of light can be viewed as a result of microscopic heterogeneities within the illuminated volume. When a volume of a homogeneous sample is illuminated with a beam of light, the scattered waves will have the same amp litude and interfere destructively in all directions, except in the direction of t he incident beam. If a heterogeneous sample is being analyzed, the index of refraction woul d differ from the av erage value at some location, as a consequence, the wave that is scattered at this location would not be compensated for and some light would be observ ed in directions other than the incident light and light scattering occurs, as shown in Figure 2-4. There are two ways to approach the phenomenon of light scattered by particulate matter in solution. The one is to consider the suspension as a homogeneous medium and ascribe light scattering to the spatial fluct uations of the solute. This is appropriate for solutions of small molecules in which the average distance between the center of the scatterers is small compared to the wavelength of light. The second is to consider each individual solute particle as a heterogeneity and therefore as a source of light scattering. The latter is more appropriate for solutions of large macr omolecules and colloids, where the average distance between particles center s is comparable to the wavelength of light.

PAGE 62

62 Figure 2-4. Possibilities for the interaction of a laser bea m with a liquid sample. In a homogeneous medium, waves of the scatte red light interact destructively, producing no scattering (A); while in a heterogeneous medium scattering is produced (B). In cases in which the size of the sc atterers is not small compared to the wavelength of light, the interference of the electromagnetic waves scattered by the solute is not all constructive and the phases of these waves must be taken into account. If the phase of a wave scattered at the origin is used as a reference, the phase of a wave scattered at a point with radius vector r is q.r. The vector q is called the scattering vector, which is a fundamental char acteristic of any scattering process. The length of the vector is indicated in equation 2-1.124 2 sin 4 qq (2-7) The essence of the DLS technique is to measure the temporal correlations in the fluctuations in the intensity of the scattered light and to rec onstruct from these data the physical characteristics of t he scatterers. In a suspension of particles, the scatterers are randomly distributed within the scattering vo lume. Since the size of the scattering where is the refractive index of the medium is the wavelength of light is the scattering angle

PAGE 63

63 volume is much larger than q -1 (with the exception of nearly forward scattering, where q ~ = 0), the phases of the waves scattered by different particles vary dramatically. As a result, the average amplitude of the scattered light is proportional to N1/2, where N is the number of scatterers, and the average intensity is simp ly N times the intensity scattered by an individual particle. The loca l intensity, however, fluctuates from one point to another around its average value. As the scattering particles move, the interference pattern changes in time resulting in temporal fluctuations in the intensity of light detected at the the observation point.125 In a DLS measurement, the in strument detects a random signal. The information is contained in the correlation function of this signal i(t), which in the case of DLS is the photocurrent, defined as in equation 2-8 )()(2 titiG (2-8) The notation G2( ) is introduced to distinguish the correlation function of the photocurrent from the correlation f unction of the electromagnetic field G1( ) (which is the Fourier transform of the light spectrum): )()()(* 1 tEtEG (2-9) The angular brackets denot e an average over time t This time averaging, an inherent feature of the DLS method, is necessary to extr act information from the random fluctuations in the intens ity of the scattered light. In the majority of practical cases, the scattered light is a sum of waves scattered by many independent particles and therefore displays Gaussi an statistics. This being the case, there is a relation between the intensity correlation function G2( ) and the electromagnetic field correlation function G1( ) :

PAGE 64

64 2 1 2 2)(1()(gItGo (2-10) Here g1( )= G1( )/ G1(0) is the normalized field correlation function, Io is the average intensity of the scattered light and is the intensity factor.125 Temporal fluctuations of the intensity of the scattered light are caused by the Brownian motion of the scatterers. The speed of the particles is related to the size, small particles move faster than large particles. Though each particle moves randomly; in a unit time more particles leave regions of high concentration than region of low concentration. This results in a net flux of particles along the concentration gradient. Brownian motion is then responsible for the diffusion of the solute, and is quantitatively characterize by the diffusion coefficient, D Rigorous mathematical analysis of the process of light scattering by Brownian particles leads to the following expression for the correlation function of the scattered light: )exp(2 1 Dq g (2-11) The diffusion coefficient in an infinitely dilute solution is determined by particle geometry. For spherical particles, the relation between the radius R and its diffusion coefficient D is given by the Stokes-Einstein equation: R Tk Db6 (2-12) Hence, from equation 2-11 the diffusion coefficient can be obtained from the correlation function g1( ) Assuming that the scatterers are spherical, Equation 2-12 can be used to obtain the hydrodynamic radius of the particles. We used the imaging techniques to investigate whether the shape of the particles matched this requirement. where: kb is the Boltzman constant is the viscosity of the solution T is the absolute temperature

PAGE 65

65 In this study, the suspensions obtained where not monodisperse but rather, made of particles a range of sizes. This m ade the analysis of the data obtained more complicated. In the case of polydispersed samples, a different analysis of the correlation function is required. For a continuous distribution of scattering particle size, the correlation function is obtained from the following equation dTDqDI I go 2 1exp 1 (2-13) where I(D)dD = N(D)Io(D)dD is the intensity of light scattered by particles having their diffusion coefficient in the interval [D,D+dD] [N(D)dD] is the number of these particles in the scattering volume, Io(D) is the intensity of light scattered by each of them.126 The main goal of the software is to reconstruct as precisely as possible the distribution function I(D) from the experimenta lly measured function G2 exp( ). The main problem encountered is that dramatically different distributions I(D) lead to nearly identical correlation functions of the scattere d light and therefore ar e equally acceptable fits of the experimental data. There are thr ee typical approaches to solve this problem: i) the direct fit method, in which a functional form of I(D) is assumed a priori and the parameters that lead to the best fit of G2 theor( ) to G2 exp( ) are determined, ii) the method of cumulants. This approach focuses on the stable characteristics of the distribution, which ar e the moments of the distri bution, or closely related quantities called cumulants. The first cumula nt of the distribution that gives the average diffusion coefficient D*, can be determined from the initial slope of the field correlation function, as shown below 2* 2 )1()( 1 ln qDdDDqDI I g d do o (2-14) The second cumulant, the width of the di stribution, is obtained from the curvature of the initial part of the correlation function iii) regularization, a combi nation of the first previous methods mentioned assumes that the distribution I(D) is a smooth function and seeks a non negative

PAGE 66

66 distribution producing the best fit to the exper imental data. This method is used in different approaches used to reconstruct the scattering particle distribution from DLS data. The key point is the selection of the smoothness of t he distribution. If the smoothing is too strong, the distribution will be stabl e but will lack details. If the smoothing is too weak, false spikes will appear in the distribution. The moments of the distribution reconstruct ed by the regularizat ion procedure gives unbiased (apart from smoothing) informati on on the shape of the distribution.126 DLS was used as the primary technique to determine the sizes of the prepared lipid vesicles due to its ease of operation and the small am ount of samples needed. As shown in Figure 2-4, the difference in size between different species can be visualized qualitative by a comparison of the normalized correlation function obtained. Our lipid samples were extruded through polycarbonate membranes with different pore sizes to manipulate the final sizes of vesicles. Ex trusion also served to eliminate dust or aggregated particles that could not be separ ated by sonication. This is especially important due to the dependence of the intensity of the light scattered by a particle to the sixth power of its radius which would result on the signal of the particles being obscured by the aggregates, producing an ov erestimation of their diameter. Experimental settings Throughout this work two instruments were used, a PDDLS/CoolBatch 90T detector with a PD2000DLSplus correlator, m anufactured by Precision Detectors Inc, and a 90 Plus/BI-MAS detector with a BI 9000AT digital correlator manufactured by Brookhaven Inc. In both cases the detector was placed at 90 from the incident beam. Each instrument works with its own softwar e for size determination, the Precision Deconvolve 32 in the first case, and the BI-ISDAW advanced size distribution software, in the second case. The vesicles prepared were sonicated fo r approximately 5 mi nutes and filtered through 0.45 m filters before DLS anal ysis. Sonication was in tended to break any

PAGE 67

67 aggregates that were not hydr ated from the dry lipid film obtained after removal of chloroform from lipids. These aggregates we re especially observed in samples with greater than 50% molar concentrations of DPhPE vesicles. The samples needed to be diluted to avoid multiple scattering. The extent of dilution differed from sample to sample due to the differences in sizes and, as a consequence, on the intensity of light they scattered. As indicated by the manufacturers of the instruments, a to tal intensity of the scattered light between 100 to 300 kilocounts per second (kcps) was used; being the concentration of scatterers adj usted until these values were reached. Subsequently, the sample time was adjusted after a quick exam ination of the correlation function. The sample time is directly related to the size of the particles, so it can be adjusted to obtain the proper fit of the correl ation function. After these setting were adjusted, the measurements were performed in 3 minute runs. This time was observed to be optimal to obtain a good signal to noise ratio, resulting in stable and reproducible results. The final results reported were obtained by the average of three measurements. Negative Staining Transmi ssion Electron Microscopy Theory Transmission electron microscopy is particu larly useful in analyzing particles with diameters similar to the wavelength of the light. The transmission electron microscope uses a beam of electrons instead of light, resulting in better re solution, due to the shorter wavelength of the electrons, compared to photons (Figure 2-5).

PAGE 68

68 Figure 2-5. A simple representation of t he basic concept of a transmission electron microscope operating in the bright field mode. In the electron microscope, electromagnetic lenses are used to focus the beam of electrons into a tight coherent beam, which is then focused on the sample. Data is collected after the beam has passed through the sample. Our imaging has been performed in the bright field mode, meaning t hat after the beam of electrons has passed the sample deposited on the substrate, an obj ective aperture has been inserted. This aperture allows the electrons in the trans mitted field to pass and contribute to the resulting bright field image, rejecting t he electrons that had been scattered by the sample as shown in Figure 2-5. The la ck of electron density with our samples necessitated the use of staining with the dye of choice being uranyl acetate which stains hydrophilic parts of t he vesicle samples. Experimental settings The instrument used in this study was a Hitachi H-7000, with a maximum resolution of 0.2 nm and a maximum a cceleration voltage of 125 kV. Imaging the

PAGE 69

69 vesicles was made more challenging due to the low electron density, as explained above. Moreover, due to their size, especially t he templates, getting a decent focus that can be used to measure the size of the particles was diffi cult. Images of our vesicle samples were taken with a voltage of 100 kV. We used this technique to image our vesicle samples before and after the dialysi s procedure we use for the formation of proteoliposomes. Similarly to what was observed with DLS, the particles needed to be diluted before loading them on the grids. If the concent ration was too high, the particles were deposited as multilayer, making impossible their visualization. It was determined that a concentration of about 0.1 mg/mL was optimal in most of the cases to load the grids. Before insertion of the samples containing the grids in the microscope, the samples were stained with a methanolic solution of ur anyl acetate (UA) for 1 minute. This is a common staining agent for biological samples, used primary to stain hydrophilic regions. Images were taken at different spots of the same grid to confirm that the species observed were present in most of the su spension and that they were not an unusual finding. Size analysis was performed by comparis on of the diameter of the particles and the scale bar in the image. This wa s done in the Adobe photoshop software. Atomic Force Microscopy Theory Atomic force microscopy, since its introduction in 1986, has been used in many different applications, especia lly because of the improved re solution compared to other microscopy techniques. In case of the study documented here, the main reason to use this technique is the ability of AFM to obt ain information on three dimensions without the need of any coatings or stains, an advantage over negative staining TEM. Additionally,

PAGE 70

70 AFM enabled the obtaining of information on the surface morphology before and after vesicle fusion, thereby clearly observing changes in the lipid assemblies as they interacted with solid substrates. The technique is based on the interaction of the tip of a cantile ver with the sample deposited on a substrate. The in strument measures the forces between the tip of the flexible cantilever and the sample. The basic i dea is that the local or attractive forces between the tip and the sample are converte d into a bending, or deflection, of the cantilever. The key feature is that t he force between the probe and the sample is maintained constant while the probe is ra ster scanned across the surface. In order to detect the probe bending, a laser beam is reflected from the back of the cantilever onto a detector, in such a way that a small c hange in the bending angle of the cantilever is converted to a measurable lar ge deflection in the position of the reflected spots. The deflection observed then is converted into an electrical signal, to produce an image of the surface. A simple represent ation of the instrument setup is shown in Figure 2-6. In order to avoid any non desired interacti on between the probe and the particles the instrument was used in the t apping mode. In this mode, the cantilever is oscillated close to its resonance frequency and the tip taps the surface only periodically, reducing significantly the lateral force. This means that the probe is free to oscillate up and down at its resonance frequency as a consequence of the interaction with the substrate when it comes extremely close to it. In tapping mode, the ima ge is obtained by imaging the force of the oscillating contacts of the tip with the sample surface.127, 128

PAGE 71

71 Figure 2-6. A simple representation of the basic setup of the atomic force microscope. Experimental settings The instrument used in this study was a Nanoscope III, manufactured by Digital Instruments, Inc. Imaging was perform ed in tapping mode, using silicon probes (Nanosensors, with dimensions: T=3.8-4.5 m, W= 26-27 m, L= 128 m). The zcalibration was performed with a silicon grati ng (TGZ01, Mikromash), with a step height of 20 nm (accuracy 1 nm). Images were analyzed with the Nanoscope III software provided by the manufacturer. To determine how accurate the z-measurem ents were in the instrument, analysis was done of a silicon grating calibration grid with a step height of 19.66 +/1 nm, as reported by the manufacturer. The instrum ent was used in contact mode. The area was scanned at 2.44 Hz and at 256 samples per li ne. The image was flattened with a first order fit, as suggested by the instru ment manual. Figure 2-7 shows the bearing analysis, in which the software determines the distribution of data points in the Z-axis. The two spikes in the histogram correspond to two elevations on the surface: the bottom and top surface. The red box (Hist rel depth) indicates the distance between the two

PAGE 72

72 cursors on the histogram ( 19.93 nm), which represent the height determined by the instrument. The result show ed that the height was measur ed with an accuracy of 1.4 %, which confirmed that the measurements performed were accurate. Figure 2-7. Data analysis of a Z-axis calibration grid performed in contact mode. The samples were deposited on gold substrates by adding a couple of drops of the suspension and the solvent left to evaporate at room temperature. No other treatment was performed before analysis. The images were obtained with the scanner E, which offers a maximum imaging area of 15 x 15 m. The exact conditions for the different parameters used varied from sample to samp le, although average values or ranges can be reported. To image the samples, first, a fast scan was performed to identify areas were the sample was deposited, after which t he tip was taken to one of this spots to zoom in and acquire an image of proper qual ity. Typically, we used 512 samples per

PAGE 73

73 line, the best quality possible with our inst rument. Next, the amplitude setpoint was decreased slightly to optimize the force use to scan the surface. This was stopped when the trace and retrace profiles in the section analysis looked similar. The scan rate was adjusted based on the dimension of the image, for areas bet ween 2 x 2 to 2.5 x 2.5 m, a typical scan rate between 1-1.5 Hz was used, and this value had to be increased for smaller areas scanned, typically to 2-2.5 Hz This was done to maintain good track of the tip on the z-axis, which a llowed the visualization of t he changes in the height profile. Finally, the integral and proportional gains we re adjusted to improv e the quality of the images, with typical values of 0.35-0.45 and 0.4-0.6, respectively. The section analysis was performed t hen with the images without any posttreatment. The particles used to measure the diamet er and height profile were selected from each image taken after zooming in the area of interest and confirmation that both dimensions were possible to determine wit hout ambiguities. Unless otherwise stated, the height images were used for the charac terization of the types of assemblies obtained when lipid vesicles at different c oncentrations interact with gold substrates. Because lipid bilayer membranes are very th in unless deposited as multi-layers, it was easily discernible whether deposition of vesicl es occurred rather than rapture and fusion to form bilayers. Vesicles being spherical would naturally have a much greater height profile in comparison to bilayer membranes. Experimental Procedures Molecular Biology DNA manipulation The murine BKCa channel ( mSlo ) gene was provided by the laboratory of Lawrence Salkoff, Washington University, St. Louis, MO. The maxi-K gene is contained

PAGE 74

74 in the plasmid: pcDNAOX, a pcDNA3 derivativ e which contains flanking 5and 3-UTR sequence from Xenopus Beta-globin. Site-directed -mutagenesis was performed with the QuickChange II XL kit (S tratagene). The competent E.coli cells used were purchased from Stratagene (XL-10gold), Invitrogen (Top-10, INV-110, BL21 (DE3), or Novagen/EMD Bioscience (BL21 (DE3), Rose tta-2 (DE3). Restriction enzymes were purchased from New Engl and Biolabs (NEB), Calf intest inal alkaline phosphatase was either from Promega or NEB, and T4 DNA-ligase was from NEB. Gel extraction and purification was done with the QIAquick gel extraction kit (Qiagen) from agarose gels. Other media and reagents were from Sigma or Fisher Sc ientific. Sequencing was performed by the fluorescent dideoxy termi nator method at either the Whitney Laboratory for Marine Bioscience, St. Augustine, FL or at the ICBR sequencing core, University of Florida, Gainesville, FL. A ll deoxyoligonucleotide prim ers were synthesized and PAGE purified by Integrated DNA Technologies, Coralvil le, IA. To allow deletion of the maxi-K C-terminal domain and fusion with mRFP1 in pcDNAOX, an insertion coding for a BamHI site flanked by a BstB I site was added in place of the BKCa stop codon by site directed mutagenesis. A second BamHI site was added by site directed mutagenesis in a position such that digest with BamHI followed by gel purification and re-ligation would result in the deletion of the C-terminal domain encoded between the two BamHI sites. As there was no stop c odon added at this point of truncation, an additional 12 amino acid residues (ALRTPRRPELFF347-stop) are coded for at the Cterminus of maxi-K mSlo CTD construct. The monomeric red fluorescent protein gene was provided by the laboratory of Roger Tsien, University of California, San Diego, CA. The mRFP1 gene was contained

PAGE 75

75 in the plasmid pRSET-B. Initially, the mR FP1 gene was cut away from the pRSET-B plasmid with BamHI and BstBI enzymes, th en purified by gel extraction (mRFP1insert, above). XT-PCR was used to modify the mRFP1 gene by addition of a Cterminal extension containing a hexa-histidine tag, a new stop codon, BstBI restriction site, and unique SpeI restriction site. The following primers were utilized for XT-PCR: (extension sequence is underlined) RFPhis-forward: (5TAAG GATCCGATGGCCTCCTCCGAGG) RFPhis-reverse: (5TATTTCGA ACTAGAGATGGTGGTGATGATGTGCGGCGCCGGTGGAGTGGCG). This C-terminally His-tagged mRFP1 gene (mRFP1His) was subsequently fused to the C-terminally deleted BKCa in the pcDNAOX vector resulting in the preparation of the C TD2-mRFP1His construct. CTD2-mRFP1His gene cassettes were anneal ed and ligated to a linearized pRSET-M vector to complete the directional cloning. The ligated vector was transformed into competent cells ( E.coli Top-10, Invitrogen), and grown on selective agar. Plasmid DNA purified from transformed colonies was screened by double restriction digest with an enzyme unique to the His-tag sequence (SpeI) and an enzyme unique to BKCa channels (PmlI). Positive subclones were confirmed by sequencing. RNA synthesis and injection into oocytes Plasmid DNA was transcribed using the Ambion mMessage mMachine T7 Ultra kit. RNA was precipitated with lithium chlo ride, washed and centrifuged in 70% ethanol, and then dissolved in diethylpyr ocarbonate (DEPC) treated wa ter (RNase free). Stage V or VI oocytes were harvested from adult Xenopus laevis (Xenopus Express, Florida) and their follicular layer removed by expos ure to 2mg/ml collagenase for 45min.The transcribed RNA was injected with a Drummond Nanoject microinjector at a volume of

PAGE 76

76 46nl per oocyte (approximat ely 50ng RNA) and the oocytes were incubated at 19 degrees in ND96 media consisting of 96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 10 mM HEPES, 1.8 mM CaCl2, adjusted to pH 7.4 with Na OH, enriched with 2.5% sodium pyruvate, 1% penstrep, and 5% horse seru m. Functional channel expression was determined by use of two-electrode voltage clamping with in 3 days for recombinant BKCa channels. Sodium dodecyl sulfate polyacrylamide gel electrophoresis SDS-PAGE is a technique used to separ ate proteins according to their electrophoretic mobility which is a function of the length of the polypeptide chain or molecular weight as well as higher order protein folding, posttranslational modifications and other factors. SDS is an anionic deterg ent which denatures secondary and non disulfidelinked tertiary structures, and applies a negative charge to each protein in proportion to its mass. Dissolved molecule s in an electric field migrate within polyacrylamide gel at a speed determined by their charge: mass ratio upon the application of an electric field resulting in separation by size. Smaller proteins travel faster through the pores of the gel while larger protei ns move slower. Ladders with known protein molecular weights are used to identify the target pr otein. The molecular weight of the recombinant BKCa channel with a truncated C-terminus and mRFP1 and hexahistidine fusion tags is around 70 kDa. Afte r the purification process, samples were analyzed by SDS-PAGE in a 4-20% gradien t Tris-HCl polyacrylamide gel (Bio-Rad Hercules, CA) in Tris-glycine running buffer. Each well was loaded with 13 l of Laemmli sample buffer, 2.0 l of Dithiothr eitol (DTT) reducing agent and 10 l of the protein sample and the gel was run for 1 hour wit h power supply settings at 100 mV and 100

PAGE 77

77 mA. Staining of the gel was done using S YPRO Ruby Coomassie dye and imaging by a Biorad GelDoc system. Western blotting Western Blot was performed afte r gel fractionation of the BKCa channel to visualize the protein and theref ore confirm that t he histidine-tagged BKCa channel was eluted from chromatography columns. This immunol ogical technique attaches antibodies to the target protein in a given sample. T he proteins fractionated by SDS-PAGE are transferred to a membrane sheet. Primaryantibodies are next added to the membrane sheet to bind the specific proteins follow ed by enzyme-tagged antibod ies that attach to the primary-antibody. Binding a substrate to the enzym e allows visualization of the selected protein by either colorime tric or fluorimetric detection. In this study, BKCa channels were electro-transferred to Immobilon Polyvinylidene Fluoride (PVDF) membranes (Millipore Corpor ation, MA) at 105 mV for 70 minutes. Blots were blocked in 2 % milk/TBS for an hour at room temperat ure. Western Blot analysis was performed using the: primar y-antibodies penta-His-tag antibody or HRPHis-Probe (Thermo Fisher Sic. Pierce Prot ein Research, Rockford, IL) against the BKCa channel C-terminally fused hist idine-tag. The secondary anti body used for the channel was Goat-anti-Rabbit HRP (BioRad). W hen protein expression was too low for recognition with the penta-His -antibody, we used the HRP-His-Probe (HRP-His-Probe was used for the majority of the study). The HRP-His-pr obe uses a Ni-chelating group that binds to the hexa-histid ine tag present on the protein. The HRP-His-probe reacts with the substrate giving off a chemiluminesc ent signal that was q uantified using X-ray sensitive film (Kodak). The film was dev eloped using the dipping method whereby the film was submerged in each of the following for 90 second intervals: developer, water,

PAGE 78

78 and fixer (Developer and Fixer purchased from Kodak). The film was then rinsed under running water for 2 minutes and quantified. Vesicle Formation Using DP hPC and DPhPE Lipids Archaea analogue phosphol ipids 1,2-diphytanoylsn -glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE) in chloroform were purchased from Avanti Pola r Lipids Inc. (Alabaster, AL, USA), and were used without further purification. Chlo roform solutions of the lipids (50 mg/mL) were mixed to a 7:3 molar ra tio (DPhPC/DPhPE) followed by rotary evaporation until a dry lipid film was obtained at the bottom of the vessel. The dry lipid film was then placed under vacuum in a dessicator overnight to e liminate any residual chloroform. The lipids were then lyophilized before hy dration for 1 hour at 50 C in buffer (5 mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4) to a final concentration of 2 mg/mL. After being cooled to room temperature the su spension was sonicated for 5 min, and extruded (21 passes) through a 100 nm polycarbonate membrane in a mini -extruder (Avanti Pola r Lipids Inc.). Reconstitution of Recombinant Proteins in Liposomes A detergent-mediated proc edure was used for reconstitution of BKCa channels into lipid vesicles. A 1 ml suspension of pre-formed liposomes prepared using DPhPC:DPhPE lipids in a 7:3 molar ratio as detailed above were solubilized by the addition of a 1 ml solution of 10 mM -octyl glucoside. A 500 l sample of protein solubilized in 10 mM OG detergent that was eluted fr om the AKTA protein purifier is added to the solubilized liposomes resulting in co-micellization. The final mixture is composed of purified me mbrane proteins in an excess of liposomes and -octyl glucoside detergent, forming a solution of mixed lipid-protein-detergent and lipiddetergent micelles. The next step involves removal of detergent from these micellar

PAGE 79

79 solutions which leads to a progressive forma tion of closed lipid bilayers in which the proteins eventually incorporate. We remove the detergent by injecting the micellar solutions in 10,000 kDa membrane MW cut-o ff dialysis cassettes (Pierce) and these are immersed in 1 liter dialysis buffer (5 mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4) for 2 hours at room temperature, the dialysis buffer is changed and dialysis continued for another 2 hours at room temperature and t hen a final change of dialysis buffer made and this is let to continue for 18 hours at 4 C. All dialysis buffers have pre-cleaned polystyrene beads outside the dialysis bags and the process proceeds with constant stirring.

PAGE 80

80 CHAPTER 3 BKCA CHANNEL EXPRESSION, PURIFICATION AND FUNCTIONAL RECONSTITUTION IN LIPID VESICLES Introduction The demand for high yields of mammalian proteins for routine in vitro applications necessitates the use of heterologous systems, which are non-native systems for protein expression as compared to isolation from native tissue. In 1971 Gurdon and his coworkers established that upon injection wit h complementary RNA (transcribed from complementary DNA coding for the protein of interest) and after a brief incubation period for translation, oocytes extract ed from the South-African clawed frog Xenopus laevis could appropriately express the RNA encoded proteins with high efficiency.115 Since then, Xenopus oocytes have become very popular for electrophysiological analysis of channels, receptors and transporters, although several alternative expression systems for ion c hannels have been discovered over the years, including mammalian cells,129, 130 insect cells e.g. Spodoptera frungiperda (Sf9 cells),24, 131 yeast 132-134 and bacterial cell lines.135, 136 The criteria for selection of the most appropriate expression system are based on factors such as the desired level of protein expression, presence of homologous endogenous receptor s in the host cell, the experimental design in use, among others. Heterologous expression of proteins in oocytes can be achieved by two methods. The most commonly used approach involves in vitro transcription, followed by the microinjection of complementar y RNA coding for the protein of interest into the oocyte cytoplasm where translation occurs. Subsequent ly, docking and functional insertion of the protein into the oocyte plasma membra ne is effected. An alternative approach involves a direct injection of complementary DNA into the oocyte nucleus. The two

PAGE 81

81 methods of introducing genetic material into the oocyte are summarized in Figure 3-1. Insertion of cDNA into the nucleus is le ss favored because of the potential risk of damaging the nuclear membrane during the mi croinjection process. Furthermore, a visual localization of the nucleus is required, making direct injection of cDNA technically challenging and less efficient. cDNA cRNA cDNA cRNAtranscription CYTOPLASM NUCLEUS translation modification sorting membrane insertion Functional Protein microinjection microinjection in vitro transcription capping Figure 3-1. Summary of the approaches used to introduce genetic material into oocytes. In this study, Xenopus oocytes were used for the expression of BKCa channels because oocytes express a low number of endogenous membrane transporters and channels, and they are virtually independent from exogenous nutrients.137 The downstream application for the expressed BKCa channels here involves the incorporation of reconstituted channels in a bilayer lipid membrane that is tethered to a gold substrate. The overall goal of the project is the devel opment of a biosensor based

PAGE 82

82 on the direct interfacing of the sensor element the BKCa channel to the microelectronics of a detector, thereby creating a real-time sensory platform. The gold substrate onto which the bilayer membrane is tethered serves as an electrode which can be used to make electrical measuremen ts. The tethered region of the tBLM is limited in terms of ionic mobility, therefore, in order to optim ize on the ionic reservoir in the tethered region beneath t he tBLM, a genetic modification of the channels to be incorporated is necessi tated. The full-length BKCa channel is tetrameric in structure with each of the four domains having 1200 amino acid residues, majority of which are located intracellularly towards the C-terminu s, forming a bulky structure referred to as the calcium bowl as illustr ated in Figure1-3. For ease of incorporation into the tBLM, C-terminally deleted constructs were prepared, with truncations after the amino acid residue at position 347 located just af ter the S6 transmembrane segment, thus eliminating most of the intracellular domain as illustrated in Figure 3-2. NH3COOSOS1S2S3S4S5S6 S7S8S9 S10 SOS1S2S3S4S5 S6 S7S8S9S10 Point of truncationCalcium bowl NH3 Full Length BKCaChannel Truncated BKCaChannelS0S1S2S3S4S5S6Genetic Modification Figure 3-2. Illustrati on of the full length BKCa channel showing the point of truncation and the truncated product.

PAGE 83

83 Fusion tags constitute a valuable com ponent for the expression of biologically active recombinant proteins and they are us eful for enhancing solubility of proteins, detection, aiding of purification and for many other roles. Two different tags were fused to the BKCa channel, a hexa-histidine tag and the monomeric red fluorescent protein (mRFP1) derived from Discosoma species. The hexa-histidine tag was added to facilitate a simple, one-step affinity purific ation of the ion channels based on the high selective affinity of the hi stidine-tagged protein to the Ni2+ chelated as an affinity ligand in the chromatography columns used. The mRFP1 on the other hand was used as a localization marker, to allo w visualization of channels in oocytes for monitoring expression, and in vesicles to confirm successful reconstitution. The excitation and emission peaks for mRFP1 are 584 nm and 607 nm respectively and these confer spectral separation from autofluoresc ence and other fluorescent proteins.138 Other features that make mRFP1 suitable as a fluorescent fusion protein are its rapid maturation relative to other fluorescent fusion proteins and minimal emission when excited at wavelengths optimal for the mo re commonly used green fluorescent protein (GFP) from the jellyfish Aequorea Victoria; mRFP1 can therefore be used for multicolor labeling in combination with GFP.139 Figure 3-3 shows the chimeric BKCa channel construct expressed for this study with the numbers of amino acid residues as well as positions of fusion tags. Figure 3-3. Illustration depicting the chimeric C-terminally deleted BKCa channel with positions of fusion proteins and numbers of amino acid residues. mRFP1 His 6 -tag 347 aa 225 aa 6 aa C-terminally Deleted BKCa Channel

PAGE 84

84 In general, the purification of membrane pr oteins involves removal from the membrane, using methods such as sonica tion, use of organic solvents, chemical modifications of proteins, extraction usi ng chaotropic reagents, alkaline or Ethylenediaminetetraacetic acid (EDTA) treatment, fr actional digestion, solubilization with acetic acid, treatment with urea or by detergent solubilization.140 The procedure utilized with the greatest success for isolation of acti ve transport proteins from the plasma membrane is detergent solubilization, t hough the conditions needed to solubilize functionally active transport proteins hav e to be found empirically for each transport system.141 Detergents can generally be classified into different groupings based on their charge into nonionic, anionic and cationic detergents and the zwitterionic detergents. Although some detergents have specific inte ractions with membrane proteins, some general principles of detergent solubilization can be outlined. Foremost, detergents bind to membrane proteins and partition between t he lipid bilayer and the aqueous phase to mimic the lipid bilayer of biological membr anes. Incorporation of detergents into the lipid bilayer may perturb the structure and func tion of transport protei ns. The binding of detergents may occur at charged protein residues if ionic de tergents are employed or to hydrophobic protein groups if ionic or nonionic detergents are used.142 After incubation of the membrane with sufficiently high detergent concentrations exceeding the critical micelle concentration (CMC), the biologi cal membranes are lysed and detergentprotein-lipid mixed micelles form. Excessive amounts of detergents are usually used for complete dissolution of the membrane and to pr ovide for a large number of micelles to give one protein per micelle molecule.

PAGE 85

85 A common problem during so lubilization and purificat ion is the tendency of hydrophobic membrane proteins to form complexes after solubilization. Ionic detergents are more efficient in dissociating these co mplexes but in many cases they lead to denaturation and cannot be used when purification is done by ion exchange chromatography and by isoelectric focusing. Nonionic detergents are known to better solubilize membrane proteins while retaining native characteristics in comparison to ionic detergents or bile salts.143 For further characterization of me mbrane proteins and other downstream applications especially those involving incor poration into lipid membranes, it is often necessary to remove unbound detergents. De tergents with high CMC can be readily removed from protein-lipid-det ergent mixed micelles by dialysi s or gel filtration, whereas low CMC detergents dialyze slowly and are o ftentimes insufficiently eliminated. The detergent CMC is therefor e a major consideration to be made when choosing solubilization detergents depending on final use of proteins. In this study, the nonionic detergent octyl glucoside was used fo r solubilization and purification of BKCa channels. Materials and Methods Lipids and Chemicals Archaea analogue phosphol ipids 1,2-diphytanoylsn -glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE) whose structures are illustrated in Figure 1-12, were purchased in an already diluted form in chloroform from Avanti Polar Lipids Inc. (Alabaster, AL, USA), and were used without further purification. 2,3-di-O -phytanoylsn -glycerol-1-tetraethylene glycol-D,L-lipoic acid ester (DPTL) was obtained from the Sch iller group (Max Planck Institute, Mainz Germany). NaCl, MgSO4, Ca(NO3)2, NaHCO3, CaCl2, NaOH, KCl, MgCl2, HEPES,

PAGE 86

86 CaCl2 and sodium pyruvate were purchased fr om Thermo scientific (Waltham, MA). Penicillin, Streptomycin and horse serum we re purchased (Sigma-Aldrich, St. Louis, MO). All buffers were prepared in ultrapure water filtered in a Milli-Q water purification system (Millipore Corporati on, Billerica, MA). Oocytes Adult Xenopus laevis frogs were purchased from Xenopus Express, Tampa, Florida. The surgical removal of oo cytes from the frogs has been previously described.144 Briefly, frogs are anesthetized by pu tting them in a solution of tricaine methanesulfonate (MS222) anesthetic, placed on ice to prolong the anesthetic effect and then a small incision of between 5-10 mm made in lower quarter of abdomen and oocytes can be pulled out in their fascia from the ovarian lobes. Oocytes are placed in a 2 mg/ml collagenase in dival ent ion-free OR-2 solution (82.5 mM NaCl, 2 mM KCl, 5 mM HEPES, 1 mM MgCl2 adjusted to pH 7.4 using NaOH ) at room temperature for 45 minutes to dissolve the follicular layer and then the largest oocyt es of the stage V and VI developmental stage sele cted under a light microsco pe. The oocytes are then washed in Barth solution (88 mM NaCl, 1.68 mM MgSO4, 10 mM HEPES, 0.47 mM Ca(NO3)2, 2.4 mM NaHCO3, 0.41 mM CaCl2 adjusted to pH 7.4 using NaOH) to stop the enzymatic digestion. Oocytes are kept at 19 C in ND96 oocyte culture medium (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 5 mM HEPES buffer, 1.8 mM CaCl2, 2.5 mM Sodium pyruvate, 100 g/ml Penicillin, 100 g/ml Stre ptomycin, 5 % horse serum). Plasmids The gene encoding the murine ( mslo ) BKCa channel was provided by the laboratory of Lawrence Salkoff, Washingt on University, St. Louis, MO. The BKCa channel ( mslo ) gene is contained in the plasmid: pcDNAOX, a pcDNA3 derivative which

PAGE 87

87 contains flanking 5and 3-UTR sequence from Xenopus Beta-globin. The gene encoding the monomeric red fluorescent protein was provided by the laboratory of Roger Tsien, University of California, San Diego, CA. The mRFP1 gene was contained in the plasmid pRSET-B prior to fusion wi th maxi-K in the pcDNAOX vector. The pXOOM vector, used for co-expression of a target gene and (non-fused) green fluorescent protein in either Xenopus oocytes or mammalian tissue culture, was provided by the laboratory of Demitri Budko, The Whitney Laboratory for Marine Bioscience, St. Augustine, FL. The bacterial expression plasmid, pET46 EK/LIC, was purchased from Novagen/EMD Biosciences. DNA Preparation and Manipulation All site-directed-mutagenesis was per formed with the QuickChange II XL kit (Stratagene). Competent E.coli cells were from Strata gene (XL-10gold), Invitrogen (Top-10, INV-110, BL21 (DE3), or Novagen/ EMD Bioscience (BL21 (DE3), Rosetta-2 (DE3). Restriction enzymes were purchas ed from New England Bi olabs (NEB), Calf intestinal alkaline phosphatase was either from Promega or NEB, and T4 DNA-ligase was from NEB. Gel extraction and pur ification was done with the QIAquick gel extraction kit (Qiagen) from agarose gels. Other media and reagents were from Sigma or Fisher Scientific. Sequencing was performed by the fluorescent dideoxy terminator method at either the Whitney Laboratory for Marine Bioscience, St. Augustine, FL or at the ICBR sequencing core, University of Florida, Gainesville, FL. All deoxyoligonucleotide primers were synthesized and PAGE purified by Integrated DNA Technologies, Coralville, IA. To allow dele tion of the maxi-K C-terminal domain and fusion with mRFP1 in pcDNAOX, an insertion coding for a Bam HI site flanked by a Bst BI site was added in place of the maxi-K stop codon by site directed mutagenesis. A

PAGE 88

88 second Bam HI site was added by site directed mutagenesis in a position such that digest with Bam HI followed by gel purification and religation would result in the deletion of the C-terminal domain encoded between the two Bam HI sites. As there was no stop codon added at this point of truncation, an additional 12 amino acid residues (ALRTPRRPELFF347-stop) are coded for at the C-terminus of BKCa mSlo CTD construct. Initially, the mRFP1 gene was digested aw ay from the pRSET-B plasmid with Bam HI and Bst BI enzymes, then purified by gel ex traction (mRFP1-insert, above). XT-PCR was used to modify the mRFP1 ge ne by addition of a C-terminal extension containing a hexa-histidi ne tag, a new stop codon, Bst BI restriction site, and unique Spe I restriction site. The following primer s were utilized for XT-PCR: (extension sequence is underlined) RFP his-forward: (5TAAG GATCCGATGGCCTCCTCCGAGG) RFPhis-reverse: (5TATTTCGA ACTAGAGATGGTGGTGATGATGTGCGGCGCCGGTGGAGTGGCG). This C-terminally His-tagged mRFP1 gene (mRFP1His) was subsequently fused to the C-terminally deleted maxi-K in the pcDNAOX vector resulting in the preparation of the C TD2-mRFP1His construct. CTD2-mRFP1His gene cassettes were annealed and ligated to a linearized pRSET-B vector to complete the directional cloning. The ligated vector was transformed into competent cells ( E.coli Top-10, Invitrogen), and grown on selective agar. Plasmid DNA purified from transformed colonies was screened by double restriction digest with an enzyme unique to the His-tag sequence ( Spe I) and an enzyme unique to BKCa (Pmi I). Positive subclones were confirmed by sequencing. Expression in Oocytes After harvesting of the oocytes, oocytes were selected that were in either maturation stage IV which show separation into animal and vegetal poles or stage V

PAGE 89

89 oocytes which are used for microinjection of the cRNA. In vitro transcriptions were performed to obtain cRNA coding for the desired recombinant BKCa channels. The mMessage mMachine T7 Ultra kit (Ambion, Austin, TX) was used for in vitro transcriptions as follows; 2 hour incubation at 37 C was done of a mixture of 4 L BKCa channel miniprep DNA, 2 L 10 transcription buffer, 10 L ribonucle otide mix, 2 L 10 enzyme mix and 2 L 30 mM GTP. 30 L of RNA se-free water and 25 L of lithium chloride were added and the mix chilled for 30 min at -20 C to terminate the reaction. The cRNA was pelleted by centrifugation at 14,000 rpm for 15 min and a temperature of 4 C. The supernatant solution was remov ed, 90 L of cold 70% ethanol added and spun again at 14,000 rpm for 5 min at 4 C and the final supernatant removed. The cRNA pellet was resuspended in 20 L of RNAse-free water. A microinjector (Drummond Nanoject, Drummond scientific Co ., Bromall, USA) attached to a three dimensional micromanipulator was used to inject 46 nl (approximately 50 ng) of cRNA into the oocyte cytoplasm. Inject ed oocytes were incubated at 18 Celsius for 72 hours in ND-96 oocyte culture medium fo r translation, expression of BKCa channels and trafficking of functional channels to t he membrane surface of oocytes. Analysis of Expression: Elect rophysiological Recordings A variety of techniques can be used to analyze the levels of expression including biochemical techniques, physical and electro physiological techniques. Analysis of the successful expression of functional ion chan nels by electrophysiological recording is significantly more sensitive than either of the other approaches. For analysis of expression by two-electrode voltage clamping, oocytes were placed in a 1ml plexiglass chamber, impaled with two electrodes fire polis hed to resistances of 0.3 to 2 megaohm (M ) fabricated from Corning 7052 borosilicate glass (War ner, Hamden, CT, USA) and

PAGE 90

90 the membrane potential was clamped at -90 mV using a Warner OC725C oocyte clamp. Data was collected and analyzed using pClamp 8.0 software (Molecular Devices Corporation, Union City, CA, USA). Current /voltage relationships were obtained by stepping to voltage clamp to -70 mV and t hen increasing in 10 mV steps of 100ms duration to +120 mV. The maximum cu rrent at each voltage was recorded. Membrane Extraction of Expressed Channels The expressed BKCa channels from Xenopus laevis oocyte membranes were solubilized using the nonionic detergent -octyl glucoside ( -OG). -OG was chosen because it has a high CMC of 25 mM at 20-25 C and low molecular weight 292.4 g/mol, properties that facilitate rapi d removal from mixed micellar complexes by dialysis or gel filtration after solubilization. For our purposes, a mild, easily removable detergent was necessary because the is olated and purified BKCa channels were to be reconstituted into lipid membranes for functional analysis. Ground oocytes were solubilized using a buffer solution at a final concentration of 10 mM -octyl glucoside containing 20 mM Tris buffer, 500 mM KCl and 20 mM imidazole, equi librated to pH 7.5 using KOH, and the mixture gently agitated at 200 rpm for 10 minutes at 4 C or in ice. The -subunit of BKCa channels is known to be pro ne to protease degradation;44 therefore 5 L of mammalian protease inhibitor cocktail (Sigma -Aldrich, St. Louis, MO) was added to the solubilization buffer. The suspension was th en agitated gently for 1 hour at 200 rpm on a platform shaker followed by centrifugation at 14,000 rpm at 4 C to separate the solubilized mixed micelles from cellular debr is. The supernatant containing the lipidprotein-detergent mixed micelles was colle cted and a portion of it assayed by SDSPAGE followed by Co omassie staining.

PAGE 91

91 Immobilized Metal Ion Affinity Chromatography Supernatant solutions at volumes of 2 mL were obtained after so lubilization by OG and loaded onto 1 mL Histrap FF (GE Healthcare Life Sciences, PA, USA) columns prepacked with Ni2+ Sepharose media made from highl y cross-linked agarose matrix. Elution was performed with columns equi librated with binding buffer (10 mM -octyl glucoside, 20 mM Tris buffer, 500 mM KCl and 20 mM imidazole, adjusted to pH 7.5 using KOH) at a flow rate of 1 mL min 1 and washed with the same buffer. Unbound material was washed off using 5 column volu mes of binding buffer, and then fractions of 1.5 mL eluted in a stepwise manner with an elution buffer containing 500 mM imidazole and all other components of the bi nding buffer at pH 7.4 and a flow rate of 1 mL min 1. IMAC purification was performed both manually in the cold room (4 C), using syringes for applying pressure and under automat ion using the KTA prime system (GE Healthcare Life Sciences, PA, USA). Western Blot Analysis Immunoblotting was used to determine the identity of purified protein samples and the procedure is described in greater detail in Chapter 2. SD S-polyacrylamide gel electrophoresis and Western blotting were performed by standard techniques.145 Briefly, samples were loaded on 4-20% gradi ent Tris-HCl polyacrylamide gels followed by electrophoretic separation, and then by tank electro-transfer, proteins were transferred to Immobilon Polyvinylidene Fluoride (PVDF) membranes (Millipore Corporation, MA) at 105mV for 70 minutes. After transfer, membranes were blocked with 2% nonfat dry milk (NFDM) in tris-buffered saline (TBS) with 0.1% Tween 20 (TTBS) for 1 hour at room te mperature. Hexa-histidine tag primary antibodies were diluted with 1% NFDM in TTBS and incubated with membranes overnight at 4 C. This

PAGE 92

92 was followed by three washes in TTBS and re-blocking of membranes with 1% NFDM in TTBS. The goat-anti-rabbit HR P secondary antibodies (Bio-rad) were diluted in 2 % NFDM (1:100,000) and applied to membranes for 1 hour at room temperature. The membranes were washed in TTBS prior to incubation with Immobilon Western Chemiluminescence HRP substrate (Millipore). The blots were exposed to X-ray sensitive film (Kodak). When the signal was weak or protein expression too low for distinct recognition by the hexa-His-ant ibody, we used the HRP-His-Probe (Pierce, Rockford, IL, USA). The HRP-His-probe uses a Ni-chelating group that binds to the 6His-tag present on the protein. The HRP-Hisprobe reacts with the substrate giving off a chemiluminescent signal that was quantified using X-ray sensitive film (Kodak). Lipid Vesicle Formation Chloroform solutions of the lipids (50 mg /mL) were mixed to a 7:3 molar ratio (DPhPC/DPhPE) followed by rotary evaporati on until a dry lipid film was obtained at the bottom of the vessel. The dry lipid film was then placed under vacuum in a dessicator overnight to eliminate any residual chloro form. The lipids were then hydrated for 1 hour at 50 C in buffer (5 mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4) to a final concentration of 2 mg/mL. After being cooled to room temperature, the suspension was bath sonicated for 5 min, and extruded (21 passes) at room temperature through 100 nm polycarbonate membranes in a mini-extruder (Avanti Polar Lipids Inc.). The vesicle size profile was determined by the use of a dynamic light scattering (DLS) instrument (Brookhaven Instruments Corporation, Holtsville, New York, USA). Typically vesicle sizes of (140) nm were obtained as computed by the BI-DLSW Dynamic light scattering software.

PAGE 93

93 Reconstitution of Recombinant Proteins in Artificial Liposomes A detergent-mediated proc edure was used for reconstitution of BKCa channels into lipid vesicles. A 1 ml volume suspension of pre-formed large unilamellar vesicles (LUVs) of 2 mg/mL DPhPC:DPhPE lipids in a 7:3 molar ratio extruded through 100 nm membranes prepared as detailed above, were solubilized by the addition of a 1 ml solution of 10 mM -octyl glucoside. A 500 l sample of protein solubilized in 10 mM OG detergent that wa s eluted from the Ni2+ chelated purification column was added to the solubilized liposomes resulting in pr otein mixed micelles. The final mixture composed of purified me mbrane proteins in an excess of liposomes and -octyl glucoside detergent was incubated for 30 min at room temperature, forming a solution of mixed lipid-protein-detergent and lipid-detergent micelles. The next step involved removal of detergent from thes e micellar solutions which l ed to a progressive formation of closed lipid bilayers in which the proteins would eventually in corporate. Detergent was removed by dialysis when the micellar solutions were placed in 10,000 kDa membrane MW cut-off dialysis cassettes (Pie rce) immersed in 1 liter dialysis buffer (5 mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4) for 2 hours at room temperature. The dialysis buffer was changed and the dialysis process continued for another 2 hours at room temperature and then a final exchange of dialysis buffer made and this was let to continue for ~18 hours at 4 C. All dialysis buffers had pre-cleaned polystyrene beads added outside the dialysis bags to reduc e on the number of buffer exchanges necessary for complete removal of detergent. The polystyrene beads absorb octyl glucoside as it is removed fr om the mixed micelle solution in the dialysis bag. The whole process of dialysis proceeds with constant stirring.

PAGE 94

94 Negative-Staining Transmi ssion Electron Microscopy Negative-staining TEM was used to image the proteoliposome products of the dialysis reconstitution and to characterize their morphologies. The instrument used was a Hitachi H-7000 transmission electron microscope with a maximum resolution at 0.2 nm with a magnification range of 110 to 600,000 operated at 75-100 kV with a SoftImaging System MegaViewIII with AnalySIS di gital camera (Lakewood, CO). Prior to imaging, samples required negative staining using 2 % uranyl acetate. A detailed staining procedure is provided in Chapter 2 of this document. Briefly, portions of sample solution obtained from the dialysis bags were incubated for 2 minutes and allowed to dry for an additional 2 minutes on a 400-mesh formvar-coated copper grid (Electron Microscopy Sciences, PA). 2 % ur anyl acetate was added to the dry sample for staining for 2 minutes before drying for another 2 minutes, followed by insertion onto the TEM specimen holder for imaging. Results and Discussion Expression of Recombinant BKCa Channels The use of a C-terminal mRFP1 fusion t ag provided an easy means by which to confirm expression of fluorescently tagged recombinant BKCa channels through bright field microscopy. A major advantage of using mRFP1 is that intact cells can be imaged without significant preparative pr ocedures, therefore allowing for the possibility of further incubation of imaged oocytes in case expressi on is not satisfactory. Oocytes expressing the chimeric protein exhibited a bright red glow emanating from the mRFP1 fusion tag on the BKCa channels when observed under the microscope as can be seen in panel A of Figure 3-4. Control experiments of oocyt es not injected with cRNA encoding for the recombinant BKCa channel show autofluorescence characteristic of oocytes when

PAGE 95

95 exposed to transmitted light. However, a si gnificant difference in the contrast and brightness can be observed, clearly di fferentiating oocytes expressing BKCa channels from those that do not express channels as observabl e in panel B of Figure 3-4. Figure 3-4. Bright field microscopy images of oocytes for analysi s of expression. A) Oocytes injected with cRNA coding for recombinant BKCa channels showing brightness of fused mRFP1. B) Uninje cted oocytes without the characteristic fluorescence observed from mRFP1. A major limitation of bright field microscopy when used to analyze expression in oocytes is that it does not provide unequivoca l information regarding the localization of the fluorescent-tagged proteins Proteins localized in the cytoplasm which have undergone complete translation but have not ye t trafficked to the oocyte membrane still fluorescence, although the signal w ould not correspond to functional BKCa channels. Hence, function is tested by the two-electrode voltage clamping (TEVC) electrophysiology technique that has the c apability to detect membrane expressed ion channels. TEVC is used in combination with t he bright field microscopy to demonstrate expression of functional chimeric mRFP1-tagged BKCa channels. Analysis of Expression by Tw o-Electrode Voltage Clamping TEVC is a method capable of detecting as few as 105 channel molecules on a single oocyte (less than 10 -18 mol) as compared to biochemical techniques which have A B

PAGE 96

96 a detection limit of 10 -12 mol, a value dependent upon experimental parameters. Uninjected oocytes (without the cRNA coding for channels) were voltage clamped as a control for comparison with oocytes expressing BKCa channels as shown in Figure 3-5. 10 0 Time ( ms ) Current (A) 0 10 20 30 Voltage (mV) -100 0 100 Figure 3-5. TEVC of uninjected oocytes as a control for monitoring expression levels of BKCa channels. Potentials applied in steps from -60 mV to 60 mV. Potentials were applied stepwise from 60 mV to + 60 mV in 10 mV increments and any current flow recorded. The control experiment showed no discernable current flow as can be observed in the upper panel on Figure 3-5 showing the recorded current levels in microamperes (A). Ion channels expressed on t he oocytes would typically be expected to allow current flow across the oocyte membrane, and because cRNA coding for BKCa channels was not injected into these oocytes, the obtained results were not surprising. Xenopus laevis oocytes are known for their endogenous expression of BKCa channels which exhibit gating behavior that is quantitatively similar to murine ( mslo ) BKCa channels and hslo channels cloned from the human brain.146 However, because of

PAGE 97

97 the low expression levels of endogenously ex pressed channels, currents through them are usually in the picoamper e range, therefore negligible and cannot be accounted for in the recording shown in Figure 3-5 above. Oocytes injected with cRNA coding for the mRFP1-tagged BKCa channel and incubated for expression were analyzed for expression in a similar experiment as the control. The data of the voltage clamping of these injected oocytes is shown in Figure 3-6. 0 5 10 Time ( ms ) Current (A) -500 0 500 1000 Voltage (mV) -40 0 40 Figure 3-6. TEVC currents show ing expression of mRFP1-tagged BKCa channels. Potentials were applied stepwise between -40 mV and 40 mV. Potentials were applied in steps just like in the control clamping of uninjected oocytes, however, voltage was reduced and the st eps in this case ranged from -40 mV to + 40 mV as can be observed in the lowe r panel in Figure 3-6 showing applied voltage in millivolts (mV). Current flowi ng through the expressed mRFP1-tagged BKCa channels

PAGE 98

98 ranged between -700 A to 700 A as can be obs erved in the upper panel in Figure 3-6 showing current amplitudes in microamperes (A). Such current amplitudes are associated with high levels of expression of t he expressed channels, in this case, that of BKCa channels encoded for by t he injected cRNA. Expression associated with the observed currents in the microampere range is generally acceptable as being from heterologous origins,147 and reflects on the success of expression in this study. All the macroscopic currents observed in the record ings of the injected oocytes shown in Figure 3-6 are assigned to those passi ng through the heterologously expressed mRFP1-tagged BKCa channels and were not from any other source. Figure 3-7 shows currents elicited when ooc ytes injected with cRNA encoding the mRFP1-tagged BKCa channel were voltage clamped to test for voltage dependence of expressed channels and analyzed by TEVC. 0 20 40 60 Time (ms)Current (A) 0 200 400 600 Voltage (mV) -100 0 100 Figure 3-7. Currents re corded from mRFP1-tagged BKCa channels to test for voltage dependence of expressed chann els. Potentials were applied stepw ise from 20 mV to + 120 mV.

PAGE 99

99 It was imperative to demonstrate the voltage dependence of the channels expressed in this study, to unequivocally show that currents observed during analysis of expression by TEVC were indeed from mRFP1-tagged BKCa channels because Xenopus oocytes are known to endogenously ex press stretch-activated channels with non-selective permeability to K+ and Na+.148, 149 The current flow observed through expressed channels in Figure 3-7 was elicited in response to voltages ranging from -20 mV to + 120 mV in 10 mV increments from a holding potential of -80 mV as shown in the voltage panel. Current at levels of up to + 450 A show the high levels of the expression of mRFP1-tagged BKCa channels achieved during this study. Increased membrane depolarization results in greater current amplitudes, and vice versa, as can be noted in Figure 3-7, a characteristic behavior of voltage dependent channels. The data attests to the success of expression of mRFP1-tagged BKCa channels. Investigations were performed to analyze the expressed mRFP1-BKCa channels in comparison to native proteins (full-length BKCa channels) to determine whether the Cterminally deleted constructs tagged with fusion proteins maintained functional integrity. A plot was made relating voltage across t he oocyte membrane (membrane potentials) against current flow elicited by ions traver sing the membrane for the expressed mRFP1BKCa channels. The current-voltage relationship or I-V curve as plotted in Figure 3-8 can be used to determine if the clamped channels ex hibit their characteristic conductances, hence optimal functional activity by the equation: = I / ( V Vrev) (3-1) Where is the single channel conductance, I the unitary current for a single channel, V is the membrane potential, and Vrev is the reversal potential.

PAGE 100

100 Unitary current was measured from the first current step from the basal level off the macroscopic currents. Figur e 3-8 below is the current-vol tage (I-V plot) relationship for mRFP1-BKCa channels. Voltage (mV) -20 20 40 60 80 100Current (A) -40 40 80 120 160 Figure 3-8. The current-voltage curve for the expressed mRFP1-tagged BKCa channel showing the voltage-dependence of this channel in oocytes. The I-V curve in was obtained from macroscopic currents flowing through clamped mRFP1-BKCa channels at a holding potential of -80 mV and applied membrane potentials ranging from -20 mV to + 120 mV in 10 mV increments. As expected the expressed channels were voltage-dependent a characteristic feature of BKCa channels indicated by the non-linearity of the I-V curve. Furthermore the single channel conductance obtained by linear approximation using E quation 3-1 was within the expected range for fully functional BKCa channels. The unitary conductance derived from TEVC macroscopic currents and calcul ated as described abov e for the expressed mRFP1-tagged BKCa channels was 129 pS, a value which was within the expected conductance range for full-length BKCa channels. BKCa channels have been shown in literature to have a conductance range between 90 pS to 300 pS.

PAGE 101

101 Isolation and Solubiliza tion of Expressed mRFP1-BKCa Channels After the screening process to determine expression levels, oocytes expressing the functional mRFP1-BKCa channels were isolated and aliquots made containing 100 oocytes each. These were ground up for st orage as described in the materials and methods section in this chapter. Prior to purification, the ex pressed channels were isolated from the oocyte membrane by deter gent solubilization and centrifugation. Two detergents were tested for solubilization and these were the non-ionic detergents -octyl glucoside (OG) and dodecyl maltoside (D DM). Both these detergents are well documented and have been extensively used for solubilization of membrane proteins over the years. Dodecyl maltoside is known to enhanc e protein delipid ation during the solubilization of membrane-a ssociated proteins, as well as improve protein mobility when performing electrophoretic assays.150 Dodecyl maltoside solubilized membrane extracts were run through chromatography co lumns for purificati on but DDM could not be removed sufficiently by dialysis to allow in corporation into lipid membranes. In this study, it was necessary to remove the detergents, because the expressed mRFP1-BKCa channels were to be incorporated in synthet ic lipid bilayer membranes. DDM is known to form large micelles of greater than 50 kDa at low concentrations of 0.17 mM (suboptimal for the present study) which makes it difficult to remove by dialysis of gel filtration;151 therefore, Octyl-glucoside a high critical micellar concentration (CMC) detergent, was used because it could be eas ily and conveniently removed by dialysis. Octyl-glucoside concentrations ranging fr om 10 mM to 30 mM were tested on trial and error basis to find the optimal concen tration for solubilizat ion without wastage, and a 10 mM concentration was settled for. Solubilized membrane extracts were then

PAGE 102

102 centrifuged to pelletize and remove ce llular debris and separate these from the supernatant containing mRFP1-BKCa channels for purification. Immobilized Metal Ion Affinity Chromatography Purification of the mRFP1-tagged BKCa channel proteins was performed under non-denaturing conditions by immobilized metal ion affinity chromatography using 1 mL Histrap FF (GE Healthcare Life sciences) pre-packed columns charged with Ni2+ affinity resin. Figure 3-9 summarizes the purific ation scheme used for these his-tagged recombinant mRFP1-BKCa channels. 20 mM Imidazole 10 mM Octyl glucoside 20 mM Tris 500 mM KCl Binding to Ni2+affinity media WashWash with 5 mL of Equilibration Buffer (5 Column Volumes)Elution500 mM Imidazole 10 mM Octyl glucoside 20 mM Tris 500 mM KCl Solubilization Purified His6-tagged Protein10 mM Octyl glucoside 5 mM Protease inhibitor 20 mM Tris 500 mM KCl 2 mL Volume pH 7.4 Equilibration Buffer Elution Buffer 1 mL Column Volume pH 7.4 pH 7.41 mL Fractions Collected Figure 3-9. Purification scheme for the histidine-tagged mRFP1-BKCa channel.

PAGE 103

103 The octyl glucoside used for solubilizati on helped to maintain the conformational integrity of mRFP1-tagged BKCa channels after de-lipidati on. Volumes of 2 mL of supernatants from solubilized and centri fuged oocyte membrane extracts were loaded on Ni2+ affinity chromatography columns for adsor ption of the histidine-tagged protein sample; unbound material was allowed to flow through the column. For the removal of unbound material, including proteins that ma y have been non-specifically bound, the Ni2+ column was washed with 5 mL of equilibrati on buffer (5 column volumes) at a flow rate of 1 ml/min. For the gravity (manual) pu rification, a flow rate of 1 mL/min was equivalent to an elution of 30 drops per minute. The desired his-tagged protein was eluted in one step with buffer containing 50 0 mM imidazole, a competitive ligand to Ni2+. In all buffers, 10 mM octyl glucoside was pr esent to provide a micellar environment for the membrane proteins. Three 1 mL volume fractions were collected after the manual purification and analyzed by UV absorbance 280 nm and data plotted in Figure 3-10. 100150200250300350400450500550600 0.00 0.07 0.14 0.21 0.28 AbsorbanceWavelength (nm) fraction 1 fraction 2 fraction 3 Figure 3-10. UV spectra of fractions from manual mRFP1-BKCa channel purification.

PAGE 104

104 The three eluted fractions were loaded in 1 cm pathlength cuvettes and into a UVVIS spectrometer followed by the collection of absorbance readings. The three fractions all yielded spectra with protein peaks at a wavelength of 280 nm corresponding to purified mRFP1-BKCa channels as can be observed from the spectra in Figure 3-10 and as expected of proteins. For the determination of the molecular weights of the proteins corresponding to the peaks observed, theref ore confirm purificat ion of the correct protein, an SDS-PAGE gel was run. Figure 3-11 shows the image obt ained from the gel documentation system. 90 kDa 50 kDa 85 kDa 4 321Coomassie Blue StainingEach lane contained 10 L Sample 5 L Laemmli Buffer 2 L Reducing agent (DTT) Lanes 1.MW Standard 2.Control (uninjected oocytes) 3.Fraction 1 4.Fraction 2 5.Fraction 3 5 Figure 3-11. Image of a 420% gradient Tris-HCl gel stained with Coomassie Blue showing lanes with purified fractions of mRFP1-BKCa channels from the Ni2+ affinity column. The protein fractions were analyzed by SDS-PAGE in a 4-20% gradient Tris-HCl polyacrylamide gel and imaged using a UV transilluminator as described in the materials and methods section in this C hapter. Lane 1 on the gel was loaded with a molecular weight ladder, lane 2 loaded with control membrane extracts isolated from oocytes which were not injected with cRNA of the mRFP1-BKCa channel, and lanes 3-5 were loaded with the purified fractions 1-3 respectively, eluted from the Ni2+ column. All

PAGE 105

105 the lanes were loaded with sample volumes of 20 L, 5 L Laemmli buffer and 2 L of the reducing agent dithiothreitol (DTT). As expected, the lane with uninjected controls had no band at 70 kDa corresponding to t he molecular weight of mRFP1-BKCa channels. Lane 3 with fraction 1 and lane 4 with fraction 2 both have discernible bands appearing below the 90 kDa band corres ponding to purified mRFP1-tagged BKCa channels. Fraction 3 yielded a very faint band that is not very clearly visible on the gel image. Most of the protein pur ification in this study wa s performed manually; however, the AKTA prime automated system was also used under similar conditions as the manual purification to give a comparison of results obtained from the two methods. The chromatogram obtained fr om the AKTA purifier is shown in Figure 3-12. Column Equilibration Sample injection Wash of unbound material Elution buffer injection Elution peak of bound protein Sample adsorption & elution of unbound material Figure 3-12. Chromatogram showing the purification stages of mRFP1-BKCa channels on the AKTA Prime aut omated protein pur ification system

PAGE 106

106 The process of automated purification on the AKTA prime system performed in this study is described in great detail in Chapter 2 of this document. The separation of the target protein from ot her oocyte membrane components and the progression of the purification were tracked by the detection of UV-Vis abs orbance at 280 nm, which is represented by the blue peaks on the chromatogr am in Figure 3-12. Fractions 3-6, each 0.4 ml in volume, contained the flow th rough after the super natant containing the solubilized mRFP1-tagged BKCa channels was loaded on the column. Wash of the column was performed in the next step in the purification with fractions 7 and 8 containing all the unbound oocyte membrane co mponents. Fractions 8, 9 and 10 were obtained after injection of the elution buffer for desorption and these fractions contained the histidine-tagged BKCa channels. Aliquots of each fraction were analyzed by SDSPAGE and western blotting to veri fy protein purification and fo r identification of the histagged constructs. Figure 3-13 s hows images of the protein gel and western blot film. Coomassie Blue Staining Each lane contained: 10 L Sample 5 L Laemmlibuffer 2 L Reducing agent Lanes 1.MW Standard 2.Fraction 8 3.Fraction 9 4.Fraction 10 1 2 3 4 90 kDa 50 kDa 1 2 3 4 Figure 3-13. Electrophoresis and immunoblotting of purified BKCa channels. A) SDSPAGE using 4-20% Tris-HCl polyacrylami de gels. B) Western blot on PVDF membrane. A B

PAGE 107

107 Panel A in Figure 3-13 shows the image of the SDS-PAGE gel loaded with protein aliquots from fractions 8-10 collected fr om the AKTA Prime purifier. The lane assignments are indicated in the figure and were as follows; lane 1 was loaded with the molecular weight ladder and l anes 2-4 loaded with fractions 8-10 respectively. Each of the lanes was loaded with 10 L of the protein sample, 5 L of Laemmli sample buffer and 2 L of the reducing agent dithiothreitol (DTT). The protein gel image is expected to show a band corresponding to a molecular mass of 70 kDa for the recombinant BKCa channel with a C-terminal truncation at positi on 347 and attachment of a 225 amino acid mRFP1 fusion tag and a six residue histidine tag. However, analysis of the gel image in panel A indicates that the BKCa channel protein migrated on the SDS/Tris-glycine gel at a molecular mass represented by a band obs erved at approximately 85 kDa. Additionally, an unidentified band with a mole cular mass of approximately 45 kDa can be observed. This unidentified band is attributable to non-specifically bound oocyte membrane components with amino acid resid ues that had greater affinity for Ni2+ than the 20 mM imidazole which was added to t he wash buffer. The apparent discrepancy of 15 kDa between the expected and actual mo lecular weights of the recombinant BKCa channel is probably as a result of post-transla tional modifications of the protein during expression, specifically N-linked glycosylation resulting in higher molecular weight of the glycosylated protein than the wild type channel. It has been demonstrated that protein expression in Xenopus laevis oocytes is associated with glycosylation, and that inhibition of N-glycosylation blocks the incorporation of channels into t he surface membrane of oocytes.152-154 It is known that the electrophoretic mobility of glycosylat ed proteins is anomalous when analyzed by

PAGE 108

108 SDS-PAGE and that migrati on of these proteins occurs with an apparent molecular weight greater than the actual protein molecular weight. The effect is not simply due to the additional mass contributed by the oligos accharide, but due to the interference of the carbohydrate with the SDSprotein association (alter ation of the mass-to-charge ratio) and effects of the glycan moieties on the migration of the denatured protein through the separating gel.155 Panel B in Figure 3-13 shows the image of the western bl ot film obtained when the SDS-PAGE gel in panel A was transferr ed to a PVDF membrane and targeted by antibodies specific to the hexa-histidine tag. Bands can be observed corresponding to those associated with the mRFP1-tagged BKCa channel on the protein gel. Western blotting provided confirmation that the correct protein was purified. Fractions 8, 9 and 10 eluted from the purifier were pooled together and the total pr otein yield quantified by the Coomassie (Bradford) method, a well known dye-binding method for determining protein concentrations. The method is based on the alteration of the absorption spectrum of the Coomassie Bril liant Blue G-250 dye when it binds to proteins in acidic media. Coomassie G-250 dye binds primarily to the basic (mainly lysine, arginine and histidine) and aromatic amino acid re sidues and generates a color change of the solution in response to various concentration s of the protein. Hy drophobic interactions and Van der Waals forces also participate in the binding of the dye by the protein. The absorbance maximum of the dye shifts from a wavelength of 465 nm to 595 nm when binding to the protein occurs. The number of Coomassie dye ligands bound to each protein molecule is approxim ately proportional to the number of positive charges found on the protein. Increase in the protein conc entration results in a linear increase and

PAGE 109

109 decrease in the absorbance of light at 595 nm and the absorb ance of Coomassie Blue dye at 595 nm is proportional to the amount of protein bound. Protein standards were prepared from bovine serum albumin (BSA) to establish a correspondence between absorbance values and known BSA concentra tions. The absorbance spectra for the protein standards are plo tted in Figure 3-14. 300350400450500550600650700750 0.0 0.2 0.4 0.6 0.8 1.0 1.2 AbsorbanceWavelength (nm) Figure 3-14. Absorbance spectra for bovi ne serum albumin standards in Coomassie blue G-250. Inset is a color code for the spectra representing different protein concentrations in g/ml. The BSA standards prepared were of c oncentrations of 30 g/mL, 20 g/mL, 15 g/mL, 10 g/mL, 5 g/mL and 1 g/mL. A s pectral shift was observed in the absorbance maximum from 465 nm to 595 nm upon binding to the protein as expected. As can be observed in Figure 3-14, there is an inverse relationship between protein concentration and absorbance below 525 nm and this inverse relationship has an absorbance maximum at about 465 nm There are possibilities of interfering substances

PAGE 110

110 such as surfactants to resu lts obtained from the Bradford assay, but these were a nonfactor in this study since all protein standards and samples were treated under identical buffer conditions. A standard curve was cons tructed of absorbance at 595 nm as a function of the known protein content of each standard. The standard curve was used to estimate the amount of t he purified recombinant BKCa channel proteins corresponding to the measured absorbance values. Figure 3-15 illustrates the standard curve constructed from the absorbance of BSA standards. 051015202530 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Absorbance (595) nmConcentration (g/mL) Figure 3-15. Standard curve plotted from A5 95 nm values from BSA for the estimation of protein concentration. The dotted line represents the absorbance of the extracts of the purified recombinant mRFP1-tagged BKCa channel. The relationship between concentration and absorbance was best described by a straight line; theref ore, a linear regression line was plotted through the set of standard points to estimate the conc entration of purified mRFP1-BKCa channel proteins. The test sample resulted in an absorbance of 0.95 interpolated by the dotted line in Figure 3-15

PAGE 111

111 and this yielded the concentration of the pur ified samples. A value of 23 g/mL was determined to represent the c oncentration of the mRFP1-tagged BKCa channels purified and quantitated as described above based on the plotted standard curve. Reconstitution of BKCa Channels into Liposomes After purification, channels were incor porated into model membranes for further functional studies. The mechanism of octyl glucoside-mediated reconstitution involves, first, a direct incorporation into deter gent-saturated liposomes, followed by a gradual removal of detergent.156 Because of the relatively hi gh CMC of octyl glucoside and its high monomer solubility, it was depleted from the protein sample by dialysis. The mRFP1-tagged BKCa channels were reconstituted in 7:3 DPhPC: DPhPE liposomes at a lipid-protein ratio of 10:1 (w /w). The reconstitution procedure is detailed in the methods section of this chapter. Figure 3-16 shows negative-staining TEM imaging of DPhPC: DPhPE lipid dispersions prepared by hydrati on in 5 mM MOPS buffer at pH 7.4. 1000 nm 200 nm Figure 3-16. Negative-staini ng of 100 nm extruded DPhPC: DPhPE lipid vesicles at a 7:3 molar ratio. A) Image of dispersions of the lipid sample. B) Magnification of one of the vesicular struct ures observed from panel A. A B

PAGE 112

112 TEM shows that the form ed structures were not spherical, and had an average diameter size of 200 nm as can be observed in panel A of Figure 3-16. Magnification of the image shown in Figure 3-15 panel B reve als an apparent aggregation of vesicles that would explain why the objects had la rger sizes than the ~100 nm diameter expected for liposomes extruded though 100 nm polycarbonate membranes. Polystyrene beads were added in the dialysis tank outside the dialysis bags to keep the external concentration of di alyzed detergent at a minimum, thus decreasing the time of dialysis by reducing on the number of required buffer changes. Figure 3-17 shows negative-staining TEM images of proteoliposomes obtained after completion of dialysis. 1000 nm 200 nm Figure 3-17. Negative-stai ning TEM images after dialysis of octyl glucoside for proteoliposome formation. A) Prot eoliposomes of a homogenous size distribution representing t he whole dialyzed batch. B) Magnification of TEM image to provide details of surface morphology. It can be observed in Figure 3-17 panel B that the resultant proteoliposomes obtained after dialysis had homogenous sizes of approximately 400 nm diameters. These proteoliposomes were morphologically very spherical with ample internal volume in which incorporated BKCa channels could accumulate ions. The confirmation of A B

PAGE 113

113 successful insertion of channels into li posomes and functional analysis of the reconstituted mRFP1-tagged BKCa channels is investigated in Chapter 5 of this study. Conclusions This chapter details the heterologous expression, purification and reconstitution of recombinant BKCa channels in Xenopus laevis oocytes. Successful expression of the desired ion channels was demonstrated by T EVC in two ways. Foremost, the voltage clamping of oocytes injected with mRFP1-tagged BKCa channel cRNA showed a flow of macroscopic currents which are consistent with those produced by voltage-dependent ion channels expressed on oocyte membrane s and not the stretch-activated channels that are typically express ed endogenously in oocytes. Secondl y, the possibility of the currents originating from endogenously expressed BKCa channels was ruled out by the currents observed of 10-6 picoampere (pA), 6 orders of magnitude higher, compared to the 10-12 picoampere (pA) currents obtained fr om the expression of endogenous BKCa channels in oocytes.146 The red glow of fluorescencing oocytes observed under a bright field microscope originated from the mRFP1 fusion tag incorporated in the heterologously expressed BKCa channels, further supporting the conclusion that the expressed channels were translat ed from the injected cRNA. The TEVC results suggest that the inco rporation of mRFP1 and histidine fusion tags to form the chimeric BKCa channel does not prevent function. Additionally, the results demonstrate that the C-terminal truncation yiel ds channels that produce macroscopic currents and electrical properties characteristic of BKCa channels.38 SDSPAGE reveals that C-terminally deleted recombinant mslo mRFP1-tagged BKCa channels expressed in Xenopus laevis oocytes have a molecular weight of 85 kDa which is approximately 15 kDa greater than expected, howev er, Western blotting using

PAGE 114

114 antibodies specific for histid ine-tag fusion proteins conf irmed that the expressed and purified channels were indeed the chimeric BKCa channels. The higher than expected molecular weight is associ ated with probable N-gl ycosylation of the channels, since use of X. laevis oocytes as an expression system require s glycosylation whic h facilitates the effective trafficking of expressed channels to the oocyte membrane.153 Glycosylation alters the mass-to-charge ration of the SD S-protein complex resu lting in an apparently greater molecular weight of proteins analyzed by SDS-PAGE. The expressed mRFP1tagged BKCa channels were successfully reconstituted in model membranes for functional analysis. Analysis by electrophysiol ogy is investigated in Chapter 5 of this study and it served as a confirmation of su ccessful reconstitution and also allowed investigations into whether the functional integrity of expressed channels was retained after purification.

PAGE 115

115 CHAPTER 4 DIPHYTANOYLPHOSPHATIDYLCHOLINE AND -ETHANOLAMINE LIPID MIXTURE CHARACTERIZATION OF VESICLES AND PLANAR BILAYER FORMATION Introduction Lipid molecules form the bilayer structures that constitute the primary component of biological membranes. Biological membranes are highly complex with regard to their constituents and the study of any single me mbrane component in isolation typically involves removal from the membrane and reconsti tution in simple synthetic bilayer lipid matrices. Lipids can be used to form synt hetic bilayers as cell membrane mimics in different model membrane configurations fo r a diversity of applications such as membrane protein studies,77, 94 pharmaceutical drug discovery,157-159 biosensor development97, 98, 160 membrane dynamics studies161, 162 and studies of protein-lipid interactions.163, 164 There is a great diversity of membrane lipids, each with specific designated roles in cells in which they are localized. The most commonly found membrane lipids are glycerophos pholipids, most of which have one of the glycerol hydroxyls linked to a polar phos phate-containing group. Each of the other two hydroxyls is attached to a hydrophobic group through ester or ether linkages, with the phosphate group being at the sn-3 position of glycerol as illustrated in Figure 1-6.66 However, in archaebacteria that live in areas of extr eme environmental conditions such as high temperatures, high acidity or high alkalinity, are found lipids with the stereoconfiguration of glycerophospholipids reversed such that the phosphoryl groups are in the sn-1 position of glycerol.67 Lipids in many of these bacteria have hydrophobic constituents that are branched, saturated isopranyl glycol ethers rather than fatty acid esters.165 These branched chains referred to as phytanoyl chains have unique acyl

PAGE 116

116 chain packing properties which confer bilaye r stability to membranes, as well as low permeability to water and other sm all ionic and nonionic molecules.166 Phytanoyl lipids exhibit high mechanical and chemical stability, and are therefore ideal for the formation of m odel membranes for protein reconstitution. Translational and rotational motions of headgroups are slower in lipids with branched acyl chains resulting in a reduction of conformational and wobbling mo tions of these lipids, and dynamics that are similar to those observed when cholesterol molecules are added to lipid bilayers.167 Phytanoyl lipids form model membranes with high resistivity to permeation by electrical currents, making them suitable for electr ophysiological applications involving ionconduction measurements through recons tituted pore-forming peptides and ion channels.168 The data for electrophysiologi cal measurements of ion channels reconstituted in model membranes containing phytanoyl lipid mixtures are discussed in Chapters 5 and 6 of this dissertation. This chapter focuses on the characte rization of the formation of model membranes of the phytanoy l lipids, 1,2-diphytanoylsn -glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE), as well as specific mixtures of the two. The chemical structures of the two lipids are presented in Figure 1-12. Choline and et hanolamine lipids are generally the most abundant in the membranes of a majority of organisms. Wh en a small amount of pure DPhPE is added to pure DPhPC, the average size of the headgroup becomes smaller while the acyl chain cross section remains the same, ma king it easier to accommodate membrane proteins in lipid mixtures of the two. Pure DPhPC, as we ll as DPhPC mixed with DPhPE in a 7:3 molar ratio, have previously been studied in our laborator y and shown to form

PAGE 117

117 vesicles, which upon fusion with modified se lf-assembled monolayer (SAM) surfaces formed tethered bilayers, creating a suitabl e environment for fully functional ion channels.99, 169, 170 For the study documented her e, particular emphasis was placed on how these lipids behave when hydrated in aqueous me dia and the assemblies they form when they interact with solid substrates. The specific conditions and compositions that favor bilayer formation were also analyzed. As a platform for electroph ysiology studies, lipid bilayers can be used either as thin barri ers separating two aqueous solutions across small apertures 171 or supported on solid substrates The reconstitution of purified recombinant BKCa channels as documented in this dissertation is performed in model membranes prepared from PC: PE lipids in a 7:3 molar ration, which is closely related to the native composition of the two lipids in Xenopus laevis oocytes, used here as the expression system for these ion channels. The percentage of total phospholipids in oocyte membranes includes 65 % PC and 19 % PE while the rest is composed of phosphatidyl inositol (PI), phosphatidyl se rine (PS), sphingomyelin and cholesterol.172 Evidence shows that specific phospholipid h eadgroups and structural f eatures of sterols play a more essential role in functional prot ein-lipid interactions than the effects of bulk lipid components.60 Dynamic light scattering (DLS), diffusion nuclear magnetic resonance (NMR-D), and cryogenic transmission electron microscopy (cryo-TEM) were used to characterize the lipid assemblies formed by different com positions of DPhPC and DPhPE mixtures in aqueous solution. Each technique gave insight into the physical properties of lipid assemblies, and these properties include size, phase, and diffusivity. The lipid

PAGE 118

118 compositions that formed vesicl es in aqueous solution were further investigated in terms of the interactions of their vesicles wit h solid surfaces. The interactions between plasma-treated gold and silica with lipids were examined using quartz crystal microbalance with dissipation monitoring (QCM-D). The effects of lipid concentrations on either vesicle fusion or vesicle adsorption were monitored by atomic force microscopy (AFM) to give visual analysis of the resulting structures formed on solid substrates. Furthermore, studies were perform ed to examine the temp erature effects on sizes of formed vesicles of different lipid compositions of DPhPC and DPhPE. Materials and Methods Lipids and Chemicals Archaea phospholipids, 1,2-Diphytanoylsn-Glycero-3-Phosphocholine (DPhPC) and 1,2-Diphytanoyl-sn-Glyce ro-3-Phosphoethanolamine (DPh PE), purity >99%, were used as purchased from Avanti Polar Lipids (Alabaster, AL). Buffer solutions were prepared with 5 mM 3-Morpholinopropanesulf onic acid, MOPS, (>99.5%, ultra grade, from Fluka), 250 mM potassium chloride (A CS certified grade), and 0.1 mM calcium chloride (ACS certified grade). The buffer was titrated to pH 7.4 using potassium hydroxide (ACS certified grade) All the salts were purchased from Fisher Scientific and used as received. MilliQ filtered water (>18 M cm) was used for all sample preparations and studies. Gold Surfaces For the AFM investigations, ultra smoot h gold surfaces were prepared on silicon wafers by evaporation of a 3 nm thick laye r of Ti followed by deposition 500 nm thick layer of 60% Au to 40% Pd alloy. A 200 nm gold layer was final ly deposited on top of the alloy. The gold surfaces were cleaned by rinsing with hexane, acetone, ethanol and

PAGE 119

119 water respectively then dried under a continuous stream of nitrogen. This was followed by UV/ozone treatment of the gold surface in a Harrick PDC-32 G plasma cleaner/sterilizer at high radio-frequency power for 20 minutes under a flow of oxygen for oxidation, and for the removal of both chemical and or ganic contaminants. Vesicle Preparation The vesicles were prepared by mixing lipi d solutions (2 mg/ml in chloroform) of DPhPC and DPhPE lipids in the desired molar ratios. The chloroform was evaporated under vacuum and the resultant f ilm of lipid mixtures hydrat ed in water or buffer at the desired concentration, and then heated at 50C for one hour under vigorous stirring until a clear solution was obtained. After allo wing the suspension to cool to room temperature, it was bath sonicated for 5 min, followed by extrusion through an 80 nm polycarbonate membrane (21 times) using a mi ni-extruder from Avanti Polar Lipids. Extrusion was performed no longer than 24 hours prior to any investigation. Cryogenic Transmission Electron Microscopy (Cryo-TEM) Cryo-TEM was performed on a Phillips CM1 20 BioTWIN cryo electron microscope operated at 120 kV using an Oxford CT3500 cryo -holder. Specimens were kept in the microscope and imaged at a te mperature of about 93.5 K us ing liquid nitrogen cooling. Images were recorded digitally with a CCD camera (Gatan MSC791), and under focus conditions to improve the phase contrast. Specimens for electron microscopy were prepared in a controlled environm ent vitrification system (CEVS)173 to ensure a fixed temperature (26-28C) and high hum idity in order to minimize evaporation. In brief, a drop of sample was put on a glow discharge pretreated Pelco grid (lacy carbon film, supported by a copper grid). Then, excess solution was removed by blotting with filter paper, leaving a thin meniscus of the solution in the holes of the carbon film. The

PAGE 120

120 blotting was done first on the side opposite to the sample drop and then gently on the same side as the placed sample. The grid wa s then rapidly plunged into liquid ethane at its melting temperature. T he vitrified specimens were stored under liquid nitrogen. Dynamic Light Scattering (DLS) DLS measurements were perform ed on a Precision Detectors PDDLS/CoolBatch+90T instrument and the data were anal yzed using the Precision Deconvolve32 Program. The measurements were performed using a 683 nm laser source and a 90 degree scattering angle, set at 20 C. When necessary, additional dilutions were performed until the count rate was between the recommended 200 and 400 k counts per second. CONTIN analysis was used for calculating the size distributions.174, 175 NMR Diffusion The NMR diffusion experiments were per formed on a Bruker 750 MHz (17.6 T) instrument equipped with a diffusion probe wi th 24 T/m actively-shielded gradients (Diff60). A Bruker 10 mm diffusion probe was used but with 5 mm NMR tubes. All NMR experiments were performed at 25 C. St imulated Echo with Pulsed Field Gradients (PFG-STE) was used. Diffusion time ( ) of 100 ms and gradient time ( ) of 1.1 ms were used and the gradient strength (g) was arrayed in 32 steps with 16 scans in each step. To monitor the diffusion of the vesicles, a small amount (less than 1% of the lipids) of hexamethyl-disiloxane (HMDS) was added. HMDS is essentially insoluble in water, so the characteristic HMDS signal (0 ppm) will aris e from HMDS that is dissolved into the lipid membrane. The echo-decays obtained from the vesicles gave curved plots in a Log Y-axis of signal intensity vs. k, k= g2, where is the gyromagnetic

PAGE 121

121 constant. Echo-decays were fitted to a Log Normal distributed function, which can be described by: 2 2 ln ln2 )ln()ln( exp 2 1 )( mDD D DP (4-1) where Dm is the mass weighted medi an diffusion coefficient and ln is the standard deviation of the logarithmic distribution of diffusion coefficients. A mean diffusion coefficient, D, can be calculated from: 2exp2 lnmDD (4-2) and the normal standard deviation, D, of the diffusion coefficients with 1exp2 ln 2 2 DD (4-3) The diameters of the vesicles were obt ained through the Stokes-Einstein relation.176 Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D) The QCM-D measurements were performed on a QCM-Z500 from KSV Instruments (Helsinki, Finland) equipped wit h a temperature control unit from Oven Instruments. AT-cut crystals, coated with ei ther an evaporated silica or gold layer, had a resonance frequency of 5 MHz and were also purchased from KSV. Prior to the measurements, the crystals were cleaned us ing either piranha solution for the goldcoated crystals (3:1 concentrated sulphuric ac id to 30% hydrogen peroxide solution) or an SDS solution for the silica-coated crystals, followed by rinsing with water and ethanol, and drying with a gentle stream of nitrogen gas. All rinse cleaning procedures were followed by UV-oxygen plasma cleani ng for 20 min. This sequence of pretreatment led to clean and oxidized surfaces that had contact angles of less than 6 degrees. The instrument was calibrated us ing the standard procedure as described by

PAGE 122

122 the instrument provider.99 The fluids were introduced batch-wise via a reservoir, and then proceeded to a bypass chamber (500 l volume), located above the crystal. After loading the instrument with the crystal, it was allowed to stabilize in the buffer solution for approximately 30 min until a stable baseli ne was obtained. The measurements were recorded at the fundament al frequency, the 3rd overtone, and the 5th overtone. The measured resonant frequency, f depends on the mass of the oscillating crystal including any adsorbed species. For rigid film s there is a relation between the change in frequency f and the adsorbed mass, m, according to the Sauerbrey equation f n C m (4-4) where C is the mass sensit ivity constant (17.7 ngcm-2Hz-1) at f=5 MHz and n is the overtone number. The dissipation factor, D is related to the energy lost ( Elost) to the surroundings in relation to the stored energy ( Estored) upon oscillation and is defined as RC L R f E E Dstored lost 2 (4-5) where is the angular frequency, R is the resistance, L is the inductance, and C is the capacitance of the crystal. Atomic Force Microscopy (AFM) AFM analysis was performed in buffer solu tion using a Nanoscope III MultiMode Scope (Veeco) equipped with a 13 m E-scanner. The instrument was calibrated in the z-direction using a silicon grating (TGZ01, Mikromash), with a step height of 20 nm (accuracy 1 nm). A fluid cell was used, whic h was cleaned in ethanol and water prior to use. The cell was used without o-ring. Images were obtained using silicon cantilevers (Nanosensors, Neuchatel, Switzerland with dim ensions: T = 3.8-4.5 m, W = 26-27 m,

PAGE 123

123 and L = 128 m). The cantilevers were mounted in the fluid cell and a drop of the solvent solution was placed on the tip to ens ure wetting. The solution was placed on the sample surface, followed by attachment of t he fluid cell. No addition or removal of liquid through the hoses that are attached to the sample holder was performed. Prior to the measurements, the sample was left undist urbed for ~20 min to let the solution equilibrate. AFM analysis was performed by sc anning in contact mode and the images were processed using a secondorder parameter flattening. Results Cryogenic Transmission Electron Microscopy (Cryo-TEM) To visualize the mesophases of vesicles, cryo-TEM was utilized. In Figure 4-1, TEM micrographs of the vesicles are present ed for the different lipid compositions. Figure 4-1. TEM micrographs of hydrated dispersions composed of extruded vesicles. A) Pure DPhPC and B) 7:3 DPhPC: DPh PE. The lipid concentration was 0.2 g/l in 5 mM MOPS buffer at pH 7.4. All scale bars correspond to 100 nm. In the sample containing pure DPhPC, uni lamellar vesicles were observed (Figure 4-1A). The mean diameter of these vesicles was 92 nm (n=8) and they were often found A B

PAGE 124

124 adhered to each other. In Figure 4-1 panel B, a micrograph of vesicles composed of 7DPhPC:3DPhPE is presented. These vesicles were also unilamellar, though slightly smaller in size compared to the pure DPhP C (diameter ~ 60 nm (n =7)). Micrographs of samples of 5DPhPC:5DPhPE are shown in Figure 4-2 panels A, B, and C. Figure 4-2. TEM micrographs of samples of varying lipid compos ition: (A, B, C) 5DPhPC:5DPhPE, and (D) 3DPhPC:7DPhP E. The white squares in B and C were Fast Fourier Transformed (FFT), re sulting in the inserted FFT patterns. The sample contained both unilamellar vesi cles (Figure 4-2 A) and more denselypacked structures, which appear to be compos ed of close packed lipid bilayers, likely B A C D

PAGE 125

125 discrete lamellar liquid crystalline phases (Figure 4-2 panels B and C). The repeat distance was calculated from the Fourier transform of selected parts of the image (Figure 4-2 C) and was found to be around 6. 5 nm, which corresponds well to the proposed lamellar phase based on the height of a lipid bilayer. However, the specific phase of the liquid crystal cannot be determi ned solely from the micrographs. In Figure 4-2 D, a micrograph obtained from the 3DPhPC:7DPhPE sample is presented. In these samples, small flower-like, vesicles were observed and both the larger unilamellar vesicles and the densely-packed lipid bila yer aggregates were absent. For the sample consisting of pure DPhPE, no ob jects could be observed with the cryo-TEM technique. DLS and NMR Diffusion DLS and NMR diffusion were utilized to invest igate the size of the lipid structures in aqueous solution. For DLS, the time-dependent fluctuation in the scattering intensity is observed. The scattering intensity is highly dependent on the size of the scattering entity which can cause larger objects to be overrepresented in the measured size distribution of a polydisperse sample. However, this overrepresentation can be corrected for by using a weighting function t hat transforms the mean intensity value to the mean number value. In Figur e 4-3, both the intensity dat a and the adjusted data are presented as the samples average effective di ameter versus the lipid composition. As seen in the figure, the size s of the aggregates calculated diameters are independent of the lipid compositions rangi ng from pure DPhPC to 30% DPhPE. The samples average effective diameters were approximately 100 nm (mean value weig hted by the total number of object). Increasing the DPhPE relative molar concentration from 30% to 70%, caused a linear increase of the sample di ameter to approximately 200 nm. Pure DPhPE samples had a relatively smaller diameter In Figure 4-3, t he aggregate sizes obtained

PAGE 126

126 by NMR diffusion using Dm and D extracted from Equation 41 are also presented. The sizes measured by the NMR diffusion technique follow the same trend as the sizes measured by DLS. For t he pure DPhPC and 7DPhPC:3 DPhPE samples, smaller diameters were obtained by NMR diffusion in comparison to the results obtained by DLS. For the sample composed of 5DPhPC :5DPhPE, the two techniques gave sizes that were in good agreement. For higher c oncentrations of DPhPE, including pure DPhPE, the sample diameters obtained by NMR diffusion were larger than the numberaveraged DLS sizes. --10:09:18:27:36:45:53:70:10-0 50 100 150 200 250 300 350 Diameter (nm)Vesicle composition DPhPC:DPhPE NMR (Dm) DLS (Intensity) NMR (D) DLS (Number) Figure 4-3. Average effective diameter as a function of lipid ratio. The sizes were obtained using DLS and NMR-diffusion and the lipid concentration was 0.2 g/l. The lines in the figure are only a guide to the eye.

PAGE 127

127 Quartz Crystal Microbalance with Dissipation Monitoring (QCM-D) Figure 4-4 shows the measured change in frequency and dissipation as a function of time that was recorded while two different concentrati ons of pure DPhPC vesicles interacted with a silica substrate. -250 -200 -150 -100 -50 0 -70 -60 -50 -40 -30 -20 -10 0 10 0 20000400006000080000 -1 0 1 2 3 4 5 6 7 8 0100020003000400050006000 -1 0 1 2 3 4 5 6 7 8 0.15 g/l 0.1 g/lf(Hz)(B) (A)D(E-6)Time (S) Time (S) Figure 4-4. Changes in resonant frequency and dissipation versus time for adsorption of pure DPhPC vesicles onto silica substr ates. (A) 0.1 g/l and (B) 0.15 g/l, The presented QCM-D data was recorded at the fifth overtone and renormalized such that f = fn=5/5. When the lipid concentration was relatively low, 0.1 g/l, a decrease in frequency together with an increase in dissipation over the course of time were observed as shown in Figure 4-4 A. The corresponding frequen cy shift indicates that vesicles were adsorbed on the silica surface, forming a high ly viscoelastic film as evidenced by the

PAGE 128

128 high energy dissipation. When the lipid concent ration was increased to 0.15 g/l (Figure 4-5 B), the decrease in frequency change is initially the same as that observed for lower concentrations, but after ~1150 seconds the fr equency starts to increase, indicating a decrease in adsorbed mass, followed by fr equency stabilization (Figure 4-4 B). This adsorption behaviour is typical of vesicle fu sion, where vesicles first adsorb on the surface until a critical concentration is attained and then rupture to form a solidsupported lipid bilayer.90 When a vesicle ruptures, the hydrodynamically-coupled solvent in the vesicles interior is released, re sulting in an increased frequency (mass decrease) and a decreased dissipation (lower degree of viscoelasticity). The total frequency change was ~37 Hz, which corresponds to an adsorbed mass of 655 ng/cm2 (according to equation 4-4). Assuming that a bilayer is formed, based on two-step vesicle fusion kinetics, a surface area per molecule of 43 2 (Mw=846 D) results. However, this area is smaller than that of 77 2, which was previously reported for DPhPC.167 Figure 4-5 shows the measured change in frequency and dissipation as a function of time that was recorded while various concentrations of pure DPhPC vesicles at various lipid concentrations interacted with a plasma-treated gold substrate. When the lipid concentration was low, 0.02 g/l, a decrease in the measured frequency together with an increase in dissipation over the course of time was observed (Figure 4-5 A). The corresponding frequency shift indicates that the vesicles adsorbed on the gold surface. When the lipid concentration was increased to 0.1 g/l, a more pronounced frequency decrease and dissipation increase as a functi on of time was observed (Figure 4-5 B). This indicates a higher concentration of adsorbed vesicles and a higher degree of film viscoelasticity (likely due to the increased vesicle surface coverage and the related

PAGE 129

129 hydrodynamically-coupled solvent) in compar ison to the 0.02 g/l sample. In Figure 4-5 C, the measured data obtained fr om a 0.15 g/l sample is shown. Here, the frequency decreases initially, followed by a frequency increase after ~1100 seconds, indicating a decrease in adsorbed mass. The total change in frequency was ~59 Hz, which corresponds to an adsorbed mass of 1044 ng/cm2 (according to equation 4-4). Assuming that a bilayer is formed based on t he characteristic two-step vesicle fusion kinetics, the result is a surface area per molecule of 27 2, which is even smaller than the one obtained using t he silica surface. -220 -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0 20 0200040006000 -1 0 1 2 3 4 5 0200040006000 0200040006000 f(Hz)0.02 g/l 0.1 g/l 0.15 g/lD(E-6) Time(s) Time(s) Time(s) Figure 4-5. Changes in resonant frequency and dissipation versus time for the adsorption of pure DPhPC vesicles onto oxidized gold with lipid concentrations of (A) 0.02 g/l, (B) 0. 1 g/l, and (C) 0.15 g/l. The data was recorded at the fifth overt one and renormalized such that f = fn=5/5. For each concentration, the data are plotted on an identical scale. A B C

PAGE 130

130 It is important to note t hat the Sauerbrey equation us ed here to calculate the molecular surface areas is only valid for non-vi scoelastic films, which is not the case in this situation. This can be seen in the di ssipation curves (Figure 4-5 panels B and C), which show that the dissipation is relative ly high even after the bilayer has formed, indicating either formation of a highly visco elastic bilayer due to the unique structural features of phytanoyl lipids (s ee discussion) or incomplete bilayer formation such that intact, adsorbed vesicles coexist with r uptured vesicles. The QCM-D experimental results presented here were all obtained in bu ffer at pH 7.4. Expe riments were also performed in pure water, which resulted in no vesicle adsorption or fusion regardless of lipid concentrations used. Furthermore, additional QCM-D measurements were performed on all of the afor ementioned lipid com positions. Mixtures that had DPhPE lipid concentrations of less than 50% showed the same behaviour as for pure DPhPC. However, with increasing DP hPE concentration, no adsorption or vesicle fusion was observed, for the various lipid compositions used. Atomic Force Microscopy (AFM) AFM was performed on ultra smooth gold surf aces in the same buffer solution as that used for performing the QCM-D measurement s. In Figure 4-6 A, an image of a pure gold surface without deposition of molecules on the surface is shown. As seen, the surface is smooth even on the nanometer length scale. In Figure 4-6 B, the same surface is shown two hours after the addi tion of a DPhPC vesicle solution at a concentration of 0.2 g/l. The surface is still smooth, however much less sharper images were obtained. This decreased sharpness is most probably due to the presence of a lipid bilayer on the gold surfac e, which would results in a viscoelastic surface that can be disturbed by the sc anning probe.

PAGE 131

131 Figure 4-6. AFM images showing vesicle fu sion on gold substrates. A) The pure gold surface (B) the same surface after vesicle addition and subsequent fusion, leading to solid-supported bilayer formation. The measurements were obtained in contact mode using a liquid cell. Figure 4-7 shows both 2 dimensional and 3 dimensional renderings of AFM images, as well as a cross section analysis. These images were obtained two hours after addition of a 0.1 g/l vesicle solution. Figure 4-7. AFM images of pure DPhPC vesicles depos ited on an ultra flat gold surface. The lipid concentration was ~0.1 g/l. The measurements were performed in contact m ode using a liquid cell. A B

PAGE 132

132 In the bottom left image in Figure 4-7, t he surface is viewed from above and there are round objects that have diameters between 100-250 nm which is within expected range for adsorbed vesicles. Due to subs trate-vesicle interactions, the adsorbed vesicles are not spherical but rather fla ttened in a pancake-like mo rphology, hence the larger diameters of t hese entities. These 3 dimensional f eatures can be seen in the right figure, where a topographical im age of the surface is presented. On the upper left, a cross section of the lower left figure is presented together with some relevant distances. Discussion In the present study, assemblies cons isting of two phytanoyl lipids, 1,2diphytanoylsn -glycero-3-phosphocholine (D PhPC) and 1,2-diphytanoylsn -glycero-3phosphoethanolamine (DPhPE) in a polar en vironment were examined. From the results obtained by Cryo-TEM, DLS and NMR diffusion, it was determined that lipid compositions of single-com ponent DPhPC and mixtures containing up to ~50% DPhPE form unilamellar vesicles, having sizes that are directly related, though not identical, to the membrane pore size of the polycarbonate membrane that was used for extrusion (~100 nm vesicle diameters vs. 80 nm membr ane pore size). However, the vesicle sizes varied depending on the analytical technique us ed to measure them. For example, in the case of pure DPhPC vesicles, the diam eters obtained were 122 nm (DLS), 43 nm (NMR-D), and 92 nm (Cryo-TEM). These differences in size are characteristic of using these different techniques and are due to th e different physical properties that are analyzed with each technique. In the case of DLS and NMR-D the hydrodynamic radius is measured since they measure the fr ee diffusion of particles. Even though the techniques give rise to differ ent sizes for the same sample, the sizes follow an identical trend when the lipid ratio was varied. Cryo-T EM allows for a more direct method to

PAGE 133

133 measure vesicle size. The technique, however, suffers from artifact s arising from the blotting process, which induces shearing on t he sample. This shearing might affect the size and structure of the observed objects.177 The specimen preparation technique might also result in objects that are forced closer to each other re sulting in an apparent adhesion of imaged objects. This could for example be the reason why the pure DPhPC vesicles seem to be adhered to each other even though DLS and NMR-D results show that the vesicles diffuse totally separ ate from one another. The cryo-TEM technique, as with all microscopy techniques, suffers from poor statistics due to the low number of observed objects. At higher DPhPE concentrations, (>50%) both DLS and NMR revealed that the sizes of the measured objects increased. This was accompanied by an increase in si ze distribution, indicating a change in physical properties of the obj ects. There was further confirmation by cryo-TEM, with solutions containing 5DPhPC:5DPhPE, showing the appearance of several new structures in addition to unilamellar vesicl es. These new structures were composed of lipid bilayers, which were less solvated in comparison to the vesicles, forming liquid crystalline structures of pr esumably hexagonal phase, even though this has not been confirmed in our studies. For the 3DPhPC:7DPhPE lipid compositions, only small flower-like vesicles were observed, although much larger objects were expected to be seen based on the results from t he DLS and NMR-D measurements. The absence of cryo-TEM images of the larger objects is likely due to the sample preparation process during which these larger objects may have been removed. These larger structures are likely si milar to the dispersed liquid cr ystalline structures that were observed in the 5DPhPC:5DPhPE sample. Unilamellar vesicles were not observed in

PAGE 134

134 the samples containing predominantly DPhPE, which would explain why our QCM-D results showed no vesicle adsorption with thes e lipid compositions. The seemingly small headgroup differences between these two lipids pl ays a role in the lipids propensity to form or not form vesicles when in different mixtures. This difference significantly affects the critical packing parameter, result ing in a change of t he lipids spontaneous curvature. Also, the electrostatic energies are significantly different between the two head groups, with DPhPC having a mu ch higher electrostatic energy.35 As a consequence, the crystal to liquid crystalline transition temperature is lower for lipids with choline head-groups as compared to th e amine head-groups. Studies performed on non-phytanoyl lipids with choline a nd ethanolamine head-groups, similar to investigations here, shows that a lamellar to reverse hexagonal phase transition occurs when the mixture contains more than 60% PE.85, 178 The QCM-D technique is an excellent tool to study vesicle adsorption and vesicle fusion since it reveals binding kinetics, calculates adsorbed film mass including hydrodynamically-coupled solvent, and provi des viscoelastic information about the adsorbed film. In comparison to QCM-D st udies performed on vesicles composed of straight chained lipids, phytanoyl lipids show slightly different behavior, especially as regards the analysis of dissipation. Usually, in the case of lipid bi layer formation, the dissipation increases when vesicles first ads orb to the surface but then decreases to near zero as vesicles rupture and the non-viscoelastic lipid bilayer forms.90 In the case of phytanoyl lipids, the dissipat ion remains relatively high throughout the entire process, which has been observed in other studies fo r these types of lipids, performed on tethered supports.99, 179 This result indicates that the bilayer formed has relatively high

PAGE 135

135 viscoelastic character as compared to other bilayers, though the dissipation is still much less than that of an adsorbed vesicle film. Also, the final change in frequency after the vesicle fusion is larger than what has been obtained for DHPE, 37 Hz compared to 26 Hz.90 In these investigations, the dissipati on continues to be high even after the frequency has leveled out, indicating that the bilayer formed has significant viscoelastic character. A possible explan ation for this finding coul d be a combination of high viscoelastic character of phytanoyl lipid bila yers, caused by the relatively high amounts of hydrodynamically-coupled solvent in t he branched chains, together with incomplete bilayer formation. The branched chain structure of the lipids hydrophobic tails leads to reduction of the conformational and wobbling motions of the alkyl chains, in contrast, the head groups do experience a higher degr ee of freedom and as a consequence, more solvent will be hydrodynamically-coupl ed around the lipid head groups, especially for DPhPC.180 This increased bound water content together with the less dense packing of the hydrophobic tails will increase bilayer viscoelasticity, which could partly explain the QCM-D dissipation response. Despite th e fact that phytanoyl membranes couple more solvent, excess chemical potential calc ulations have demonstrated that the lipid chain branching does not significantly c hange the partition of water through the hydrophobic part of the membrane.167 Also, the alkyl branching inhibits the formation of cavities in the membrane, which could explain why the membranes have higher electrical resistances and low ion conduction.167 The viscoelasticity of the formed bilayer could also be an explanation of the observati ons made using the AFM. When the lipid concentration was high enough resulting in lipi d bilayer formation no sharp well resolved AFM micrographs could be obtained. This is something which is typical for AFM

PAGE 136

136 investigations on low surf aces having low viscosity.181 The QCM-D results showed that vesicle fusion was dependent on lipid concentr ation and the presence of ions, since no adsorption was observed in pure water. Even though the vesicle concentration is sufficient for full surface coverage, the c oncentration may not be great enough to cause vesicle fusion and bilayer formation. An excess of vesicles is needed to induce the bilayer formation, indicating t hat the interaction between the vesicles is essential. This phenomenon has been observed earlie r for other lipid systems.182 The AFM studies yielded results which are in agreement with observations made from the QCM-D data; that a relatively low lipid concentration resu lts in vesicle adsorption and that there is a threshold in the lipid concentration when a bilayer starts to fo rm, which would at approximately 0.15 g/l. The fact that the vesicles fused independently on the two tested surfaces, plasma-treated silica and plasma -treated gold is unusual. Vesicle fusion resulting in bilayer formation is often obser ved on silica surfaces but rarely on pure gold.90 It is believed that the used plasma treatment, which reduced both of the surfaces contact angles to ~6 degrees, played a major role in causing vesicle fusion to occur on the gold surface. Analysis of the AFM results and the resulting cross sectional calculations presented in Figure 4-7 indicate that the adsorbed vesicles appear to be flattened out and deformed. This vesicle deformation has been observed earlier as a result of vesicle-surface interactions and depends on both the surface and vesicle composition, among many other factors.183 Conclusions In this study, it was de monstrated that the phyt anoyl lipids, 1,2-diphytanoylsn glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3phosphoethanolamine (DPhPE) can be used to form supported lipid bilayers on

PAGE 137

137 hydrophilic surfaces as well as at the liquid-air interface. DLS, NMR diffusion, and cryoTEM showed that pure DPhPC and mixtures of DPhPC and DPhPE containing less than 50% of DPhPE formed unilamellar vesicles. These vesicles have the ability to fuse on hydrophilic surfaces of both plasma-treated gold and silica, forming solid-supported lipid bilayers as monitored by QCM-D and AFM. The bilayer formation, however, was concentration dependent and only initiated when t he lipid concentration was above 0.15 g/l and only in buffer solution. The lipid bila yers formed had high viscoelasticities, based on the QCM-D dissipation measur ements. This viscoelasticity is probably because of the ability of phytanoyl lipids to imbibe relatively high amounts of water or due to incomplete vesicle rupture and bilayer formation. No supported bilayers could be formed from solutions containing more than 50% DPhPE. As a whole, t he results demonstrate the ability to form lipid bilayers of cert ain compositions of DPhPC and DPhPE at the solid-liquid and liquid-air interfaces.

PAGE 138

138 CHAPTER 5 INCORPORATION OF RECOMBINANT BKCA CHANNELS IN A TETHERED LIPID MEMBRANE AND FUNCTIONAL ANALYSIS Introduction High conductance calcium-activated potassium (BKCa) channels are found in most organisms where they regulate a large variety of physiological processes including smooth muscle tone, neurosecretion and hearing.184-186 BKCa channels are typically gated by membrane depolarization and an increase in cytosolic Ca2+, however, channel opening has been observed in the absence of Ca2+. This observation suggests that calcium acts as a modulator to decrease t he energy necessary for channel opening and that the BKCa channel is actually a voltage-dependent ion channel.17 The Ca2+ bowl region of the channel, located intracellularly between the S9 and S10 segments (refer to Figure 1-3), contains an Asp-rich s equence motif (-QDDDDDP-) associated with Ca2+ binding and sensitivity.41 The S4 segment of the BKCa channel is positi vely charged and forms part of the intrinsic voltage sensor of t he channel, similar to t hat found in the other members of the voltage-dependent K+ channel family.37 The BKCa channel characteristic features are single-channel conductances r anging from 90 300 pS, and ionic currents which do not inactivate. Additionally, numer ous peptide blockers from scorpion venom have been isolated and characte rized and these elicit peculiar pharmacological effects and selective blockade of the channel.187 These properties make BKCa channels ideal as candidates for biosensor development when incorporated in suitable membrane configurations. Bilayer lipid membranes supported on solid substrates such as glass, metal and silica substrates have been used as models of biological membranes for fundamental studies of membrane dynamics and lipid-protein interactions.188 Numerous studies have

PAGE 139

139 been performed involving the incorporation of pore-forming peptides in supported lipid bilayers (SLBs) integrated to systems capabl e of rapid and reliable detection of analytes at the molecular level via stochastic sensing.189-191 However, SLBs suffer the drawbacks of restricted fluidity and space limitations between the membrane and substrate. In most cases in SLBs, the membrane is separated from the substrate by a thin film of water of just about 10 20 73, 126 which compromises the lateral mobility and function of incorporated proteins.94, 162, 192 Attempts to incorporate f unctional membrane proteins have been met with limited success due to denaturat ion, as a result of contact between extramembranous regions of pr oteins with the substrates.77 Tethered bilayer lipid membranes are lipid bilayers with the proximal layer coupled to the substrate by a covalently attached spacer group which provides a reservoir beneath the membrane.193 These tethers have extremel y high electrochemical sealing properties, comparable to those of natural membranes,96 and therefore, allow electrical monitoring of the incorporation and function of channel proteins within the membrane. The tBLM can be characterized by surface s ensitive techniques such as atomic force microscopy (AFM) and surface plas mon resonance spectroscopy (SPR).85-87 When metal surfaces are used as substrates for tethering bilayer memb ranes, such as the commonly used technique of membrane tether ing based on gold-sulfur interactions,7, 93, 96 the substrate acts as an electrode and can therefore be used for making electrical measurements of ion channel cu rrents. Like supported lipid bilayers, tBLMs have the advantage of enhanced resistance to mechan ical and vibrational perturbation.194 Presently, most of the work done on the inco rporation of pore-fo rming proteins in tethered bilayer membranes has invo lved only short peptides such as -hemolysin96, 113

PAGE 140

140 gramicidin7, 195, 196 and valinomycin.93 The relatively simple structures of these peptides tends to be limiting in terms of their sensing c apacity. This, therefore, calls for studies to be performed on more complex proteins that have direct biological sensing capabilities.99 BKCa channels have served as models for ion permeation, gating kinetics and pharmacology because of their robustness, inherently high conductance and relative ease of reconstitution from native membranes into lipid bilayers,197, 198 making them ideal candidates for incorporation in to tBLMs as stochastic sensors. Possible applications for biosensors based on transme mbrane peptides and ion channels include the high throughput screening of drug candidates in the pharma ceutical industry, realtime environmental diagnostics, detection of heavy-metal contaminants in ground water, detection of biological warfare agents and many other possibilities.96, 113 The ion channels to be incorporated into tBLMs c an also be genetically re-engineered to develop sensor elements that are selectivel y sensitive to other different classes of compounds that they woul d naturally not detect.112, 199 For the studies presented here, recombinant BKCa ion channels were interfaced to a gold surface by fusing proteoliposomes containing reconstituted channels into tethered bilayer lipid membranes (tBLMs), which in turn were formed at multiple individual pixels of a microelectrode array dev ice (MEA) illustrated in Figure 5-1. Singlechannel activity recordings of the BKCa channels were performed using a modified patch clamp electrophysiology set-up, with elec trodes from the sensor pad and probe pads on the MEA connected to an amplifier which processed current signals and gave a computer read-out. A silver and silver chlori de electrode in a conventional patch clamp microelectrode pipette was used to measure currents from incorporated ion channels.

PAGE 141

141 BK Channel Figure 5-1. The tethered bilayer memb rane on a microelectrode array device. A) Optical microscope image of the sensor pad showing pad dimensions of 100 m2 and tungsten electrode. B) The microelectrode array device. The graphics representing sensor pad show an incorporated channel in the tBLM. (Adapted from Keizer, H. M, et al ChemBioChem 1246-1250 (2007) One of the main advantages of bilayer membranes on solid supports, both tBLMs and SLBs, is that they can be easily characterized by surface-specific techniques in real-time. The process of t he formation of the tBLM on t he MEA was done in stages, the first of which was the deposition of the tethering moiety of the self-assembled monolayer 2,3-di-O-phytanylsn -glycerol-1-tetraethylene glycol-D,L-lipoic acid ester lipid (DPTL) by gold-sulfur interactions. This was followed by the formation of the bilayer from a mixture of t he two archaeal lipid s, 1,2-diphytanoylsn -glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE) in a 7:3 molar ratio to yield the co mplete tBLM. Finally to pe rmit surface plasmon resonance (SPR) spectroscopy of the final configuration, the recombinant BKCa channel was inserted into a membrane tethered to a gold substrate that was not part of an MEA. The

PAGE 142

142 SPR analysis showed that t he observed channel activity was indeed as a result of gating events, and not membrane def ects allowing current flow. Because of the limited volume of the ioni c reservoir in the sub-membrane space, we used a recombinant mslo BKCa channel with a C-terminal deletion after the S6 segment for incorporation into the tBLM. T he truncation eliminated the bulky intracellular domain and the Ca2+ bowl region of the channel to a llow better ionic flow within the reservoir and to maintain fluidity in the already densely packed reservoir region. Singlechannel recordings were obtained within th e experimental configuration described above and analysis of the pharmacological responses of the ion channels to externally applied tetraethylammonium (TEA) solutions was performed. Materials and Methods The Microelectrode Array (MEA) Device The MEA device was fabricated by Daniel Fine in Dr. Ananth Dodabalapurs laboratory at the Microelectronics Research Cent er at the University of Texas at Austin, Austin, TX. The devices were fabricated on silicon wafers each containing 66 pixels. Each sensor pad was formed by evaporation of 3 nm of Ti on the silicon substrate followed by a 500 nm thick alloy of Au (60 %)/Pd (40%). A 200 nm thick layer of pure gold was then deposited on the alloy to form t he surface of the pad, and a polyimide resist was photolithographically applied to define the pad size. The sensor and probe pads formed on the gold device were linked by a thin line of gold and electrically isolated from each other by the polyimide resist. Before tBLM formation, the device was rinsed with hexane, acetone, ethanol and water respectively, followed by UV/ozone treatment of the gold surface in a Harrick PDC-32 G plasma cleaner/ sterilizer at high

PAGE 143

143 radio-frequency power for 20 minutes under a flow of oxygen for oxidation, and for the removal of both chemical and organic contaminants. Lipids Archaea analogue phosphol ipids 1,2-diphytanoylsn -glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE) in chloroform were purchased from Avanti Pola r Lipids Inc. (Alabaster, AL, USA), and were used without further purification. 2,3-diO -phytanoylsn -glycerol-1-tetraethylene glycol-D,L-lipoic acid ester (DPTL) was syn thesized as described in literature.101 BKCa Channel Plasmid DNA encoding mut ant constructs of the C-terminal truncated BKCa channel with mRFP1 and histidine fusion tags was transcribed by the Ambion mMessage mMachine T7 Ultra kit. RNA was precipitated with LiCl, washed, and centrifuged in 70 % ethanol, dissolved in DEPC water and injected (46 nL per oocyte,~50 ng RNA) into defolliculat ed stage V or VI oocytes from Xenopus laevis (Xenopus Express, Florida, USA). Inje cted oocytes were maintained at 19 C in ND-96 media (96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 10 mM HEPES, 1.8 mM CaCl2, pH 7.4) enriched with sodium pyruvate (2.5 %), peni cillin/streptomycin (1 %), and horse serum (5 %). Functional channel ex pression was verified by twoelectrode voltage clamping (TEVC). Oocytes expressing the chimeric BKCa channels were rinsed in high K buffer (400 mM KCl, 5 mM PIPES, pH. 6.8) supplemented with 100 M phenylmethylsulfonylfluoride, 1 M pepstatin, 1 g/ml aprotinin 1 g/ml leupeptine, 1 M p-aminobenzamidine and transfe rred to a 1 ml ground glass tissue grinder (Kontes Duall). After grinding, oocyte memb rane extracts were stored at 80 C. Purification of the BKCa channels was performed by the use of immobilized metal ion affinity

PAGE 144

144 chromatography as described in chapter 3 of th is dissertation, followed by reconstitution by dialysis into liposomes formed from 7:3 DPhPC: DPhPE lipid mixtures. DLS and negative-staining TEM showed that the obtained proteolipos omes had diameters of 270 nm. A drop of pure lipid vesicle susp ension was deposited, and incubated for >8 hours at 4 C on the sensor pad containing a DPTL monolayer on the MEA device, to allow for vesicle fusion. A droplet of buffer solution (5 .0 mm MOPS, 250 mm KCl, 0.1 mm CaCl2) was applied to the sensor pad, and t he bilayer quality was determined from membrane resistance. If a gigaohm resistanc e seal was obtained (~80% of trials) BKCa proteoliposomes were added to the sensor pad for single-channel analysis. Electrophysiology Recordings of single-channel activity were obtained using two electrodes. The active electrode, which was a conventional patch pipette containing an Ag/AgCl electrode and filled with a buffer solu tion (5mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4), was inserted into a buffer solution dropl et over the tBLM. The reference electrode consisted of a tungsten-ti pped electrode that was posit ioned onto the probe pad that extended from the sensor pad. Both electr odes were connected to an Axon patch-clamp amplifier (Axopatch 200B, Molecular Devices Corporation, Union City, CA, USA). The signal was passed through a low-pass 5 kHz and digitized at a sampling rate of 50 kHz by using a Digidata 1322A (Mole cular Devices Corporation). Pharmacology In the pharmacology measurements, st able control channel recordings were obtained before a drop of tetraethylammonium (TEA) in buffer solu tion (5.0 mm MOPS, 250 mm KCl, 0.1 mm CaCl2) was added. Modulations of single channel currents in response to applied solutions of TEA were then recorded.

PAGE 145

145 Lipid Vesicle Formation Chloroform solutions of the lipids (50 mg /mL) were mixed to a 7:3 molar ratio (DPhPC/DPhPE) followed by rotary evaporati on until a dry lipid film was obtained at the bottom of the vessel. The dry lipid film was then placed under vacuum in a dessicator overnight to eliminate any residual chlo roform. The lipids were then lyophilized before hydration for 1 hour at 50 C in buffer (5mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4) to a final concentration of 2 mg/mL. After being cooled to room temperature the suspension was sonicated for 5 min, and extruded (21 passes) through a 100 nm polycarbonate membrane in a mini-extruder (Av anti Polar Lipids Inc.). The vesicle size profile was determined by the use of a dynamic light scatteri ng (DLS) instrument (Brookhaven Instruments Corporation, Holtsville, New York, USA). Typically vesicle sizes of (140) nm were obtained as computed by the BI-DLSW Dynamic light scattering software. Preparation of the Tethered Bilayer The device was immersed into a DPTL-ethanol solution (0.3 mg/ml) overnight in a saturated ethanol environment (closed ethan ol bath). The wafer was then rinsed with ethanol and millipure water. A drop of the lip id vesicle suspension was deposited on the DPTL treated sensor pad and left for >8 hours at 4 C allowing for the vesicle to fuse therefore forming the tethered bilayer memb rane. Prior to taking electrophysiological measurements a 5 l droplet of buffer solution (5 mM MOPS, 250 mM KCl and 0.1 mM CaCl2, pH 7.4) was applied to the treated s ensor pads. Due to evaporation of water during the measurements, the droplet was regularly refreshed with pure water to keep a constant volume of the buffer.

PAGE 146

146 Characterization of tBLM Formation and BKCa Membrane Insertion The tBLM formation on gold substrates wa s investigated by surface plasmon resonance (SPR) enhanced ellipso metry and visualized by atomic force microscopy (AFM). The SPR instrument used was from Nanofilm technologie GmbH, Goettingen, Germany and for analysis of t he kinetic fits, AnalysR softw are was used. Linearly plane-polarized light wa s directed through a 60 equilateral prism coupled to a gold coated glass slide via diodomethane oil as an index matching fluid in the Kretschmann configuration. The gold sli de had a self-assembled monolayer of DPTL deposited on the surface by use of a similar methodology as on the MEA devices. Vesicles were injected into the flow-cell and the flow stopped for 30 mi nutes to 1 hour to allow for vesicle fusion to occur on the surface. The formation of the lipid bilayer was monitored by recording psi ( ) data versus time. Buffer solution (5 mM MOPS, 250 mM KCl and 0.1mM CaCl2, pH 7.4) was flowed to remove free vesicles. A solution containing BKCa proteoliposomes was then flowed into the sample cell and up on stabilization of the SPR signal, more buffer solution was allowed to flow to remove any unincorporated proteoliposomes. Kinetic information associated with channel incorporation was then obtained. Results Membrane Insertion of BKCa Channels The assembly of a tethered bilayer membrane on a gold substrate and subsequent incorporation of the BKCa channel involves three steps which include formation of the DPTL self-assembled monolayer, fusion of vesicles to form a tethered bilayer membrane and finally the memb rane insertion of the BKCa channel. The process of the formation of a functionalized tBLM on the go ld substrate was analyzed by SPR and the fit illustrated in Figure 5-2 was obtained.

PAGE 147

147 9 11 13 15 17 19 21 0 20 40 60 80100120140160180200 Time (min)Psi ()Buffer rinse Rinse BKCa Channel addition Liposome addition Figure 5-2. Fitted Kinetic data for incorporation of the BKCa Channel in the tBLM. The Y-axis shows the value and changes resulting in bilayer formation and incorporation of the channel. The X-axis shows the time -scale for the process. The Y-axis shows the response in the SPR signal with changes occurring at the surface. Buffer solution was allowed to fl ow through the sample cell for ~30 min to establish a stable baseline signal. Liposomes prepared from 7/3 DPhPC: DPhPE lipid vesicles were then injected into the flow ce ll and the flow stopped for ~1 hour for vesicle fusion to occur on DPTL. A 7.5 change in the value was observed, corresponding to adhesion of vesicles followed by rupture to form a bilayer. Free vesicles were washed away leaving a tBLM; a minimal reduction in the value was observed. The SPR kinetic data, suggests that vesicle fusion to form a bilayer required about 1 hour for completion. The bilayer was rinsed, and judging by the stability of the SPR signal as can be observed in Figure 5-2 between 105 120 min it was clear that a stable membrane was formed. Proteoliposomes were injected in the flow cell leading to an increase in

PAGE 148

148 thickness equivalent to 2.5 degrees. A final rinse resulted in 40 % reduction of the value as the channels that do not partition in to the bilayer were washed off. Figure 5-3 illustrates the steps involved in tBLM formation and membrane insertion of channels. Figure 5-3. Schematic showing tB LM formation and incorporation of BKCa channels. A) Injection of vesicles on the SAM. B) Vesicle fusion to form tBLM and rinse of unbound vesicles. C) Addition of BKCa proteoliposomes on the tBLM. D) Incorporation of BKCa channels into tBLM. A D C B

PAGE 149

149 Electrophysiology In order to perform electroph ysiology studies of ion channels within the tBLM, it is imperative that a stable, high electr ically-resistant membrane be formed with resistances of 1 gigaohm (G ). Single-channel events ar e represented by unitary currents in the picoampere range and leakage currents between the recording electrode and ground had to be minimized. The tethering of the bilayer established an ionic reservoir that allowed the flow of ions across the membrane through the ion channel from the bulk solution above the membrane, to the reservoir below, and vice versa, in response to changes in transmembrane potential. The SPR experiments above served as a c ontrol for tBLM formation and channel incorporation on the MEA. A drop of vesicle suspension was added on the MEA device, and after fusion of vesicles to form a b ilayer over a period of 2 hours at room temperature, buffer solution (5 mM MOPS, 250 mM KCl, 0.1 mM CaCl2, pH 7.4) was added to the sensor pad contai ning the tethered membrane. The limited size of the gold sensor pad surface (100 m2) and the electrical stability of the overlying lipid bilayer membrane made each pixel on the M EA sensitive enough to measure single ionchannel currents in the picoampere r ange. A comparison was made between the membrane resistance before and after BKCa channel incorporation into the tBLM. In both cases, the membrane resistance was above 1 G therefore suitable for electrophysiological measurements. As shown in Figure 5-4, although changes in membrane potential gener ated ohmic changes in membr ane potential consistent with the resistance of the membrane as a whole, no transient current steps consistent with single channel activity were recorded in the absence of BKCa. Figure 5-4 shows a stable trace recorded testing the bilayer memb rane with different applied potentials.

PAGE 150

150 Figure 5-4. Recording from an electrica lly stable membrane formed by fusion of 7:3 DPhPC: DPhPE vesicles. The reco rding was done for approximately 6 minutes with applied voltages of between -100 mV to +120 mV. Membrane potentials ranging from -100 mV to +120 mV were applied to the tBLM and stable traces were observed throughout t he measurement duration. Pore-formation (electroporation) as a result of applied membrane potentials during electrophysiological experiments can complicate single-channel analysis. Electroporation is a phenomenon observed especially in bilayers supported at the air-water interface in experimental techniques with the tip-dip conf iguration. Within the tBLM system, stable traces were observed in several runs of experiment s with giga-ohm seals ranging from 2.3 G to 4.6 G at applied potentials of 120 mV without evidence of membrane defects. The control recording in Figure 5-4 was obtaine d under recording conditions identical to those used during actual sing le-channel measurements. The ti ght sealing and electrical stability of the tBLM without evidence of l eakage currents demonstrat es that the system

PAGE 151

151 is well suited for protein incorporation. Functional studies under a defined electric field are possible for inserted channel proteins and other pore-forming peptides within the tBLM. This experimental set-up therefore shows electrical pr operties similar to those of SLBs but with enhanced stability. Single-Channel Analysis of the BKCa Channel Proteoliposomes of the BKCa channel were incorporated into the tBLM and singlechannel activity observed following changes in transmembrane potential. Fluctuations and single-channel records reflect the time course of gating steps. The single-channel activity exhibited by ion channels is a st ochastic process; ther efore, the order and duration of gating events are random variables Statistical distributions were therefore employed to measure and quantify the parameters representative of gating events. The open and closed probabilities observed illustrated the voltage dependence of the channels. Figure 5-5 shows single channel activity of the chimeric mRFP1-tagged BKCa channel at applied membr ane potentials of 120 mV. 02000400060008000 -6 -4 -2 0 2 4 6 8 10 12 14 Current (pA)Time (ms) Open Closed -4-2024681012 0 1000 2000 3000 4000 Counts #Time (s) Current (p A ) Open Closed Figure 5-5. Single-channel activity of BKCa channels at 120 mV. A) Single-channel activity in the tBLM configuration. B) Histogram of open/close probabilities. A B

PAGE 152

152 When potentials of +120 mV were applied, open and closed events were observed with current amplitudes of 8 pA upon normalizati on of the closed state to 0 pA as can be observed in Panel A of Figure 5-5. Singl e-channel histograms quant ifying the number of events were fitted to a Gaussian distributi on as can be observed in Figure 5-5 B and recombinant BKCa channels were found to be open ~89 % of the time at +120 mV. A reduction in applied potential resulted in a commensurate reduction in the open probability and vice versa; a characteristic property of voltage-dependent ion channels like the BKCa channel. For the experimental run represented by the trace in Figure 5-6 the applied potentials were reduced to 80 mV and open and closed events analyzed. Open Closed Closed Open Figure 5-6. Single-channel activity of BKCa channel in tBLM at 80 mV applied voltage. A) A single-channel trace showing open/close events. B) Histogram showing the open/close probabilities at 80 mV. After normalization of the closed state to 0 pA, the current amplitude in Figure 5-6 A is at the 4.5 pA level, corresponding to a reduction of ~ 43% in comparison to currents observed at applied potentials of +120 mV Additionally, the open probability of the channel was reduced. From a total of approximately 9 3 single-channel events, the open interval observed corresponds to ~ 3.8 103 events (29.6%) at this state, as can be observed in the Gaussian distributi on fit illustrated in Figure 5-6 B. A B

PAGE 153

153 The recordings of single-channel event s observed in Figure 5-5 and Figure 5-6 demonstrate voltage dependence of the open probability of the BKCa channel. Singlechannel recordings were also performed at negative voltages eliciting response of the BKCa channel with current amplitude exhibi ting voltage dependence similar to that observed at positive potentials. The data in Figure 5-7 below shows fluctuation measurements of a single BKCa channel incorporated in to the membrane with recordings obtained at applied potentials of -50 mV. Open Closed -2 -1 0 0 400 800 1200 Counts (#)Current (pA)Open Closed Figure 5-7. Single-cha nnel activity of the BKCa channel in tBLM at 50 mV applied potential. A) Single-channel open/close events. B) Open and closed probabilities. There is a clear difference in the magnit ude of open times at negative voltages in comparison to the experiments above run at +120 mV and + 80 mV. From panel B in Figure 5-7, it can also be noted that the number of open and shut events is significantly lower (1.2 3) in comparison to the counts of ev ents during recordings performed at positive potentials (4 3 and 9 3). This observation is consistent with the space limitation within the hydrophilic volume of the sub-membrane region in the tethered bilayer membrane system resulting in a reduc tion in the volume of ions that can be contained in this region and available to traver se the channel at negative potentials. The A B

PAGE 154

154 percentage of open time at app lied potentials of -50 mV was ~10%. In all experimental recordings, the channels responses to applied voltages consisted of unitary currents in the picoampere (pA) range and a time domain response at frequencies in the kilohertz (KHz) range, with the time-scale of channel openings in the millisecond range. BKCa Channel Open Probability The single-channel current histograms fitt ed to Gaussian distributions in Figure 55, Figure 5-6 and Figure 5-7 show a volt age dependence of the open probability. Being voltage dependent, the open probability of the BKCa channels could be fitted with a Boltzmann distribution by the equation below and represented in the Figure 5-8: Popen = 1/(1 + exp{V1/2 V)/k (5-1) Where Popen is the fraction of open channels, V1/2 is the half-activation potential, k is the e-fold increase of the open probability. -40-20020406080100120140 0.0 0.2 0.4 0.6 0.8 1.0 Voltage (mV)P(open) Figure 5-8. Open probability versus applied voltage fitted to a Boltzmann distribution, Popen = 1/(1 + exp{V1/2 V)/k, where V1/2 = 70.2 16.9 mV and k = 12 5 mV. The Boltzmann equation dictates the ra tio of open to closed channels at equilibrium in terms of energy change, and explicitly gives the voltage dependence of

PAGE 155

155 gating in the system.1 The half-activation potential (V1/2) observed here was 70.2 16.9 mV representing approximately 85 mV of a sh ift compared to that observed in native membranes where V1/2 values of -19 mV were measured.200 The e-fold increase of the open probability k = 12 5 mV is comparable to the k = 10 15 mV obtained for channels in native membranes under comparable conditions.38 Event detection was performed by the half-amplitude threshold met hod and curves fitted for the investigation of channel kinetics and measurement of amplitudes. When fitted with three exponentials, the open dwe ll-time distribution (Vapplied = 100 mV) resulted in the lifetimes, o,1 = 0.6, o,2 = 7.5 and o,3 = 12. The closed state fitted with three exponentials yielded lifetimes of c,1 = 0.9, c,2 = 4.2 and c,3 = 56 ms as determined by Clampfit 9.1 data analysis software. The lifetimes for the mRFP1-BKCa channel constructs all fell within the millisecond (ms ) time range. Figure 5-9 illustrates the logarithm histograms of square root ordinates that bin dwell times of all intervals and the fitted lifetimes from three exponentials. -0.50.00.51.01.52.02.5 0 5 10 15 20 25 Log Time (ms)Counts / Bin -0.50.00.51.01.52.0 0 5 10 15 20 25 Log Time (ms)Counts / BinOpen lifetimes O,1= 0.6 ms O,2= 8.0 ms O,3= 12 ms Closed lifetimes C,1= 0.9 ms C,2= 4.2 ms C,3= 56 ms Figure 5-9. Logarithm histogram of square root ordinate that bins dwell-times of all intervals and fitted lifetimes from three exponentials.

PAGE 156

156 BKCa Conductance Among the major characteristic electrical properties of the BKCa channel is the large conductance of 90-300 pS in 100 mM symmetrical K+.17 The slope of the linear fit of the I-V curve in Figure 5-10 repres ents the conductance of the BKCa channel. -150-100-50050100150 -10 -5 0 5 10 Current (pA)Applied Voltage (mV) Figure 5-10. Current-volt age relationship for the BKCa channel. The data in the figure was obtained from five experiments and the straight line is the linear fit through origin. The data reflects the apparent non-linearity of the current-voltage relation in the channel as opposed to the perfectly linear slope expected from the Ohms law. The curvature observed in this case likely stem s from the asymmetry in concentrations of bathing ions which would explain differences in ion flow between large positive voltages and large negative voltages through the BKCa channel incorporated in the tBLM. The current-voltage relationship in Figure 5-10 illustrates the linear function of the BKCa channel current amplitude with ap plied potential. Similar exper imental runs with different membrane resistances (seals) were fitt ed and the Ohms law used to calculate

PAGE 157

157 conductance. Calculations of this data using the Ohms law yield a unitary conductance value of ~ 50 pS which shows an apparent reproducibility over similar repeated experimental runs. The apparent mean conduc tance (n = 5) after fitting similar experiments from different seal resistances was 39 2 pS, a value which is considerably lower than the conductance typi cally obtained from the wild type full-length BKCa channel in other planar systems.200 The unitary conductance observed here for the chimeric BKCa channel was ~ 40 % greater t han the mean conductance, with the discrepancy comparable to that obser ved in patch clamped ion channels.201 Pharmacology Blockers of the BKCa channel have long been used as experimental tools for elucidating structural properti es of the channel and for probing the mechanisms of their action under various physiological conditions.202, 203 It is generally accepted that in selective ion channels, ions must interact with specific chemical sites which are responsible for the bias in the permeation of some ions and not others. Some of the important sites for ion interaction are energy wells, where ions may bind stably, or energy barriers which are points of maximum resistance to ion passage.204 All channel blockers tend to have a common mechanism of action whereby they completely obstruct channel current for a short durati on of time. The tetraethylammonium (TEA+) ion belongs to a group of compounds know n as quaternary ammoniums (QAs) which have been known to block K+ conductance in voltage-gat ed and inward rectifying potassium channels by binding to two dist inct sites which are accessible from the intracellular and extracellular space.205-207 For studies of the response of the BKCa channel in a tBLM to tetraethy lammonium, solutions with vari ous concentrations of TEA were used and observations of electrophysiol ogical responses made. Representative

PAGE 158

158 data are documented in Figure 5-11 below, of t he control current traces in the absence of TEA, and the response obtained after ap plying solutions of TEA of different concentrations (100 M, 250 M and 500 M). Membrane potentials of 100 mV were applied for each experimental run. 400 ms5 pA Figure 5-11. Single channel traces of channels at 100 mV applied voltage under different concentrations of TEA soluti ons. The control experiment had plain buffer without TEA and the others had TEA at 100 M, 250 M and 500 M as shown on the legend. Three categories of channel blockers can be described based of the appearance of single-channel recordings that they elic it. Slow blockers produce clearly resolved interruptions to channel opening that clos ely resemble the closing of channels, intermediate blockers produce rapid fluctuations in current that are too brief to resolve as individual events, and fast blockers pr oduce frequent, extremely brief interruption of channel current that can only be detected as an apparent reduction in the level of open channel current.204 From the results doc umented here after repeated experiments, it is clear that reconstituted BKCa channels in tBLMs demonstrate typical decrease of current amplitude upon the addition of micromolar and millimolar concentrations of CONTROL 100 M 500 M 250 M

PAGE 159

159 tetraethylammonium solutions. These results correspond to the known characteristics of TEA which exhibits fast kinetics theref ore demonstrates blockade evident by an apparent reduction in open current amplitudes as in Figure 5-11. A dose-response curve was plotted to represent the relati onship between the perc entage block of the mRFP1-tagged BKCa channel by TEA versus the concentrations of applied solutions of TEA as illustrat ed in Figure 5-12. 0100200300400500 0 20 40 60 80 100 TEA Concentration (M)% Blockade Figure 5-12. Dose-response curve of the BKCa channel mutant. Plot of percentage block of the BKCa channel versus concentrati on of TEA fitted to the Hill equation (solid line). Response of TEA block was plotted in a dose-response curve and fitted to the Hill equation: (5-2) The half-activation concentration K1/2 at 100 mV of applied membrane potential was determined to be 60 10 M (n= 3) wh ich is lower than the half-activation concentration for TEA of K1/2 = 158 M determined for the BKCa channel under patchclamp conditions.200 The location of the TEA binding site relative to the tBLM and hence %100* 1 %Block N DTEA K

PAGE 160

160 the apparent orientation of the channel in the membrane can be inferred from the voltage dependence of block. Voltage-dependence of blockade of channels at applied potentials of 120 mV, 95 mV, and 80 mV in 500 M TEA solutions was investigated. Figure 5-13 shows the volt age-dependence of TEA block. Figure 5-13. Voltage-dependence of block ade by 500 M tetraethylammonium (TEA). A) Block by TEA at 120 mV with a current amplitude of 8.5 pA. B) Block by TEA at 95 mV and current amplitude of 6 pA. C) Block by TEA at 80 mV with current amplitude of 4.5 pA. A B C

PAGE 161

161 When membrane potentials of 120 mV were applied on the tBLM, current amplitudes corresponded to levels of approxim ately 8.5 pA. At memb rane potentials of 95 mV the current amplitude was 6 pA, whil e at 80 mV the curr ent amplitude was approximately 4.5 pA (Figure 5-13). The observed voltage dependence of blockade is associated with an internal TEA binding site, since external TEA block would usually show almost no voltage dependence.208 Tetraethylammonium ions do not partition through the membrane and any elicited responses observed are strictly as a result of antagonist bound in the respective site of app lication, in this case internally on an inside-out BKCa channel. Evidence of an internal bind ing site being responsible for the pharmacological responses observed here suggests that the mutant BKCa channel was incorporated into the tBLM in the inside-out configuration; that is, with the C-terminal region oriented towards the external bulk so lution and the N-terminus oriented proximal to the gold substrate within the tethered domai n constituting the ionic reservoir beneath the membrane. Discussion The results in this chapter detail the incorporation of recombinant BKCa channels into tethered bilayer lipid membranes on microelectrode array devices and the subsequent functional characterization in these novel membrane systems. The BKCa channels under study here had a C-terminal delet ion just after the S6 segment and an mRFP1 fluorescent tag attached for purposes of tracking expression and determining the configuration of the channel within the membrane. A histidine tag was added to facilitate purification, as discussed in great detail in Chapter 2 of this dissertation. Control experiments using the surface plasmon resonance spectroscopy techniques were performed to demonstrate the viability of the incorporat ion of channel proteins in

PAGE 162

162 microelectronic devices. A modified patch cl amp electrophysiology technique was used to perform single-channel recordi ngs of the channels in the tBLM. The findings in this study indicate that tethered bilayer membranes were successfully formed on the MEA devices by fusion of DPhPC: DPhPE vesicles on the DPTL self-assembled monolayer and recombinant BKCa channels incorporated in the tBLM as evidenced by SPR ki netic experimental results. The bilayer membranes formed were electrically stable with resistances in the gigaohm range, ther efore suitable for electrophysiological measurements. Functi onal analysis probed by measurements of single-channel activity in the tBLM syst em showed that for the most part, the recombinant BKCa channel had electrical properties that were comparable to those observed in the wild-type (full length) BKCa channel. The channel also exhibited fast kinetics and lifetimes in the microsecond ti me-scale as observed in similar channels studied under patch clamp techniques.38 However, analysis of the recorded data showed a significant reduction in the conduct ance of this channel in the tBLM in comparison to published data obtained using other measurement c onfigurations. The Cterminal deletions of the recombinant BKCa channel and the atta chment of mRFP1 and histidine tags are ruled out as possible causes for this sub-optimal conductance because similar conductance ranges of 35 pS 50 pS have been observed in wild-type BKCa channels in the tBLM system.99 Additionally, in Chapter 3 of this study, the conductance calculated from macroscopic currents obtained by TEVC of oocytes expressing mRFP1-tagged BKCa channels were ~ 129 pS, well within the typical range (90 300 pS) expected for BKCa channels.

PAGE 163

163 Incidentally, the mechanosensitive channel of large conductance (MscL) and the synthetic ion channel based on the pore-lining domain of t he nicotinic acetylcholine receptor (nAChR) of Topedo californica the M2 channel, studied independently in our laboratory under the same tBLM model me mbrane system also showed significantly lower conductances than observed in under patch clamp and whole cell systems.99,97 The exception to this trend was observ ed in gramicidin (gA), a 2 kDa linear pentadecapeptide consisting of 15 hydrophobic alternating D, L-amino acids known to induce a high permeability for small monovalent cations in natural and artificial lipid membranes.160 Within the tBLM, the four ion channels, recombinant and wild type BKCa, M2 and MscL exhibited conductances that we re between 3 times lower to an order of magnitude lower than those observed in pat ch clamp or whole cell measurement configurations, while the gramicidin (g A) channel had compar able conductance ranges in the tBLM relative to other measurement configurations, the data shown in Table 5-1. There is an apparent correlation between the relative sizes of the ion channels incorporated into the tBLM and their conduct ances as calculated from the Ohms law based on obtained data. The full-length BKCa channel and the C-terminally deleted recombinant BKCa channel with mRFP1 and histidine fusion tags have molecular weights of 120 kDa and 70 kDa respectively.209 The native E. coli MscL channel has a molecular weight of 60-80 kDa and monomeric molecular weight of 15 kDa while the M2 ion channel with a molecular weight of 65 kDa.210, 211 These four ion channels have much higher molecular weights than the gram icidin (gA) channel and all of them have relatively bulky domains extending both in tracellularly and extracellularly which increases on the packing dynamics in t he reservoir region below the membrane

PAGE 164

164 subsequent to incorporation in the tBLM. These structural aspects may well be responsible for the observed low conductances in the channels with high molecular weights. The ionic reservoir region within the te thers in the tBLM is known to be densely packed and with a higher viscosity than the bulk solution above the membrane.93 Table 5-1. Comparison of the different i on channels incorporated in the tBLM and their molecular weights Ion Channel Conductance (pS) (Literature) Conductance (pS) (tBLM) Molecular Weight (kDa) (Literature) BKCa Channel 90 300 30 50 (99) 120 kDa (99) MscL Channel 100 3000 (212, 213) 50 300 (97) 17 kDa (213) M2 15 45 (214) 13 15 (99) 15 kDa (214) Gramicidin (gA) 20 90 (88, 215, 216) 50 90 (160) 2 kDa (102) The additional volume within this region as a result of the incorporation of the bulky ion channels limits ionic mobility even furt her and prevents the free flow of ions in both directions. Because the gramicidin (gA) channel is much smaller, and has a significantly lower molecular weight in com parison to the other three channels, it does not affect ionic mobility, hence its conductance compares favorably with values obtained from experiments done using conventi onal patch clamp setups. In previous work performed in the Duran laboratory,99 the BKCa mutant construct studied here was inserted into the tBLM directly from Xenopus laevis oocyte membrane vesicles, and therefore differs from the c onfiguration examined her e, in terms of the immediate lipid environment of the channel protein: native lipids in the former case and phytanoyl model membranes in the latter. T he results show that the purified mRFP1tagged BKCa channel construct studied in the tBLM showed electrical properties comparable to channels in native membranes within this measurem ent configuration. Analysis of open probabilities showed that the BKCa channel in native membranes had a

PAGE 165

165 half-activation potential of 69 5 mV while the mutant cons truct had a half-activation potential of 70.2 16.9 mV. Both channels had lifetimes in the millisecond range and pharmacological profiles that were statistically indistinguishable. BKCa channels in native membranes and the mutant constructs had mean conductances of ~ 40 pS, both being at least 55 % less that expected for BKCa channels studied by patch clamping.27, 200 These similarities in electrical properti es are rather intrig uing considering the distinct differences in compositio n between native membranes and the model membrane system (tBLM). The phytanoyl li pids used for preparation of model membranes in this study constituted 70% phosphatidylcholine (PC) and 30 % phosphatidylethanolamine (PE) On the other hand, the percentage of total phospholipids in X. laevis oocyte membranes includes 65 % PC and 19 % PE, while the rest is composed of phosphatidyl inositol (PI), phosphatidyl serine (PS), sphingomyelin and cholesterol.172 A possible explanation for similariti es in electrical properties between these channels in two contrasting lipid envir onments would be because of the nature of the phytanoyl lipids used in the model membranes. Phytanoyl lipids have branched, saturated acyl chains that induce me mbrane packing dynamics similar to those observed when cholesterol molecule s are added to lipid bilayers.167 Because native X. laevis oocyte membranes have some cholesterol, it is likely that channels in native membranes and phytanoyl model membranes woul d show similar functional properties. The characteristics of the reservoir within the tBLM and its effects on the electrical properties of the membrane possibly a ccount for the reduced conductance of incorporated BKCa channels. Of great significance is the interfacial capacitance at the gold substrate. The use of tetraethylene glycol chains as tethering moieties for lipid

PAGE 166

166 bilayer membranes to gold surfaces results in a polar layer between the membrane and gold surface which may sequester ions and act as a reservoir for ions translocated across the tBLM.102 The gramicidin (gA) channel exhibits its known conductance unlike the other bulkier channels despite the effects of the interfacial capacitance as shown in Table 5-1. This is so because, unlike with BKCa ion channels, the ionic conduction of gA channels is not strongly dependent on membrane potential but is nonli nearly dependent on concentration, therefore, t he mobility and density of the ions within the reservoir can directly limit the ion flux through the gramicidin channel.217, 218 Because the ionic concentration of buffer solutions used is high, the conductance of gA in the tBLM remains unaffected. A possible approach to be considered in order to correct the anomaly observed as regards conductance of BKCa ion channels in the tBLM w ould be to experiment with different types and lengths of polar tethers for the membrane. The conductivity of the channels to various ion types depends on t he type of tether us ed for anchoring the membrane to the gold substrate, which in turn affects the region in the reservoir which is in close proximity to the gold surface and would help maintain the ionic reservoir properties to mimic those of the bulk elec trolyte solution as much as possible. Conclusions In this chapter, proteoliposomes containing BKCa channels were fused to a tethered bilayer lipid membrane and functi onal analysis of incorporated channels performed. Control experiments showed the viability of incorporation of channels in this model membrane system and functional studies showed typical results as far as the electrical properties of t he channels are concerned albeit with a deviation in expected conductance. The results highl ighting the function of BKCa channels in a tBLM are in

PAGE 167

167 good agreement with data obtaine d using other measurement configurations such as whole-cell and other planar patch -clamp systems. Lifetime and kinetic studies of the channel are comparable to those obtai ned and documented elsewhere and these suggest that the BKCa channel in the tBLM functions as expected and is suitable for any envisaged applications that would require a so lid supported receptor. However, further studies are required to improve the system a nd provide a close mimic as possible to the natural membrane by altering the tethering units to yield a reservoir environment similar to the bulk electrolyte soluti on above the tethered membrane. Pharmacological studies performed on the channel in the tBLM focused on probing the response to micromolar concentra tions of tetraethylammonium show that the mRFP1 tagged BKCa channel within this configurati on is as sensitive to TEA as those channels in native membranes studied using conventional patch clamp techniques. Furthermore, response to TEA also reveals the apparent preferred orientation of the channel in the tBLM with signals recor ded showing characteristic features of an inside-out BKCa channel evidenced by t he voltage dependence of TEA blockade. It is clear that besides possibly influencing the orientatio n of the recombinant BKCa channel in the tBLM, the mRFP1 and hist idine fusion tags do not compromise the activity of the channels incorporated in t he tBLM. Such mutant channels can therefore be modified to detect an even greater variet y of chemical analytes and purified for incorporation in these novel model me mbrane systems for numerous applications limited only by imagination.

PAGE 168

168 CHAPTER 6 VOLTAGE-INDUCED GATING OF THE MECHANOSENSITIVE CHANNEL OF LARGE CONDUCTANCE (MSCL) IN TETHERED BILAYER LIPID MEMBRANES Introduction Mechanosensitive channels (MS channels) are pore-forming proteins found in prokaryotic and eukaryotic cell membranes where they open for ion conduction in response to mechanical stress. MS channels react to mechanical stimuli triggering a shift in equilibrium from a closed to open conformation of the protein.97 Stimulation of MS channels is responsible for physiologic al processes such as pain sensation, micturition, touch, hearing, salt and fluid balance, and turgor pressure changes in cells.219 MS channels can generally be classified based on conductance into three categories, the 3 pS mechanosensitive chan nel of large conductance (MscL), and two smaller conducting MS channels, the mechan osensitive channel of small conductance, MscS (~1 nS), and the MS channel of mini conductance, MscM (0.3 nS). The most studied MS protein is the Escherichia coli MscL which typically serves as the model system for mechanosensory transductions. The E. coli MscL Structure and Function When a bacterial cell is exposed to hypoosmotic conditions, MscL senses the tension in the membrane causing it to open a conducting pore releasing osmolytes. MscL opening therefore allows the channel to serve as an emergency release valve preventing membrane rupture during an osmotic downshock.123, 220, 221 X-ray crystallography with a resolution of 3.5 by the Douglas Reess group revealed the MS channel from M. tuberculosis to be a 17 kDa homopentamer with a width of approximately 50 in the plane of the me mbrane and a height of 85 Furthermore, MscL was demonstrated to have two transmembrane domains (TM1 and TM2) joined

PAGE 169

169 by a periplasmic loop and bearing cytoplasmic N and C termini.222-224 Figure 6-1 shows the crystal structure of the MscL from M. tuberculosis adopted from the protein data bank showing the N and C termini and the transmembrane domain. linker Figure 6-1. Crystal structur e of MscL from M. tuberculo sis adopted from the protein data bank (PDB) showing the two transmembrane domains and positions of the N and C termini. The TM1 from each peptide forms the 30-40 pore in response to tension in the membrane bilayer, releasing ions and other solutes from the cytoplasm while the TM2 domain faces the lipid membrane.222 Gating in the MscL Bacterial MS channels gate by sensing tension in the memb rane independent of other proteins or artifacts allowing them to serve as a simple, easy-to-gate molecular model for functional studies. The structural changes that occur during gating are not known, but the crystal structure of a nearly closed state of the M. tuberculosis homologue and computer simulations have resu lted in models for the protein in its

PAGE 170

170 closed and open state.213, 225-227 Mutagenesis studies have indicated that the periplasmic loop plays the role of a torsional spring, inhibiting the channel from gating under non-osmotic conditions.228 The role and conformati on of the two cytoplasmic domains, the N-terminal and the C-terminal during gating are not well understood. However, the C-terminal region consists of a linker area starting just after TM2, and the -helical bundle termed S3-bun dle which is involved in conformational changes during gating. A detailed evaluation of the S3-helical bundle obtained from the crystal structure revealed a conforma tion in which the hydrophobic side-chains were exposed outside the S3 helix with the hydrophilic and charged side-chains packed inside the bundle. Gating of MscL has been successfully achieved using patch clamping techniques using a pressure stimulus to induce tension in the membrane.97, 229 Although gating has been shown to be voltage dependent,97, 229 the amount of voltage required to induce a gating membrane tension (~300 mV) is not pr actically achievable under normal patch clamp conditions without c ausing membrane defects, thus gating of the channel solely by voltage-induced membrane tension had not been previously reported. Presented here is the novel technique util izing tethered bilaye r lipid membranes (tBLM) to measure single channel activity of the MscL channe l. The tBLM is part of an engineered microelectronic array chip and is the same system used for incorporation and functional analysis of BKCa channels in Chapter 5 of this dissertation and illustrated in Figure 5-1. The tethers work as stab ilizing anchors and serve as spacers to the underlying gold surface thereby forming an ionic reservoir allowing for ionic flow through the channel. The tBLM system here is also assembled using the phytanoyl lipids 1,2-

PAGE 171

171 diphytanoylsn -glycero-3-phosphocholine (D PhPC) and 1,2-diphytanoylsn -glycero-3phosphoethanolamine (DPhPE) illustrated in Figure 1-12 of this dissertation. Phytanoyl lipids have branched acyl chains that have been demonstrat ed to confer electrical stability to the membrane, and have been us ed in previous investigations for electrophysiological studies where a membrane with stability to mechanical shocks and vibrations is desirable. The work in this chapter represents analysis of MscL channel activity for the first time in a supported bilayer membrane. Materials and Methods Lipids and Chemicals Archaea analogue phosphol ipids 1,2-diphytanoylsn -glycero-3-phosphocholine (DPhPC) and 1,2-diphytanoylsn -glycero-3-phosphoethanolamine (DPhPE) in chloroform were purchased from Avanti Polar Lipids Inc. (Alabaster, AL, USA), and were used without further purification. 2,3-diO -phytanoylsn -glycerol-1-tetraethylene glycol-D,L-lipoic acid ester (DPTL) was synthes ized as described in literature and supplied by a proj ect collaborator.101 Morpholinopropanesulfonic acid (MOPS) (>99.5%, Fluka, ultra grade). Potassium chloride (certifi ed ACS grade), calcium chloride (certified grade) and potassium hydroxide (ACS cert ified grade) were purchased form Fisher Scientific and used as received. The Microelectrode Array Device The microelectrode array device was fabr icated by Daniel Fine in Dr. Ananth Dodabalapurs laborator y at the Microelectroni cs Research Center at the University of Texas at Austin, Austin, TX. The fabricati on of the MEA devices was on silicon wafers each of which contained 66 pixels. Each s ensor pad was formed by evaporation of 3 nm of Ti on the silicon substrate followed by a 500 nm thick alloy of Au (60 %)/Pd (40%). A

PAGE 172

172 200 nm thick layer of pure gold was then depos ited on top of the alloy to form the surface of the pad, and a polyimide resist wa s photolithographically ap plied to define the pad size. The sensor and probe pads formed on the gold device were linked by a thin line of gold and were electrically isolated from each other by the polyimide resist. Before the formation of the tBLM, the device was rinsed with hexane, acetone, ethanol and water respectively. This was followed by UV/ozone treatment of the gold surface in a Harrick PDC-32 G plasma cleaner/sterilizer at high radio-frequency power for 20 minutes under a flow of oxygen for oxidation, and for the removal of both chemical and organic contaminants. MscL Isolation and Purification Expression and purification of the MscL was done in Dr. Paul Blounts laboratory at the University of Texas Southwestern M edical Center. In order to isolate and purify MscL, the protein is first generated with a Cterminal 6His-tag and expressed in the mscL-null strain PB104 using the pET21a expr ession vector. The pET vector is able to mass-produce proteins due to its T7-promoter site that binds T7 RNA polymerase and promotes transcription of the ta rget protein. Secondly, cells were grown to an OD of about 0.6 and induced with 1mM IPTG for 1 hour at 37C. The cells were then centrifuged at 4000 rpm and the pellets were re-suspended in 40 ml of base buffer (10 mM NaPi and 300 mM pH 8.0) and 100 L of protease inhibitor cocktail (His-tag Protease Inhibitor from Sigma) was added to prevent the denaturing of the proteins. A few crystals of DNase and lysozyme was added and allowed to mix for 45 minutes. The samples were then disrupted by French pr essing at 16K PSI at 4C and membrane fractions were separated and resuspende d in extraction buffer (50 mM Na2HPO4/NaH2PO4, pH 8.0, 300 mM NaCl, 20 mM imi dazole plus 2% (v/v) Triton X-

PAGE 173

173 100).202 The resulting suspension was incubat ed with 500 L of pre-equilibrated with extraction buffer Ni-NTA agarose (Qiagen) for 30 mi n at room temp to bind the proteins. To prepare for the protein pur ification by affinity chromatography, the NTA-agarose matrix was loaded into a 10 cm column with the protein bound. The column was washed with 10 mL of extraction buffer in the presence of 50mM of imidazole and 1% Triton X100. Proteins were then eluted in 1.5 mL extraction buffer with 200mM imidazole and 0.2% Triton X-100. They were lastly analyzed by SDS-Page gel for purity and stored at 80C by flash freezing. Preparation of the Tethered Bilayer The device was submersed into a DPTLet hanol solution (0.3 mg/mL) and left overnight in a saturated ethanol environment (closed ethanol bath). The wafer was then rinsed with ethanol. A chloroform solution (2 mg/mL) containing DPhPC:DPhPE 7:3 molar percentage was mixed and the so lvents were evaporated under vacuum and hydrated in 1mL of water and heated at 50 C until a clear solution was obtained (ca. 1 hour). After cooling to room temperature the suspension was sonicated for 5 min, followed by filtration through a 0.45 m filter. The vesicle size was determined by dynamic light scattering (DLS) instrument using a 90 Plus/BI-MAS detector with a BI 9000AT digital correlator (Brookhaven Inc.). Ty pically vesicle diameters of 150 nm were obtained. A drop of the vesicle solution was deposited on the DPTL treated sensor pad and left for >8 hours at 4 C allowing for the vesicle to fuse and form the tethered bilayer. Prior to the electroph ysiology examinations a drop (5 L) of buffer solution (5 mM MOPS, 250 mM KCl and 0.1 mM CaCl2, titrated to pH 7.4 using KOH) was applied to the treated sensor pads.

PAGE 174

174 Electrophysiology Recordings of single channel activity we re obtained using two different types of electrodes. The active electrode, which was in serted into the buffer drop over the tBLM was a conventional patch pipette containi ng an Ag/AgCl electrode and filled with buffer solution. The reference electrode consist ed of a tungsten-tipped electrode that was positioned onto the probe pad that extended from the sensor pad. Both electrodes were connected to an Axon Patch-Clamp amplif ier (Axopatch 200B Molecular Devices Corporation, Union City, CA). The signal was passed through a low-pass 5 kHz filter and digitized at a sampling rate of 20 kHz using a Digidata 1322A (Molecular Devices Corporation). After a giga-seal was obtained, 2 L of the protein solution was added for MscL incorporation into the tBLM and subsequent single channel analysis. Results and Discussion The detection of single channel activity, r epresented by currents in the picoampere (pA) range, requires a high signal-to-noise ra tio. As a consequence, low leakage current in the pathway between the tw o electrodes is a necessity. This requires a stable and low conducting lipid membrane having an el ectrical resistance in the gigaohm (G ) range. In this study, the resistance betw een the two pure gold pads connected to the two electrodes ranged from 1.5 G to approximately 17 G Figure 6-2 shows a plot of current as a function of applied voltage for one of the obtained membrane seals prior to incorporation of the MscL channel for functi onal studies. The particular presented seal had a resistance of 11 G as calculated from the Ohms law: V = IR and R = V/I (6-1) Where V is the applied voltage, I is the current and R is the resistance.

PAGE 175

175 Figure 6-2. Current plotted against the appli ed voltage of a giga-seal. The solid line is a linear fit forced through origin. Figure 6-3 below illustrates a trace of a stable bilayer formed by the 70PC/30PE lipid composition within the tBLM. The calculated resistance is 4.08 G and the applied voltage is 300 mV. The duration of recording as can be observed from the trace is over a 10 minute time scale with no evidence of memb rane defects or membra ne instability. Figure 6-3. A stable trace recorded from a 7:3 DPhPC/DPhPE bilayer membrane over a duration of 10 minutes.

PAGE 176

176 MscL channels were incorporated into stable tBLMs and single channel activity recorded. In Figure 6-4 typical current traces obtained from the wildtype MscL are shown. These recordings were obtained at 300mV, which was the voltage-gating threshold, i.e. no signals were observ ed at lower potentials for the MscL. Figure 6-4. Single channel behavior of MscL presented as m easured current at an applied transmembrane potential of + 300 mV. (A) The trace shows the activity of one MscL channel having two conductivity states and it current distributions. The trace in Figure 6-4 shows two distinct conductivity levels for the incorporated MscL channel: a lower current level corresponding to a unitary conductance of ~23 pS ((21) pA/0.3 V) and a higher current level corresponding to a unitary conductance of ~203 pS ((75) Pa/0.3 V). The succe ss rate of achieving stable G seals was ca. 60% and the lifetimes (generally due to loss of seal) varied between a few seconds up

PAGE 177

177 to 1.5 hours, depending on the appl ied voltage in these unop timized experiments. The lower conductance was not always present as seen in Figure 6-5 below which is from a recording of single channel activity under identical conditions as those for the experiment that yielded t he data in Figure 6-4. Figure 6-5. A trace for one M scL showing only one conductivi ty state and B) its current distributions. The conductance of the MscL observed here under a tethered bilayer membrane (tBLM) system was 0.2 nS which is lower in comparison to 3 nS reported in the literature under patch clamp cond itions. The reason for this is not very clear, however, the characteristics of the ionic reservoir in the sub-membrane space are believed to contribute to this atypical conductance. Another possible explanation for this sub-

PAGE 178

178 optimal conductance could be the lack of an ex ternal pressure to induce gating. In the reservoir the ion mobility is decreased due to t he lack of space, pres ence of tethers, and the low ionic concentration in comparison to that in the free buffer on the opposite side of the membrane.102 However, this concentration difference is not believed to be large enough to induce much of an osmotic pressure. This is based upon earlier findings made using gramicidin A, in which the observed conductance was the same on the device as when utilizing tip dipping, a mo re conventional meas uring technique in electrophysiology.160 Furthermore, the lipid environment hosting the protein is also crucial, which also can be a reason for the low observed conductivity. The tethers that decouple the tBLM from the su bstrate also form a somewhat different composition in the lower leaflet of the membrane compared to the upper one. The lifetime distributions determined from probability dens ity functions displayed fast kinetics with the channels open for 1 = 0.64 ms, 2 = 1.17 ms and 3 = 2.31 ms ( n = 4), which are comparable with values found in literature when patch clamping methods were utilized.221 As given by the Maxwell stre ss tensor, an electrical field applied over a lipid bilayer induces a perpendicu lar stress in the me mbrane. This stress can be viewed as an increase in membrane area and can cause changes in the curvature in a phenomenon referred to as flex oelectricity that has been demonstrated in both lipid bilayers230 and cell membranes.147 In the present work, an electrical field induced membrane stress ( ) strong enough to trigger an opening of the MscL channel was observed. According to Needham and Hoch muth the tension as a function of the transmembrane potential ( V ) can be expressed as: = 0d ( 2/ de 2) V2 (6-2)

PAGE 179

179 where is the dielectric const ant of the surrounding medium (typically ca. 60), 0 is the permittivity of free space, d is the bilayer thickness (~4 nm), and de is the dielectric thickness (~3 nm).231 The thicknesses are derived from capac itance measurements on similar lipids.64 The membrane tension needed to gate the MscL ion channel, in a lipid environment, as used in this work, has been shown to be approximately 12 dyne/cm, which has been derived from pressure-induced stretching of the bilayer.97 According to Equation 6-1 this tension corresponds to a transmembrane potent ial of ~300mV, which is in very good agreement with our findings, that is, no si gnals were observed for voltages below 300mV. This strongly suggests that the gating of the MscL is tension dependent. Precautions were taken during this study to avoid true electroporation of the membrane. A stable giga-seal was always obtained before the addition of the pr otein and no signals were detected before the MscL addition. The seal was stable at 300mV for approximately 1.5 hours. Conclusions In conclusion, a novel approach that examines a functional mechanosensitive channel reconstituted on a tethered memb rane surface has been developed. The channel was gated upon the applicat ion of potential, which a ccording to calculations give rise to a stress on the lipid bilaye r similar to pressure-induced tensions. This strengthens the fact that the Ms cL ion channel is gated in response to stress in the lipid membrane as opposed to pressure across it Furthermore, these findings show the possibility of using MscL as a release va lve for engineered membrane devices; one step closer to mimicking the true function of the living cell.

PAGE 180

180 CHAPTER 7 CONCLUSION AND FUTURE DIRECTIONS The long term objective of this projec t was to develop a biosensor based on the BKCa channel for the detection of chemical and biological analytes. Several parameters needed to be considered to achieve the goal of incorporating channels within bilayer lipid membranes and realize functional activi ty. Foremost, the bilayer lipid membrane had to exhibit the appropriate resistances requi red for electrical studies of incorporated ion channels and be stable to mechanical sh ocks and vibrations. Additionally, the bilayer membrane had to have curvature that would allow optimal insertion of the channel in the correct functional conformati on within a system capable of converting the biological events of gating into measurable and recordable electr ical signals. In order to demonstrate the potential for the detection of analytes by this system, it was imperative that we obtain pharmacological responses of incorporated BKCa channels to known antagonists. Investigations were organized into specific goals that would collectively build up to the overall objective. The first step was to obtain channels and this was achieved through heterologous expression of the mslo mouse gene encoding for the BKCa channel. For the expression system, Xenopus laevis oocytes were found to be convenient and channels were expressed on the membranes of these oo cytes. A mutant construct of the BKCa channel with a truncation of the C-terminus was express ed because of the potential difficulty of inserting the full-length channel with a bulky in tracellular domain into a tethered bilayer membrane. We developed an affinity chromatogr aphy purification protoc ol to isolate the channels from other solubilized membrane components that would potentially perturb functional activity upon insertion in synthet ic membranes. The properties of non-native

PAGE 181

181 lipids underlie the success of reconstitution of purified channels into vesicles and formation of tethered bilayer membranes. Among the characteristics of lipids critical for our applications would be curvature, me mbrane packing dynamics and the propensity to form bilayers on solid substrates. We used atomic force microscopy (AFM) and quartz crystal microbalance with dissi pation monitoring (QCM-D) to investigate the kinetics of lipid fusion for the formation of bilayers on solid substrates. Among the different concentrations of the phytanoyl lipids DPhPC and DPhPE studied for their ability to form bilayers on gold and silica substrates, a minimum concentration of 0.15 g/ml was found to be the threshold concentration required for fusion on solid substrates. A number of compositions of the two phytanyl lipids were also studied to determine the most suitable combination required for the formation of stable bilayers, and 7/3 DPhPC:DPhPE was most ideal. A size analysis of vesicles extruded through polycar bonate membranes of varied pore sizes was performed by dynamic light scattering and a comparison was done by NMR. Imaging by cryogenic transmission electron microscopy was used to give an idea of the structural morphology of the vesicles. Purified BKCa channels were reconstituted in vesicles of DPhPC: DPh PE in a 7:3 molar ratio through a dialysis procedure and then fused on self-assembled monolayers of DPTL deposited on microchip devices. Numerous studies have been done on poreforming peptides incorporated in tethered bilayer membranes; however, all of the studies have involved use of simple peptides which cannot not give qualitative pha rmacological data as ion channels would. Besides offering the advantage of yielding pharmacological responses, ion channels also exhibit characteristic stochastic signatures in contrast to pores, and are therefore

PAGE 182

182 natural biological sensing units. The studi es documented here represent the first successful incorporation of a functional ion channel within a tethered bilayer lipid membrane environment. The in corporation of the BKCa channel within the tBLM was confirmed by surface plasmon resonance-e nhanced ellipsometry and functional studies were performed by use of a modified pat ch clamp electrophysiology technique. The gold surface on the microchip device that serves as a substrate for tethering of the bilayer membrane allows the study of surface molecules under a defined electric field, therefore is ideal for investigations of field sensitive processes such as ion translocation through tBLM incorporated channels. Demonstration of a response to activators, blockers or modulators by ion channels is a major milestone towards realizing the goal of developing a bios ensor. Quaternary ammonium compounds have been shown to block potassium channels therefore; we used derivatives of tetraethylammonium (TEA) to investigate modulation of BKCa channel activity within the tBLM. Being an open channel blocker with diffe rences in stochastic signals depending on whether internally or externally applied, TEA responses by the BKCa channel gave insight on channel orientati on within the membrane. Future experimental work can be focused on the application of selective BKCa channel blockers such as iberiotoxin (IbTX) and monitoring subsequent current modulations through channels in the tether ed bilayer membrane system. Such a step would go a long way in demonstrating the potent ial of this system as a biosensor for specific target molecules. The beauty about biological sensing elements such as ion channels is the ability to re-engineer them an d create novel binding sites for a variety of agents they would naturally not respond to. Biochemical modifications can be made to

PAGE 183

183 channels at locations that would alter normal function, ther eby producing defined changes to stochastic signatures and allow dete ction to a high level of specificity. The electrical properties of BKCa channels in the tBLM showed a deviation from those observed under other m easurement configurations such as patch clamping. These deviations could be attributed to a few factors relating to both the truncated channel as well as the experimental setup. Piskorowski and his co-workers performed studies of truncated BKCa channels and recorded currents that were modulated by alterations in calcium concentrations, howev er, they noted that t he kinetics of channel opening were altered to some degree. The obser vation they made regarding variation in electrical properties of channels due to tr uncation of the C-terminus as well as the differences in experimental configurations could partly explain why our results show a deviation from studies under the patch clamp c onfiguration. It is also possible that the differences between our results and those docum ented in literature st em from the lipid environment within which reconstitution is done and bilayers formed for our investigations. Synthetic phytanoyl lipids DPhPC and DPhP E which are derivatives of archaeal lipids were used for formation of vesicles and bilayer membranes. Despite the ubiquity of BKCa channel distribution in most living organi sms, these channels are not found in bacteria and the unique membrane packing dynamics and fluidity of phytanoyl lipids could have conformational implications to proteins that are inserted in membranes formed of such lipids. These conformational implications might be responsible for differences in electrical properties between c hannels in the tBLM system in this study, and those channels that are reconstituted in eukaryotic lipid membranes. Among the

PAGE 184

184 electrical properties that show a deviation fr om literature values would be conductance. Conductances of BKCa channels in the tBLM were f ound to be approximately 40 pS, nearly an order of magnitude lower than the 300 pS recorded under patch clamp conditions. The same pattern was obser ved during investigations performed and documented in Chapter 6 of this study involving the incorporation of the mechanosensitive channel of large conductance (MscL), which had recordings of 300 pS in the tBLM system, compared to 3 nS recorded under patch clamp conditions, an order of magnitude hi gher. The interesting observ ation made here points to the molecular sizes of the channels incorporated in the tBLM as being responsible for the recording of anomalously lower conductances than expected. The conclusion regarding the molecular weights of incorporated channel s explaining the atypical conductance is further supported by observations of normal conductance levels in the gramicidin (gA) peptides inserted in tBLMs. The gramicidin channel is comparably a very small peptide when contrasted against true ion channels like the BKCa channel or the MscL channel. Apparently, high molecular weight ion channel s compromise on the volume of the submembrane space of the tBLM which is al ready densely packed with DPTL tethering units, hence limiting ion mobility critical for maintaining free ionic flow hence normal conductance levels. Future work could involve changes to the overall configuration of the tBLM system which exhibits asymmetry between the visco sity of the ionic reservoir below the membrane and the bulk solution above the tB LM that collectively would affect conductance. On-going work targeted at impr oving the tBLM system, in particular the thiol-modified tethering unit, for the last few years by Wolfgang Knoll and Ingo Koepper

PAGE 185

185 has resulted in advances in tBLM properties and the movement of ions in this region. 100, 148, 232-236 Among possibilities expl ored include increasing the volume of the submembrane space by synthesis of longer tet hering units with the goal of increasing the fluidity of the ionic reserv oir within the tBLM system.236 Additionally, Knoll and coworkers combined polymer supported bilayer and the tethered bilayer to create the polymer-tethered lipid bilayer system on modified substrates.148 This alternative membrane configuration significantly incr eased the decoupling of the membrane from the metal substrate hence improved the fluidity in the ionic reservoir. Further improvements of t he measurement configurat ion could be considered towards improving the system. In the present tBLM system, there is an interfacial capacitance associated with the membrane and the solution interfaces above and below the bilayer, as well as resistance in t he bulk solution and within the membrane. An alternative equivalent circuit model which w ould allow for a more accurate account of the actual transmembrane potential within the tBLM system could be considered for study of electrical properties of ion channel s. Nonetheless, the system studied in this work demonstrates the potential of using purified BKCa channels incorp orated within a tBLM successfully for biosensor applications.

PAGE 186

186 APPENDIX A DYNAMIC LIGHT SCATTERING ANALYS IS OF TEMPERATURE EFFECTS ON SIZES OF PHYTANOYL LIPID VESICLES OF DIFFERENT MIXTURES Figure A-1. Diameters of 100 % DPhPC vesicles at 25 C, 45 C and 65 C as a function of intensity of scattered light. 4080120160200240280320 0 20 40 60 80 100 Intensity (au)Diameter (nm) 25 C 45 C 65 C Figure A-2. Diameters of vesicles of DPhPC: DPhPE lipid mixtures at a 7:3 molar ratio at 25 C, 45 C and 65 C as a function of intensity of scattered light. 4080120160200240280 0 10 20 30 40 50 60 70 80 90 100 Intensity (au)Diameter ( nm ) 25 C 45 C 65 C

PAGE 187

187 200400600800100012001400 0 20 40 60 80 100 Intensity (nm)Diameter (nm) 25C 45C 65C Figure A-3. Diameters of vesicles of DPhPC: DPhPE lipid mixtures at a 5:5 molar ratio at 25 C, 45 C and 65 C as a function of intensity of scattered light.

PAGE 188

188 APPENDIX B X-RAY PHOTOELECTRON SPECTROSCOPY USED FOR SURFACE ANALYSIS OF PLASMA-TREATED GOLD Figure B-1. XPS results showing gold, oxyg en and carbon as the only elements present on plasma-treated gold.

PAGE 189

189 APPENDIX C DNA AND AMINO ACID SEQ UENCES OF PROTEINS Table C-1. Full Length mslo Amino Acid Sequences (BKCa Channel) 1 MELEHPKSPP YPSSSSSSSS SSVHEPKMDA LIIPVTMEVP CDSRGQRMWW 51 AFLASSMVTF FGGLFIILLW RTLKYLWTVC CHCGGKTKEA QKINNGSSQA 101 DGTLKPVDEK EEVVAAEVGW MTSVKDWAGV MISAQTLTGR VLVVLVFALS 151 IGALVIYFID SSNPIESCQN FYKDFTLQID MAFNVFFLLY FGLRFIAAND 201 KLWFWLEVNS VVDFFTVPPV FVSVYLNRSW LGLRFLRALR LIQFSEILQF 251 LNILKTSNSI KLVNLLSIFI STWLTAAGFI HLVENSGDPW ENFQNNQALT 301 YWECVYLLMV TMSTVGYGDV YAKTTLGRLF MVFFILGGLA MFASYVPEII 351 ELIGNRKKYG GSYSAVSGRK HIVVCGHITL ESVSNFLKDF LHKDRDDVNV 401 EIVFLHNISP NLELEALFKR HFTQVEFYQG SVLNPHDLAR VKIESADACL 451 ILANKYCADP DAEDASNIMR VISIKNYHPK IRIITQMLQY HNKAHLLNIP 501 SWNWKEGDDA ICLAELKLGF IAQSCLAQGL STMLANLFSM RSFIKIEEDT 551 WQKYYLEGVS NEMYTEYLSS AFVGLSFPTV CELCFVKLKL LMIAIEYKSA 601 NRESRILINP GNHLKIQEGT LGFFIASDAK EVKRAFFYCK ACHDDVTDPK 651 RIKKCGCRRL IYFEDEQPPT LSPKKKQRNG GMRNSPNTSP KLMRHDPLLI 701 PGNDQIDNMD SNVKKYDSTG MFHWCAPKEI EKVILTRSEA AMTVLSGHVV 751 VCIFGDVSSA LIGLRNLVMP LRASNFHYHE LKHIVFVGSI EYLKREWETL 801 HNFPKVSILP GTPLSRADLR AVNINLCDMC VILSANQNNI DDTSLQDKEC 851 ILASLNIKSM QFDDSIGVLQ ANSQGFTPPG MDRSSPDNSP VHGMLRQPSI 901 TTGVNIPIIT ELVNDTNVQF LDQDDDDDPD TELYLTQPFA CGTAFAVSVL 951 DSLMSATYFN DNILTLIRTL VTGGATPELE ALIAEENALR GGYSTPQTLA 1001 NRDRCRVAQL ALLDGPFADL GDGGCYGDLF CKALKTYNML CFGIYRLRDA 1051 HLSTPSQCTK RYVITNPPYE FELVPTDLIF CLMQFDHNAG QSRASLSHSS 1101 HSSQSSSKKS SSVHSIPSTA NRPNRPKSRE SRDKQNATRM TRMGQAEKKW 1151 FTDEPDNAYP RNIQIKPMST HMANQINQYK STSSLIPPIR EVEDEC Table C-2. Truncated BKCa Channel at position 335 with 12 additional residues coded for at the C-terminus to introduce a stop codon 1 MELEHPKSPP YPSSSSSSSS SSVHEPKMDA LIIPVTMEVP CDSRGQRMWW 51 AFLASSMVTF FGGLFIILLW RTLKYLWTVC CHCGGKTKEA QKINNGSSQA 101 DGTLKPVDEK EEVVAAEVGW MTSVKDWAGV MISAQTLTGR VLVVLVFALS 151 IGALVIYFID SSNPIESCQN FYKDFTLQID MAFNVFFLLY FGLRFIAAND 201 KLWFWLEVNS VVDFFTVPPV FVSVYLNRSW LGLRFLRALR LIQFSEILQF 251 LNILKTSNSI KLVNLLSIFI STWLTAAGFI HLVENSGDPW ENFQNNQALT 301 YWECVYLLMV TMSTVGYGDV YAKTTLGRLF MVFFIALRTP RRPELFF

PAGE 190

190 Table C-3. Red fluorescent protei n (mRFP1) DNA and protein sequences DNA ATGGCCTCCT CCGAGGACGT CATCAAGGAG TTCATGCGCT TCAAGGTGCG CATGGAGGGC TCCGTGAACG GCCACGAGTT CGAGATCGAG GGCGAGGGCG AGGGCCGCCC CTACGAGGGC ACCCAGACCG CCAAGCTGAA GGTGACCAAG GGCGGCCCCC TGCCCTTCGC CTGGGACATC CTGTCCCCTC AGTTCCAGTA CGGCTCCAAG GCCTACGTGA AGCACCCCGC CGACATCCCC GACTACTTGA AGCTGTCCTT CCCCGAGGGC TTCAAGTGGG AGCGCGTGAT GAACTTCGAG GACGGCGGCG TGGTGACCGT GACCCAGGAC TCCTCCCTGC AGGACGGCGA GTTCATCTAC AAGGTGAAGC TGCGCGGCAC CAACTTCCCC TCCGACGGCC CCGTAATGCA GAAGAAGACC ATGGGCTGGG AGGCCTCCAC CGAGCGGATG TACCCCGAGG ACGGCGCCCT GAAGGGCGAG ATCAAGATGA GGCTGAAGCT GAAGGACGGC GGCCACTACG ACGCCGAGGT CAAGACCACC TACATGGCCA AGAAGCCCGT GCAGCTGCCC GGCGCCTACA AGACCGACAT CAAGCTGGAC ATCACCTCCC ACAACGAGGA CTACACCATC GTGGAACAGT ACGAGCGCGC CGAGGGCCGC CACTCCACCG GCGCCTAA PROTEIN MASSEDVIKE FMRFKVRMEG SVNGHEFEIE GEGEGRPYEG TQTAKLKVTK GGPLPFAWDI LSPQFQYGSK AYVKHPADIP DYLKLSFPEG FKWERVMNFE DGGVVTVTQD SSLQDGEFIY KVKLRGTNFP SDGPVMQKKT MGWEASTERM YPEDGALKGE IKMRLKLKDG GHYDAEVKTT YMAKKPVQLP GAYKTDIKLD ITSHNEDYTI VEQYERAEGR HSTGA Table C-5. E. coli Mechanosensitive Channel of Large Conductance (MscL) DNA sequences 1 TATGGTTGTC GGCTTCATAG GGAGAATAAC ATGAGCATTA TTAAAGAATT 51 TCGCGAATTT GCGATGCGCG GGAACGTGGT GGATTTGGCG GTGGGTGTCA 101 TTATCGGTGC GGCATTCGGG AAGATTGTCT CTTCACTGGT TGCCGATATC 151 ATCATGCCTC CTCTGGGCTT ATTAATTGGC GGGATCGATT TTAAACAGTT 201 TGCTGTCACG CTACGCGATG CGCAGGGGGA TATCCCTGCT GTTGTGATGC 251 ATTACGGTGT CTTCATTCAA AACGTCTTTG ATTTTCTGAT TGTGGCCTTT 301 GCCATCTTTA TGGCGATTAA GCTAATCAAC AAACTGAATC GGAAAAAAGA 351 AGAACCAGCA GCCGCACCTG CACCAACTAA AGAAGAAGTA TTACTGACAG 401 AAATTCGTGA TTTGCTGAAA GAGCAGAATA ACCGCTCTTA ACAAGCGCCT 451 GAAAGCAGAA GACCAGTGGT AAAAAAGTGA TTTACTTTCT TGCCACTGGC 501 CTCCCAGTTC CCCCGATTGC CATG Table C-4. E. coli Mechanosensitive Chann el of Large Conductance (MscL) amino acid sequences 1 MSIIKEFREF AMRGNVVDLA VGVIIGAAFG KIVSSLVADI IMPPLGLLIG 51 GIDFKQFAVT LRDAQGDIPA VVMHYGVFIQ NVFDFLIVAF AIFMAIKLIN 101 KLNRKKEEPA AAPAPTKEEV LLTEIRDLLK EQNNRS

PAGE 191

191 APPENDIX D PLASMID MAP FOR THE PCDNAOX AND MSLO GENE INSERTS KpnI SacIII BstXI NotI NheI Cla I XhoI XbaI Cla I Met End Xba Imslo coding sequence Figure D-1. Map for the p CDNAOX plasmid containing the mslo BKCa channel gene insert and the restriction sites.

PAGE 192

192 LIST OF REFERENCES 1. Hille, B. Ion Channels of Excitable Membranes (Sinauer, Sunderland, Massachusetts, USA, 2001). 2. Willumsen, N. J. et al. High throughput electrophysiology: new perspectives for ion channel drug discovery. Receptors and channels 9 3-12 (2003). 3. Owen, D. & Silverthorne, A. Channeling drug discovery Current trends in ion channel drug discovery research. Drug Discovery World 53, 48-61 (2002). 4. Ashcroft, F. M. Ion Channels and Disease Channelopathies (Academic Press, San. Diego CA, USA, 2000). 5. Steizle, M., Weismuller, G. & Sackmann, E. On t he application of supported bilayers as receptive layers for biosensors with electrical detection. J. Phys. Chem. 97, 2974-2981 (1993). 6. Reiken, S. R. et al. Bispecific anti body modification of nicotinic acetylcholine receptors for biosensing. Biosensors and Bioelectronics 111 99-102 (1996). 7. Cornell, B. A. et al. A biosensor that uses ion-channel switches. Nature 387 580 583 (1997). 8. Neher, E. & Sakmann, B. Single-channel currents re corded from membrane of denervated from muscle fibres. Nature 260 799-802 (1976). 9. Salkoff, L., Butler, A., Ferreira, G., Santi, C. M. & Wei, A. High conductance Potassium Channels of the Slo Family. Nature Reviews Neurosci. 5 921-931 (2006). 10. Miller, C. An overview of the potassium channel family. Genome Biol. 1 0004 (2000). 11. Papazian, D. M., Schwarz, T. L., Tempel, B. L., Jan, Y. N. & Jan, Y. L. Cloning of genomic and complementary DNA from S haker, a putative potassium channel gene from Drosophila. Science 237 749-753 (1987). 12. Ho, K. et al. Cloning and expression of an inwardly rectifying ATP-regulated potassium channel. Nature 362 31-38 (1993). 13. Heginbotham, L., Lu, Z., Abramson, T. MacKinnon, R. & Mutations in the K+ channel signature sequence. Biophys J. 66, 1061-1067 (1994). 14. Pongs, O. et al. Shaker encodes a family of putative potassium channel proteins in the nervous system of Drosophila. EMBO J. 7 1087-1096 (1988).

PAGE 193

193 15. Hille, B. & Schwarz, W. Potassium channels as multi-ion single-file pores. J. Gen Physiol. 72, 409-442 (1978). 16. Gouaux, E. & Mackinnon, R. Principles of selective ion transport in channels and pumps. Science 310 1461-1465 (2005). 17. Orio, P., Rojas, P., Ferreira, G. & Lat orre, R. New guises for an old channel: maxik channel beta subunits. News Physiol. Sci. 17 156-161 (2002). 18. Marty, A. Ca-dependent K channels with large unitary conductance in chromaffin cell membranes. Nature 291 497-500 (1981). 19. Park, Y. B. Ion sele ctivity and gating of small conductance calcium-activated potassium channels in cultured rat adrenal chromaffin cells. J. Physiol. (Lond.) 481 555-570 (1994). 20. Blatz, A. L. & Magleby, K. L. Corre cting single channel data for missed events. Biophys. J. 49, 967-980 (1986). 21. Logsdon, N. J., Kang, J., Togo, J. A., Chri stian, E. P. & Aiya r, J. A novel gene, hKCa4, encodes the calcium-activ ated potassium channel in human T lymphocytes. J. Biol. Chem. 272 32723 32726 (1997). 22. Ishii, T. M. et al. A human intermedi ate conductance calcium-activated potassium channel. Proc Natl Acad Sci USA 94, 11651 (1997). 23. Adelman, J. P. et al. Calcium-act ivated potassium channels expressed from cloned complementary DNAs Neuron 9 209 216 (1992). 24. Atkinson, N. S., Robertson, G. A. & Ganetzky, B. A component of calciumactivated potassium channels encoded by the Drosophila slo locus. Science 253 551-555 (1991). 25. Meera, P., Wallner, M. Song, M. & Toro, L. Lar ge conductance voltageand calcium-dependent K+ channel, a distinct member of voltage-dependent ion channels with seven N-terminal transmembrane segments (S0-S6), an extracellular N terminus, and an in tracellular (S9-S10) C terminus. Proc Natl Acad Sci USA 94 14066-14071 (1977). 26. Wallner, M., Meera, P. & Toro, L. Determinant for bet a-subunit regulation in highconductance voltage-activated and Ca(2+)-sensitive K+channels: an additional transmembrane region at the N terminus. Proc Natl Acad Sci USA 93 14922 14927 (1996). 27. Toro, L., Wallner, M., Meer a, P. & Tanaka, Y. Maxi-K(C a), a unique member of the voltage-gated K channel superfamily. News Physiol Sci 13, 112-117 (1998).

PAGE 194

194 28. Knaus, H. G. et al. Primary sequence and immunological characterization of betasubunit of high conductance Ca(2+)-act ivated K+channel from smooth muscle. J. Biol. Chem. 269 17274 17278 (1994). 29. Knaus, H. G. et al. Tremorgenic indole alkaloids potently inhibit smooth muscle high-conductance calcium-activated potassium channels. Biochemistry 33, 5819 5828 (1994). 30. Jiang, Y., Pico, A., Cadene, M., Chait, B. T. & MacKinnon, R. Structure of the RCK domain from the E. coli K+ channel and demonstration of its presence in the human BK channel. Neuron 29, 593-601 (2001). 31. Wei, A., Solaro, C., Lingle, C. & Salk off, L. Calcium sensitivity of BK-type KCa channels determined by a separable domain. Neuron 13, 671-681 (1994). 32. Stefani, E. et al. Voltage-controlled gating in a lar ge conductance Ca2+-sensitive K+ channel (hslo). Proc Natl Acad Sci USA 94 5427 5431 (1997). 33. Bezanilla, F., Perozo, E., Papazian, D. M. & Stefani, E. Molecular basis of gating charge immobilization in Shaker potassium channels. Science 254 679 683 (1991). 34. Cui, J., Cox, D. H. & Aldrich, R. W. Intrinsic voltage dependance and calcium regulation of mslo large conductance calcium-activated K+ channels. J. Gen. Physiol. 109 647-673 (1997). 35. Latorre, R., Oberhauser, A., Labarca, P. & Alvarez, O. Varieties of calciumactivated potassium channels. Ann. Rev. Physiol. 51, 385-399 (1989). 36. Pallotta, B. S. N-br omoacetamide removes a ca lcium-dependent component of channel opening from calc ium-activated potassium channels in rat skeletal muscle. J. Gen. Physiol. 86, 601-611 (1985). 37. Papazian, D. M., Timpe, L. C., Jan, Y. N. & Jan, Y. L. Alteration of voltagedependence of shaker potassium channels by mutations in the S4 sequence. Nature 349 305-310 (1991). 38. Rothberg, B. S. Allost eric modulationof ion channe ls: the case of maxi-k. Sci STKE 2004 (2004). 39. Tian, L., Coghill, L. S., MacDonald, S. H., Armstrong, D. L. & Shipston, M. J. Leucine zipper domain targets cAMPdependent protein kinase to mammalian BK channels. J. Biol. Chem. 278 8669 8677 (2003). 40. Wei, A., Jegla, T. & Salkoff, L. Eight potassium channel families revealed by the C. elegans genome project. Neurophar macology 35, 805-829 (1996).

PAGE 195

195 41. Schreiber, M. & Salkoff, L. A novel calcium-sensing domain in the BK channel. Biophys. J. 73, 1355 1363 (1997). 42. Xia, X. M., Zeng, X. & Lingle, C. J. Multiple regul atory sites in largeconductance calcium-activated potassium channels. Nature 418 880-884 (2002). 43. Piskorowski, R. & Aldrich, R. W. Calcium activation of BK(Ca) potassium channels lacking the calcium bowl and RCK domains. Nature 420 499-502 (2002). 44. Knaus, H. et al. Characte rization of Tissue-expressed -Subunits of the High Conductance Calcium-activated Potassium Channel. Journal of Biological Chemistry 270 22434-22439 (1995). 45. Jin, P., Weiger, T. M. & Levitan, I. B. Reciprocal modulation between the alpha and beta 4 subunits of hSlo calc ium-dependent potassium channels. J. Biol. Chem. 277 43724 43729 (2002). 46. Zeng, X. H., Xia, X. M. & Lingle, C. J. Redox-sensitiv e extracellular gates formed by auxiliary beta subunits of calcium-activated potassium channels. Nat. Struct. Biol. 10, 448-454 (2003). 47. Jin, P., Weiger, T. M., Wu, Y. & Levitan, I. B. Phosphorylation dependent functional coupling of hSlo calciumdependent potassium channel and its beta 4 subunit. J. Biol. Chem. 277 10014 (2002). 48. Egan, T. M., Dagan, D. & Levitan, I. B. Properties and modulation of a calciumactivated potassium channel in rat olfactory bulb neurons. J. Neurophysiol. 69, 1433 1442 (1993). 49. Galvez, A. et al. Purification and char acterization of a unique, potent, peptidyl probe for the high conductance calcium-act ivated potassium channel from venom of the scorpion Buthus tamulus. J. Biol. Chem. 265 11083 (1990). 50. Giangiacomo, K. M., Garcia, M. L. & McManus, O. B. Mechanism of iberiotoxin block of the large-conductance calciumactivated potassium channel from bovine aortic smooth muscle. Biochemistry 31, 6719 (1992). 51. Candia, S., Garcia, M. L. & Latorre, R. Mode of action of iberiotoxin, a potent blocker of the large conduct ance Ca2+-activated K+ channel. Biophys. J. 63, 583590 (1992). 52. Kaczorowski, G. J., Knaus, H.-G., Leonard, R. J., McManus, O. B. & Garcia, M. L. High-conductance calcium-activated potassium channels; Structure, pharmacology, and function. J. Bioenergetics & Biomembranes 28 255-267 (1996). 53. Blatz, A. L. & M agleby, K. L. Calcium-activated potassium channels. Trends in Neurosciences 10, 463-467 (1987).

PAGE 196

196 54. Yellen, G. Voltage gated potassi um channels and their relatives. Nature 419 3542 (2002). 55. Tristani-Firouzi, M., C hen, J. & Sanguinetti, M. C. Interactions between S4-S5 linker and S6 transmembrane domain modul ate gating of HERG K+ channels. J. Biol. Chem. 277 18994-19000 (2002). 56. Doyle, D. A. et al. The structure of the potassium channel: molecular basis of K+ conduction and selectivity. Science 280 69-76 (1998). 57. Zagotta, W. N., Hoshi, T. & Aldrich, R. W. Restoration of inactivation in mutants of Shaker potassium channels by a peptide derived from ShB. Science 250 568-571 (1990). 58. Hoshi, T., Zagotta, W. N. & Aldrich, R. W. Biophysical and molecular mechanisms of Shaker potassium channel inactivation. Science 250 533-538 (1990). 59. Yellen, G. The moving parts of voltage-gated ion channels. Quart. Rev. Biophys. 31, 239-295 (1998). 60. Opekarov, M. & Tanner, W. Specific lipid requirements of membrane proteins--a putative bottleneck in het erologous expression. Biochim Biophys Acta Biomembranes 1610 11-22 (2003). 61. Findlay, H. E. & Booth, P. J. The biological significance of lipidprotein interactions. J. Phys. Condens. Matter 18, 1281-1291 (2006). 62. Vilayaveeti, F. I., Zhou, Y. & MacKinnon, R. Lipids in the structure, folding and function of the KcsA K+ channel. Biochemistry 41, 10771-10777 (2002). 63. Fernandis, A. Z. & Wenk, M. R. Me mbrane lipids as signaling molecules. Current opinion in lipidology 18 121-128 (2007). 64. Simons, K. & Toomre, D. Li pid rafts and signal transduction. Nat. Rev. Mol. Cell. Biol. 1 31-39 (2000). 65. Yeagle, P. L., 822 (1985) 267-287. Cholesterol and the cell membrane. Biochimica et Biophysica Acta (BBA) Reviews on Biomembranes 822 267-287 (1985). 66. Gennis, R. Biomembranes: Molecular structure and function (Springer-Verlag New York Inc., New York, 1989). 67. Schleifer, K. H. & Stackenbrandt, E. Molecular systematics of prokaryotes. Ann. Rev. Microbiol. 37, 143-187 (1983). 68. Silvius, J. R. Role of cholesterol in lipid raft formation: lessons from lipid model systems. Biochim Biophys Acta 1610 174-183 (2003).

PAGE 197

197 69. Yeagle, P. Structure of Biol ogical Membranes (CRC, New York, 2005). 70. Nelson, L. & Cox, M. M. Lehniger Principles of Biochemistry (Sara Tenney, New York, 2005). 71. Madden, T. D. Current concepts in membrane protein reconstitution. Chem. Phys. Lipids 40, 207-222 (1986). 72. Chan, Y.-H. M. & Boxer, S. G. M odel membrane systems and their applications Current Opinion in Chemical Biology 11, 1-7 (2007). 73. Tamm, L. K. & McConnell, H. M. Supported phospholipid bilayers. Biophys J. 47, 105-113 (1985). 74. Cremer, P. S. & Boxer, S. Formation and Spreading of Lipid Bilayers on Glass Surfaces. J. Phys. Chem. B. 103 2554-2559 (1999). 75. Israelachvili, J. Intermolecular and Surface Forces (Academic Press, New York, 1991). 76. Parsegian, V. A. Long range physical forces in the biological milieu. Annu. Rev. Biophys. Bioeng. 2 221-225 (1973). 77. Sackmann, E. Supported membranes: Scientific and practical applications. Science 271 43-48 (1996). 78. Miller, C., Cuendet, P. & Grtzel, M. K+ sensitive bilayer supporting electrodes. J. Electroanal. Chem. 278 175-192 (1990). 79. McConnell, H. M., Watts, T. H., Weis, R. M. & Br ian, A. A. Supported planar membranes in studies of cell-cell recognition in the immune system. Biochim Biophys Acta. 864 95-106 (1986). 80. Elender, G., Khner, M. & Sackmann, E. Functionalisation of Si/SiO2 and glass surfaces with ultrathin dextran film s and deposition of lipid bilayers. Biosens. Bioelectron. 11, 565-577 (1996). 81. Ariga, k. & Okahata, Y. Polymerized monolayers of single-, double-, and triplechain silane amphiphiles and permeation control through the monolayerimmobilized porous glass plate in an aqueous solution. J. Am. Chem. Soc. 111 5618-5622 (1989). 82. Fare, T. L. Electrical characteriza tion of dipalmitoylphosphatidylethanolamine and cadmium stearate films on platinum surfaces in aqueous solutions. Langmuir 6 1172-1179 (1990).

PAGE 198

198 83. Stelzle, M. & Sa ckmann, E. Sensitive detection of protein adsorption to supported lipid bilayers by frequency-depen dent capacitance measurements and microelectrophoresis. Biochim. Biophys. Acta 981 135-142 (1989). 84. Nikolelis, D. P., Siontorou, C. G., Krull, U. J. & Katrivanos, P. L. Ammonium Ion Minisensors from Self-Assembled Bilayer Lipid Membranes Using Gramicidin as an Ionophore. Modulation of Ammonium Select ivity by Platelet-Activating Factor. Anal. Chem. 68, 1735-1741 (1996). 85. Shao, Z., Mou, J., Czajko wsky, J., Yang, J. & Yuan, J. -Y. Biological atomic force microscopy: what is achieved and what is needed. Adv. Phys. 45, 1-86 (1996). 86. Kung, L. A., Kam, L., Hovis, J. S. & Bo xer, S. G. Patterning Hybrid Surfaces of Proteins and Supported Lipid Bilayers. Langmuir 16, 6773-6776 (2000). 87. Knoll, W. Interfaces and thin film s as seen by bound electromagnetic waves. Annu. Rev. Phys. Chem. 49, 569-638 (1998). 88. Richter, R. P., Brat, R. & Brisson., A. R. The forma tion of solid-supported lipid bilayers an integrated view. Langmuir 22, 3497-3505 (2006). 89. Zhdanov, V. P. & Kasemo, B. Comment s on Rupture of Adsorbed Vesicles. Langmuir 17 3518-3521 (2001). 90. Zhdanov, V. P., Keller, C. A., Glasma star, K. & Kasemo, B. Simulation of adsorption kinetics of lipid vesicles. J. Chem. Phys. 84, 5443-5446 (2000). 91. Tanaka, M. & Sackmann, E. Polymer s upported membranes as models of the cell surface. Nature 437 656-663 (2005). 92. Kiessling, V., Crane, J. M. & Tamm, L. K. Transbilaye r effects of raft-like lipid domains in asymmetric planar bilayers measured by single molecule tracking. Biophys. J. 91, 3313-3326 (2006). 93. Raguse, B. et al. Tethered Lipi d Bilayer Membranes: Formation and Ionic Reservoir Characterization. Langmuir 14, 648-659 (1998). 94. Salafsky, J., Groves, J. T. & Boxer, S. G. Architecture and function of membrane proteins in supported bilayers: a study with photosynthetic reactive centers. Biochemistry 35, 14773-14781 (1996). 95. Wagner, M. L. & Tamm, L. K. Tether ed Polymer-Supported Planar Lipid Bilayers for Reconstitution of Integral Membrane Proteins: Silane-Poly(ethylene glycol)Lipid as a Cushion and Covalent Linker. Biophys. J. 79, 1400-1414 (2000). 96. Vockenroth, I. K., Atanasova, P. P., T oby, A., Jenkins, A. & Kper, I. Incorporation of -Hemolysin in Different Tethered Bila yer Lipid Membrane Architectures. Langmuir 24 496-502 (2008).

PAGE 199

199 97. Andersson, M. et al. Voltage-induced gating of the mechanosensitive MscL ion channel reconstituted in a tethered lipid bilayer membrane. Biosensors and Bioelectronics 23, 919-923 (2008). 98. Andersson, M. et al. Detection of si ngle ion channel activity on a chip using tethered bilayer membranes. Langmuir 23, 2924-2927 (2007). 99. Keizer, H. M. et al. Functional Ion Channels in Tethered Bilayer Membranes Implications for Biosensors. ChemBioChem 8 1246-1250 (2007). 100. Knoll, W. Functional tethered lipid bilayers. Reviews in Molecular Biotechnology 74, 137-158 (2000). 101. Schiller, S. M., Naumann, R., Lovejoy, K., Kunz, H. & Knoll, W. Archaea Analogue Thiolipids for Tethered Bilayer Lipid Membranes on Ultrasmooth Gold Surfaces. Angew. Chem., Int. Ed. 42, 208-211 (2003). 102. Krishna, G., Schulte, J., Cornell, B. A ., Pace, R. & Osman, P. Tethered bilayer membranes containing ionic reservoirs: Selectivity and conductance. Langmuir 19, 2294-2305 (2003). 103. Dubois, L. H. & Nuzzo, R. G. Synthesis, Structure, and Properties of Model Organic Surfaces. Annual Review of Physical Chemistry 43, 437-463 (1992). 104. Woese, C. R. & Fox, G. E. Phylogenetic structure of the prokaryotic domain: the primary kingdoms. Proc Natl Acad Sci USA 74 5088 (1977). 105. Sten-Knudsen, O. Passive transport processes in membrane transport in biology (Springer-Verlag, New York, 1978). 106. Tien, H. T. & Ottova-Leitmannova. Planar lipid bilayers (BLMs) and their applications (Elsevier, Amsterdam, 2003). 107. Cole, K. S. Membranes, ions and impulses: A chapter of classical biophysics (University of California Press, Berkeley, 1968). 108. Almers, W. Gating currents and char ge movements in excitable membranes. Rev. Physiol. Biochem. Pharmacol. 82 96-190 (1978). 109. Coronado, R. & Latorre, R. Phospholipid bilayers made from monolayers on patchclamp pipettes. Biophys. J. 43, 231-236 (1983). 110. Wiegand, G., Arribas-Layton, N., Hil lebrandt, H., Sackmann, E. & P. Wagner. Electrical properties of supported lipid bilayer membranes. J. Phys. Chem. B. 106 4245-4254 (2002).

PAGE 200

200 111. Hladky, S. B. & Haydon, D. A. Discr eteness of conductance change in bimolecular lipid membranes in the presence of certain antibiotics. Nature 225 451-453 (1970). 112. Bayley, H., Braha, O. & Gu, L.-Q. Stochastic sensing with protein pores. Adv. Mater. 12, 139-142 (2000). 113. Bayley, H. & Cremer, P. S. Stochastic sensors inspired by biology. Nature 413 226-230 (2001). 114. Kodadek, T. Protein microa rrays: prospects and problems. Chem Biol. 8 105-115 (2001). 115. Gurdon, J. B., Lane, C. D., Woodland, H. R. & Marbaix, G. Use of frog eggs and oocytes for the study of messenger RNA and its translation in living cells. Nature (Land.) 233 177-182 (1971). 116. Vargo, M. A. & Colman, R. F. Heterodimers of wild-type and subunit interface mutant enzymes of glutathione S-transfe rase A1: Interactive or independent active sites? Protein Sci. 13, 1586-1593 (2004). 117. Hochuli, E., Dobeli, H. & Struber, A. New metal chelate adsorbents selective for proteins and peptides containi ng neighbouring histidine residues. J. Chromatogr. 411 177-184 (1987). 118. Hochuli, E., Banworth, W., Dobeli, H., G entz, R. & Struber, A. Genetic approach to facilitate purification of recombinant prot eins with a novel metal chelate adsorbent. BioTechnol 6 1321-1325 (1988). 119. Sakmann, B. & Neher, E. Single-channel recording (Plenum Press, New York, 1995). 120. Subrahmanyam, S., Piletsky, S. & Turner, A. Applicat ion Natural Receptors in Sensors and Assays. Anal. Chem. 74, 3942-3951 (2002). 121. Delcour, A. H., Martinac B., Adler, J. & Kung, C. Modified Reconstitution Method Used in Patch-Clamp Studies of Escherichia-Coli Ion Channels. Biophys. J. 56, 631-636 (1989). 122. MacKinnon, R. Potassium Chanels and Atom ic Basis of Selective Ion Conduction. Angewandte Chemie 43 4265-4277 (2004). 123. Sukharev, S. I., Blount, P. Martinac, B. & Kung, C. Mechanosensitive channels of Escherichia coli: The MscL gene, protein, and activities. Annual Review of Physiology 59, 633-657 (1997). 124. Chu, B. Laser Light Scattering (Academic Press, Inc, San Diego, 1991).

PAGE 201

201 125. Precision Deconvolve32 User Manual (2003). 126. Johnson, S. J. et al. Structure of an adsorbed dimyristoylphosphatidylcholine bilayer measured with specular reflection of neutrons. Biophys. J. 52, 289-294 (1991). 127. Ebenstein, Y., E. Nano Letters 2 945-950 (2002). 128. Digital Instruments, User manual (2000). 129. Ikeda, S. R., Soler, F., Zuhlke, R. D., Joho, R. H. & Lewis, L. D. Heterologous expression of the human potassium channel Kv2.1 in clonal mammalian cells by direct cytoplasmic microinjection of cRNA. Pflgers Archiv 422 201-203 (1992). 130. Capecchi, M. R. High efficiency transforma tion by direct microinjection of DNA into cultured mammalian cells. Cell 22 479-488 (1980). 131. Kamb, A., Korenbrot, J. I. & Kitajewski, J. Expression of ion channels in cultured cells using baculovirus. Methods in Enzymology 207 423-431 (1992). 132. Fulita, N. et al. Biosynthesis of the Torpedo californica acet ylcholine receptor a subunit in yeast. Science 231 1284-1289 (1986). 133. King, K., Dohlman, H. G., Thorner, J., Car on, M. G. & Lefkowitz, R. J. Control of yeast mating signal transduction by a mammalian beta 2-adrener gic receptor and Gs alpha subunit. Science 250 121-123 (1990). 134. Strosberg, A. D. & Marullo, S. Functional expression of receptors in microorganisms. Trends Pharmacol. Sci. 13, 95-98 (1992). 135. Barthel, F. et al. Gene regulation analysis by lipopolyamine-mediated DNA transfer in primary neurons. Methods in Neuroscience 9 292-312 (1992). 136. Holt, C. E., Garlick, N. & Cornel, E. Lipofection of cDNAs in the embryonic verterbrae central nervous system. Neuron 4 203-214 (1990). 137. Bianchi, L. & Driscoll, M. Heterologous expression of C. elegans ion channels in Xenopus oocytes (Wormbook, ed. The C. elegans Research Community, 2006). 138. Campbell, R. E. et al. A m onomeric red fluorescent protein. PNAS 99, 7877-7882 (2002). 139. Bevis, B. J. & Glick, B. S. Rapi dly maturing variants of the Discosoma red fluorescent protein (DsRed). Nature Biotechnology 20, 83-87 (2002). 140. Ohlendieck, K. Protein Purification Protocols (Humana Press, New Jersey, 2008).

PAGE 202

202 141. Kusaka, l. & Kanai, K. Purification and C haracterization of Alanine Carrier Isolated from H-Proteins of Bacillus subtilis. European Journal of Biochemistry 83, 307 311 (1978). 142. Riepl, R. & Vidaver, G. A. Effects of Triton X-100 Treatments on the Composition and Activities of Membrane Vesicl es from Pegion Erythrocytes. Biochimica et Biophysica Acta 507 94 -106 (1978). 143. Mimms, L. T., Zampighi G., Nozaki, Y., Tanford, C. & Reynolds, J. A. Phospholipid vesicle formation and trans membrane protein incorporation using octyl glucoside. Biochemistry 20, 833 (1981). 144. DiChiara, T. J. & Reinhart, P. H. Distinct effects of Ca2+ and voltage on the activation and deactivation of cl oned Ca2+-activated K+ channels. J. Physiol. (Lond.) 489 403-418 (1995). 145. Ausubel, F. M. et al. Current Protocols in Molecular Biology (Wiley Interscience, New York, New York, 1988). 146. Krause, J. D., Foster, C. D. & Reinhart, P. H. Xenopus laevis oocytes contain endogenous large conductance Calciu m-activated potassium channels. Neuropharmacology 35 1017-1022 (1996). 147. Zhang, Y. L., Dunlop, J. & Dalziel, J. E. Recombinant human voltage-gated sodium channels are pharmacologically functional in planar lipid bilayers. Biosensors and Bioelectronics 22, 1006-1012 (2007). 148. Lu, L., Montrose-Rafizadeh, C., Hwang, T. -C. & Guggino, W. B. A delayed rectifier potassium current in Xenopus oocytes. Biophys. J. 57, 1117-1123 (1990). 149. Methfessel, C. et al. Patch clam p measurements on Xenopus laevis oocytes: Currents through endogenous channels and implanted acetylcholine receptor and sodium channels. Pflugers Arch. 407 577-588 (1996). 150. Witzmann, F., Jarnot, B. & Parker, D. Dodecyl maltoside detergent improves resolution of hepatic membrane prot eins in two-dimensional gels. Electrophoresis 9 687-688 (1991). 151. Bernajee, P. Differentia l solubilization of lipids al ong with membrane proteins by different classes of detergents. Chem. Phys. Lipids 77, 65-78 (1995). 152. Sumikawa, K. & Miledi, R. Assembly and N-glycosylation of all ACh receptor subunits are required for their efficient insertion into plasma membranes. Molec. Brain Res. 5 183-192 (1989). 153. Sumikawa, K., Parker, I. & Miledi, R. Effect of tunicamicyn on the expression of functional brain neurotransmitter rec eptors and voltageoperated channels in Xenopus oocytes. Molec. Brain Res. 4 191-199 (1988).

PAGE 203

203 154. Sigel, E. Properties of single sodium channels translat ed by Xenopus oocytes after injection with messenger ribonucleic acid. J. Physiol., Lond. 386 73-90 (1987). 155. Wheatley, M. & Hawtin, S. R. Glycosyl ation of G-protein-co upled receptors for hormones central to normal reproductive functioning: its occurence and role. Human Reproduction Update 5 356-364 (1999). 156. Rigaud, J.-L. & Lvy, D. Reconstituti on of membrane proteins into liposomes. Methods in enzymology 372 65-86 (2003). 157. Berquand, A., Mingeot-Leclercq, M. P. & Dufrene, Y. F. Real-t ime imaging of drugmembrane interactions by atomic force microscopy. Biochim. Biophys. Acta, Biomembr. 1664 198-205 (2004). 158. Deshayes, S., Morris, M. C., Divita, G. & Heitz, F. Interactions of amphipathic carrier peptides with membrane components in relation with their ability to deliver therapeutics. J. Pept. Sci. 12, 758-765 (2006). 159. Rodrigues, C., Gameiro, P., Reis, S., Lima, J. L. & Castro, B. d. Derivative spectrophotometry as a tool for the determination of drug partition coefficients in water/dimyristoyl-L-alpha-phosphatid ylglycerol (DMPG) liposomes. Biophys. Chem. 94, 97-106 (2001). 160. Andersson, M. et al. Detection of si ngle ion channel activity on a chip using tethered bilayer membranes. Langmuir 23, 2924-2927 (2007). 161. Tamm, L. K. & Kalb, E. Microspectrofluorometry on supported planar membranes (John Wiley and Sons, New York, 1993). 162. Poglitsch, C. L., Sumner, M. T. & Thompson, N. L. Bi nding of IgG to MoFc gamma RII purified and reconstituted into suppor ted planar membranes as measured by total internal reflection fluorescence microscopy. Biochemistry 30, 6662-6671 (1991). 163. Silvestro, L. & Axelsen, P. H. Infrared spectroscopy of supported lipid monolayer, bilayer and multibilayer membranes. Chem. Phys. Lipids 96, 69-80 (1998). 164. Tamm, L. K. & Tatulian, S. A. Infrared spectroscopy of proteins and peptides in lipid bilayers. Q. Rev. Biophys. 30, 365-429 (1997). 165. Kagawa, Y. Energy transducing pr oteins in thermophilic biomembranes. J. Memb. Biol. 55, 1-8 (1980). 166. Driessen, A. J. M., Vo ssenberg, J. L. & Konings, W. N. Membrane composition and ion-permeability in extremophiles. FEMS Microbiol. Rev. 18, 139-148 (1996). 167. Shinoda, W., Mikami, M., Baba, T. & Ha to, M. Dynamics of a highly branched lipid bilayer: a molecular dynamics study. Chem. Phys. Lett. 390 35 (2004).

PAGE 204

204 168. Hsieh, C.-H., Sue, S.-C., Lyu, P.-C & Wu, W.-G. Membrane packing geometry of diphytanyolphosphatidylcholine is highly se nsitive to hydration: Phospholipid polymorphism induced by molecular r earrangement in the headgroup region. Biophys. J. 73, 870-877 (1997). 169. Atanasov, V. et al. Membrane on a Ch ip. A functional tethered lipid bilayer membrane on silicon oxide surfaces. Biophys. J. 89, 1780-1788 (2005). 170. Naumann, R. et al. Te thered lipid Bilayers on ultraflat gold surfaces. Langmuir 19, 5435-5443 (2003). 171. Suzuki, H., Tabata, K. V ., Noji, H. & Takeuchi, S. Highly reproducible method of planar lipid bilayer reconstitution in polym ethyl methacrylate microfluidic chip. Langmuir 22 1937-1942 (2006). 172. Santiago, J. et al. Probing the effect s of membrane cholesterol in the Torpedo californica acetylcholine receptor and the novel lipid-exposed mutation alpha C418W in Xenopus oocytes. J. Biol. Chem. 276 46523-46532 (2001). 173. Gustafsson, J., Oradd, G. Nyden, M., Hansson, P. & Almgren, M. Defective lamellar phases and micellar polymorphism in mixtures of glycerol monooleate and cetyltrimethylammonium br omide in aqueous solution. Langmuir 14 4987-4996 (1998). 174. Provencher, S. W. CONTIN A gener al-purpose constrained regularization program for inverting noisy linear algebraic and integral equations. Comput. Phys. Commun. 27, 229-242 (1982). 175. Provencher, S. W. A c onstrained regularization pr ogram for inverting data represented by linear algebraic or integral equations. Comput. Phys. Commun. 27, 213-227 (1982). 176. Einstein, A. Investigations on the t heory of Brownian movement. (Dover Publications Inc. USA 1956). 177. Danino, D., Talmon, Y. & Zana, R. Cryo-TEM of thre ad-like micelles: on-the-grid microstructural transformations i nduced during specimen preparation. Colloid Surf. A-Physicochem. Eng. Asp. 169 67-73 (2000). 178. Rand, R. P. & Fuller, N. L. Structural dimensions and their changes in a reentrant hexagonal-lamellar transition of phospholipids. Biophys. J. 66, 2127-2138 (1994). 179. Naumann, R. et al. Te thered lipid Bilayers on ultraflat gold surfaces. Langmuir 19, 5435-5443 (2003). 180. Gauger, D. R., Binder, H., Vogel, A., Se lle, C. & Pohle, W. Comparative FTIRspectroscopic studies of the hydrati on of diphytanoylphosphatidy lcholine and ethanolamine. J. Mol. Struct. 614 211-220 (2002).

PAGE 205

205 181. Scott, W. W. & Bhushan, B. 4th International Conference on Scanning Probe Microscopy, Sensors and Nanostructures (Elsevier Science, Las Vegas, Nv, 2002). 182. Seantier, B., Breffa, C. Felix, O. & Decher, G. Dissipation-enhanced quartz crystal microbalance studies on the experimental parameters controlling the formation of supported lipid bilayers. J. Chem. Phys. 109 21755-21765 (2005). 183. Reviakine, I. & Brisson, A. Formati on of supported phospholipid bilayers from unilamellar vesicles investigated by atomic force microscopy. Langmuir 16, 18061815 (2000). 184. Brenner, R. et al. Vasoregulation by t he beta1 subunit of the calcium-activated potassium (BK) channel. Nature 407 870-876 (2000). 185. Ramanathan, K. & Fuchs, P. A. Modeling hair cell t uning by expression gradients of potassium channel beta subunits. Biophys. J 82, 64-75 (2002). 186. Robitaille, R., Garcia, M. L., Kaczorowski, G. J. & Charlton, M. P. Functional colocalization of calcium and calcium-gated potassium channels in control of transmitter release. Neuron 11, 645-655 (1993). 187. Gao, Y.-D. & Garcia, M. L. Interaction of Agitoxin2, Charybdotoxin and Iberiotoxin with Potassium Channels: Selectivity between cause blockade of the channel Voltage-gated and Maxi-K channels. Proteins: structure function and genetics 52, 146-154 (2003). 188. Terrettaz, S., Mayer, M. & Vogel, H. Highly electrically insulating tethered lipid bilayers for probing the function of ion channel proteins. Langmuir 19, 5567-5569 (2003). 189. Steinem, C., Janshoff, A., Ulrich, W. P., Sieber, M. & Ga lla, H. J. Impedance analysis of supported lipid bilayer membr anes: a scrutiny of different preparation techniques. Biochim Biophys Acta. 1279 169-180 (1996). 190. Vallejo, A. E. & Gerv asi, C. A. Impedance analysi s of ion transport through gramicidin channels in supported lipid bilayers. Bioelectrochemistry 57, 1-7 (2002). 191. Schuster, B., Pum, D., Br aha, O., Bayley, H. & Sleytr, U. B. Self-assembled alphahaemolysin pores in an S-laye r supported lipid bilayer. Biochim. Biophys. Acta. 1370 280-288 (1998). 192. Hinterdorfer, P., Baber, G. & Tamm, L. K. Reconstitution of Membrane Fusion Sites. J. Biol. Chem. 269 20360-20368 (1994). 193. Atanasov, V., Atanasova, P. P., Vockenroth, I. K., K norr, N. & Koper, I. A molecular toolkit for highly insulating tet hered bilayer lipid membranes on various substrates. Bioconjugate Chem 17, 631-637 (2006).

PAGE 206

206 194. Tien, H. T. Bilayer Lipid Membranes (BLM): T heory and Practice. The Quarterly Review of Biology (Elsevier, Amsterdam 1976). 195. Woodhouse, G. E., King, L. G., Wieczorek, L. & Corne ll, B. A. Kinetics of the competitive response of receptors imm obilized to ion-channels which have been incorporated into a tethered bilayer. Faraday Discuss 111 247-258 (1998). 196. Schmidt, E. K. et al. Incorporation of the acetylcholine re ceptor dimer from Torpedo californica in a pept ide supported lipid membrane investigated by surface plasmon and fluorescence spectroscopy. Biosensors & Bioelectronics 13, 585-591 (1998). 197. Horrigan, F. T. & Aldrich, R. W. Coupl ing between voltage sensor activation, Ca2+ binding and channel opening in large c onductance (BK) potassium channels. J. Gen. Physiol. 120 267-305 (2002). 198. Magleby, K. L. Gating mechanism of BK (Slo1) channels: so near, yet so far. J Gen Physiol. 121 81-96 (2003). 199. Braha, O. et al. Designed protein pores as components for biosensors. Chem Biol. 7 497-505 (1997). 200. Prez, G., Lagrutta, A., Adelman, J. P. & Toro, L. Reconstitution of expressed BKCa channels from Xenopus oocytes to lipid bilayers. Biophys. J. 66, 1022-1027 (1994). 201. Fenwick, E. M., Marty, A. & Neher, E. Sodium and calcium channels in bovine chromaffin cells. J. Physiol. (Lond.) 331 599-635 (1982). 202. Liu, A. L. Advances in Planar Lipid Bilayers and Liposomes (Elsevier, London, 2006). 203. Hanner, M. et al. The beta subunit of the high c onductance calcium-activated potassium channel. Identificat ion of residues involved in charybdotoxin block. J. Biol. Chem. 273 16289-16296 (1998). 204. Yellen, G. Ionic permeation and blockade in calcium-activated K+ channels of bovine chromaffin cells. J. Gen. Physiol. 84, 157-186 (1984). 205. Stanfield, P. R. Tetraethylammonium ions and the potassium permeability of excitable cells. Rev. Physiol. Biochem. Pharmacol. 97, 1-67 (1983). 206. Mackinnon, R. & Yellen, G. Mutations affecting TEA blockade and ion permeation in voltage-activated K+ channels. Science (Washington D.C.) 250 276-279 (1990). 207. Naumann, R. et al. Tethered lipid bilayer on ultraflat gold surface. Langmuir 19 5435-5443 (2003).

PAGE 207

207 208. Taglialatela, M. et al. Patterns of inte rnal and external tetraethylammonium block in four homologous K+ channels. Molecular Pharmacology 40, 299-307 (1991). 209. Garcia-Calvo, M. et al. Purification and reconstitution of the high-conductance, calcium-activated potassium channel from tracheal smooth muscle. Journal of Biological Chemistry 269 676-682 (1994). 210. Hase, C. C., Alexander C. Le Dain & Martinac, B. Purification and Functional Reconstitution of the Recombinant Large Mechanosensitive Ion Channel (MscL) of Escherichia coli. J. Biol. Chem. 270 18329-18334 (1995). 211. Stroud, R., McCarthy, M. & Shuster, M. Nicotinic acetylcholine receptor superfamily of ligand-gated ion channels. Biochemistry 29, 11009-11023 (1990). 212. Blount, P., Sukharev, S. I., Schroeder, M. J., Nagle, S. K. & Kung, C. Single residue substitutions that change the gat ing properties of a mechanosensitive channel in Escherichia coli. Proceedings of the National Academy of Sciences of the United States of America 93, 11652-11657 (1996). 213. Chang, G., Spencer, R. H., Lee, A. T., Barclay, M. T. & Rees, D. C. (1998) Structure of the MscL homolog from Mycobacterium tuberculosis: A gated mechanosensitive ion channel. Science 282 2220-2226. (1998). 214. Montal, M. O., Iwamoto, T., Tomich J. & Montal, M. Design, synthesis and functional characterization of a pentameric channel protein that mimics the presumed pore structure of the ni cotinic cholinergic receptor. Federation of European Biochemical Societies 320 261-266 (1993). 215. Goennenwein, S., Tanaka, M., Hu, B., Moroder, L. & Sackmann, E. Functional Incorporation of Integrins into Solid Supported Membranes on Ultrathin Films of Cellulose: Impact on Adhesion. Biophys. J. 85 646-655 (2003). 216. Matsuno, Y., Osono, C., Hirano, A. & S ugawara, M. Single-Channel Recordings of Gramicidin at Agarose-Supported Bilayer lipid Membranes formed by the Tip-Dip and Painting Method. Analytical Sciences 20 1217-1221 (2004). 217. Hianik, T., Passechnik, V. I., Paltauf, F. & Hermetter, A. Nonlinearity of currentvoltage characteristics of t he gramicidin channel and stru cture of the gramicidin molecule. Bioelectrochem.Bioenerg. 34, 61-68 (1994). 218. Koeppe, R. E. & Andersen, O. A. Engineering the Gr amicidin Channel. Annu. Rev. Biophys. Biomol. Struct. 25, 231-258 (1996). 219. Hamill, P. O. & Martinac, B. Molecular basis of mechanotransduction in living cells. Physiological Reviews 81, 685-740 (2001).

PAGE 208

208 220. Blount, P., Shroeder, M. J. & Kung, C. Mutations in MscL, a bacterial mechanosensitive channel, can change the cellular response to osmotic stress. Biophysical Journal 74 A239-a239 (1998). 221. Blount, P., Sukharev, S. I., Moe, P. C., Nagle, S. K. & Kung, C. Towards an understanding of the stru ctural and functional properties of MscL, a mechanosensitive channel in bacteria. Biology of the Cell 87, 1-8 (1996). 222. Cruickshank, C. C., Minchin, R. F., LeDain, A. C. & Martinac, B. Estimation of the pore size of the large-conductance mechano sensitive ion channel of Escherichia coli. Biophysical Journal 73, 1925-1931 (1997). 223. Kung, C. & Blount, P. Channels in microbes: so many holes to fill. Molecular Microbiology 53, 373-380 (2004). 224. Martinac, B., Buechner, M., Delcour, A. H., Adler, J. & Kung, C. PressureSensitive Ion Channel in Escherichia-Coli. Proceedings of the National Academy of Sciences of the United States of America 84, 2297-2301 (1987). 225. Bartlett, J. L., Levin, G. & Blount, P. An in vivo assay identifies changes in residue accessibility on mechanosensitive channel gating. Proceedings of the National Academy of Sciences of t he United States of America 101 10161-10165 (2004). 226. Iscla, I., Levin, G., Wray, R., Reynol ds, R. & Blount, P. Defining the physical gate of a mechanosensitive channel, MscL, by engineering metal-binding sites. Biophysical Journal 87 3172-3180 (2004). 227. Levin, G. & Blount, P. Cysteine scanni ng of MscL transmembrane domains reveals residues critical for mechanosensitive channel gating. Biophysical Journal 86, 2862-2870 (2004). 228. Ou, X. R., Blount, P., Hoffman, R. J. & Kung, C. One face of a transmembrane helix is crucial in mechanosensitive channel gating. Proceedings of the National Academy of Sciences of t he United States of America 95 11471-11475 (1998). 229. Blount, P. & Moe, P. C. Bacterial mechanosensit ive channels: integrating physiology, structure and function. Trends in Microbiology 7 420-424 (1999). 230. Todorov, A., Petrov, A. G. & Fendler, J. H. Flexoelectricity of charged and dipolar bilayer lipid membranes studied by stroboscopic interferometry. Langmuir 10, 2344-2350 (1994). 231. Needham, D. & Hochmuth, R. M. Elec tro-Mechanical Permeabilization of Lipid Vesicles Role of Membr ane Tension and Compressibility. Biophysical Journal 55, 1001-1009 (1989).

PAGE 209

209 232. Nauman, C. A. et al. The polymer-suppor ted phospholipid bilayer: tethering as a new approach to substrate-membrane stabilization. Biomacromolecules 3 27-35 (2002). 233. Schiller, S. M., Naumann, R., Lovejoy, K., Kunz, H. & Knoll, W. Archaea Analogue Thiolipids for Tethered Bilayer Lipid Membranes on Ultrasmooth Gold Surfaces. Angewandte Chemie International Edition 42, 208-211 (2003). 234. Knoll, W., Koper, I., Naumann, R., Schiller, S. & Duran, R. S. Functional tethered bimolecular lipid membranes. Abstracts of Papers of the American Chemical Society 227, U853-U853 (2004). 235. Breffa, C. Chemistry 150 (Max Planck Institut fur Polymerforschung and Johannes Gutenberg Universitat Mainz, Geboren, 2005). 236. He, L. et al. Tethered Bilayer Lipid Me mbranes Based on Monolayers of Thiolipids Mixed with a Complementary Dilution Mole cule. 1. Incorporation of Channel Peptides. Langmuir 21 11666-11672 (2005).

PAGE 210

210 BIOGRAPHICAL SKETCH George Okeyo was born in Nairobi, Keny a. He graduated from the University of Nairobi in November 1998 with his B.S. degree having majored in chemistry with a minor in botany. He then worked as a phar maceutical representative and later as Regional sales manager for Bristol-Myers Squibb in Western Kenya. He proceeded on to graduate school enrolling in the chemistr y graduate program at the University of Florida in January 2005, working under the di rection of Dr. Randolph Duran. He later worked under the directi on of Dr. Gail Fanucci.